The idea of injecting low salinity water into a petroleum reservoir is not novel and was often used in the 70s prior to the injection of surfactant. Recently it was shown that simply injecting sufficiently low salinity water improves oil recovery. Many possible mechanisms concerning low-salinity waterflood have been proposed in the literature. This paper describes an experimental investigation into some of the factors control ling the increased oil recovery observed when low salinity brine is injected into oil saturated reservoir core samples. Extensive chemical analyses were per formed on the effluent showing the extent of inter action between the injected brine, the oil and the rock matrix.
It has been 10 years since Yildiz and Morrow (1996) pushed forward the research started by Jadhunandan (Jadhunandan, 1990; Jadhunandan and Morrow, 1991; Jadhunandan and Morrow, 1995) and published their paper on the influence of brine composition on oil recovery, which showed that changes in injection brine composition can improve recovery. Tang and Morrow (1999) advanced the research on the impact of brine salinity on oil recovery, followed by other researchers such as Webb et al. (2004; 2005) and McGuire et al. (2005), which carried out an extensive research program on low salinity injection. This program included numerous core floods at ambient and reservoir conditions (e.g. temperature and pressure with live fluid) both in secondary and tertiary mode, single well tracer tests and log inject log, which resulted in a series of publications (McGuire et al., 2005; Webb et al., 2004; Webb et al., 2005) and the registration of the LoSal™ EOR process trademark.
Numerous hypotheses have been devised to explain the increase in oil production associated with low salinity water injection, including increasing pH leading to in-situ saponification and interfacial tension reduction, emulsion formation, clay migration, and wettability alteration (McGuire et al., 2005; Tang and Morrow, 1999; Yildiz and Morrow, 1996). In this paper the fines migration and pH increase mechanism are reviewed and discussed in the light of new data obtained during ambient and reservoir condition low salinity floods. Also a new mechanism based on the extended DLVO theory and cation exchange is discussed.
All corefloods discussed in this paper were per formed on nominally 3" long by 1.5" diameter plug samples from sand stone oil reservoirs. Plug samples were loaded into hydro static coreholders and miscibly solvent cleaned with cycles of toluene and methanol, prior to saturation with simulated formation brine. Initial water saturation was acquired by constant pressure oil flood. Initial water saturation values were found to be matched to those corresponding to the height of the sample above the oil water con tact. From analyses of sister samples using in-situ saturation data, the distribution of water was uniform across the length of the core samples. In all cases simulated formation water was used for the initial water.
Samples were then loaded into hydro static coreholders, prior to taking the samples to the conditions of the test.
