Superconducting qubits offer an unprecedentedly high degree of flexibility in terms of circuit encoding and parameter choices. However, in designing the qubit parameters one typically faces the conflicting goals of long coherence times and simple control capabilities. Both are determined by the wavefunction overlap of the qubit basis states and the corresponding matrix elements. Here, we address this problem by introducing a qubit architecture with real-time tunable bit-flip protection. In the first, the `heavy' regime, the energy relaxation time can be on the order of hours for fluxons located in two near-degenerate ground states, as recently demonstrated in Ref. [Hassani et al., Nat.~Commun.~14 (2023)]. The second, `light' regime, on the other hand facilitates high-fidelity control on nanosecond timescales without the need for microwave signals. We propose two different tuning mechanisms of the qubit potential and show that base-band flux-pulses of around 10 ns are sufficient to realize a universal set of high-fidelity single- and two-qubit gates. We expect that the concept of real-time wavefunction control can also be applied to other hardware-protected qubit designs.
Superconducting qubits offer an unprecedentedly high degree of flexibility in terms of circuit encoding and parameter choices. However, in designing the qubit parameters one typically faces the conflicting goals of long coherence times and simple control capabilities. Both are determined by the wavefunction overlap of the qubit basis states and the corresponding matrix elements. Here, we address this problem by introducing a qubit architecture with real-time tunable bit-flip protection. In the first, the heavy' regime, the energy relaxation time can be on the order of hours for fluxons located in two near-degenerate ground states, as recently demonstrated in Ref. [Hassani et al., Nat.~Commun.~14 (2023)]. The second, light' regime, on the other hand facilitates high-fidelity control on nanosecond timescales without the need for microwave signals. We propose two different tuning mechanisms of the qubit potential and show that base-band flux-pulses of around 10 ns are sufficient to realize a universal set of high-fidelity single- and two-qubit gates. We expect that the concept of real-time wavefunction control can also be applied to other hardware-protected qubit designs.
Superconducting qubits offer an unprecedentedly high degree of flexibility in terms of circuit encoding and parameter choices. However, in designing the qubit parameters, one typically faces the conflicting goals of long coherence times and simple control capabilities. Both are determined by the wavefunction overlap of the qubit basis states and the corresponding matrix elements. Here, we address this problem by introducing a qubit architecture with real-time tunable bit-flip protection. In the first, the “heavy” regime, the energy relaxation time can be on the order of hours for fluxons located in two near-degenerate ground states, as recently demonstrated in Ref. [Hassani ., (2023)]. The second, “light” regime, facilitates high-fidelity control on nanosecond timescales without the need for microwave signals. We propose two different tuning mechanisms of the qubit potential and show that base-band flux-pulses of around 10 ns are sufficient to realize a universal set of high-fidelity single- and two-qubit gates. This realizes a hardware-efficient and strongly noise-biased qubit platform with analogies to cat qubits with the potential for efficient quantum error correction. We expect that the introduced concept of fast and real-time wavefunction control can also be applied to other hardware-protected qubit designs.
The heavy-fluxonium circuit is a promising building block for superconducting quantum processors due to its long relaxation and dephasing time at the flux-frustration point. However, the suppressed charge matrix elements and low transition frequency make it challenging to perform fast single-qubit gates using standard protocols. We report on new protocols for reset, fast coherent control,and readout that allow high-quality operation of the qubit with a 14 MHz transition frequency, an order of magnitude lower in energy than the ambient thermal energy scale. We utilize higher levels of the fluxonium to read out the qubit state and to initialize the qubit with 97% fidelity corresponding to cooling it to 190 μK. Instead of using standard microwave pulses, we control the qubit only with fast-flux pulses, generating control fields much larger than the qubit frequency. We develop a universal set of gates based on nonadiabatic Landau-Zener transitions that act in 20-60 ns, less than the single-qubit Larmor period. We measure qubit coherence of T1,T2e∼300 μs for a fluxonium in a 2D architecture and realize single-qubit gates with an average gate fidelity of 99.8% as characterized by randomized benchmarking.
Fast, high-fidelity control and readout of protected superconducting qubits are fundamentally challenging due to their inherent insensitivity. We propose a flux qubit variation which enjoys a tunable level of protection against relaxation to resolve this outstanding issue. Our qubit design, the double-shunted flux qubit (DSFQ), realizes a generic double-well potential through its three-junction ring geometry. One of the junctions is tunable, making it possible to control the barrier height and thus the level of protection. We analyze single- and two-qubit gate operations that rely on lowering the barrier. We show that this is a viable method that results in high-fidelity gates as the noncomputational states are not occupied during operations. Further, we show how the effective coupling to a readout resonator can be controlled by adjusting the externally applied flux while the DSFQ is protected from decaying into the readout resonator. Finally, we also study a double-loop gradiometric version of the DSFQ which is exponentially insensitive to variations in the global magnetic field, even when the loop areas are nonidentical. Published by the American Physical Society 2024
Quantum bits (qubits) are prone to several types of errors due to uncontrolled interactions with their environment. Common strategies to correct these errors are based on architectures of qubits involving daunting hardware overheads. A hopeful path forward is to build qubits that are inherently protected against certain types of errors, so that the overhead required to correct remaining ones is significantly reduced. However, the foreseen benefit rests on a severe condition: quantum manipulations of the qubit must not break the protection that has been so carefully engineered. A recent qubit - the cat-qubit - is encoded in the manifold of metastable states of a quantum dynamical system, thereby acquiring continuous and autonomous protection against bit-flips. Here, in a superconducting circuit experiment, we implement a cat-qubit with bit-flip times exceeding 10 seconds. This is a four order of magnitude improvement over previous cat-qubit implementations. We prepare and image quantum superposition states, and measure phase-flip times above 490 nanoseconds. Most importantly, we control the phase of these quantum superpositions without breaking bit-flip protection. This experiment demonstrates the compatibility of quantum control and inherent bit-flip protection at an unprecedented level, showing the viability of these dynamical qubits for future quantum technologies.
Achieving fast and high-fidelity qubit operations is crucial for unlocking the potential of quantum computers. In particular, reaching low gate errors in two-qubit gates has been a long-standing challenge in the field of superconducting qubits due to their typically long duration relative to coherence times. To realize fast gates, we use the hybridization between fixed-frequency superconducting qubits with a strongly interacting coupler mode that is tunable in frequency. To reduce population leakage during required adiabatic passages through avoided level crossings, we employ a sensitivity-adaptive closed-loop optimization method to design complex pulse shapes. We compare the performance of Gaussian-square, Fourier-series, and piecewise-constant-slope pulse parametrizations and are able to reach 99.9% controlled- Z -gate fidelity using a 64-ns-long Fourier-series pulse defined by only seven parameters. These high-fidelity values are achieved by analyzing the optimized pulse shapes to identify and systematically mitigate signal-line distortions in the experiment. To improve the convergence speed of the optimization, we implement an adaptive cost function, which continuously maximizes the sensitivity. The demonstrated method can be used for tune-up and recalibration of superconducting quantum processors.
This dissertation explores three types of protected superconducting qubits: the rhombusqubit, the 0-π qubit, and the gyrator qubit. For the rhombus qubit, we analyze both symmetric and asymmetric circuit variants, quantizing the system and comparing their energy spectra. We find that the symmetric rhombus exhibits exponential protection against both depolarization and charge noise-induced dephasing. Importantly, we show that introducing asymmetry yields a firstorder flux noise sweet spot, which can significantly extend qubit coherence and improve gate depth. For the soft 0-π qubit, we study single- and two-qubit gates using microwave drives. We demonstrate that driving nθ achieves higher-fidelity single qubit X gate than driving nφ, with simulated fidelities ∼ 99.9% and gate time ∼ 30 ns for T1 = Tφ = 30µs. We also develop two kinds of two-qubit gates, CZ and CNOT gates. These gates are realized using both direct and Raman-type transitions, with fidelities ∼ 99.9% and gate times around 160 ns for T1 = Tφ = 30µs. We also propose a method to generate GKP states using parametrically induced gyration between coupled oscillators. We describe a circuit implementation using a modulated inductor and derive the corresponding quantum Hamiltonian. Importantly I show that having an imperfect gyration ratio degrades the code space of gyrator qubit. Finally I develop a graph-theoretic approach to enumerate superconducting circuits and identify candidates with favorable coherence properties. Applying this framework, we compare transmon, fluxonium, and rhombus qubits in terms of relaxation time, dephasing, and expected gate count, finding that the asymmetric rhombus offers competitive, and potentially superior, performance.