Effect of Oil Composition on Miscible-Type Displacement by Carbon Dioxide

Abstract

Abstract This paper presents additional data related to the correlation between minimum miscibility pressure (MMP) for CO2 flooding and to the composition of the crude oil to be displaced. Yellig and Metcalfe have stated that there is little or no effect of oil composition on the MMP. However, their conclusion was based on experiments with one type of reservoir oil that was varied in C through C6 content and in the amount of C7 + present but not varied in composition of the C7 + fraction. We have found that the Holm-Josendal correlation, which is based on temperature and C5 + molecular weight, predicts the general trend of the MMP's required for CO2 flooding of various crude oils. MMP's were predicted with this correlation and then tested for several crude oils using oil recovery of 80% at CO2 break through and 94% ultimate recovery as the criteria. We now have data showing that miscible-type displacement is also correlatable with the amount of C5 through C3O hydrocarbons present in the crude oil and with the solvency of the CO2 as indicated by its density. Variations from such a correlation are shown to be related to the C5 through C 12 content and to the type of these hydrocarbons. The MMP data were obtained from slim-tube floods with crude oils having gravities between 28 and 44 degrees API (0.88 and 0.80 g/cm3) and C5 + molecular weights between 171 and 267. The crude oils used varied in carbon residue between 1 and 4 wt% and in waxy hydrocarbon content between 1 and 17%. The required MMP for these crude oils at 165 degrees F (74 degrees C) varied between 2,450 and 4,400 psi (16.9 and 30.3 MPa) for an oil recovery of 94% OIP. The MMP was found to be a linear function of the amount of C5 through C30 hydrocarbons present and of the density of the CO2. Introduction Our 1974 paper, "Mechanisms of Oil Displacement by Carbon Dioxide," discussed the various mechanisms by which oil is displaced from reservoir rock using CO2. One conclusion of this study was that multiple-contact, miscible-type displacement of oil occurs through extraction of C5 through C30 hydrocarbons from the reservoir oil by COB when a certain pressure is maintained at a given flood temperature. The mechanism of oil recovery was described as follows. The CO2 vaporizes or extracts hydrocarbons from the reservoir oil until a sufficient quantity of these hydrocarbons exists at the displacement front to cause the oil to be miscibly displaced. At that point, the vaporization or extraction mechanism stops until the miscible front that has been developed breaks down through the dispersion mechanism. When miscibility does not exist, the vaporization or extraction mechanism again occurs to re-establish miscibility. The miscible bank is formed, dispersed, and reformed throughout the displacement path; a small amount of residual oil remains behind all along the displacement path. Also, an optimal flooding pressure at a given temperature for a given oil was defined in that paper as when oil recovery of about 94% OIP was achieved and above which point essentially no additional oil was recovered. This pressure has since been termed the "minimum miscibility pressure" by others. We further determined in our previous study thatthis miscible-type displacement does not depend on the presence of C2 through C4 in the reservoir oil and thatthe presence of methane in the reservoir oil does not change the MMP appreciably. Those findings have been confirmed by Yellig and Metcalfe with the qualification that the CO2 MMP must be greater than or equal to the bubble-point pressure of the reservoir oil. SPEJ P. 87^