Find in Library

The objective of this study is to evaluate the predictive performance of three semi-theoretical approaches for modeling viscosity of binary liquid mixtures composed of 1-alcohols. For the mixtures that contain 1-alcohols with a similar number of carbon atoms, three schemes including the Rough Hard-Sphere model, the McAllister model, and UNIFAC-VISCO were demonstrated to be able to reproduce the mixture viscosity with an average deviation of 5%. For highly asymmetrical mixtures without the presence of small molecules, only McAllister-type model successfully reproduced the experimental values. The prediction done by the Rough Hard-Sphere model on the basis of using simple mixing rules for characteristic parameters indicated that this model should be further modified to improve its predictive capability.