Tests of Predictive Viscosity Models for Pure Liquids

It is of considerable importance to be able to predict accurately the viscosity of liquids over a wide range of conditions. In the present work, the ability of the three-parameter generalized corresponding states principle (GCSP) for the prediction of the viscosity of pure liquids is demonstrated. The viscosity of six different classes of pure liquids, vi:., alkanes (19 compounds; 207 data points), cycloalkanes (6 compounds; 74 data points), alkenes (9 compounds; 146 data points), aromatics (4 compounds; 123 data points), alkanols (8 compounds; 89 data points), and esters (4 compounds; 28 data points) have been predicted over a wide range of temperatures using the three-parameter (TQ, Pc, /)) GCSP. Five options for the third parameter (0) were studied, vi:., Pitzerʹs acentric factor to, molar mass M, characteristic viscosity rj*, critical compressibility factor Zc, and modified acentric factor Q, in addition to groups toZc and £2ZC being treated as composite third parameters. Pressure effects were neglected. Good agreement between experimental and predicted values of viscosity was obtained, especially with either to or //* being used as the third parameter. Furthermore, the viscosities of alkanes predicted by the TRAPP method and an empirical, generalized one-parameter model for liquid hydrocarbons provide comparisons with the more accurate GCSP method. The GCSP provides a simple and yet a powerful technique for the correlation and prediction of viscosities of a variety of pure liquids over a wide range of temperatures.