Application of multiparameter corresponding states methods to predict the thermodynamic properties of nonpolar, polar, and associating pure fluids.

Abstract

Sensitivity of thermodynamic-property predictions to the values of polar/associative parameters are examined. It is shown that the parameters can be effectively determined using a small number of vapor-pressure and density data. As a result, the correlation can be applied readily to pure compounds with few experimental data. For undefined pseudocomponent fractions, the parameters are determined from the average normal-boiling point, molecular weight, density (at two temperatures), and, if available, the vapor pressure at a temperature lower than the normal-boiling point.

The corrrelation is generalized for other polar and associating compounds by using a temperature- and density-dependent factor to characterize their polar/association effects relative to those for water. Parameters in the polar/association factor are determined via multiproperty analysis of vapor-pressure and density data. The correlation is applied to a wide variety of compounds, including water, ammonia, alcohols, ethers, ketones, mercaptans, amines, phenol, cresols, xylenols, indanols, quinoline, acridine, dibenzofuran, and many other nitrogen-, oxygen-, and sulfur-containing model coal compounds. The average absolute deviations for vapor-pressure, density, and enthalpy and enthalpy-of-vaporization predictions are 1.8%, 2.6%, and 9.5 kJ/kg (4.1 Btu/lb), respectively.

The correlation is used to predict with reasonable accuracy the densities for several wide-boiling coal-fluid fractions from the Exxon-Donor-Solvent process.

A multiparameter corresponding-states correlation has been developed to accurately predict the thermodynamic properties of nonpolar, polar, and associating pure fluids. The nonpolar contribution to the correlation, developed using multiproperty regression analysis of vapor-pressure, density, and enthalpy-departure data for methane through n-decane, is due to Starling et al. (1978). The polar/associating contribution is developed in this research using multiproperty regression analysis of extensive vapor-pressure, density, and enthalpy-departure data for water, covering all fluid states. The correlation predicts vapor pressure and density for water with percent average absolute relative deviations of 0.6% and 1.4%, respectively. For enthalpy departure, the average absolute deviation is 9.3 kJ/kg (4 Btu/lb).