Abstract
A promising correlation has been developed that can be used to predict miscible or near-miscible residual oil saturation, Sorm, for a wide range of injected gases, crude oils, temperature, and pressure conditions. The correlation is based on representation of the chemical and physical properties of the crude oil and the injected gas through Hildebrand solubility parameters. This approach has the advantage that characteristics of both the injected gas and crude oil are included in the correlation, in contrast to correlations based solely on properties of the injected gas.
The correlation was developed using available experimental data for tertiary recovery of eight crude oils in carbonate and sandstone cores with common EOR gases (CO2, N2, CH4, CH4 + liquefied petroleum gas). Results for 45 coreflood tests at reservoir conditions collapsed along a band when Sorm was plotted as a function of the difference in solubility parameter between the injected gas and the crude oil. Results for a pure oil, decane, with CO2 lay along the same band. The success of this correlation scheme may be due to the basic characterization of the fluids and to a relationship between solubility parameters and interfacial tension. Use of the correlation requires knowledge of only injected gas composition, injected gas density, oil average molecular weight, and temperature.
This empirical correlation should have utility in screening studies or in process simulation as a simple means to forecast residual oil saturations as measured in coreflood tests. The correlation can be used to predict roughly the effects of changes in pressure, temperature, or injected gas composition on residual oil saturation. A new method to predict minimum miscibility pressure based on the solubility parameter concept is also described.
Introduction
The miscible residual oil saturation, Sorm, is a key property for simulation and screening studies of gas injection EOR processes. This property represents the oil saturation remaining in a porous media after injection of a large bank of a high pressure gas, such as CO2, N2, or CH4, after a waterflood. The miscible residual oil saturation thus represents the local displacement efficiency of oil by the injected gas in a ternary system of oil, gas, and water. Injected gases are frequently supercritical fluids, and proposed mechanisms of oil recovery include low interfacial tension displacement, extraction, and oil swelling. Within the industry, a common parameter used in design of these processes is the minimum miscibility pressure (MMP) or minimum miscibility enrichment (MME) level for hydrocarbon gases as determined from sandpack slim-tube tests. Recent work has suggested use of reservoir-condition coreflood data in design of gas injection EOR processes instead of MMP or MME levels.
Miscible recovery processes have been studied extensively, and a variety of schemes have been developed to predict MMP. In contrast to the large number of predictive schemes for MMP, few methods have been proposed to predict Sorm. Use of a capillary number correlation has been suggested, but this approach requires knowledge of interfacial tension between equilibrated phases. A correlation of residual oil saturation with pore structure in carbonates has been suggested as well as correlations of Sorm with reduced density of the injected gas for one crude oil with several hydrocarbon gases. Although interesting, these approaches do not meet the need for a general method to predict Sorm for any injected gas and any crude oil, and laboratory coreflood tests at reservoir conditions are usually recommended to determine this important measure of local displacement efficiency.
Laboratory and Field Measurements of Residual Oil Saturation from Gas Injection at the Oseberg Field
