Displacement of oil by water-effect of wettability, rate, and viscosity on recovery

Abstract

The displacement of oil by water in typical porous media has been studied using new equipment developed to handle flooding experiments on 1-in core plugs with a precision of about one pore per cent. Long core floods using cores up to 10 ft in length were used to supplement the short core studies. From the results of several thousand flooding experiments on a variety of porous media, we have found that the wettability is the single most important variable affecting the recovery history of a water flood. Cores are, therefore, usefully described as being (l) water-wet, (2) oil-wet, or of (3) intermediate wettability. This classification scheme has provided a sound basis for the interpretation of laboratory waterflooding data. For water-wet cores it can be concluded: (1) 1-in core plugs show the same recovery history as long cores when flooded under the same conditions of rate, viscosity and wettability; (2) recovery is practically independent of rate of flooding from reservoir rates up to 0 ft/day; (3) at extremely high rates (over 100 ft/day) significant recovery increases were observed; (k) displacement by imbibition alone displaces the same quantity of oil as water drives do with rates up to -0 ft/day; (5) lowering the inter-faclal tension will reduce residual oil if the lowered value is maintained at the flood front. An explanation for these conclusions based on the interplay of the viscous and capillary forces at the flood front (called "Viscap" concept ) is presented, and from this we may conclude that laboratory floods of water-wet core plugs offer no particular scaling problems. The intermediate wettability case is more complicated-, and because of limitations in measuring the wettability of cores and the reservoir, this group needs more study. The behavior of oil-wet cores is markedly different from the water-wet case. Generally, oil-wet cores show a considerable amount of oil production after water breakthrough even for a viscosity ratio of unity. Also, they exhibit a saturation gradient at the flood front (sometimes called an end effect). This end effect invalidates laboratory results unless the floods are scaled using a relationship. involving length, viscosity, rate, interfacial tension and contact angle.