First order phase transitions and the thermodynamic limit

Abstract

We consider simple mean field continuum models for first order liquid-liquid demixing and solidliquid phase transitions and show how the Maxwell construction at phase coexistence emerges on going from finite-size closed systems to the thermodynamic limit. The theories considered are the Cahn-Hilliard model of phase separation, which is also a model for the liquid-gas transition, and the phase field crystal model of the solid-liquid transition. Our results show that states comprising the Maxwell line depend strongly on the mean density with spatially localized structures playing a key role in the approach to the thermodynamic limit.However, in a system of finite size or when a finite time horizon is considered, metastable states often play an important role and even unstable states may be crucial, as transient states, for extended time periods. The full set of states and their dependence on the various control parameters is conveniently presented in the form of bifurcation diagrams, well known in the context of dynamical systems and pattern formation theory [3][4][5]. The place of thermodynamic phase diagrams is taken by 'morphological phase diagrams' or state diagrams and stability diagrams [3,4,6]. The notion of a Maxwell point is often used in the context of pattern formation in nonconserved systems [7][8][9][10][11] to indicate equal energy states because of its dynamical significance [12,13]. In this context this notion applies equally to finite and infinite systems [7,8] although for finite systems it lacks the thermodynamic relevance as the condition for phase coexistence. In the context of buckling the corresponding concept is the Maxwell load [14].In this paper we show and discuss how the discontinuities in the TL represented by the Maxwell construction arise from the bifurcation diagrams relating stable, metastable and unstable steady states in finite-size systems. We focus on two systems: (i) phase decomposition of a binary liquid mixture and (ii) the liquid to crystalline solid phase transition. We investigate the transitions that occur in the context of the most basic mean-field continuum models for these two different phase transitions, namely, the Cahn-Hilliard equation [15][16][17] and the phase field crystal (PFC) model (or conserved Swift-Hohenberg equation) [18][19][20].Some aspects related to this question have been considered previously, in particular in relation to the nature of some of the states that can arise in finite-size systems in the two-phase region. References [21-24] describe theory and computer simulation results for atomistic models exhibiting gas-liquid, liquid-hexatic and hexaticsolid phase transitions that indicate how the Maxwell construction develops as the system size increases or the temperature decreases. For example, figures 3-5 of [23] compare Monte-Carlo computer simulation results in a finite three-dimensional domain (see also figures 2 and 3 of [22]) with the results from a capillary drop type model, also in a finite domain, with a mean-field expression for the chemi...

Resting and traveling localized states in an active phase-field-crystal model

“…The branch undergoes a series of folds corresponding to the addition of a pair of layers of density peaks, symmetrically with respect to y = 0 (Fig. 2); this is not the case in larger domains, however [106]. At each fold, the stability of the branch changes.…”

Efficient calculation of phase coexistence and phase diagrams: application to a binary phase-field-crystal model

“…Refs. [38,39]. domain size brings the secondary bifurcation where the localised states emerge from the periodic one closer to [further away from] the primary bifurcation.…”

“…Ref. [39] explains in detail how the increase in the number of folds ultimately results in the emergence of the Maxwell construction when approaching the thermodynamic limit. However, here, in contrast to in Ref.…”

Why a Large-Scale Mode Can Be Essential for Understanding Intracellular Actin Waves

“…It is convenient to first consider total conservation of an observable, described by the continuity equation where u is a scalar field, M is a mobility function, and F is a free energy. If the free energy contains an intrinsic length scale, like in the phase field crystal model or in wetting, stationary periodic and localized patterns may emerge [ 88 , 89 , 90 , 91 , 92 , 93 , 94 , 95 ]. The mutual aspect is coupling between the large-scale mode ( ), which is model independent and remains always neutral due to conservation (also known as the Goldstone mode), and the pattern forming instability of finite wavenumber (Turing) type [ 96 ], as shown in Figure 2 .…”