We use Extended Irreversible Thermodynamics (EIT) to describe transport phenomena in membranes. The basic assumptions of EIT are summarized and the membrane is regarded as a passive phase where diffusive and viscous flow of a binary solution may occur. It is shown that in contrast to Linear Irreversible Thermodynamics (LIT), EIT is adequate to describe noninstantaneous transport processes across membranes. This is illustrated by showing that EIT predicts that the concentration profile inside the membrane obeys the telegraph equation instead of the usual diffusion equation. As a result we find that for dynamic osmosis the initial pressure gradient should grow as t 3 instead of the linear dependence implied by LIT. This and other differences between LIT and EIT are pointed out and the sense in which the latter may be considered as a generalization of the former is also discussed.
IntroductionThe most widely used theoretical tools in the study of the rich variety of transport phenomena through membranes have been linear irreversible thermodynamics [1], Navier-Stokes hydrodynamics (NSH) [2], and nonequilibrium statistical mechanics [3]. In fact, in spite of the undeniable success of LIT and NSH to deal with many irreversible processes in open, semipermeable or leaky membranes, they are limited to the description of linear processes due to the fact that both formalisms are constructed upon the local equilibrium assumption. But there are transport phenomena in membranes, such as flow through very narrow interstices, that do not fit into the scheme of LIT [2] and for which the local equilibrium assumption is no longer valid.One theory aiming at overcaming the limitations of LIT and NSH is extended irreversible thermodynamics. The basic idea of this approach is to modify the local equilibrium assumption by including the fluxes of the system as independent variables [4]. This theory has been successfully applied to many systems such as diffusion in a binary mixture [5], viscoelastic fluids [6], polar and electrical solids [7], systems with internal degrees of freedom [8], etc. The basic purpose of this work is to exhibit how BIT may fte used to describe transport across a membrane. We show that this theory is specially appropriate to describe phenomena associated with a transient response of the membrane. We exhibit how the structure of the theory incorporates in a unified and systematic way several features of transport through membranes that have been proposed in the literature using a variety of different approaches. In particular, it will be shown that the results of LIT can be derived from EIT and other new results will be discussed.