A simplified method for computing phase behavior of crude oil - carbon-dioxide mixtures

Phase behavior plays a fundamental role in oil recovery processes, ranging from the production of gas condensate and volatile oil reservoirs to the injection of CO{dollar}sb2{dollar} and N{dollar}sb2{dollar} for enhanced oil recovery processes. In phase behavior methods, equilibrium ratios are used to predict compositional changes in the reservoir fluids, particularly when using compositional simulators.; Literature search and experience in the phase behavior of CO{dollar}sb2{dollar}-reservoir oil systems have shown that equations of state and available correlations give acceptable results in some areas, but in general, they are not satisfactory due to lack of accuracy, large computational time, or sometimes yielding trivial solutions. Therefore, accurate, faster and more reliable new methods are needed, particularly for actual compositional studies.; In this study, a K-value method is developed. This method, expressed in a simple mathematical form, relates the equilibrium ratios of each component with its boiling temperature, critical temperature and pressure, and the mixture's pressure, convergence pressure and overall compositional changes. This method uses some experimental data for the mixture under study to adjust the form of the K-value expression. These experimental data are obtained from some of the routine PVT laboratory tests. A least squares-linear programming optimization routine is adopted to adjust the correlation to match the actual behavior of the mixtures with the calculated.; Nine reservoir fluid samples were simulated, for retrograde gas condensate and five oil systems. The K-value method demonstrated good matches with the experimental data for all systems, including crude oil-carbon dioxide mixtures. This method worked well compared to Peng-Robinson and Soave-Redlich-Kwong equations of state for matching the saturation pressures and swollen volumes for four cases of CO{dollar}sb2{dollar} and N{dollar}sb2{dollar} injections into crude oil systems. The K-value method was faster than the equations of state by a factor of 7-20. In addition, the K-value method required less computer memory, less input data and fewer parameters to adjust than the equations of state. An approach for characterizing the heavy hydrocarbon fractions was developed as a combination of Ahmed's method, Whitson's method and Kay mixing rules. This approach was successfully used to split the heptanes-plus into pseudo components and estimate their properties.