Experimental results are presented that demonstrate the effect on residual oil, under water-wet conditions, of particle size, particle-size distribution, macroscopic particle size, particle-size distribution, macroscopic and microscopic heterogeneities, microscopic dimensions such as ratio of pore-body to pore-throat size, and pore-to-pore coordination number. Experiments were pore-to-pore coordination number. Experiments were performed in random packs of equal spheres, heterogeneous performed in random packs of equal spheres, heterogeneous packs of spheres with microscopic and macroscopic packs of spheres with microscopic and macroscopic heterogeneities, two-dimensional (2D) capillary networks having various pore geometries, and Berea sandstone. Detailed information on residual oil structure is presented, including blob-size distributions of residual presented, including blob-size distributions of residual oil. Major conclusions areresidual saturations are independent of absolute pore size, per se, in systems of similar pore geometry;well-mixed two-component aggregates of spheres gave virtually the same residual saturations as random packings of equal spheres;clusters of large pores accessible through small pores will retain oil;high aspect ratios tend to cause entrapment of oil as a large number of relatively small blobs, each held in single pores; andthe role of pore-to-pore coordination number is generally secondary; pore-to-pore coordination number is generally secondary; hence, correlations that have been proposed between residual oil and coordination number are unreliable. Introduction In recent years, there has been increased interest in the factors that determine the magnitude of residual oil and its microscopic distribution. Residual oil remaining in the swept zone of a waterflood is often taken as the target oil for enhanced recovery processes. Oil saturations remaining in these zones typically can occupy 15 to 35% of the pore space, but values outside this range are often measured. For the reservoir, it can be expected that the pore structure, the initial water content, and the superimposed effects of wettability determine recovery behavior and residual oil distribution under normal waterflood conditions. Salathiel has presented examples of the manner in which pore geometry, wettability, and volume throughput of floodwater can interact to affect oil recovery characteristics and final oil saturation. The likely complexity of trapping phenomena is indicated by the work of Wardlaw and Cassan, who investigated possible correlations between residual oil and 27 petrophysical parameters. Rocks with similar macroscopic properties often differed markedly in their residual oil saturations, and no significant correlation was observed between displacement efficiency and permeability. A tendency for residual nonwetting-phase permeability. A tendency for residual nonwetting-phase saturations to increase as porosity decreased was noted. This was related to a strong relationship between trapping and aspect ratio (ratio of pore-body to pore-throat size). A theory of residual oil trapping has been proposed by Larson et al. that provides an alternative explanation of the relationship between residual oil and porosity. It was reasoned that the trapped nonwetting-phase saturation will correspond reasonably well to the percolation threshold i.e., to the oil saturation at which oil continuity through the pore space is lost. SPEJ p. 311

Morrow, Norman R., SPE, New Mexico Petroleum Recovery Research Center Petroleum Recovery Research Center Lim, Hau T., SPE, New Mexico Petroleum Recovery Research Center Ward, Jill S., SPE, New Mexico Petroleum Recovery Research Center Summary Displacements in strongly water-wet cores; are compared with results of similar tests for a mixed wettability condition induced by a selected crude oil. Cores exposed to crude oil showed weakly water-wet imbibition behavior and 30 to 65% improvement in microscopic displacement efficiency. Other characteristics included clean breakthrough and low relative permeability to water at residual oil saturation. Flow visualization experiments with crude oil in micromodels showed improved oil recovery from pore bodies with oil trapping in pore throats. Capillary numbers for mobilization of residual oil from weakly water-wet systems were higher than for strongly water-wet systems. Contact-angle measurements showed that adsorbed transition metal ions at a high-energy surface could have a dominant effect on wetting behavior. Introduction Determination of reservoir wettability and its effect on oil recover are long-standing problems in reservoir engineering. Many researchers maintain that waterflood oil recovery is greatest under strongly water-wet conditions. Owens and Archer predicted steadily decreasing oil recovery as the water advancing contact angle increases. Other work has shown that wetting conditions other than strongly water-wet are frequently encountered and may actually be preferable. In 1955, Richardson et al. reported low residual saturations of kerosene to waterflooding in fresh cores from the East Texas field. Repeated alteration of oil- and waterfloods resulted in increased waterflood residual oil saturation (Sor) that was further increased by extraction between displacement tests. Water imbibition results indicated that a change in wetting behavior had occurred and that the extracted cores were more strongly water-wet than the fresh samples. Salathiel explained the phenomenon of fresh core behavior by postulating a mixed-wettability condition. Strongly oil-wet paths through the rock are generated at those parts of the pore surface in contact with crude oil, while the remainder stays strongly water-wet. Salathiel concluded that these paths are connected in consolidated media and allow oil to continue to flow even at very low oil saturations. Laboratory-prepared, mixed-wettability systems gave low Sor by extended waterflooding. Salathiel noted that the low Sor reported for the East Texas cores was achieved after flooding with 40 PV of water, a point not mentioned by Richardson et al. Salathiel did not report imbibition results for his mixed-wettability cores; thus, comparison with those of the fresh East Texas cores is not possible. The anomalously low Sor found for the East Texas field was ascribed to long-term gravity drainage rather than direct displacement by water-flooding. Low Sor for crude-oil displacements was reported by Rathmell et al. Imbibition tests established that most of the systems that gave decreased all trapping were weakly water-wet. Clean water breakthrough with little or no subsequent oil recovery was observed. Amott also found that Sor for weakly water-wet Berea cores was lower than for strongly wetted cores. The variation in oil recovery with wettability change is of interest for a number of reasons. Three of the most important are:prediction of oil recovery in the course of making economic evaluations of waterflooding,determination of the magnitude of Sor which is critical to the economic evaluation of tertiary oil recovery processes; andimprovement of waterflood recovery and perhaps tertiary recovery that may be achieved by wettability alteration. However, systematic and unambiguous studies of the effects of wettability are not easy to design or carry out. A major difficulty in working with oil/water interfaces at high-energy mineral surfaces is creation of well-characterized wetting situations. Several attempts have been made to relate wettability change to extent of reaction with silanizing agents, but none of the published procedures has gained general acceptance as a means of procedures has gained general acceptance as a means of providing a range of known wettability conditions. providing a range of known wettability conditions. Relationships between oil/water contact angles at solid surfaces and concentration levels of known compounds other than silanes are also available. However, difficulties in reproducing reported contact-angle measurements indicate that these systems are not likely to provide reliable wettability control in porous media, especially considering the added complexity that the pore surfaces may not correspond well to the carefully selected and prepared smooth mineral surfaces generally used for prepared smooth mineral surfaces generally used for contact angle measurements. Several studies have shown the importance of crude-oil components, especially the asphaltenes, in altering the wettability of mineral surfaces. SPEFE p. 89