Lyapunov equation in open quantum systems and non-Hermitian physics

TL;DR

This paper establishes the Lyapunov equation as a fundamental and efficient formalism for linear open quantum systems, providing new insights and broader applicability beyond traditional quantum master equations, especially in non-Hermitian physics.

Contribution

The authors derive the Lyapunov equation for general linear quantum systems, connecting it to non-Hermitian physics and offering a less complex alternative to quantum master equations.

Findings

Derived Lyapunov equations for general linear systems in arbitrary dimensions

Clarified the relation between Lyapunov formalism and quantum master equations

Provided insights into non-Hermitian Hamiltonians and quantum fluctuations

Abstract

The continuous-time differential Lyapunov equation is widely used in linear optimal control theory, a branch of mathematics and engineering. In quantum physics, it is known to appear in Markovian descriptions of linear (quadratic Hamiltonian, linear equations of motion) open quantum systems, typically from quantum master equations. Despite this, the Lyapunov equation is seldom considered a fundamental formalism for linear open quantum systems. In this work we aim to change that. We establish the Lyapunov equation as a fundamental and efficient formalism for linear open quantum systems that can go beyond the limitations of various standard quantum master equation descriptions, while remaining of much less complexity than general exact formalisms. This also provides valuable insights for non-Hermitian quantum physics. In particular, we derive the Lyapunov equation for the most general number conserving linear system in a lattice of arbitrary dimension and geometry, connected to an arbitrary number of baths at different temperatures and chemical potentials. Three slightly different forms of the Lyapunov equation are derived via an equation of motion approach, by making increasing levels of controlled approximations, without reference to any quantum master equation. Then we discuss their relation with quantum master equations, positivity, accuracy and additivity issues, the possibility of describing dark states, general perturbative solutions in terms of single-particle eigenvectors and eigenvalues of the system, and quantum regression formulas. Our derivation gives a clear understanding of the origin of the non-Hermitian Hamiltonian describing the dynamics and separates it from the effects of quantum and thermal fluctuations. Many of these results would have been hard to obtain via standard quantum master equation approaches.