Linear Oil Displacement by the Emulsion Entrapment Process

Abstract

Abstract Lack of mobility control is a major impediment to successful EOR, especially for high-viscosity oils. This paper presents experimental and theoretical results for continuous, linear, secondary oil displacement using dilute, stable suspensions of oil drops. The major hypothesis is that the oil/water (O/W) emulsion provides microscopic mobility control through entrapment or local permeability reduction not through viscosity-ratio improvement. To describe the displacement process, previous emulsion filtration theory is extended to longer cores and to two-phase flow. Agreement between theory and experiment is satisfactory for continuous secondary oil displacement with 1- to 2-µm [1- to 2-micron] diameter drops of volume concentrations up to 5% in unconsolidated sand packs with permeabilities ranging from 1 to 3 µm2 [1 to 3 darcies]. Dilute suspensions of stable oil drops in water also are successful in diverting flow in parallel-core flooding to the lower-permeability core; therefore, they provide macroscopic mobility control. Introduction To date, two alkaline displacement processes employing stable emulsions have been suggested to improve oil recovery.1 In one process, emulsification with entrainment, oil drops are generated in situ upon reaction of alkali with acidic crude oil. Oil production occurs as an O/W emulsion. In emulsification with entrapment, the other process, oil drops that are generated in situ, or which are externally injected, aid in oil recovery by providing mobility control. These two processes are based on opposing views of how emulsions behave in porous media. According to the entrainment view, oil drops do not interact with the reservoir medium, and recovery of tertiary oil is a possibility.1 Conversely, according to the entrapment view, oil drops interact strongly with the reservoir medium, and improvement only in secondary recovery is sought. Recent work by Soo2 on silute emulsion flow in unconsolidated porous media shows that oil drops clog in pore constrictions and on pore walls, thereby restricting flow. Once captured, there is negligible particle reentrainment. Even drops smaller than the pore throats have a significant capture probability. Soo's study supports the entrapment picture as a more viable description of emulsion flow. However, in spit of field applications of the entrapment technique,3,4 no current methodology exists to predict quantitatively possible mobility-control improvement. This paper presents a theoretical framework for calculation of secondary oil displacement in linear systems with injection of dilute, stable O/W emulsions. Although we focus mainly on microscopic mobility control with dilute emulsions, some attention is given to macroscopic flow redistribution or sweep improvement in parallel cores. The basic premise is that dilute emulsions lower the mobility of the displacing phase through local permeability reduction, not through increasing the viscosity of the displacing phase. We rely heavily on filtration theory, which is successful in describing transient emulsion flow in water-saturated cores.2 The significance of the mathematical treatment is not restricted to the emulsion entrapment technique. It is well known that certain polymers, notably polyacrylamides, establish more mobility control than can be accounted for by bulk rheology.5–11 Large permeability reductions sometimes are observed following polymer injection. Adsorption does not appear to be the main cause of this flow restriction but rather mechanical entrapment - i.e., trapping of high-molecular-weight polymer molecules or, as likely, gels in pore constrictions. Willhite and Dominguez11 recognized the analogy between polymer mechanical entrapment and deep-bed filtration of liquid or solid particulate suspensions. However, they did not explore this analogy quantitatively. Polymer, solid particulate, and emulsion droplet entrapment are directly analogous. Hence, any theory devised for one phenomenon should, in principle, be applicable to the other. Moreover, macroemulsions, as distinguished from microemulsions, sometimes form in surfactant/polymer flooding. Larson et al.12 outline how such emulsion formation might be modeled in displacement calculations. They consider the emulsified oil drops to be retarded in percolating through the porous medium. Permanent capture is not envisioned. This study focused on the filtration and mobility-control aspects of emulsified oil flow. It, therefore, provides an alternative treatment to that of Larson et al. To model EOR with dilute emulsions requires extension of the filtration theory of Soo2 to long cores and to two-phase flow. Combination with classical Buckley-Leverett water flooding theory then permits transient displacement calculations. Before outlining the theory, we present the experimental procedures.