Extending the UNIFAC-VISCO Model and Introducing the UNIFAC-THERMO Model for Improved Viscosity Prediction of Binary Liquid Mixtures

The accurate prediction of liquid mixture viscosity is
essential
for the design and optimization of chemical processes. This study
extends the UNIFAC-VISCO (UVM) model and introduces a new group-contribution
framework, the UNIFAC-THERMO (UTM) model, which eliminates the need
for experimental density data of mixtures, particularly beneficial
for ambient applications and cases lacking such data. Eighteen new
group interaction parameters (αnm) involving aromatic
alcohols, carboxylic acids, and cyclic ethers were determined to broaden
the applicability of UVM. Both models were validated using 335 binary
systems across 21 chemical categories. The Grand Average Relative
Deviation improved from 3.21% (UVM) to 2.75% (UTM) for dynamic viscosity
and from 3.21% (UVM) to 2.72% (UTM) for kinematic viscosity. A user-friendly
Excel tool implementing both models is provided to facilitate application.
Overall, the UTM establishes a more versatile and transferable framework
for viscosity prediction, reinforcing the role of group-contribution
methods in thermophysical property estimation.