Keynote: Superconducting qubits for quantum computation: transmon vs fluxonium

Michelle Davore begins by expressing gratitude to the organizers for inviting her to present a review on quantum computation hardware design. She acknowledges her time at Google's Quantum AI team during her sabbatical and outlines her talk, which includes an explanation of superconducting qubits as part of a larger structure known as superconducting artificial atoms. Davore emphasizes the importance of understanding these atoms to grasp the creation of qubits. She introduces the concept by comparing a simple hydrogen atom to a superconducting circuit, highlighting the differences in scale and behavior between electrons in an atom and the electron condensate in a superconducting circuit.

Davore delves into the details of superconducting circuits, comparing them to a harmonic oscillator with controllable energy levels determined by inductance and capacitance values. She discusses the significance of dissipation in the circuit, which can be managed to optimize the quality factor, a measure of the circuit's energy retention. The talk also touches on the necessity of some dissipation for resetting the circuit to a known state. Davore highlights the advantage of superconducting circuits' ease of coupling, which allows for Hamiltonian engineering. She introduces the concept of non-linearity in the circuit, which is crucial for isolating specific energy transitions for qubit operations, and discusses the non-linearity ratio as a key design parameter.

The speaker discusses the introduction of non-linearity in superconducting circuits through the use of Josephson junctions, which act as non-linear inductances. She explains how these junctions, along with capacitances and inductances, form the basis of superconducting artificial atoms designed for qubits. Davore emphasizes the importance of optimizing the product of the quality factor and the non-linearity factor in these artificial atoms. She also addresses the physical aspects and challenges in designing these circuits, such as managing dissipative elements and controlling parasitic radiation.

Davore compares two types of artificial atoms used in quantum computing: the transmon and the fluxonium. She positions these on a parameter space diagram, highlighting their optimization to minimize noise influences. The transmon, shunted by a large capacitance, and the fluxonium, shunted by a large inductance, are described as the current favorites due to their noise resistance. She presents spectroscopic data comparing the quality factor of the hydrogen atom to that of the transmon, noting the progress in achieving high quality factors in人造 atoms. Davore also mentions the ongoing development and potential of the fluxonium.

The speaker addresses the challenges in engineering quantum circuits, focusing on the need for optimizing not just individual qubits but also the coupling between them, readout resonators, and amplifier chains. She discusses the importance of various characteristics such as readout fidelity and time, reset fidelity and time, and multi-qubit operations. Davore points out that while the transmon can be coupled to multiple other qubits, the fluxonium has so far only been coupled to one, suggesting potential for future development. She also raises concerns about design constraints such as leakage, the impact of cosmic rays, and the need for a multi-dimensional parameter optimization approach.

In the final part of her talk, Davore reflects on the potential for further optimization of superconducting circuits and the challenges of achieving higher precision in qubit operations. She suggests that while current technology can be optimized to reach very low infidelity rates, breakthroughs in materials and fabrication methods may be necessary for further advancements. Davore proposes the idea of sealing devices in vacuum as a potential innovation. She concludes by thanking her team at Yale, her colleagues, and funding agents for their support and acknowledges the audience's attention.