Modeling the density, solubility and viscosity of bitumen/solvent systems using PC-SAFT

• A new and universal bitumen characterization using distillation and molar mass data.
• 8 Bitumen pseudo-components defined by boiling point temperatures.
• A proposed kij correlation to improve solubility matching.
• Viscosity modeled using Expanded Fluid Model with density predicted by PC-SAFT.
• The overall AARDs within 2.3% for density, 6.6% for solubility, and 14.6% for viscosity.

Stemming from statistical thermodynamics, Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT) shows promising results for predicting phase equilibrium and density for light and heavy oil. In this work, the simplified version of PC-SAFT was applied to model the density and solubility of bitumen/solvent systems and was integrated with the Expanded Fluid Viscosity Theory to predict and match the viscosity of such systems. A simple new eight-pseudo-component characterization method was developed for different Alberta Bitumens using distillation and molar mass data. The proposed characterization method for the simplified PC-SAFT was applied to match the density and solubility data of Athabasca Bitumen with gas alkanes and CO2 and was used to predict the density of Athabasca Bitumen and Peace River Bitumen with liquid solvents (alkanes and aromatics), without further adjustment. Finally, the viscosity of the bitumen and solvent mixtures was predicted and matched using PC-SAFT coupled with the Expanded Fluid Viscosity Model. The overall average absolute relative deviations were within 2.3% for the density of the 13 bitumen/solvent systems, within 6.6% for the solubility of three systems, and 14.6% for the viscosity of eight systems.