The Signal
Chalmers Team Uses Parasitic Capacitances to Cut Qubit Crosstalk
A study of a two-qubit superconducting processor finds that interference among parasitic coupling paths can suppress unwanted microwave drive signals. The authors say appropriately chosen parasitic capacitances could help guide low-crosstal
Quantum.world Team · Written with AI · fact-checked ·
Researchers affiliated with Chalmers University of Technology report that parasitic electrical paths can help suppress unwanted microwave signals in a superconducting quantum processor. Their study examines two fixed-frequency transmon qubits connected by a tunable coupler and finds that crosstalk suppression involves interference among coupling paths, not only cancellation of the effective interaction between the qubits.
More than one route for a microwave drive
A microwave control line is intended to address a particular qubit, but signals can also reach other parts of a processor. The paper describes these unwanted effects as drive crosstalk, which can contribute to coherent errors. The researchers focus on how parasitic electrical couplings affect the signal reaching a qubit that is not the intended target.
The processor in the study contains two fixed-frequency transmons and a coupler whose frequency can be tuned. The paper’s circuit model includes paths from each drive line not only to its target qubit, but also to the coupler and the other qubit. The authors analyze how these pathways contribute to crosstalk as the coupler is tuned.
That distinction matters because the unwanted drive is not described by a single route. The paper’s model treats the contributions from different paths as capable of interfering. Under suitable conditions, that interference can reduce the effective drive reaching the non-target qubit. The study therefore considers both coupling between circuit elements and the direct influence of the control lines on those elements.
Experiments and a circuit model
The researchers report experimental crosstalk measurements in their two-qubit processor and compare them with a circuit model. They measured both self- and cross-Rabi frequencies as a function of the coupler frequency. The model is designed to capture how crosstalk changes with that frequency and to account for multiple parasitic coupling paths.
The paper’s central experimental finding is that suppression does not arise solely from cancellation of the effective qubit-to-qubit coupling. Interference among other paths also contributes. That result connects the observed reduction to the way the drive lines, qubits and coupler are electrically linked, rather than to the qubit interaction alone.
The authors write:
“We experimentally demonstrate suppression of crosstalk, finding that the suppression does not originate solely from cancellation of the effective qubit-qubit coupling, but also is a result of interference between other coupling paths.”
Anuj Aggarwal and co-authors, paper authors affiliated with Chalmers University of Technology
Capacitances as design choices
Parasitic capacitances are often treated as unwanted imperfections in circuit design. The study instead proposes that choosing them appropriately could help suppress drive crosstalk. Its model is intended to provide design guidance for superconducting processors by making those pathways part of the design analysis rather than ignoring them.
The findings concern a processor with two qubits. The authors present the work as guidance for device design, not as a demonstration across processors of different sizes or architectures. Their broader proposal is to engineer parasitic coupling paths where useful, alongside approaches that seek to minimize unwanted interactions.