Multimode RF Reflectometry Achieves 98% Fidelity Spin Qubit Readout and Device Characterization

Spin qubits represent a promising pathway towards quantum computing, but reliably reading their state remains a significant challenge. Joffrey Rivard, Alexis Morel, and Olivier Romain, all from Université de Sherbrooke, alongside their colleagues, now present a new approach using multimode radio-frequency reflectometry. Their work overcomes limitations of traditional methods by employing a specially designed inductor architecture that operates at multiple frequencies, enabling faster and more accurate spin readout. The team demonstrates single-shot spin readout with 98% fidelity, a substantial improvement in speed and accuracy, and also provides a powerful new tool for comprehensively characterizing the quality of quantum devices. This advance establishes multimode inductance as a scalable and versatile component for both rapid spin readout and detailed device analysis, bringing practical quantum computation a step closer to reality.

Multiple discrete frequencies up to 2GHz are addressed, overcoming limitations inherent in conventional single-mode designs. The spiral inductor’s distributed inter-turn capacitance yields distinct resonant modes, each with varied impedance-matching conditions. By probing a quantum dot across several of these modes, researchers extract tunneling rates over a broad frequency range and identify signatures of nearby charge defects. Utilizing one of the higher-order modes, the team demonstrates single-shot spin readout via a radio-frequency single-electron transistor (RF-SET), achieving singlet, triplet readout with an integration time of 8μs and a readout fidelity of 98%. These results establish multimode inductance as a scalable and flexible component.

Fast Spin Qubit Readout via RF Reflectometry

This research details the development and application of a multimode radio-frequency (RF) reflectometry technique for fast, sensitive, and versatile characterization of spin qubits in silicon. The team addressed the need for faster and more sensitive qubit readout, a significant challenge in quantum computing, and the importance of detailed device characterization for optimization and fabrication improvements. Their solution integrates a radio-frequency single-electron transistor as a highly sensitive electrometer and employs a range of RF frequencies to probe different aspects of the quantum dot and qubit behavior, allowing for simultaneous measurement of multiple parameters. Key findings demonstrate high-fidelity, single-shot readout of spin qubit states and effective characterization of key device parameters like tunnel couplings, alongside the identification of defects. Its compatibility with CMOS fabrication suggests potential for scaling up to larger qubit arrays, and the method proves effective at identifying defects in nanoscale transistors, improving device quality. This research presents a powerful new tool for both reading out qubit states and characterizing the devices that host them, paving the way for more advanced and scalable quantum computing architectures in silicon.

Multimode Inductor Boosts Reflectometry Performance

Scientists have developed a new superconducting multimode inductor architecture that significantly advances radio-frequency reflectometry. This work establishes a platform capable of supporting multiple discrete resonance frequencies, each offering distinct impedance-matching conditions, and provides increased flexibility for characterizing diverse device impedances. The team modeled the spiral inductor as a one-dimensional transmission line with distributed parameters, effectively realizing a ladder network that supports multiple standing-wave resonances. Simulations, using a discretized transmission-line model, accurately predicted observed behavior and were validated by measurements at 4 Kelvin.

Experiments demonstrate frequency-multiplexed charge sensing and extraction of tunneling dynamics across a broad 600MHz frequency range. By utilizing one of the higher-order resonant modes, the researchers achieved single-shot spin readout of a singlet-triplet qubit with an integration time of 8 microseconds and a remarkable readout fidelity of 98%. The inductor, fabricated from a 100 nanometer niobium nitride film deposited on a sapphire substrate, consists of a 150-turn spiral with 1 micrometer line width and spacing. Measurements confirm the simulations, showing well-separated impedance-matching conditions for radio-frequency reflectometry. These results highlight the potential of multimode inductors as a powerful and scalable tool for fast, high-fidelity spin-qubit readout and broad-spectrum device characterization.

Multimode Reflectometry Characterizes Quantum Dot Spin States

This research demonstrates a new architecture for radio-frequency reflectometry, employing a multimode inductor to achieve measurements at multiple discrete frequencies up to 2GHz. Conventional single-mode designs present limitations that this approach overcomes by utilizing the distributed inter-turn capacitance within a spiral inductor, creating distinct resonant modes with varied impedance-matching conditions. By probing a quantum dot across these multiple modes, scientists successfully extracted tunneling rates over a broad frequency range and identified signatures of nearby charge defects, significantly enhancing device characterization. Furthermore, the team demonstrated single-shot spin readout using one of the higher-order modes of the inductor, achieving an impressive singlet-triplet readout fidelity of 98% with an integration time of only 8 microseconds. These results establish multimode inductance as a scalable and flexible platform for both fast spin readout and detailed device characterization, offering advantages in speed, bandwidth, and flexibility. While acknowledging that further optimization of amplifiers and measurement stability could improve performance, this work confirms the potential of higher-order modes for high-fidelity spin readout and positions multimode inductance as a promising component for future quantum processor architectures.