Multiport Impedance Quantization

TL;DR

This paper introduces a comprehensive multiport impedance quantization method for superconducting quantum systems, enabling accurate modeling, Hamiltonian derivation, and qubit lifetime prediction for complex multi-junction and cavity structures.

Contribution

It extends previous single-port models to multiport systems using impedance fitting and state-space synthesis, improving accuracy and numerical stability in quantum circuit analysis.

Findings

Able to model complex superconducting circuits with multiple ports.

Provides a stable, accurate Hamiltonian derivation method.

Enables computation of qubit T1 times across wide frequency bands.

Abstract

With the increase of complexity and coherence of superconducting systems made using the principles of circuit quantum electrodynamics, more accurate methods are needed for the characterization, analysis and optimization of these quantum processors. Here we introduce a new method of modelling that can be applied to superconducting structures involving multiple Josephson junctions, high-Q superconducting cavities, external ports, and voltage sources. Our technique, an extension of our previous work on single-port structures [1], permits the derivation of system Hamiltonians that are capable of representing every feature of the physical system over a wide frequency band and the computation of T1 times for qubits. We begin with a black box model of the linear and passive part of the system. Its response is given by its multiport impedance function Zsim(w), which can be obtained using a finite-element electormagnetics simulator. The ports of this black box are defined by the terminal pairs of Josephson junctions, voltage sources, and 50 Ohm connectors to high-frequency lines. We fit Zsim(w) to a positive-real (PR) multiport impedance matrix Z(s), a function of the complex Laplace variable s. We then use state-space techniques to synthesize a finite electric circuit admitting exactly the same impedance Z(s) across its ports; the PR property ensures the existence of this finite physical circuit. We compare the performance of state-space algorithms to classical frequency domain methods, justifying their superiority in numerical stability. The Hamiltonian of the multiport model circuit is obtained by using existing lumped element circuit quantization formalisms [2, 3]. Due to the presence of ideal transformers in the model circuit, these quantization methods must be extended, requiring the introduction of an extension of the Kirchhoff voltage and current laws.