Viscosity prediction from a modified square well intermolecular potential model
In the chemical and process industries, viscosity of pure components and mixtures is an important property in hydraulics, heat and mass transfer calculations. As was indicated by an extensive literature review, there is a need for an accurate analytical predictive method for calculating the viscosities of pure component gases and liquids. In this work, a theoretically based viscosity model was developed by modifying a statistical mechanics based viscosity model for dense fluids based on the square well intermolecular potential. The original theory was corrected to account for the assumptions of only two-body interactions and molecular chaos for velocities and for the inadequacy of the square well potential. The model was modified so that it approaches the theoretically correct low density limit and to improve the dilute gas temperature dependence. The resulting three parameters were regressed from both gas and liquid data simultaneously utilizing a consistent, reliable database of 191 organic and inorganic fluids. Regressed parameters were subsequently generalized using group contributions. The model requires molar mass, critical temperature, acentric factor and density as input parameters. The modified square well viscosity model is applicable across the entire density range from gas to liquid. Ultimately, for gas and liquid viscosities of a wide variety of nonpolar, polar and hydrogen-bonding fluids, data are correlated within about 1 % and 2%, respectively, and predictions with generalized parameters are within about 2% and 6%, respectively. These results are considerably better than those from existing models.
Bibliography: p. 188-198.
Monnery, W. D. (1995). Viscosity prediction from a modified square well intermolecular potential model (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/20596