PREDICTION OF MULTICOMPONENT VAPOR-LIQUID EQUILIBRIA FROM BINARY HEATS OF MIXING.

In order to design industrial distillation equipment, accurate vapor-liquid equilibrium (VLE) data are needed. In many cases such data are not available for the multicomponent mixtures with which one frequently must work. A method is described which allows the designer to predict binary or multicomponent VLE data at design conditions of temperature and pressure from only binary heat of mixing data and pure component vapor pressure and compressibility data. No measurements of VLE data are necessary. Heat of mixing data may be measured at convenient laboratory conditions for each of the possible binary pairs involving the constituents of the desired multicomponent mixtures. Such data may be obtained with great precision and relative ease at temperatures from 25-75 degree C and pressures from 1-400 atm using a new isothermal flow calorimeter developed by the authors. The heat of mixing data are curve-fitted using an h**E model, derived by means of the Gibbs-Helmoltz relation from a g**E model, appropriate to the system. The method is demonstrated for nonideal systems involving various hydrocarbon-hydrocarbon and hydrocarbon-alcohol mixtures under both isothermal and isobaric conditions.