Interrelationship of Wettability, Initial Water Saturation, Aging Time, and Oil Recovery by Spontaneous Imbibition and Waterflooding Xianmin Zhou,* SPE, and N.R. Morrow, SPE, U. of Wyoming, and Shouxiang Ma,** SPE, Western Research Inst. Summary Previous studies of crude oil/brine/rock ~COBR! and related sys- tems showed that wettability and its effect on oil recovery depend on numerous complex interactions. In the present work, the wet- tability of COBR systems prepared using Prudhoe Bay crude oil, a synthetic formation brine, and Berea Sandstone was varied by systematic change in initial water saturation and length of aging time at reservoir temperature ~88°C!. All displacement tests were run at ambient temperature. Various degrees of water wetness were achieved and quantified by a modified Amott wettability index to water, the relative pseudowork of imbibition, and a newly defined apparent advancing dynamic contact angle. Pairs of spontaneous imbibition ~oil recovery by spontaneous imbibition of water! and waterflood ~oil recovery vs. pore vol- umes of water injected! curves were measured for each of the induced wetting states. Several trends were observed. Imbibition rate, and hence, water wetness, decreased with increase in aging time and with decrease in initial water saturation. Breakthrough recoveries and final oil recovery by waterflooding increased with decrease in water wetness. Correlations between water wetness and oil recovery by waterflooding and spontaneous imbibition are presented. Introduction Wettability is an important factor in the performance of waterfloods. 1 Final recovery of oil by waterflood is expressed in this paper as R wf , the fraction of the original oil in place ~OOIP! recovered. Review of the literature showed variations in R wf could be very large and were strongly dependent on how wetta- bility was established. 2,3 The wettability state induced by adsorp- tion in the presence of an initial water saturation S wi is referred to as mixed. The main characteristic of mixed wettability is that only the areas of rock surface exposed to crude oil undergo changes in wetting that are significant to the displacement process. The term mixed wettability was adopted by Salathiel 4 to describe this form of wettability distribution with strongly water-wet and strongly oil-wet surfaces. However, a wide range of wetting states can be induced by contact with crude oil and adsorption of oil compo- nents does not necessarily result in a strongly oil-wet surface. Nevertheless, the geometric description of mixed wettability en- visaged by Salathiel is still appropriate and in this paper is applied to all possible types of mixed wetting that depend on the location of bulk water in the rock at the time of adsorption. The term bulk water is used here to identify water retained by capillary forces as distinct from water retained by adsorption at solid surfaces as thin films. The fraction of S wi existing as surface films is very small except at extremely low water saturations. Jadhunandan and Morrow 5 showed that wettability can be var- ied in Berea Sandstone by changing the conditions for adsorption from crude oil. Variations in the crude oil/brine/rock ~COBR! ensembles that were studied included the crude oil, brine, initial water saturation S wi , aging temperature of the rock in crude oil T a , and rate of flooding. The aging time t a , was standardized at 10 days. From the results for over 50 waterfloods, a correlation was obtained between oil recovery and wettability defined by the Amott-Harvey wettability index 6,7 I w- o , with the oil recovery peaking at close to neutral wettability ~see Fig. 1!. For all wetta- bility conditions below I w- o 50.9, with two exceptions, produced water was accompanied by a small and decreasing fraction of oil. This mode of production was postulated by Salathiel 4 to result from a bi-continuous distribution of oil and water within the rock. Jadhunandan and Morrow 8 also investigated factors influencing the rate of oil recovery by spontaneous imbibition ~SI!. Two sets of imbibition and waterflood ~WF! data measured under compa- rable conditions showed that the systems with the lower, but still finite, imbibition rates gave higher waterflood oil recoveries. De- termination of the circumstances under which there are systematic relationships between oil recovery by spontaneous imbibition and by forced displacement ~waterflooding! would be of special value in evaluating possible methods of improved oil recovery. The objective of the present work was to obtain data for a COBR combination in which wettability was varied systemati- cally. The selected system was Prudhoe Bay crude oil/synthetic formation brine/Berea Sandstone. In laboratory studies related to prediction of reservoir performance, choice of t a is mainly gov- erned by the need to provide sufficient time to achieve adsorption equilibrium. This condition is sometimes assumed to match the reservoir wettability. Values of t a on the order of two to six weeks are commonly adopted. 9,10 For investigation of relationships be- tween wettability and oil recovery, variation of t a provides a con- venient method of obtaining systems with graded wettability, 11 while holding other variables constant. This approach was used in the present work to obtain 23 sets of spontaneous imbibition and waterflood data for different t a at three levels of S wi ~nominally, 15, 20, and 25%!. Special attention is given to using spontaneous imbibition curves ~oil recovery vs. imbibition time! to quantify wettability. This requires that the effect of wettability on imbibition rate be separated from the effect of interfacial tension, liquid viscosities, differences in permeability and porosity, and the size, shape, and boundary conditions of the core samples. These effects have been correlated for very strongly water-wet conditions by a definition of dimensionless time. 12 Reduction in imbibition rate relative to the correlation for very strongly water-wet conditions can result from decrease in water wetness. In the present work, relationships are presented between wettability and oil recovery by spontaneous imbibition and waterflooding. Experiments Fluids. A Prudhoe Bay, Alaska, crude oil ~A’93! was used as the oil phase. The crude oil was evacuated before use to reduce the possibility of gas evolution during the course of an experiment. The evacuated crude oil had a viscosity m o of 39.25 cP and den- sity r o of 0.895 g/cc at 25°C. A refined oil, Soltrol 220, m o 53.98 cP, was used to obtain a set of imbibition and waterflood data for very strongly water-wet conditions. A synthetic formation ~SF! brine was prepared: 21.3 g/L NaCl, 0.6 g/L CaCl 2 "6H 2 O, 0.1 *Now at Westport Technology Center Intl. **Now at Exxon Production Research Co. Copyright © 2000 Society of Petroleum Engineers This paper (SPE 62507) was revised for publication from paper SPE 35436, presented at the 1996 SPE/DOE Improved Oil Recovery Symposium held in Tulsa, Oklahoma, 21–24 April. Original manuscript received for review 22 April 1996. Revised manuscript received 14 January 2000. Manuscript peer approved 24 January 2000. SPE Journal 5 ~2!, June 2000 1086-055X/2000/5~2!/199/9/$5.0010.50 199