Interpolation and extrapolation of thermodynamic data for liquid mixtures are common necessities in chemical engineering. The model of ideal solutions is useful for providing a first approximation and a reference, but deviations from ideality are frequently large. These deviations are expressed by excess functions which depend on the concentrations of the components and on the temperature.As Gibbs energy cou gl d be conveniently expressed by an algeresults calculated r rom several models. We consider both THE WILSON AND H E I L E Q U A T I O N STo take into account nonrandomness in liquid mixtures, Wilson (24) suggested a relation between local mole fraction x i 1 of molecules 1 and local mole fraction x21 of molecules 2 which are in the immediate neighborhood of molecule 1: Both Equations ( 2 ) and (5) are useful, semiempirical relations for thermodynamic excess functions; both equations contain only two adjustable parameters per binary,

No abstract

A method is presented for fast and accurate determination of the limiting activity coefficient of a solute dissolved in a liquid mixture. It is based on the study of the solute elution with time; the solute is stripped from the solution by a constant flow of inert gas. The variation of solute concentration in the gaseous phase is measured by gasliquid chromatography. Experimental results for several systems are in good agreement with data from retention time measurements and extrapolation of vapor-liquid equilibria.

International audienceA new equation of state for electrolyte solutions has been developed from an expression of the Helmholtz free energy containing a nonelectrolyte part and a part relative to ions. The nonelectrolyte part is taken from the equations of state (EOS) of Schwartzentruber et al. (1989). The ionic part is composed of an MSA long-range term to account for electrostatic interactions and a short-range interaction term specific to ions. Using correlations between parameters and experimental ionic diameters, the model reduces to a one-parameter model. It has been applied to numerous strong electrolyte systems and extended to ternary systems to test its predictability without mixing parameters for ions. Its results compare well to the results reported for other one-parameter models (electrolyte EOS). Furthermore, it was found that the cation-anion interaction parameter could also be correlated to experimental ionic diameters. Then, the osmotic coefficients of 28 alkaline and alkaline-earth halide systems may be represented with a root mean square relative deviation of 2.9% using only six correlation parameters. This result has been extended to other systems, with the conclusion that the model with all parameters correlated may also be applied to systems other than halide solutions. The resulting model is predictive. The quality of the prediction was tested by determining osmotic coefficients relative to six systems without any parameter adjustment. The deviations of the predicted values range from 2.0 to 5.4%. The quality of the representation of mixed salts systems without mixing parameters was evaluated using experimental osmotic coefficients of 30 ternary systems

No abstract