CFD and Population Balance Modeling of Crude Oil Emulsions in Batch Gravity Separation—Comparison to Ultrasound Experiments.

Water-in-oil emulsion destabilization and separation in a batch gravity separator was investigated experimentally and by numerical modeling. A multiphase computational fluid dynamics (CFD) was used with a population balance model (PBM) to model separation behavior of crude oil emulsions. The inhomogeneous discrete method is used to solve the population balance equations. Closure kernels are applied to model droplet–droplet coalescence. To describe the increase in emulsion viscosity with water concentration, an emulsion viscosity model was selected that predicted emulsion stability and the denser emulsion layer forming above the coalescing interface, otherwise known as the dense packed zone or layer (DPZ). The results from a commercial CFD code are compared to experimental data of the water fraction vertical distribution measured by low-power ultrasound in the batch separator. The predicted time-dependent profiles of water fraction in the separator were found to be in good agreement with the experimental measurements for the range of water content from 6 to 50%. The model predicts the effect of water fraction on the separation kinetics and the evolution of the DPZ. Further studies are underway to apply the models to emulsions from different types of crude oils. [ABSTRACT FROM PUBLISHER]