arXiv:cond-mat/0110630v2 [cond-mat.stat-mech] 10 Dec 2001Nonequilibrium Dynamics in Low Dimensional Systems M. R. Evans and R. A. Blythe1 Department of Physics and Astronomy, University of Edinburgh, Mayfield Road, Edinburgh EH9 3JZ, U.K. Abstract In these lectures we give an overview of nonequilibrium stochastic systems. In par- ticular we discuss in detail two models, the asymmetric exclusion process and a ballistic reaction model, that illustrate many general features of nonequilibrium dynamics: for example coarsening dynamics and nonequilibrium phase transitions. As a secondary theme we shall show how a common mathematical structure, the q-deformed harmonic oscillator algebra, serves to furnish exact results for both sys- tems. Thus the lectures also serve as a gentle introduction to thingsq-deformed. Key words:Nonequilibrium Dynamics, Stochastic Processes, Phase Transition, Asymmetric Exclusion Process, Reaction Kinetics PACS:02.50.-r, 05.40.-a, 05.70.Fh 1Introduction In these lectures we explore the subject of nonequilibrium dynamics. Before getting into any kind of detail let us first establish what we mean by a nonequi- librium system. This is best done by taking stock of our understanding of an equilibrium system. Consider the Canonical (Boltzmann) distribution for a systems with configurations labelledCeach with energyE(C): P(C) = exp(−βE(C)) Z(1) Email addresses:m.evans@ed.ac.uk(M. R. Evans),r.a.blythe@ed.ac.uk(R. A. Blythe). 1Present address: Department of Physics and Astronomy, University of Manch- ester, Manchester, M13 9PL, U.K. Preprint submitted to Elsevier Science1 November 2018
whereβ= 1/kT. The task is to calculate the partition function Z=∑ C exp(−βE(C)),(2) from which all thermodynamic properties, in principle, can be computed. The distribution (1) applies to systems in thermal equilibriumi.e.free to exchange energy with an environment at temperatureT. It can easily be generalised to systems free to exchange particles, volume etc but it always relies on the concept of the system being at equilibrium with its environment. If one were interested in dynamics, for example to simulate the model on a computer, one might choose transition rates between configurations to satisfy W(C′→ C)e−βE(C′)=W(C → C′)e−βE(C)(3) whereW(C′→ C) is the transition rate from configurationC′toC. Condi- tion (3) is known as detailed balance and guarantees (under the assumption of ergodicity to be discussed below) that starting from some nonequilibrium initial condition the system will eventually reach the steady state of thermal equilibrium given by (1). We will discuss further this dynamical relaxation process and the properties of the steady state endowed by the detailed bal- ance condition in Section 3. For the moment we note that a system relaxing to thermal equilibrium is one realisation of a nonequilibrium system. In recent years such relaxation dynamics have been of special interest, for example, in the study of glassy dynamics whereby, on timescales realisable in experiment (or simulation), the system never reaches the equilibrium state and it is a very slowly evolving nonequilibrium state that is observed. This is sometimes re- ferred to as ‘off-equilibrium’ dynamics. Also let us mention the field of domain growth whereby an initially disordered state is quenched (reduced to a tem- perature below the critical temperature for the ordered phase) and relaxes to an ordered state through a process of coarsening of domains. The interesting physics lies in the scaling regime of the coarsening process which is observed before the equilibrium (ordered) state is reached. The other meaning of nonequilibrium refers to a system that reaches a steady state, but not a steady state of thermal equilibrium. Examples of such nonequi- librium steady states are given by driven systems with open boundaries where a mass current is driven through the system. Thus the system is driven by its environment rather being in thermal equilibrium with its environment. A pragmatic definition of a nonequilibrium system that encompasses all of the scenarios above is as a model defined by its dynamics rather than any energy functioni.e.the configurations of the model are sampled through a local stochastic dynamics whicha prioridoes not have to obey detailed balance. 2
1.1Structure of these notes These notes broadly follow the four lectures given at the summer school. In addition a tutorial class was held to explore points left as exercises in the lectures. In the present notes these exercises are included in a self-contained form that should allow the reader to work through them without getting stuck or else leave them for another time and continue with the main text. The notes are structured as follows: in section 2 we give an overview of two simple models that we are mainly concerned with in these lectures. In section 3 we then set out the general theory of the type of stochastic model we are interested in and point out the technical difficulties in calculating dynamical or even steady- state properties. Section 4 is an interlude in which we introduce, in a self- contained way, a mathematical tool—theq-deformed harmonic oscillator— that will prove itself of use in the final two sections. In Section 5 we present the solution of the partially asymmetric exclusion process and amongst other things how the phase diagram (Figure 3) is generalised. In Section 6 we discuss the exact solution of a stochastic ballistic annihilation and coalescence model. 2Two simple models In this work we will focus on two exemplars of nonequilibrium systems: the partially asymmetric exclusion process and a particle reaction model. These models have been well studied over the years and a large body of knowledge has been built up [1]. We introduce the models at this point but will come back to these models in more detail in Sections 5 and 6 in which we summarise some recent analytical progress. 2.1Asymmetric exclusion process 2.1.1Model definition The asymmetric simple exclusion process (ASEP) is a very simple driven lat- tice gas with a hard core exclusion interaction [2]. ConsiderMparticles on a one-dimensional lattice of lengthNsay. At each site of the lattice there is either one particle or an empty site (to be referred to as a vacancy or hole)— there is no multiple occupancy. The dynamics are defined as follows: during each time interval ∆Teach parti- cle has probability ∆Tof attempting a jump to its right and probabilityq∆T of attempting a jump to its left; a jump can only succeed if the target site is 3
100%
Scan to connect with one of our mobile apps
Coinbase Wallet app
Connect with your self-custody wallet
Coinbase app
Connect with your Coinbase account
Open Coinbase Wallet app
Tap Scan
Or try the Coinbase Wallet browser extension
Connect with dapps with just one click on your desktop browser
Add an additional layer of security by using a supported Ledger hardware wallet