Microwave cavity quantum memories

Short answer: A quantum memory for superconducting circuits stores quantum states in a microwave resonator that loses energy far more slowly than a qubit does. A 2016 Physical Review B paper reports storage of superpositions on the millisecond scale in a 3D cavity, a 2021 Physical Review Letters paper reports single-photon lifetimes of 2 ms across 9 modes of a seamless cavity, and a 2024 Applied Physics Letters paper describes a suspended coaxial resonator that combines thin-film conductors with a 3D package. These are laboratory results from related research groups.

What a memory needs

The 2016 abstract says significant advances in coherence make superconducting circuits a viable platform for fault-tolerant quantum computing, and that highly coherent systems could act as quantum memories for them. A useful memory must be rapidly addressable by Josephson-junction-based artificial atoms while maintaining superior coherence. The 2024 abstract adds that encoding information in the bosonic excitations of microwave resonators gives long coherence times, low dephasing rates and a hardware-efficient route to quantum error correction.

2016: millisecond storage in a cavity

The Yale authors demonstrate a superconducting microwave cavity architecture that is highly robust against major sources of loss in circuit QED engineering. It allows storage of quantum superpositions in a resonator on the millisecond scale, while strong coupling between the resonator and a transmon qubit gives control, encoding and readout at megahertz rates. They say this extends the maximum available coherence time in superconducting circuits by almost an order of magnitude compared with earlier hardware, and call the design an ideal platform for coherent quantum optics. Those are the authors' characterizations.

2021: a seamless multimode cavity

The Physical Review Letters abstract describes multimode cavity QED, where a two-level system interacts with many cavity modes, as a versatile framework for quantum information processing. The authors realize a 3D multimode circuit QED system with single-photon lifetimes of 2 ms across 9 modes of a novel seamless cavity. They demonstrate protocols for universal single-mode quantum control across all modes using a single drive line. They do this with a method they call the flute method for making monolithic superconducting cavities, which reduces loss while allowing control of the mode spectrum and mode-qubit interaction. They say it provides a template for multimode systems with exceptionally low dissipation, and is an important step toward random access quantum memories. That final step is their outlook.

2024: a suspended coaxial resonator

The Applied Physics Letters paper contrasts two ways to make superconducting resonators. Machined bulk 3D resonators have few loss channels and low participation in lossy parts, but material quality is hard to achieve. On-chip thin-film resonators can use low-loss films but confine the field more tightly, giving higher participation and extra loss from the dielectric substrate. The authors present a design in which a dielectric scaffold supports a thin-film conductor within a 3D package, combining the low surface participation of bulk cavities with the material control of thin films. The abstract continues beyond that point, but the rest was not readable here and is not described.

Limits

The three papers differ in cavity design and in the figures they report, and the 2024 result figures are not available from the part of the abstract that could be read. A millisecond-scale lifetime describes the stored state in the resonator, and a full memory also needs fast, high-fidelity interaction with qubits. The abstracts describe laboratory devices.

FAQ

What is a microwave cavity memory?

A resonator that stores a quantum state as microwave photons, read and written by a coupled qubit.

How long did states last?

On the millisecond scale in 2016, and single-photon lifetimes of 2 ms across 9 modes in 2021, per the authors.

Why use a 3D cavity?

Few loss channels and small participation in lossy parts, per the 2024 abstract.

What is the flute method?

A method the 2021 authors developed to make monolithic seamless superconducting cavities with lower loss.

Is this a quantum computer memory today?

These are device demonstrations. Random access memory is described as a goal.