A Predictive Equation-of-State for Modeling Microemulsion Phase Behavior with Phase Partitioning of Co-Solvent

Abstract

Co-solvents are widely used to improve chemical flooding formulation design, but their partitioning between phases has great impacts on microemulsion phase behavior. Therefore, it is critical to accurately model microemulsion phase behavior with phase partitioning of co-solvents in a compositional chemical flooding simulator, in order to correctly predict oil recovery and better evaluate the performance of the designed formulation. We use the physics based HLD-NAC EOS implemented with the co-solvent partitioning model developed by Biais et al. (1981) and Hirasaki (1982) to model microemulsion phase behavior with cosolvents. The HLD-NAC model can determine microemulsion phase type by correlated phase behavior dependent variables and can calculate the overall interfacial area and sizes of micelle by characterized surfactant properties. To accurately estimate the interfacial area taking into account the contribution of co-solvents, an interfacial pseudophase composed of surfactant and co-solvents is defined. Partitioning coefficients of co-solvents between interfacial pseudophase and bulk pseudophase are experimentally measured. This novel model has only one fitting parameter, which is the tail length (Lsurf) of surfactant mixtures. We used five microemulsion systems to examine the developed approach. Solubilization ratios under a salinity scan of these systems are reproduced. Without using the co-solvent partitioning model, the matched Lsurf is dramatically underestimated for each system due to assuming all co-solvents are adsorbed onto the interface. With the improved model, the matched Lsurf is more physically representing the actual surfactant tail length. Microemulsion compositions predicted by the novel HLD-NAC EOS is in a good agreement with the experimentally measured microemulsions at all phase behavior types from Winsor Type I, through Winsor Type III to Winsor Type II. The results prove the physics based HLD-NAC EOS coupling with the thermodynamic co-solvent phase partitioning model can accurately simulate phase behavior of surfactant/co-solvent/brine/crude oil systems. The novelty of the present work is to exercise the physics-based HLD-NAC EOS to model the phase behavior of the aforementioned systems considering co-solvent partitioning. The new model can be used in compositional chemical flooding reservoir simulation to improve the predictability of surfactant floods.

Challenges in modeling microemulsion phase behavior

“…Besides, just changing the optimum characteristic length may not be sufficient to perform good calibration and prediction. To the best of our knowledge, the only paper that presented a comprehensive methodology for handling cosolvents with the HLD‐NAC approach through the use of pseudo‐ternary diagrams was by Basilio et al (2017) which considered the models from Biais et al (1981) and Hirasaki (1982) to model the alcohol partition coefficients. Basilio et al (2017) considered an HLD‐NAC approach similar to that presented by Acosta et al (2003) combined with the model from Jin et al (2015).…”