Effect of WAG Injection and Rock Wettability on Oil Recovery During CO2 Flooding

Abstract

Summary Previous studies to determine the extent of oil trapping by water during CO2 water-alternating-gas (WAG) flooding have shown that rock wettability strongly affects this trapping. A significant trapping occurs in water-wet rock, and less trapping occurs in oil-wet rock. This paper presents laboratory results of Devonian crude oil displacement from presents laboratory results of Devonian crude oil displacement from watered-out Berea and reservoir cores by use of continuous CO2 injection, single-slug CO2 injection (followed by water), and CO2 WAG injection at miscible reservoir conditions of 120 degrees F and 2,500 psig [49 degrees C and 17.2 MPa]. The reservoir cores used in this study were mixed-wet (Devonian and Muddy formations) and oil-wet (Tensleep formation). The Berea cores used had their wettability artificially altered to simulate these natural wettabilities. The X-ray photoelectron spectroscopic (XPS) method for measuring carbon content of rock surface was used to provide a qualitative measurement of wettability of the rock samples. The results of the study indicated that the experimental wettability-altering technique came close to duplicating reservoir rock wettability. The oil recovery data at the end of 1 PV fluid injection (continuous CO2 or WAG CO2) indicated thatin preferentially water-wet Berea cores, more than 45% of the waterflood residual oil trapped by CO2 WAG;in mixed-wettability Berea cores, 15 to 20% of the waterflood residual oil was trapped; andin oil-wet Berea cores, less than 5% residual oil was trapped. Introduction In many miscible CO2 flooding field projects, water is injected alternately with increments of the total CO2 slug. This technique developed from the concept of Dyes that simultaneous (or frequent alternate) injection of water and natural gas following a miscible liquefied petroleum gas slug would lower the combined mobility of these displacing fluids. WAG injection should be distinguished from sporadic water injection during a gas-injection project to achieve such goals as injection profile improvements or gas channeling reduction. WAG injection is a planned alternate injection program with water-to-gas injection ratios of 0.5 to 4 volumes of program with water-to-gas injection ratios of 0.5 to 4 volumes of water to 1 reservoir volume of gas at alternation frequencies of 0.1 to 2% PV slugs of each fluid. When the water is alternately injected with the miscible agent (CO2), water saturations are increased to the extent that oil production is not only delayed, but part of the oil may become inaccessible to the CO2. In a tertiary CO2 flood, the residual oil may remain trapped by water so that the displacement efficiency of the CO2 is reduced. Hence, the use of the CO2 WAG injection process in a reservoir must be carefully evaluated. Computer simulation studies have indicated an overall sweep efficiency improvement when CO2 WAG is used from the start of CO2 injection. However, one study points out the problems with gravity segregation of the CO2 and injected water. A survey of published reports of single-slug CO2 (followed by water) and CO2 published reports of single-slug CO2 (followed by water) and CO2 WAG field floods does not indicate improved oil recovery results for the WAG floods (Table 1). Only one pilot CO2 WAG flood, the Slaughter test in the San Andreas formation, had total oil recovery equivalent to single-slug floods. In this test, a large slug of a CO2/H2S solvent mixture was injected followed by an equal-size slug of N2. Both solvent and gas were injected alternately with water. Injected CO2 slug sizes varied among the projects but were greater than 15% HCPV in all projects. CO2 was projects but were greater than 15% HCPV in all projects. CO2 was continuously injected only in the Little Creek field test. The data in Table 1 indicate that the average incremental oil recovery for single-slug CO2 was 15% of the original oil in place (OOIP) whilthat for WAG CO2 was about 8% OOIP. Previous laboratory studies to determine the extent of oil trapping by water in solvent flooding have shown that rock wettability strongly affects this trapping and that less trapping occurs in oil-wet rock. However, no information was presented in the field reports cited about the wettability of the reservoir rock. There is no apparent correlation with rock composition, although the projects include sandstone, limestone, dolomite, and dolomitic sandstone formations. For water-wet cores, Raimondi and Torcaso and Stalkup demonstrated a severe trapping of oil by water. Physical model studies using CO2 as the solvent have been conducted by Tiffin and Yellig in linear corefloods. Jackson et al. used hydrocarbon liquid to simulate CO2 and to perform low-pressure coreflood tests in a quarter five-spot model. Both studies show that when the sand is made completely and irreversibly oil-wet, water blockage is minimized or eliminated. However, Jackson et al. showed that WAG gave no appreciable areal sweep efficiency improvement even in the absence of trapping. Both Tiffin and Yellig and Jackson et al. used a solution of Quilon(TM) in isopropyl alcohol to make the sand or sandstone cores oil-wet. Quilon is a chrome fatty-acid complex that reacts with negatively charged rock surfaces to make them water repellent. HCl is formed during the reaction; however, the effect of HCl on CO2 flooding was not discussed. There were no attempts in these investigations to study CO2 flood performance of intermediate- or mixed-wettability rock. Previous work by Holm and others has shown that some degree of oil-wetness exists in a reservoir rock because of the presence of crude. As shown in the current study, some degree of water-wetness also occurs in most oil reservoirs, particularly those reservoirs containing oils suitable for miscible CO2 flooding--i.e., API gravity greater than 20 degrees API [0.934 g/cm3]. In this paper, the overall effect on oil recovery of water injected with CO2 under various rock-wettability conditions was investigated. The CO2 floods were conducted at miscible conditions (2,500 psig and 120 degrees F [17.2 MPa and 49 degrees C]). The paper presents laboratory results of crude oil displacement from watered-out presents laboratory results of crude oil displacement from watered-out Berea an reservoir cores by use of continuous CO2 injection, single-slug CO2 injection followed by water, and CO2 WAG injection. For Berea corefloods, the cores were physically and chemically treated to alter the rock wettability before the CO2 floods. Also included in the paper are descriptions of the XPS used to indicate relative rock wettability qualitatively and to compare the adsorbed oil components on reservoir rock.