Experimental Investigation of Osmosis as a Mechanism for Low-Salinity EOR

The objective of this paper is to describe experiments conducted to investigate osmosis as a mechanism for low-salinity enhanced oil recovery (EOR). For this purpose, an experiment was designed to facilitate enhanced oil recovery by osmosis-induced connate water expansion, while at the same time reducing the contributions of other proposed low-salinity mechanisms. Considerations were also made to achieve osmotic water transport rates comparable to what is expected at reservoir temperature. The correlation between enhanced oil recovery and the surface-to-volume ratio was of particular interest. Because the osmotic pressure gradients occur over distances comparable to the pore size, it is plausible that fluid redistribution due to osmosis would lead to a fairly local redistribution of oil, and thereby have a small impact on overall enhanced recovery in the field. However, near exposed surfaces, this local redistribution may result in oil production. Previous investigations of osmosis as an underlying low-salinity mechanism have consisted of visualization experiments, where water transport and oil movement under influence of osmotic gradients have been observed. Our experiments are intended to increase the understanding of the relative importance of osmosis in both small-scale low-salinity experiment results, and for field-scale low-salinity flooding. In the experiments, oil-wet sandstone samples with different surface-to-volume ratios were saturated with high-salinity water and oil to irreducible water saturation. The samples were first left to spontaneous imbibe in high-salinity water and afterward in low-salinity water. Additional oil production from spontaneous imbibition of low-salinity was recorded and compared with the surface-to-volume ratio. The experiment was performed twice, at both ambient and elevated temperatures. The experiments at ambient temperature resulted in increased oil production values of 8-22% of pore volume by low-salinity spontaneous imbibition. No clear correlation was found between increased oil recovery and the surface-to-volume ratio. A correlation was, however, seen between increased oil production and the pore volume. Thus, increased oil production by low-salinity imbibition seems to be proportionate to the pore volume. The experiments at elevated temperature resulted in low values of increased oil production by low-salinity spontaneous imbibition, and the values do not seem to correlate with either surface area or pore volume. The low response is believed to be caused by thermal effects from repeated heating and cooling of the samples during the preparations. Our results cannot dismiss osmosis as an important mechanism for low-salinity EOR. Possible explanations for the correlation between increased oil production and pore volume are hysteresis and simultaneous connate water expansion throughout the core.