No abstract

No abstract

In the preceding chapters, we have considered in detail prediction of the thermodynamic properties of fluids as this is of prime importance in the design of chemical process systems. In many areas of design, quantitative knowledge of the transport properties of the process fluids is also required and it is to the theory and practice of predicting such properties that we now turn.Unlike thermodynamic quantities, which refer fundamentally to properties of a system at equilibrium, transport properties determine the rate at which processes such as heat and mass transfer occur in a system that is not at equilibrium. Practical process situations often involve a system at or near to a steady state but, neverthe less, far from true equilibrium. In particular, there may be large spatial gradients in temperature, concentration or fluid velocity which give rise to non-linear transfer processes. In contrast, the science of fluid transport properties is concerned almost exclusively with somewhat idealised systems displaced only slightly from equili brium. The well known linear rate laws of Fourier, Fick and Newton arise in this context and each is associated with a transport coefficient that relates flux to gradient. The connection between these situations is that empirical models which are often, out of necessity, adopted in the description of practical transfer processes involve so-called transfer coefficients which depend parametrically on the formal transport properties. Here we shall be concerned only with the latter.Although for engineering purposes temperature and pressure are usually specified quantities, in the case of the transport properties both theory and expe rience show that temperature and density (or molar volume) are the fundamental state variables while, in essence, pressure is of no direct importance. Indeed, a transport property X, where X may be the viscosity rj, the thermal conductivity X, the diffusion coefficients D H and Dy, or the thermal diffusivity a, is written most conveniently as the sum of three contributions: X{p m T) = X 0 (T) + AX(p n ,r> + AX c (p",r).(9.1) Thermophysical Properties of Fluids Downloaded from www.worldscientific.com by UNIVERSITY OF BIRMINGHAM LIBRARY -INFORMATION SERVICES on 03/22/15. For personal use only.

The quantitative prediction of phase behaviour is the central problem in chemicalengineering thermodynamics and a very important consideration in the design of chemical process plant. The almost infinite number of possible mixtures and wide ranges of temperature and pressure encountered in process engineering is such that no single thermodynamic model is ever likely to be applicable in all cases. Conse quently, knowledge and judgement are required to select the most appropriate methods by which to estimate the conditions under which two or more phases will be in equilibrium.In this chapter, some basic algorithms for the determination of vapour-liquid equilibrium conditions in mixtures are discussed together with their application to certain kinds of flash calculation. A short discussion of liquid-liquid equilibrium is also included. For every topic considered, a computer routine will be given and used as the basis of an example. As elsewhere in this book, the computational methods used are not the most efficient possible (instead, they are designed to be easy to understand and modify) and this may be a consideration if the routines are to be called many times as a part of a larger calculation. Elementary Phase BehaviourBefore embarking on the quantitative aspects of phase equilibrium, it would be as well to have a qualitative understanding of the most common types of phase behaviour. The subject of phase behaviour in mixtures is a complicated one and we shall consider here only the most elementary kinds of phase diagram for vapourliquid and vapour-liquid-liquid systems.The simplest kind of vapour-liquid equilibrium (VLE) in a binary system is illustrated in the phase diagrams of Figure 8.1. Here, we show either the pressures at which vapour and liquid coexist under conditions of constant temperature (the P-x-y Thermophysical Properties of Fluids Downloaded from www.worldscientific.com by UNIVERSITY OF BIRMINGHAM LIBRARY -INFORMATION SERVICES on 03/22/15. For personal use only.

In the previous chapter, a brief description of the kinetic theory was presented. We now consider appropriate applications of the theory to the prediction of transport properties of pure fluids and mixtures in the gaseous and liquid states. The methods chosen are those that have a reasonable connection with the theory and in which a transport property X is written:The critical enhancement AX C will not be considered further here. The dilute-gas term X 0 will be discussed first. Then, the estimation of the excess transport properties AX for compressed gases by the method of Thodos will be examined while, for liquids, the scheme proposed by Dymond and Assael will be presented. Finally, for the estimation of the viscosity and thermal conductivity of non-polar mixtures, the corresponding-states scheme of Ely and Hanley will be outlined.In Section 10.6, numerical examples based on these methods will be given. Dilute Gases and MixturesFor a dilute gas, the viscosity is given from Eq.(9.30) as f is generally taken as unity. This expression, although strictly applicable only to a monatomic gas, is used in practice as a correlating tool for many pure gases including those with polyatomic molecules. The reduced collision integral Q* = Q/(n