Laboratory Measurements of CO2-Foam Mobility

Abstract

Summary Mobility control of high-pressure CO2 floods by use of foam is a promising oil recovery technique. Critical information for the use of CO2 foam as a thickened displacement fluid is the ratio of the combined CO2/surfactant-solution flow rate to the pressure drop in the swept region. This foam mobility can be measured in laboratory experiments. In this work, a high-pressure apparatus was designed and carefully calibrated to measure foam mobilities at various flow rates. These measurements were made during simultaneous flow of the dense CO2 and surfactant solutions through core samples. Dependence of CO2-foam mobility on several variables was observed. The effect of aqueous surfactant concentration on CO2-foam mobility was explored thoroughly. The dependence in this case was such that foam mobility decreased steadily as surfactant concentration increased until a minimum mobility was attained at some particular concentration well above the conventional critical micelle concentration (CMC). The effect of foam volume fraction showed that mobility decreases with increasing fraction of surfactant solution. Also, it was found that the relative mobility (mobility divided by rock permeability) of CO2 foam was higher in Rock Creek sandstone than in Berea sandstone. Shear-thinning behavior was observed under some conditions in our experimental range of total flow rate. Introduction For a miscible displacement at the required reservoir conditions, CO2 must exist as a dense fluid (in the range of 0.5 to 0.8 g/cm3). Unfortunately, the viscosity of even dense CO2 ranges from 0.03 to 0.08 mPa.s [0.03 to 0.08 cp], no more than 1/20 that of crude oil. When CO2 is used directly to displace the crude, the unfavorable viscosity ratio produces inefficient oil displacement by causing fingering of the CO2 owing to frontal instability. Adding surfactant to the flowing water injected during a CO2 flood reduces mobility and should improve both areal and vertical sweep efficiencies by stabilizing viscous fingering and flow through the more permeable zones. Laboratory study has demonstrated that if contact is made with the oil, dense supercritical CO2 can develop multicontact miscibility with many crudes. Most of the time, though, oil recoveries with CO2 have been much higher in the laboratory than in the field because field conditions are more severe for all oil recovery processes, permitting much more nonuniform flow. Contributing to this nonuniformity of displacement is the adverse mobility ratio. CO2 mobility in porous rock can be decreased if it is contained in a foam-like dispersion. Such CO2 foams have been proposed as useful injection fluids in EOR. (A critical literature review on general foam rheology is given elsewhere.) The foam flooding method modifies the flow mechanism by changing the structure of the displacing fluid at the pore level. This method of decreasing the mobility of low-viscosity fluids in a porous rock requires the use of a surfactant to stabilize a population of bubble films or lamellae within the pore space of the rock. Hence, researchers have endeavored to find suitable foaming agents, especially applicable to EOR. Furthermore, because of the harsh reservoir conditions, studies on the compatibility of the surfactant with oilfield brines at reservoir temperatures and pressures have been made. The degree of thickening achieved, however, apparently depends greatly on the rock properties. These properties probably include both the distance scale of the pore space and the wettability and can be expected to differ from reservoir to reservoir and, to some extent, within a given field. For CO2 floods, a feature has been reported that may partially compensate for the unfavorable mobility ratio. Unexpectedly low mobilities have been observed during CO2 injection in both field and laboratory experiments, and it was concluded that this effect was caused by mixed wettability of the rock. The decreased mobility also might be connected with the high CO2 solubility in crude oil, which adds to the effectiveness of transverse dispersion in dissipating at least the closely spaced fingers. Despite these mitigating features, most CO2 floods show early breakthrough, indicating a higher flow rate in a CO2 finger or channel connecting the injection and production wells. So long as this finger expands laterally, thus entraining enough additional oil to make continuance of the flood economical, the produced CO2 can be reinjected. The costs of gathering, processing, and recompression are additional operating expenses that necessarily hasten abandonment. Because of this, the overall recovery efficiency is reduced, which causes oil loss that can be considered a result of the unfavorable mobility ratio. To understand the nature and mechanisms of foam flow in the reservoir, some investigators conducted a laboratory study of CO2-foam properties and displacement mechanisms. Porous micromodels also have been used to represent actual porous rock in which the flow behavior of bubble films or lamellae has been observed. Furthermore, because foaming agents often exhibit pseudoplastic behavior in a flow situation, the flow of non-Newtonian fluids in porous media has been examined from a mathematical standpoint. However, representation of such flow in mathematical models was reported to be inadequate. The apparent viscosity of foam also has been measured in capillary tubes and found to decrease rapidly with increasing ratio of bubble radius to capillary radius. A numerical approach, with the goal of computing foam mobility in a porous medium modeled by a bead- or sandpack, was attempted as well. There also has been considerable effort to calculate foam mobility in porous media from first principles with usually measured rock properties. Finally, a theoretical approach was attempted to investigate the effect of surface viscosity on mobility. It has been verified that the interface of a surfactant solution exhibits non-Newtonian behavior and its rheological properly can be measured. This paper demonstrates the laboratory measurement of CO2-foam mobility at different flow rates by steady-state experiments with real rock samples used as the porous media. The dependencies of foam mobility on surfactant concentration and CO2-foam fraction were studied. Also, CO2-foam mobility in the samples studied was found to depend greatly on the single-phase permeability of the rock sample. In addition to the results of these measurements, the operation of the apparatus, the experimental procedures, and the testing of several features of surfactant suitability at reservoir conditions are discussed briefly.