The Role of Adsorption Isotherms on Chemical-Flooding Oil Recovery

Abstract

Abstract The injection of chemical solutions plays an important role in increasing the recovery factor of mature fields. Compositional models considering adsorption phenomena describe polymer/surfactant flooding as a function of the chemical component concentration and salinity of reservoir/injection water. The concentration part of the solution for 1-D multicomponent polymer/surfactant flooding is completely defined by adsorption isotherm and is independent of relative permeabilities and phase viscosities. The one-phase displacement solution has been already developed and it is a well-known problem. The continuous injection of chemical solutions is a Riemann problem solved by introduction of a self-similar variable. In this paper the solution of the two-phase problem for chemical flooding is presented. It is considered the injection of two dissolved components considering adsorption phenomena for two different isotherms: Langmuir and Henry. A flow potential associated with the conservation of the aqueous phase is introduced as a new independent variable instead of time. This change of variables allows the system splitting in one equation and a thermodynamic system. The number of auxiliary equations is less than the number of equations in the compositional model by one. The solution of the thermodynamic part was done by the method of characteristics and it is completely determined by the adsorption isotherm. Once the two-component adsorption problem is solved, the two-phase flow behavior can be predicted. The applications include the choice of the chemical components to be injected for each reservoir, and also evaluate water compatibility. Another important utilization is testing of numerical compositional simulators by checking the independence of compositional dynamics on transport properties and by comparison of the numerical and analytical solutions. The obtained analytical solutions may be used in the development of streamline chemical EOR simulators. Introduction Enhanced Oil Recovery (EOR) methods include injection of different fluids into reservoirs to improve oil displacement. The EOR methods may be classified into the following kinds: chemical methods, solvents methods and thermal methods. The chemical fluids most commonly injected are polymers, surfactants and micellar solutions1. Oil displacement by chemical solutions is a widespread Enhanced Oil Recovery method. Analytical models for 1-D displacement of oil by water with chemical compounds have been developed since the 1960's. This problem involves complex physical-chemical processes of inter-phase mass transfer, phase transition and transport properties changes. One-dimensional displacement of oil by an aqueous solution containing several chemicals species considering adsorption is described by an (n+1)×(n+1) hyperbolic system of conservation laws, where n is the number of components in the displacing phase. Continuous polymer injection results in a Riemann problem for this hyperbolic system. The displacement of oil by a polymer slug with water drive is described by an initial and boundary value problem with piecewise constant initial data and results in wave interactions2. The Riemann problem for the displacement of oil by hot water3 is mathematically equivalent to one-component polymer flooding (n=1) for a convex sorption isotherm4. Several Riemann solutions for the case n=2 have already been found1–6 and a graphical procedure for the solution of this problem was developed5,6. The Riemann solution for n-component polymer flooding was found for the case where the i-th adsorbed concentration depends only on the concentration of the i-th component in the aqueous phase6,7. Exact solutions for non-self-similar slug problems were also published2,8. The Riemann solution for the (n+1)×(n+1) system for two-phase n-component displacement was studied in several papers 9–12 for Langmuir's type adsorption isotherm. The particular case of one-phase n-component flow leads to an (n)×(n) hyperbolic system, which was used for solving the Riemann problem. The projection and lifting procedures developed allow the calculation of any Riemann solution for the two-phase system once the associated one-phase solution is known. The theory developed is based on the fact that the Riemann problem solutions depend on a single parameter, x/t, but this theory cannot be extended to non-self-similar Cauchy/initial-boundary value problems.