A Multicell Equilibrium Separation Model for the Study of Multiple Contact Miscibility in Rich-Gas Drives

Abstract

Abstract The growing interest in miscible flooding operations demands that reliable techniques be found to enable the engineer to predict miscible performance. In the area of enriched gas drives performance. In the area of enriched gas drives there is only one correlation available for predicting enrichment compositions. This correlation, however, makes no attempt to predict either the compositional path necessary for the eventual development of path necessary for the eventual development of miscibility or the effect of phase mobilities on this path. path. A multicell equilibrium flash separation model bas been developed that can predict the composition necessary to develop miscibility in linear systems. The model computes the compositional changes that occur in the transition zone between the in-place oil and the injected fluid and can also compute effluent volumes and compositions. Furthermore, it has the capability of incorporating phase mobilities for determining the composition of phase mobilities for determining the composition of the flowing fluids. This model bas been run with both three-component and "real" system phase equilibria data. Results show that in some cases more enrichment may be required to develop miscibility than that predicted by the Benham correlation. The model bas proved to be much faster computationally than a more sophisticated model described recently in the literature. In addition, it yields results that compare favorably with the predictions of the more sophisticated model. The model being presented shows mathematically the effect that phase mobilities have on the eventual development of miscibility. Introduction Hydrocarbon miscible flooding offers the advantage of 100-percent displacement efficiency of the oil contacted in the swept region of a reservoir, but it requires a large capital outlay for the necessary solvents The solvent requirements for a successful miscible operation can be significantly reduced if reservoir conditions, including the composition of the oil in place, permit miscibility to develop through component transfer from injected fluid to the oil in a multiple contact miscible process. The success of this "enriched gas drive" miscible operation is also directly dependent upon the level of intermediate (C2 through C6) components of the injected fluid. Planning such a miscible operation requites Planning such a miscible operation requites prediction of the necessary enrichment composition prediction of the necessary enrichment composition and knowledge of the effects of rock and fluid properties. properties. The correlation of Benham et al. provides a method for determining the composition requirements for the injected fluid. In this correlation the maximum methane composition is a function of temperature, pressure, and the molecular weights of the C5+ fraction of the oil and the C2+ fraction of the injected fluid. According to Benham, this correlation is conservative in its prediction and should be confirmed experimentally. An inherent assumption of the correlation is an infinite gas-phase relative permeability. Van Quy et al. recently published the results of their work with a sophisticated reservoir model. Their model was designed to predict the changes in composition that occur when oil is contacted by an injected fluid under conditions of "mono- or diphasic flow". The model considered convection, diffusion, relative permeability, and mass transfer effects. Corteville et al. tested the model by performing flow experiments in tubes packed performing flow experiments in tubes packed with glass beads and comparing predicted performance with experimental data. They concluded performance with experimental data. They concluded that relative permeability characteristics affected the calculated compositions at a given pressure. Our purpose here is to describe a simple and inexpensive mathematical model that predicts the development of miscibility. This model is useful because many different cases can be studied in a short time. The model has been used primarily to study the effects of fluid flow and fluid properties on the development of miscibility, wherein the solvent is generated primarily from components injected into a crude-oil reservoir.