Fundamental limits of superconducting quantum computers

TL;DR

This paper investigates the impact of the CSL model on superconducting transmon qubits, showing that while direct effects are negligible, indirect effects could limit quantum computer performance and offer a way to test CSL.

Contribution

It provides a quantitative analysis of CSL-induced decoherence in transmon qubits and assesses its implications for quantum computing and experimental tests.

Findings

CSL causes negligible direct superposition reduction in transmons.

CSL-induced noise generates quasiparticles, leading to decoherence.

Large quantum devices may have their performance limited by CSL effects.

Abstract

The Continuous Spontaneous Localization (CSL) model is an alternative formulation of quantum mechanics which introduces a noise coupled non linearly to the wave function to account for its collapse. We consider CSL effects on quantum computers made of superconducting transmon qubits. As a direct effect CSL reduces quantum superpositions of the computational basis states of the qubits: we show the reduction rate to be negligibly small. However, an indirect effect of CSL, dissipation induced by the noise, also leads transmon qubits to decohere, by generating additional quasiparticles. Since the decoherence rate of transmon qubits depends on the quasiparticle density, by computing their generation rate induced by CSL, we can estimate the corresponding quasiparticle density and thus the limit set by CSL on the performances of transmon quantum computers. We show that CSL could spoil the quantum computation of practical algorithms on large devices. We further explore the possibility of testing CSL effects on superconducting devices.