New discovery to reduce residual oil saturation by polymer flooding
MetadataShow full item record
Eight coreflood experiments were conducted to investigate the effect of aqueous hydrolyzed polyacrylamide (HPAM) polymer solutions on residual oil saturation in sandstone cores. Seven of the experiments were conducted in high-permeability (~1500 mD) Bentheimer sandstones, six of the cores were saturated with a viscous oil (~120 cp), and one core was saturated with a light (10 cp) oil. The eighth experiment was performed in a Berea sandstone core using the light oil. Experiments #6 to 8 were done by Pengpeng Qi. These experiments are included in this thesis to provide more complete and convincing results. All experiments were first saturated with brine, flooded with oil to reach initial oil saturation, and then waterflooded with brine to zero oil cut. For experiments with viscous oil, a viscous glycerin solution was injected after the waterflood until the oil cut was zero. FP 3630S polymer was used in the seven Bentheimer coreflood experiments and FP 3330S polymer was used in the Berea coreflood experiment. The polymer solutions in low salinity brine had a high relaxation time. Additional hydrolysis of the polymers was done to further increase the relaxation time. The coreflood experiments were designed to maximize the effect of viscoelasticity on the residual oil saturation by flooding the cores at a high Deborah number, N [subscript De], which ranged from 30-300. The low-salinity polymer floods were followed by a second polymer flood with a similar viscosity, but higher salinity (viscosity was controlled by increasing polymer concentration). The higher salinity resulted in a much lower polymer relaxation time than the first polymer in low salinity brine, and therefore a lower N [subscript De] for the coreflood. Two of the experiments included additional polymer floods by alternating between the low and high salinity polymer solutions. The original objective of this work was to investigate the effect of polymer elasticity (measured by the dimensionless Deborah number, N [subscript De]) on residual oil saturation. The polymer flooding experiments were designed to keep the capillary number less than the capillary number of the preceding glycerin floods as well as less than the critical capillary number to avoid a reduction in the residual oil saturation caused by a high capillary number. Early in this experimental study, a surprising and remarkable discovery was made that completely changed the direction of the research. The residual oil saturation following the high-salinity polymer floods was reduced to remarkably low values. All eight experiments showed that the low-salinity polymer floods with high Deborah numbers resulted in additional oil recovery. The average reduction in oil saturation was ~10% for the seven Bentheimer corefloods, including the one with light oil (4%). There was a (weak) correlation indicating lower residual oil saturations with increasing N [subscript De] consistent with the observations by Qi et al. (2017). The most surprising observation and discovery was that the residual oil saturation decreased between 4 and 21% with an average reduction of 11% when high-salinity polymer solution was injected following the low-salinity polymer flood with the same viscosity and at the same or similar flow rates. The total reduction in residual oil saturation from both polymer floods was 21% below the residual oil saturation of the glycerin floods with the same viscosity. The lowest residual oil saturation in these experiments was only 7%. This is a truly remarkable result considering the interfacial tension between the polymer solution and oil is about the same as between water and oil. Additional measurements are needed to understand the mechanisms e.g. wettability measurements before and after the polymer floods in low and high salinity brines.