Change in Apparent Viscosity of CO2 Foam With Rock Permeability

Abstract

Summary This paper summarizes new and previous high-pressure experiments measuringthe mobility of CO2 foam in porous rock, including both sandstones andcarbonates, with permeabilities ranging from less than 1 md to hundreds ofmillidarcies. Foam mobility is defined here as the ratio of the combined flowrate per unit superficial area to the pressure drop required for simultaneousflow of dense CO2 and brine/surfactant through the sample and can be expressedin units of millidarcies per centipoise. The measured results, which can alsobe expressed in terms of the apparent viscosity of the CO2 foam, give aquantitative basis for the belief that this displacing fluid should beparticularly effective in heterogeneous formations. From a macroscopicviewpoint, the CO2 foam acts like a single fluid with a viscosity equal to theratio of the rock's brine permeability to the measured mobility. According toour results, the permeability to the measured mobility. According to ourresults, the apparent viscosity depends on rock permeability in a nonlinearfashion, increasing from a minimum of about 2 cp for the tightest rocks to anapparent maximum of about 1,000 cp for Berea. Individual mobility measurementswere made at steady-state conditions; this is justified by consideration ofoilfield operational constraints. Results of auxiliary experiments are givenand the methods described. These include measurements of essential surfactantcharacteristics, such as durability against coalescence of the dense CO2 bubbles in brine/ surfactant solution. Results given also show the influence ofother factor s, such as surfactant concentration and type, flowing fraction ofthe dense gas, and overall rate, on foam mobility. Introduction The increased use of CO2, flooding for EOR within the past decade has beenrelated directly to the availability of sufficient quantities of relativelyinexpensive CO2 and has occurred principally in the Permian Basin of west Texasand southeastern New Mexico. These facts might lead skeptics to conclude thatthe efficiency and economics of the process justify its use only under suchspecial circumstances. The inference is partly justified by experience. Although CO2 floods have been profitable and more oil has been recovered thanwould have been by waterfloods alone, things could be better. On the basis of laboratory displacements in slim tubes, at pressures abovethe minimum miscibility pressure, a developed pressures above the minimummiscibility pressure, a developed miscibility flood could be expected toproduce a large fraction of the oil remaining in the formation. Instead, although field experience with CO2 floods has been considerably better thanexperience with early liquified petroleum gas and other solvent floods, mosthave been subject to early breakthrough and have yielded (or are expected toyield) overall recoveries of only 10 to 20% of the original oil in place, representing less than half the oil remaining after waterflood. Among the reasons for these somewhat disappointing results, two stand out. First, the reservoir is heterogeneous, containing preferential" channels" along which the injected fluid can be conducted preferential" channels" along which the injected fluid can be conducted more rapidlyto the producing wells than along other paths. Second, because of the lowviscosity of even dense CO2, the mobility ratio is usually 20 or higher, andthe consequent frontal instability leads to viscous fingering that also causespreferential flow. Both factors, which may be intensified by their interaction, lead to the "bypassing" of substantial quantities of oil. Much of thisoil will remain in the reservoir at the economic limit of production andconstitutes the target for enhanced mobility-controlled CO2 floods. The data presented here show that simultaneous injection of dense CO2 and anaqueous solution of suitable surfactant will produce a foam-like dispersion inreservoir rock. The mobility of this foam is substantially lower than that of CO2 alone. Furthermore, the apparent viscosity of this CO2 foam is not constantbut depends in a favorable way on the permeability of the rock through which itis flowing. The experiments from which these data were derived were designed tomeasure the mobility of the foam in sections of the reservoir that lie farbehind the displacement front. At the front, in the displacement region itself, the dense CO2 (which issupercritical above 88 degrees F) can mobilize residual oil by generation of a Hutchinson-Braun, developed-miscibility zone. If the integrity of the zone canbe maintained by stabilizing the front against the nonuniform flow discussedabove, its graded composition can displace almost all the oil in the pathsalong which it flows. While the displacement itself is accomplished in thefrontal region, the stabilization of the front requires mobility control in themuch larger portion of the reservoir that lies between the front and theinjection well. With effective mobility control in the field, a greaterfraction of the oil remaining after either primary recovery or waterflood canbe expected to be recovered.