There is widespread interest in improved oil recovery by the low salinity effect (LSE) and a pressing need to better predict the likely response and its relation to wettability change. A LSE in kaolinite-bearing sandstones can arise from detachment of crude oil, by its peeling from rock surfaces due to increased oil/rock repulsion, and/or by detachment of mineral fines with adhering oil, due to increased fines/rock repulsion. In a mixed wet sandstone reservoir, oil is typically in close contact with an extremely small fraction of total rock surface, a key component of which are asperity tips such as at edges of kaolinite platelets. An Integrated pH Ion Surface Electrostatics (IpHISE) model is used to predict speciation and interactions of oil surfaces and kaolinite edges across NaCl and CaCl 2 solutions of variable pH in sandstones. At pH < 5, a LSE can arise by weakened oil adhesion due to fewer positively charged oil base groups adsorbed to kaolinite edges. At higher pH, the electrostatics is dictated by competition between negatively and positively charged acid groups produced by respectively deprotonation and calcium binding. The LSE is predicted to be strongest in a narrow range around pH 5−6 in which salinity reduction switches the oil/kaolinite edge interaction to repulsive. At pH > 6, the interaction becomes increasingly repulsive at all salinities. There, a LSE can only arise from the extended range of repulsion, both between oil and kaolinite edges and between the latter and the underlying rock. The existence and cutoff values of these pH ranges depend sensitively upon the oil's acid number/base number, salt concentrations, and the pH shift caused by injection of low salinity fluid.

Duplicate tests on mixed-wet sandstone cores have been shown to give closely reproducible results. For studies with reservoir cores, duplicate tests are often precluded because of the limited availability of cores and their heterogeneity. In repeat flooding tests for recovery of crude oil, it has been found that each test affects the outcome of a subsequent test. Some consistent trends of increase in recovery have been observed. In this paper, results are presented for up to four cycles of flooding between initial water saturation and residual oil for reservoir sandstone, outcrop sandstone, and outcrop oolitic limestone. A single crude oil was used for most of the tests. Sequential flooding was also tested for low-and highviscosity mineral oils. The brine was synthetic seawater for all but one of the floods. Each consecutive flood for the sandstones resulted in a systematic increase in the recovery of crude oil. Crude oil was sometimes produced as an emulsion of about 2% water. Trends for carbonate cores were comparable except that, for one of four of the tested cores, reduced recovery was observed for the second cycle. Implications for establishing baseline results for assessment of processes, such as low-salinity water flooding, are discussed.

A wide range of outcrop sandstones and carbonates have been tested for waterflood response to one twentieth dilution of synthetic seawater, using a single crude oil that gave high response to low salinity flooding for a reservoir rock. The tested outcrop rocks included 17 sandstones and 6 carbonates. Gas permeabilities ranged from 1.49 to 7,187 mD and porosities from 10 to 39%. The average reduction in residual oil for tertiary response was only 1.5% OOIP with the highest being 5.8% OOIP; some rocks showed no response. After testing, three of the outcrop rocks that responded to injection of low salinity brine were restored by cleaning and re-aging with crude oil using procedures comparable to those commonly used in restoration of reservoir cores. When re-tested, the response to low salinity flooding was eliminated for two of the cores and significantly reduced for the third core. Companion plugs for 6 of the sandstones and 3 carbonates were tested for increased oil recovery using low salinity connate and injected brine; this data provided comparison with the measurements for high salinity connate and injection brine. Six of the plugs showed increase in recovery for the low salinity waterfloods, and three showed decrease. Results are also summarized for the effects of reduction in salinity for 11 sandstone cores taken from five reservoirs and 8 carbonate cores all taken from the same reservoir. Comparison between results for outcrop and reservoir cores, including literature data, indicates that overall response to low salinity flooding is significantly higher for the reservoir cores.

The spreading of multi-component oils on water has been investigated by direct observations and predicted from measurements of the interfacial tensions and surface tensions of decane, toluene, heptane, and their mixtures. Pure decane does not naturally spread at ambient conditions, as indicated by its negative spreading coefficient. However, when decane is mixed with toluene and heptane, the mixture spreads on water over a wide range of compositions. The spreading coefficients are highly nonlinear with respect to concentration and feature a maximum. The spreading is ascribed to preferential accumulation of toluene at the oil/water interface and heptane at the oil/vapor interface. Molecular dynamics simulations corroborate the hypothesis of preferential accumulation. The accumulation of lighter alkanes at the oil/vapor interface reduces the surface tension, and the accumulation of aromatics at the oil/water interface decreases the interfacial tension. As a consequence, the oil mixture spreads over water.