A cavity-mediated reconfigurable coupling scheme for superconducting qubits

TL;DR

The paper introduces a cavity-mediated, reconfigurable coupling architecture for superconducting qubits in which a shared bus resonator enables on-demand, non-local qubit interactions via frequency-tunable couplers. By operating in an idle, decoupled regime and selectively activating couplers, the system realizes fast iSWAP and CZ gates through a cavity-mediated exchange with strong spectator isolation; reported simulated gate times are ~$45$–$58$ ns with coherent errors below $10^{-4}$ and residual $ZZ$ interactions at the kilohertz level. Extending to four qubits, the authors demonstrate selective two-qubit gates with minimal crosstalk across all pairs, supported by detailed error budgets that separate leakage, population transfer, and phase errors. The work highlights a scalable route to enhanced interaction flexibility in superconducting processors, while noting challenges like spectral crowding and shared-bus contention that motivate modular bus architectures and coordinated gate scheduling for larger systems.

Abstract

Superconducting qubits have achieved remarkable progress in gate fidelity and coherence, yet their typical nearest-neighbor connectivity presents constraints for implementing complex quantum circuits. Here, we introduce a cavity-mediated coupling architecture in which a shared cavity mode, accessed through tunable qubit-cavity couplers, enables dynamically reconfigurable interactions between non-adjacent qubits. By selectively activating the couplers, we demonstrate that high-fidelity iSWAP and CZ gates can be performed within 50 ns with simulated coherent error below $10^{-4}$, while residual $ZZ$ interaction during idling remains below a few kilohertz. Extending to a four-qubit system, we also simulate gates between every qubit pair by selectively enabling the couplers with low qubit crosstalk. This approach provides a practical route toward enhanced interaction flexibility in superconducting quantum processors and may serve as a useful building block for devices that benefit from selective non-local coupling.

Figures (9)

• Figure 1: Architecture for reconfigurable connectivity of superconducting qubits. (a) Circuit schematic showing qubits (red), tunable couplers (blue), and a shared bus resonator (black), coupled via $C_{\mathrm{QC}}$ (qubit-coupler), $C_{\mathrm{QR}}$ (qubit-resonator), and $C_{\mathrm{CR}}$ (coupler-resonator). (b) Single-excitation level diagram for a selected qubit pair ($\mathrm{Q}_i$, $\mathrm{Q}_j$), with couplers active (solid) or idle (dashed). (c) Operating frequency ranges of the qubits, couplers, and the fundamental resonator mode (black). In the idle configuration, modes are detuned to suppress interactions and brought closer to activate coupling. (d) Single-excitation spectrum versus $\omega_{\mathrm{Q}_2}$, where an avoided crossing appears only when the couplers are enabled. (e) Effective exchange coupling $\tilde{g}_{\mathrm{QQ}}$ and residual $ZZ$ interaction $\zeta$ versus coupler frequency $\omega_{\mathrm{C}}$ with the qubits on resonance. (f) Residual $ZZ$ interaction versus coupler frequency $\omega_\mathrm{C}$ and qubit detuning $\Delta_{\mathrm{Q_1Q_2}}$ ($\mathrm{Q}_2$ fixed at $4$ GHz), demonstrating robust suppression over a broad parameter range.
• Figure 2: Time evolution for activating the effective exchange coupling and simulated iSWAP gate errors. (a) Frequency trajectories during a gate of duration $t_g$: the two couplers are tuned to $\omega_{\mathrm C}^{\min}$ and $\mathrm{Q}_2$ to $\omega_{\mathrm Q_2}^{\min}$, while $\mathrm{Q}_1$ and the resonator (black) remain fixed. (b) Vacuum Rabi oscillations between $\left|10\right\rangle$ and $\left|01\right\rangle$ as a function of coupler minimum frequency $\omega_{\mathrm C}^{\min}$ at fixed $\Delta_{\mathrm{Q_1Q_2}}=0$ (left) and as a function of qubit detuning $\Delta_{\mathrm{Q_1Q_2}}$ at fixed $\omega_{\mathrm C}^{\min}$ (right); the red dashed line indicates the operating trajectory used to implement the iSWAP gate. (c) Simulated iSWAP errors: leakage (blue), total infidelity (orange), and residual controlled phase $\phi_{ZZ}$ (green).
• Figure 3: CZ gate operation and error characterization. (a) Double-excitation spectrum versus $\omega_{\mathrm{Q}_2}$, highlighting the interacting states $\left|11\right\rangle$ and $\left|02\right\rangle$. Dashed (solid) curves denote coupling off (on); the avoided-crossing splitting (purple arrow) corresponds to the effective coupling $2g_{\mathrm{CZ}}/2\pi$. (b) Extracted $2g_{\mathrm{CZ}}/2\pi$ versus coupler frequency $\omega_{\mathrm{C}}$. (c) Leakage, (d) controlled-phase deviation $|\Delta\phi_{\mathrm{CZ}}|$, and (e) total infidelity from the tune-up at $t_g=58$ ns, plotted against deviations of the minimum coupler frequency ($\delta f_{\mathrm{C}}$) and qubit detuning ($\delta f_{\mathrm{Q}}$) from the optimal operating point (red star). (f) Leakage (blue) and total infidelity (orange) versus gate time, showing leakage-limited performance (inset: $|\Delta\phi_{\mathrm{CZ}}|$ versus gate time).
• Figure 4: Selective coupling in a four-qubit system. (a) Single-excitation spectrum obtained by sweeping the frequency of $\mathrm{Q}_2$ under different coupling configurations. An avoided crossing appears only when the $\mathrm{Q}_1$-$\mathrm{Q}_2$ coupling is enabled (solid blue), indicating an effective exchange interaction. No crossings are observed when all couplers are idle (dotted gray) or when only the $\mathrm{Q}_1$-$\mathrm{Q}_3$ pair is coupled (dashed red). (b) Residual $ZZ$ interactions $\zeta_{ij}$ between qubit pairs $(\mathrm{Q}_i, \mathrm{Q}_j)$ as a function of the common coupler frequency $\omega_\mathrm{C}$ of $\mathrm{C_1}$ and $\mathrm{C_2}$, showing suppressed spectator interactions outside the selected pair.
• Figure 5: Schematic of the circuit architecture used for quantization. A transmission-line resonator serves as a common coupling bus and is restricted to its fundamental mode with flux amplitude $\Phi_{\mathrm{R}}$. Each qubit–coupler unit is capacitively connected to the resonator at position $x_j$, where the local resonator flux is $a_j \Phi_{\mathrm{R}}$, with $a_j$ determined by the spatial profile of the fundamental mode. The coupler (blue) and qubit (red) are modeled as tunable-frequency transmon modes with node flux variables $\Phi_{\mathrm{C}_j}$ and $\Phi_{\mathrm{Q}_j}$, respectively.