Superconducting Qubit Reset Improved With Novel Multi-Mode Purcell Filter

Superconducting qubits represent a promising avenue for realising practical quantum computation, yet maintaining qubit fidelity necessitates efficient methods for state preparation and measurement. Current approaches frequently rely on incorporating additional components into complex quantum circuits, increasing fabrication challenges and potential sources of error. Researchers Xu-Yang Gu, Da’er Feng, and colleagues, from the Beijing National Laboratory for Condensed Matter Physics and Northwest University, present a novel architecture utilising a multi-mode Purcell filter integrated directly into a superconducting circuit. Their work, entitled ‘Engineering a Multi-Mode Purcell Filter for Superconducting-Qubit Reset and Readout with Intrinsic Purcell Protection’, details an experimental demonstration of a device capable of both qubit reset and readout using the fundamental and second-order modes of a single coplanar waveguide resonator, thereby streamlining circuit design and enhancing performance. The device achieves rapid, high-fidelity qubit reset and significant leakage reduction, offering a pathway towards more scalable and efficient quantum processors.

Quantum computing’s progression towards practical application demands innovative strategies for qubit control and connectivity. Superconducting quantum computing, a leading platform, currently faces significant engineering challenges in scaling systems to achieve fault tolerance. Researchers pursue two primary avenues: distributed quantum networks and increased qubit density on single chips, both requiring sophisticated control methods without substantially increasing circuit complexity. Effective quantum error correction relies on rapid and precise qubit reset and leakage reduction, processes traditionally demanding dedicated hardware components.

Leakage, referring to unwanted excitation of qubits into higher energy states that hinders accurate computation, necessitates mitigation alongside reset, ensuring qubits begin in a known ground state. Implementing these functions typically involves introducing additional dissipative channels, increasing the circuit footprint. Alternative methods, such as repurposing existing readout resonators or employing parametrically activated interactions, have limitations regarding calibration, noise sensitivity, or unintended side effects. Optimising the use of available on-chip resources, moving away from a one-function-per-component paradigm, therefore represents a crucial advancement.

Coplanar waveguide resonators, commonly used in superconducting circuits, inherently support multiple resonant modes, each with a distinct frequency. These higher-order modes remain largely untapped in conventional designs, representing a potential avenue for integrating additional functionality without increasing component count. Harnessing these modes for distinct qubit operations promises a more scalable and efficient architecture. Purcell filters, a type of resonator, enhance qubit relaxation rates and improve signal clarity by carefully engineering their properties to optimise qubit performance and reduce unwanted noise.

This work demonstrates a novel technique employing a multi-mode Purcell filter, integrated into a superconducting circuit, to simultaneously address qubit reset and readout. This method leverages the inherent properties of microwave resonators, utilising both the fundamental and second-order modes for distinct operations, a departure from conventional single-mode designs. The core innovation lies in the design of a coplanar waveguide resonator, where the fundamental mode facilitates qubit readout, while the second-order mode acts as a Purcell filter, selectively attenuating the qubit’s excited state and driving it towards the ground state.

This dual functionality eliminates the need for separate reset circuitry, simplifying the chip layout and reducing potential signal interference. The implementation utilises a flip-chip architecture, optimising the coupling between the qubit and the resonator, and enhancing the overall signal strength. Crucially, the design achieves a high degree of selectivity, ensuring that the reset process primarily targets the qubit’s excited state without significantly affecting the readout signal. Experiments demonstrate an unconditional reset fidelity exceeding 99%, with residual excitation remaining below 1% within 220 nanoseconds.

The system incorporates a leakage reduction unit, specifically designed to address qubit state leakage to the second excited state, a common source of error. This unit selectively resets the second excited state within 62 nanoseconds, further improving the overall fidelity of qubit operations. Simulations predict Purcell-limited relaxation times exceeding 1 millisecond over an 800 megahertz bandwidth, indicating the potential for even faster and more efficient qubit control. This approach represents a significant step towards scalable, mode-efficient processor design, minimising circuit complexity and reducing the footprint of quantum processors.

The work builds upon established principles of circuit quantum electrodynamics, a field that explores the interaction between superconducting circuits and quantum systems. Leveraging advancements in microwave engineering and materials science, this research draws upon foundational works by Pozar on microwave engineering and Sete, Martinis and Korotkov on Purcell filtering. Further context is provided by the work of Swiadek et al. on enhancing dispersive readout and Jerger et al. on intrinsic resonator reset.

Recent advancements demonstrate control and entanglement on processors containing 78 qubits, pushing the boundaries of current technology. Concurrently, investigations into quantum error mitigation and the creation of logical qubits – qubits encoded to protect against errors – represent crucial steps towards fault-tolerant quantum computation. These developments, alongside novel control techniques like random multipolar driving, collectively advance the field towards practical quantum computing applications.

This work represents the first experimental demonstration of utilising different-order modes of a microwave resonator for distinct qubit operations. Researchers are actively exploring parametric coupling, where qubits interact through a shared resonator, and symmetry-protected topological phases, which offer inherent resilience against noise. These approaches aim to create more flexible and robust quantum circuits.

Future work will focus on scaling this mode-efficient architecture to larger qubit arrays, investigating the impact of cross-talk between adjacent qubits and optimising the resonator design for multiplexed readout. Exploring the integration of this reset and readout scheme with advanced qubit control techniques, such as those described by Swiadek et al., could further enhance the performance and scalability of superconducting quantum processors. Additionally, research into non-Hermitian phenomena, as explored by Han et al., may offer further avenues for optimising qubit control and coherence.