Quantum information science - the use of quantum mechanics to perform novel computing, simulation, communication, and sensing - is poised to revolutionize computing, biochemistry, condensed matter physics, cryptography, and astronomy, as well as a host of other fields. One of the most promising technology platforms is based on electrical circuits made of superconducting materials and operated at cryogenic temperatures. Quantum computers based on these quantum circuits have already been created and used for simple applications. However, their performance is limited by errors in the basic operations that make up quantum algorithms. For quantum processors to realize their full potential, these errors must be made inconsequential. This project aims to develop new ways to model, suppress, and correct errors in quantum circuits. The approach uses both hybrid "hardware" - different types of physical circuit designs - and hybrid "firmware" - different error correction and suppression protocols - in combination. By leveraging this hybrid approach, the research aims to create modular, scalable unit cells that can be used to create large-scale quantum processors with low error rates. The project will also train graduate students in the rapidly-expanding field of quantum information technology, growing the workforce for both academia and industry.
Several types of superconducting qubit designs exist, each with their own advantages and drawbacks. To date, no design combines the long-lived coherence and fast addressability necessary for use in large-scale quantum processors. Similarly, several methods for error suppression and correction exist, but most are only partially effective or are far too resource-intensive to be practical. This project combines different types of qubit hardware and multiple error correction schemes to realize ultra-low-error logical qubits. The key to the approach is to use hybrid error suppression protocols carefully designed to harness the detailed strengths and vulnerabilities of the basic physical qubit elements, while combining different qubit hardware to best leverage the protocols. The work includes using transmon qubits as "error detectors" for flux qubits used in quantum annealing protocols; combining dynamical decoupling with "generalized Markovian" noise for enhanced error suppression; using full quantum circuit simulations to correctly model the effects of noise channels; combining transmon and fluxonium qubits to achieve fast gates with ultra-long-lived coherence; and developing novel ways of encoding quantum information in decoherence-protected subspaces, together with other experimental and theoretical work. The eventual outcome of this project is intended to be the creation of logical qubits that function as modular, scalable unit cells. These basic modular elements can then be used in intermediate-scale quantum processors with no further error correction. Such modular elements that allow for resource-efficient error correction could be used in the full-scale, error-corrected quantum processors envisioned in the early days of quantum computing.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.