Pressure Evolution and Production Performance of Waterflooding inn-Heptane-Saturated Fired Berea Cores

Abstract

SummaryThis work presents experimental results and interpretation of injection pressure and recovery performance of waterflooding in strongly water-wet fired Berea cores saturated with n-heptane. The experiments were conducted at constant injection rate at room conditions, and the effects of injection rate and initial water saturation on the oil-recovery performance and dynamic-injection pressure were investigated. Elements of surprise were observed in the injection-pressure data. The pressure profiles showed four distinct regimes, each governed by capillary or viscous forces. At low capillary numbers (Ca=uμ/σ<10–6), capillarity governed two pressure regimes, corresponding to the core inlet and outlet. In the early part of waterflooding, pressure stayed constant for a considerable time before hydrodynamic pressure gradient could overcome the capillary pressure gradient. After viscous forces dominated, a linear increase in injection pressure over time was observed up to breakthrough time. A sudden pressure rise was observed close to breakthrough because of capillary retention at the core outlet. The pressure became constant after the breakthrough when the water- and oil-saturation distributions were stabilized. Changing the injection rate by an order of magnitude in the range from 2.2 to 22.2 pore volumes (PV)/D (equivalent to Ca = 10–7 to 10–6) did not appreciably change the oil-recovery performance; similar breakthrough time and final oil recovery were observed. The effect of initial water saturation was also investigated. When lowering the initial water saturation beyond that established in oil flooding, production performance and injection pressure were similar to those of a core without the initial water saturation. The injection pressure at breakthrough was found to decrease with increase of the initial water saturation. Waterflooding was modeled by including the capillary pressure and excellent agreement was obtained with experimental results of production and injection pressure. We find that in the absence of in-situ saturation measurements, the injection pressure is a better variable for tuning the model parameters compared with the production history alone.