High-Fidelity, 2D Noise-Resilient Superconducting Quantum Processors
Abstract
We propose to design, fabricate, and benchmark planar 2D superconducting processors with = 8 qubits and 2-qubit gate fidelities = 0.99. We leverage a toolbox containing submicron Josephson junctions/arrays and high-coherence, passive microwave circuit elements including capacitors, superinductors, and co-planar waveguide resonators. Our approach does not rely on circuit elements for which robust fabrication techniques have not been developed (eg. phase-slip elements and cos(2À) elements) or qubit architectures where a clear route to logical gate operation has not been proven (eg. 0-p or other topological qubits). We adjust the ratios of EJ/EC and EJ/EL to realize two unexplored noise-resilient qubits: The High-Coherence Fluxonium that encodes quantum information in matter and the Multi-Photon Cat where, in the presence of an intense 2-photon microwave drive, information is encoded in photonic degrees of freedom. We have thoroughly simulated both designs and have also fabricated proof-of-concept devices. We will (i) pair these next-generation matter and photonic superconducting 2D circuits with a robust, high-fidelity CZ gate (originally developed for transmons in the HiPS program); (ii) quantify and mitigate decoherence due to dielectric loss and non-equilibrium quasiparticles, integrating our recently demonstrated Q > 5M Nb co-planar resonators into qubit circuits and establishing coherence limits due to these mechanisms; (iii) use charge-sensitive sensor qubits for characterizing correlated quasiparticle errors; (iv) mitigate quasiparticle losses by fabricating circuitry on suspended membranes; (v) use both quantum-information and condensed matter methods to characterize many-body noise processesÑincluding correlated logical errorsÑin a multi-qubit processor; and (vi) execute standard quantum algorithms such as the quantum Fourier transform to validate quantum processor performance. Our integrated theory effort will support and extend these experimental efforts. We will (i) optimize gate performance and qubit lifetimes using machine learning techniques and realistic models of the environmental noise and drive cross-talk; (ii) optimize readout fidelity by accounting for nonidealities such as memory effects and basis hybridization; (iii) characterize and model the environmental noise by mapping its spectral density using correlated sensor-qubit data; and, (iv) study and develop protocols to mitigate sources of dynamical dephasing. For High-Coherence Fluxonium qubits, we will (i) simulate 2-qubit gates with a charge multi-path coupler; (ii) explore implementations of the 2-qubit fSim gate family that includes CZ and iSWAP as special cases; and (iii) design tunable 2-qubit gates that use both charge and flux couplers, as well as multi-path interference. For Multi-Photon Cat qubits, we will (i) model coherence limitations and analyze alternate circuit designs that address those limitations; (ii) design faster, noise-biased single-qubit gates; and, (iii) explore novel interactions suitable for implementing multi-qubit gates. We will also close the loop in the qubit design innovation chain by correlating the following: materials imperfections, measured coherence, logical gate fidelities, measured noise-spectra, and error reconstruction based on quantum verification/validation protocols. This is a fundamental research project that is not expected to produce any developmental items. Should any developmental items result from this work they will have both civilian and military applications.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 28, 2022
- Source ID
- W911NF2210258
Entities
People
- Irfan Siddiqi
Organizations
- Army Contracting Command
- National Security Agency
- University of California, Berkeley