Effects of CO 2 Addition to Steam on Recovery of West Sak Crude Oil M.W. Hornbrook, SPE, BP Exploration (Alaska); Kaveh Dehghanl, * SPE, U. of Alaska; Suhall Qadeer, SPE, Stanford U.; and R.D. Ostermann, SPE, D.O. Ogbe, SPE, U. of Alaska Summary. A high-pressure ID laboratory displacement study evaluated the effects of adding CO 2 to steam on the recovery of West Sak crude oil. Results of the laboratory experiments indicate that the simultaneous injection of CO 2 and steam increases recovery, reduces injection temperatures, and reduces the heat input required. Introduction A high-pressure ID laboratory displacement study was undertaken to evaluate the effects of adding CO 2 to steam on the recovery of West Sak crude oil. In addition, a run was made below the bub- blepoint pressure to assess the effects of a free-gas phase on steam- flood recovery of West Sak crude. Experiments were conducted in an unconsolidated sandpack 2 in. in diameter and 4 ft long. The sandpack was saturated with West Sak crude oil (19.2° API) that previously had been saturated with methane at a bubblepoint pres- sure of 1,690 psig. The volumetric flow rate was held constant to focus the study on the effects of CO 2 addition to steam. The tem- perature profile and pressure drop along the length of the sandpack were recorded in addition to ultimate recovery and effluent prop- erties. The West Sak reservoir is on the North Slope of Alaska, about 250 miles north of the Arctic Circle and to the west of the Prudhoe Bay Unit in the Kuparuk River Unit. The reservoir is estimated to contain 15 to 25 billion STB of crude oil at 16 to 22 °API. 1 The presence of a 2,000-ft permafrost layer and a reservoir depth that ranges from 3,000 to 4,500 ft results in an average reservoir tem- perature of 45 to lOO°F. Lower-than-expected temperatures for these depths result in a viscous in-situ crude oil. Alaska is also the site of a large reserve of natural gas. The Prud- hoe Bay field contains approximately 27 x 10 12 scf of natural gas that is approximately 12.5% CO 2 .2 Because of the large reserves of heavy oil and the potential use of natural gas to generate steam and CO 2 as an additive, it was decided to study the effects of add- ing CO 2 to steam on recovery of West Sak crude oil. While a considerable amount of work has been done in the area of solvent addition to steam, only a small percentage of this work deals with CO 2 addition to steam. The reported laboratory work on physical models,3-9 the published works on numerical studies, 10-13 and field studies 14-18 indicate a strong potential for the success of such a process. Laboratory Studies. Pursley3 conducted experiments on a cylin- drical model to investigate the effect of injecting air, methane, or CO 2 on steam stimulation. His results show a dramatic improve- ment in the oil/steam ratio as a result of injecting methane or air. He reported that addition of CO 2 was somewhat less effective be- cause of its high solubility in water. Redford 4 studied the effects of adding CO 2 and ethane to steam in a 3D physical model. His results showed that adding CO 2 or ethane to steam greatly improves the recovery of Athabasca tar sand over that recovered with other additives. Redford attributed the in- crease in recovery to improved sweep efficiency, solution-gas drive, swelling, and viscosity reduction. Harding et al. 5 presented results of a physical model study of steamflooding with nitrogen and CO 2 additives. The tests were conducted in a linear porous medium saturated with a moderately viscous refined oil and water. It was found that the simultaneous injection of the gases with steam resulted in a significant improve- ment in the ultimate recovery of the crude oil. • Now at Chevron Oil Field Research Co. Copyright 1991 Society of Petroleum Engineers 278 Briggs et at. 6 studied the effects of CO 2 and naphtha addition to steam in a cylindrical 1D physical simulator with Athabasca tar sand. Their results indicate that the use of CO 2 with steam im- proves recovery primarily by providing additional drive energy on the depletion portion of a cyclic process. Paracha 7 studied the effects of CO 2 addition to steam in aID physical model on 15, 20, and 26° API oils. On the basis of this study, he concluded that although CO 2 with steam increases the rate of recovery significantly, the overall recovery is dependent on oil viscosity and hence the API gravity. Stone and Malcolm 8 carried out high-pressure steam/C0 2 coin- jection experiments in a 1D physical model with Athabasca tar sand. The results from the physical model were compared with results from a numerical model study. Both models indicated that coinjec- tion of CO 2 and steam increases ultimate recovery. Frauenfeld et al. 9 conducted physical model experiments to study the effects of coinjection of steam and CO 2 into "dead" heavy oil and into an oil saturated with methane. For "dead" oils, the coinjection of CO 2 with steam improved oil recovery. When the oil was saturated with methane, however, the coinjection of CO 2 was not beneficial. Most sensitivity studies in these laboratory works on CO 2 /steam showed that there is an optimum CO 2 concentration where the oil recovery is maximized. 3,4,7 Numerical Studies. Bader et al. 10 and Fox et al. 11 published comparative simulation studies of steamdrive and a steam/gasdrive. Results from these simulation studies showed that recovery o<;curred more quickly and the production rate history curve peaked about 30% earlier than in the steam case. Claridge and Dietrich 12 used a 3D numerical model to study the effects of CO 2 addition to steam for stimulating bitumen-containing hydrocarbon solution gas under partially depleted reservoir condi- tions. The study concluded that adding CO 2 decreased the bitu- men recovery expected from the application of steam alone. Leung 13 conducted a numerical simulation study of steam stimu- lation and steamflood for simultaneous injection of steam and CO 2 , For an Athabasca oil sand reservoir, a 36 % increase in recovery over steam stimulation was achieved. He also found that in a 3D steamdrive simulation, CO 2 injection with steam did not improve the recovery significantly for a California-type reservoir, where the stripping effect of steam was the main recovery mechanism. The injected gas promoted vertical gravity override, and steam breakthrough occurred slightly earlier than in the case with steam only. The CO 2 was seen to concentrate at the leading edge of the steam zone. Field Studies. In one of the early single-well field tests described by Clark et at. , 14 combustion exhaust gases were injected into a viscous reservoir. Increased oil production rates were attributed to decreased oil viscosity from CO 2 absorption in the crude and to increased reservoir energy from the injected gases. Shelton and Morris 15 reported results from a field test of a huff 'n' puff proc- ess where rich gas was used to increase production rates in viscous- oil reservoirs. They concluded that the energy effects from gas in- jection lead to only short-term benefits and that unless the reser- SPE Reservoir Engineering, August 1991