Generalized Minimum Miscibility Pressure Correlation (includes associated papers 15845 and 16287 )

This paper presents a generalized correlation for predicting the minimum miscibility pressure (MMP) required for predicting the minimum miscibility pressure (MMP) required for multicontact miscible displacement of reservoir fluids by hydrocarbon, CO2, or N2 gas. The equations are derived from graphical correlations given by Benham et al. and give MMP as a function of reservoir temperature, C7+ molecular weight of the oil, mole percent methane in the injection gas, and the molecular weight of the intermediates (C2 through C6) in the gas. CO2 and N2 are represented in the current correlation by "equivalent" methane/propane- and methane/ethane-mixture injection gases, respectively. The study shows that for hydrocarbon systems, paraffinicity has an effect on MMP. In the equations, the C7+ paraffinicity has an effect on MMP. In the equations, the C7+ molecular weight of the oil is corrected to a K factor of 11.95, thereby accounting for varying paraffinicity. An additional temperature effect on N2 MMP is related to the API gravity of the oil. The N2 correlation, however, is not tested against measured MMP data other than those used to develop the equation and should be used with care. A correlation that accounts for the additional effect on CO2 MMP caused by the presence of intermediate components in the reservoir oil is presented. Predicted MMP's from the correlations developed are compared to experimental slim-tube displacement data from the literature and from our displacement tests on North Sea gas/oil systems. These displacement tests have been performed with a packed slim tube, where the effect of viscous fingering is reduced to a minimum. Introduction Multicontact miscibility is represented most easily with a ternary diagram, where the composition of the driving or displaced fluid is altered. This is obtained by vaporization of light hydrocarbon components into a driving gas or by condensation of hydrocarbon components from a driving gas into the reservoir oil. Miscibility between reservoir oils and hydrocarbon gases is achieved either by vaporization or by condensing-gas-drive mechanism, depending on the reservoir oil and injection-gas composition. With N2 and CO2, miscibility is obtained by vaporization, but with CO2, miscibility usually is achieved at lower pressure because CO2 extracts much higher-molecular-weight hydrocarbons from the reservoir oil than N2 gas. The prediction of miscibility conditions from ternary diagrams is based on experimentally determined or calculated gas and liquid compositions of a reservoir-oil/injection-gas mixture. The experimental gas and liquid equilibrium data are not easy to obtain and are often time-consuming to determine, especially near the plait point. The method for calculating gas and liquid data with point. The method for calculating gas and liquid data with equations of state to predict miscibility relies largely on gas and liquid compositions near the plait-point region. It is generally accepted that such data may not be sufficiently accurate. Flow experiments offer the most reliable method to determine the pressure required for miscibility with CO2, N 2, and hydrocarbon gas. The slim-tube method has been most widely used to determine miscibility. Different experimental procedures and interpretation criteria, however, have ted to different definitions of miscibility and have caused considerable confusion. The limitation of the slim-tube test and the problems associated with miscible displacement in porous media have been described by several authors. Phase behavior and mechanisms of miscible flooding with CO2, N2, and hydrocarbon gas have also been described by several authors. Correlations for predicting MMP have been proposed by a number of investigators and are important tools in the selection of potential reservoirs for gas miscible flooding. Therefore, the correlations must be as accurate as possible. Several CO2 MMP correlations have been published, but none of these can be used with enough published, but none of these can be used with enough confidence for final project design. They are useful, however, for screening and preliminary work. Correlations on CO2 miscible flooding have shown temperature to be the most important parameter but they disagree regarding the effect of oil type (e.g., C7+ properties of the oil). Compared with CO2 miscible flooding, very little has been published on high-pressure hydrocarbon gas miscible flooding. A recent publication gives a correlation for predicting MMP with lean hydrocarbon gases and nitrogen. In 1960, Benham et al. presented empirical curves that can estimate miscibility conditions for reservoir oils that are displaced by rich gas within a pressure range of 1,500 to 3,000 psia [10.34 to 20.68 MPa]. They assumed a limiting tie line (at the critical composition on a ternary diagram) parallel to the C1–C7+ axis and estimated mole percent methane in the injection gas from calculated percent methane in the injection gas from calculated critical points with pressure, temperature, molecular weights of C2 through C4 in the gas, and the C5+ molecular weight of the oil as variables. From Benham et al.'s data, the proposed equations have been derived for predicting MMP. SPEJ P. 927