Low-power qubit control based on ballistic SFQ and quasiparticle poisoning suppression

A utility-scale fault-tolerant quantum computing (FTQC) would likely require millions of physical qubits. Each qubit needs multiple control signals presently generated with room temperature electronics and delivered over meters of high-quality coaxial cables which dominate the cryostat heat load, cause signal distortions and crosstalk. This fundamental scaling problems can be addressed with implementing qubit control and readout using low power superconducting electronics deployed inside of the cryostat next to quantum processors. One of the objectives is to ensure the cryogenic control energy dissipation would be significantly lower that the heat load of the cables.

The successful realizations of qubit control using superconducting single-flux quantum (SFQ) circuits were demonstrated with ~nW of power per qubit dissipated at the millikelvin stage [1]. Minimization of the power dissipation would allow scaling to higher qubit count for a given heat capacity. Conventional SFQ circuits, such as ERSFQ, RQL, etc. are built with artificially introduced losses in form of resistive shunts of Josephson junctions for controllable handling of individual quanta of magnetic flux (fluxons). As a result, a constant injection of external energy is needed to enable sustainable fluxon propagation.

It is possible to significantly reduce the circuit power dissipation by using kinetic energy of the fluxons ballistically moving throughout a circuit. We minimize losses by using unshunted underdamped Josephson junctions. To preserve a full control on fluxons, the data processing circuits are unbiased. To compensate the eventual loss of kinetic energy and assure sustainable ballistic fluxon propagation, an on-demand injection of external energy is done via the quantized phase sources [2]. All these enable a reduction of power dissipation by at least an order of magnitude compared to the standard energy-efficient SFQ logics while increasing circuit operation margins and yield.

We will also discuss the suppression of potential parasitic influence of SFQ circuits via quasiparticle generation to the proximally located qubits.

[1]. J. Bernhardt, C. Jordan, et al. arXiv”2503.09879 (2025).

[2]. F. Lupo, O. Mukhanov, M Arzeo, PCT/US24/44937 patent application (2023).