Improving Well Injectivity by Surfactant Flushing - A Digital Rock Study

Abstract There are many causes for injection wells to perform poorly; in this paper, we address the effect of residual hydrocarbons near the wellbore, which reduce the rock's effective permeability to brine, thus decreasing well injectivity. Immobile hydrocarbon saturations are found around injection wells when these are drilled above the water-oil contact level, which arises when a producer well is switched to injection, or the formation's underlying aquifer is hard to reach. Furthermore, oil can accumulate around a wellbore when produced brine containing even trace amounts of hydrocarbons is used as the injection fluid. To address this problem, it is possible to flush out hydrocarbons around the wellbore by periodically injecting small amounts of surfactants for short periods. This technique leverages the capillary desaturation behavior of a multi-phase fluid mixture, wherein increasing the capillary number of a rock-fluids system beyond a threshold value will decrease the residual hydrocarbon saturation. The capillary number, which characterizes the ratio of viscous to capillary forces, can be increased by injecting surfactant loaded brine; this reduces the hydrocarbon-brine interfacial tension which reduces the capillary forces, decreases the system's residual hydrocarbon saturation, and increases the brine's effective permeability. To efficiently assess the impact of a surfactant flush, we perform digital rock multi-phase flow simulations on Berea and Fontainebleau sandstones at different capillary numbers. The simulations provide the residual hydrocarbon saturation and the brine's effective permeability as a function of capillary number, which is related to the amount and type of surfactant. These results are then used in a radial wellbore simulation to compute the attainable injection rate for a maximum allowable pressure drop. Since the largest pressure drop occurs very close to the wellbore, the surfactant flush is very effective even though it only affects a few feet from the wellbore. For the specified scenario we observe a potential well performance improvement ranging between 20% and 150%. By using digital methods, this study was performed in about two weeks, at lower cost than Special Core Analysis Laboratory (SCAL) physical testing, and with ideal reproducibilty since the capillary number can be modified without affecting the sample or any other aspect of the test procedure.