On light cone bounds for Markov quantum open systems

Israel Michael Sigal, Xiaoxu Wu · 2025

Read On light cone bounds for Markov quantum open systems by Israel Michael Sigal, Xiaoxu Wu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-20635.txt.

Introduction

Israel Sigal and Xiaoxu Wu study how effects spread in space and time in open quantum systems whose future depends only on their present state. For a large class of the equations behind such systems, they prove that an effective light cone exists. Outside the cone the leftover effect is not zero, but it shrinks at an exponential rate with distance, giving these noisy systems a clean causal shape.

Catalog record

CREATORIsrael Michael Sigal, Xiaoxu Wu
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE60 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-20635

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On light cone bounds for Markov quantum open systems Israel Michael Sigal, Xiaoxu Wu Abstract . We study space-time behaviour of solutions of the von Neumann-Lindblad equations underlying the dynamics of Markov quantum open systems. For a large class of these equations, we prove the existence of an effective light cone with an exponentially small spill-over. Published in Lett Math Physics, 2026 2020 Mathematics Subject Classification 35Q40 (primary); 35Q94, 81P45, 46N50 (secondary). 1. Introduction 1.1. Markovian quantum open systems In this paper, we study space-time dynamics of Markov open quantum systems (MOQS) on the Hilbert space \mathcal{H}=L^{2}(\Lambda) , where \Lambda is either \mathbb{R}^{n} or \mathbb{Z}^{n} . We prove the existence of an effective light cone with an exponentially small spill-over for a large class of such systems. Conceptually, our bounds are related to the celebrated Lieb-Robinson bound [ 59 ] which plays a central role in analysis of evolution of quantum information (see e.g. [ 67 , 69 ] for the Lindladian setting in quantum spin systems and [ 9 , 8 , 17 , 16 , 28 , 29 , 30 , 35 , 33 , 41 , 42 , 43 , 44 , 40 , 53 , 55 , 56 , 57 , 58 , 61 , 62 , 66 , 63 , 65 , 64 , 72 , 75 , 79 , 84 , 83 ] for the unitary evolutions). An open quantum system (OQS) is a pair (\mathcal{S}_{1}^{+},\beta_{t}) , where the state space \mathcal{S}_{1}^{+} is the space of positive trace-class operators (density operators) on a Hilbert space \mathcal{H} and the evolution \beta_{t} is a family of quantum maps (or quantum channels) on \mathcal{S}_{1}^{+} , i.e. linear, completely positive, trace preserving maps (see [ 3 , 2 , 12 , 24 , 39 , 54 ] and the references therein). The concept of OQS is an extension of that of the (closed) quantum system ( \mathcal{S}_{1}^{+},\alpha_{t} ), where \alpha_{t} is the von Neumann dynamics, (1.1) \alpha_{t}(\rho)=e^{-iHt}\rho e^{iHt}, incorporating, in a natural way, the influence of the system’s environment. OQS arise also in the quantum measurement theory where the degrees of freedom of systems under investigation (rather than of the environment) are integrated out and in studying entanglement between two or more systems. Importantly, even when the interaction with environment can be neglected, investigation of OQS is needed to determine whether properties of closed systems are robust w.r.to weak interaction with an outside environment. For instance, whether transmission of quantum information is stable w.r.to decoherence induced by such an interaction. It is shown in [ 38 , 60 ] that under the Markovian assumption that \beta_{t} is a strongly continuous semigroup, (1.2) \beta_{t}\circ\beta_{s}=\beta_{t+s},\,\qquad\forall t,s\ \geq 0,\quad\text{ and }\beta_{t}\xrightarrow{s}\mathds{1}\text{ as }t\downarrow 0, the evolution \rho_{t}=\beta_{t}(\rho_{0}) satisfies the von Neumann-Lindblad equation (vNLE) (here and in the rest of this paper, we set \hbar=1 ) (1.3) \partial_{t}\rho_{t}=-i[H,\rho]+\sum_{j=1}^{\infty}(W_{j}\rho_{t}W_{j}^{*}-\frac{1}{2}\{W_{j}^{*}W_{j},\rho_{t}\}), with the initial condition \rho_{t=0}=\rho_{0} . Here H and W_{j},j=1,\cdots , are operators on \mathcal{H} , H is a quantum Hamiltonian of the system of interest and W_{j} are operators produced by the interaction with environment, called the jump operators, and \{A,B\}:=AB+BA . Conversely, under rather general conditions (see a discussion below and in Appendix A ), solutions to the vNLE exist for any initial condition \rho_{0} in \mathcal{S}_{1} and generate Markov open quantum (MOQ) dynamics, \beta_{t}(\rho_{0})=\rho_{t} . Thus the class of MOQ semigroups is rather rich. Furthermore, equations of the form ( 1.3 ) were derived in the van Hove limit of a particle system coupled to a thermal reservoir, see [ 20 , 22 , 21 , 24 , 50 ] . Hence, ( 1.3 ) captures, at least approximately, natural physical models. Clearly, the vNLE is an extension of the von Neumann equation (vNE) (1.4) \partial_{t}\rho_{t}=-i[H,\rho_{t}], which generates evolution ( 1.1 ), describing the statistics of closed quantum systems. While the vN dynamics can be always reduced to the Schrödinger one on the corresponding Hilbert space ( L^{2}(\Lambda) , in our case), this is not true for the vNLE. Thus, vNLE is a genuine extension of the Schrödinger equation beyond QM (to OQS, or to what can be termed as quantum statistics) making it a central object of quantum physics. By virtue of its origin, the vNLE plays a foundational role in quantum information science and in non-equilibrium quantum statistical mechanics, see [ 25 , 19 , 27 , 48 , 49 , 69 , 85 ] , and [ 3 , 2 , 68 ] , respectively, and references therein. It is also used in computational physics to construct the Gibbs and ground states for given Hamiltonians, [ 14 , 15 , 17 , 26 , 51 , 76 , 82 , 86 ] and references therein. See [ 70 ] for an elementary lecture-notes exposition of a role of vNLE in quantum information theory. The vNLE also appears naturally in Fröhlich et al theory of randomness in Quantum Mechanics (ETH-Approach, see [ 36 ] and references therein). Mathematically, vNLE is a key representative of non-abelian PDEs. It is related to stochastic differential equations on Hilbert spaces, see [ 46 ] . As is standard, we assume that the operators H and W_{j},j=1,\cdots , satisfy the conditions (H) H is a self-adjoint operator; (W) W_{j},j=1,\cdots, are bounded operators s.t. \sum\limits_{j=1}^{\infty}W_{j}^{*}W_{j} converges weakly. It is shown in [ 23 ] that, under conditions (H) and (W), the operator (1.5) L(\rho)=-i[H,\rho]+\sum_{j=1}^{\infty}(W_{j}\rho W_{j}^{*}-\frac{1}{2}\{W_{j}^{*}W_{j},\rho\}), defined by the r.h.s. of vNLE, generates a OQD semigroup, \beta_{t}=e^{Lt} . This implies, in particular, that Eq. ( 1.3 ) with initial conditions in \mathcal{S}_{1}^{+} has unique weak solutions in \mathcal{S}_{1}^{+} (and strong solutions on the natural domain of L ), see [ 68 ] for a streamlined version and more references, and Appendix A below, for a brief discussion. We call a QOD \beta_{t} satisfying ( 1.2 ), the Markov QOD, or MQOD and L , the von Neumann-Lindblad (vNL) generator. For other results on vNLE ( 1.3 ), we mention the scattering theory, see [ 31 , 32 ] , and the problem of return to equilibrium, see [ 68 ] and references therein. Notation In what follows, \mathcal{H}=L^{2}(\Lambda) , where \Lambda is either \mathbb{R}^{n} or \mathbb{Z}^{n} , \mathcal{B}(X) denotes the space of bounded operators on a Banach space X , and \mathcal{S}_{1} and \mathcal{S}_{2} , the Schatten spaces of trace-class and Hilbert-Schmidt operators. The norms in \mathcal{H} and \mathcal{B}(\mathcal{H}) are denoted by \|\cdot\| , and in \mathcal{S}_{1},\mathcal{S}_{2} and \mathcal{B}(\mathcal{S}_{1}) , by \|\cdot\|_{1},\|\cdot\|_{2} and \|\cdot\|_{1}^{op} , respectively. Explicitly, \|\lambda\|_{1}=\Tr(\lambda^*\lambda)^{\frac{1}{2}} and \|\lambda\|_{2}=(\Tr\lambda^{*}\lambda)^{\frac{1}{2}} . A,B will denote bounded operators (observables), X,Y\subset\Lambda stand for subsets of \Lambda and \chi_{X} , the characteristic function of X\subset\Lambda . In what follows, \lambda,\mu\in\mathcal{S}_{1} and \rho\in\mathcal{S}_{1}^{+} , always. To fix ideas, we assume that the DO’s \rho are normalized as \Tr\rho=1 . 1.2. Light cone bound Consider on \mathcal{H}=L^{2}(\Lambda) the n -parameter group of unitary operators T_{\xi} of multiplication by the function e^{-i\xi\cdot x},\xi\in\mathbb{R}^{n} . Define the polystrip S_{a}^{n},a>0, in the complex space \mathbb{C}^{n} as (1.6) S_{a}^{n}:=\{\zeta=(\zeta_{1},\cdots,\zeta_{n})\in\mathbb{C}^{n}\,:\,|{\rm Im}\zeta_{j}|<a\quad\forall j\}. We assume the following conditions: (AH) The operators H_{\xi}:=T_{\xi}HT_{\xi}^{-1},\xi\in\mathbb{R}^{n} , have the common domain \mathcal{D}(H) and H_{\xi}(H+i)^{-1} are bounded operators for all \xi\in\mathbb{R}^{n} and the operator function \xi\to H_{\xi}(H+i)^{-1}, from \mathbb{R}^{n} to \mathcal{B}(\mathcal{H}) , has

Read the complete text