Date of Award

Fall 1996

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Chemical Engineering - (Ph.D.)

Department

Chemical Engineering, Chemistry and Environmental Science

First Advisor

Dana E. Knox

Second Advisor

R. P. T. Tomkins

Third Advisor

James M. Grow

Fourth Advisor

Norman W. Loney

Fifth Advisor

Ralph D. Gray

Abstract

This study dealt with the prediction and correlation of vapor-liquid equilibria behavior of nonideal fluids. The thermodynamic formalism of the GE-EoS models, which combines the two traditional methods γ-Φ and Φ-Φ used so far for low and high pressure phase equilibria correlations respectively, has been combined with the 1FGE model, based on one-fluid theory, to produce a more consistent approach to the phase equilibrium problem.

In the first part of our study we examine the predictive abilities of our model for vapor-liquid equilibria of highly nonideal fluids. The results establish the fact that the Huron-Vidal mixing rule with a one parameter version of the lFGE model, is able to successfully utilize available experimental information at low pressures for phase equilibria predictions of multicomponent mixtures over an extended range of pressures and temperatures.

In the second part of the study we perform an analysis of the correlative abilities of the 1FGE model, as applied to hydrogen-hydrocarbon mixtures. The results of this part suggest that the unique local composition character of the 1FGE model, along with its one size and one temperature-dependent energy parameter, make it able to adequately describe vapor-liquid equilibria behavior of multicomponent mixtures for this highly asymmetric class of mixtures. Moreover, it is shown that the model parameters for binary hydrogen-hydrocarbon mixtures can be correlated to the acentric factor of the hydrocarbon.

The important class of the refrigerant mixtures was modeled in the third part of this study. The lFGE model was introduced into the Wong-Sandler mixing rule, based on the infinite pressure state thermodynamic formalism. The results for these systems showed that a limited amount of experimental data, either at low or high pressures can be utilized to provide a parameter which is practically independent of the temperature set used. As shown from the results, this single parameter can be used to extend vapor-liquid equilibria predictions over a range of conditions for this difficult class of systems. More importantly, we set a heuristic rule able to screen multiparameter and one parameter models. A coordination temperature-parameter planet can be used as a predictive tool from a limited amount of information.

Our model comes in lieu of the GE models based on two-fluid theory, which are inconsistent with the one fluid character of an EoS. The 1FGE-EoS framework proposed in this work meets current needs in the area of Applied Thermodynamics, which require that the model's parameters can be obtained from a limited information of experimental data and can give for accurate phase equilibria predictions of nonideal mixtures from low to high pressures.

Share

COinS