Analyzing the Smart Water Core-flooding Data in Carbonates by Modelling the Oil Breakthrough Time Using a Novel Reactive Transport Model

Abstract

Summary Numerous forced and spontaneous imbibition experiments in carbonate cores have demonstrated that the injection of modified-salinity water with an ionic composition different from the formation water (also called Smart Water) accelerates oil recovery and reduces the remaining oil saturation. Different physical mechanisms are suggested based on the carbonate-oil-brine physicochemical interactions, e.g., wettability alteration due to the ion exchange and surface complexation, carbonate dissolution, and water-weakening (only in chalk). Each of these can be described by relatively accurate thermodynamic models (e.g., ion exchange and dissolution) or a combination of thermodynamics and semi-empirical models. Although there is still an ongoing discussion on the importance and/or relevance of these physical mechanisms, it is widely accepted that due to the change in ionic composition the mobilities of the oleic phase and to a lesser extent the aqueous phase are altered toward a more water-wet state, exhibiting increased capillary forces and improved sweep efficiency. This is reflected in the mathematical models as two sets of relative permeability curves, one for the formation water-oil and modified-salinity water-oil systems. The multiphase flow model switches between these relative permeabilities based on a chosen indicator in the carbonate-oil-brine system, e.g., the total salinity of the brine for simple transport models to the surface density of a complex on the carbonate surface for more complicated reactive transport models. A quick review of the literature shows that apart from the complexity of the reactive transport models and the chosen indicator for the mobility alteration, almost all the proposed models can reasonably fit the measured recovery factors in a selected set of smart water core floodings. This is due to the large number of adjustable parameters in the two sets of relative permeability curves, which makes the choice of physical mechanism for the development of a mechanistic model irrelevant. Here, we address this problem by performing a constrained history matching of the Smart Water core flooding in carbonate cores (limestone and chalk). Moreover, we give a higher priority to fitting the oil breakthrough time during the smart-water injection in tertiary mode. We use an optimized surface complexation model to accurately simulate the adsorption of ions on the carbonate surface at high temperature. We then couple it with an in-house finite volume solver and a state of the art optimization package to obtain the relative permeability parameters. Our results show that the oil breakthrough times can only be correctly obtained by accurately modeling the carbonate-brine interactions and choosing the adsorbed potential determining ions’ concentrations as a mobility-modifier indicator.