Experimental and modeling study of the phase behavior of synthetic crude oil+CO2

A full understanding of the phase behavior of CO2–hydrocarbon mixtures at reservoir conditions is essential for the proper design, construction and operation of carbon capture and storage (CCS) and enhanced oil recovery (EOR) processes. While equilibrium data for binary CO2–hydrocarbon mixtures are plentiful, equilibrium data and validated equations of state having reasonable predictive capability for multi-component CO2–hydrocarbon mixtures are limited. In this work, a new synthetic apparatus was constructed to measure the phase behavior of systems containing CO2 and multicomponent hydrocarbons at reservoir temperatures and pressures. The apparatus consisted of a thermostated variable-volume view cell driven by a computer-controlled servo motor system, and equipped with a sapphire window for visual observation. Two calibrated syringe pumps were used for quantitative fluid injection. The maximum operating pressure and temperature were 40 MPa and 473.15 K, respectively. The apparatus was validated by means of isothermal vapor–liquid equilibrium measurement on (CO2 + heptane), the results of which were found to be in good agreement with literature data. In this work, we report experimental measurements of the phase behavior and density of (CO2 + synthetic crude oil) mixtures. The ‘dead’ oil contained a total of 17 components including alkanes, branched-alkanes, cyclo-alkanes, and aromatics. Solution gas (0.81 methane + 0.13 ethane + 0.06 propane) was added to obtain live synthetic crudes with gas-oil ratios of either 58 or 160. Phase equilibrium and density measurements are reported for the ‘dead’ oil and the two ‘live’ oils under the addition of CO2. The measurements were carried out at temperatures of 298.15, 323.15, 373.15 and 423.15 K and at pressures up to 36 MPa, and included vapor–liquid, liquid–liquid and vapor–liquid–liquid equilibrium conditions. The results are qualitatively similar to published data for mixtures of CO2 with both real crude oils or and simple hydrocarbon mixtures containing both light and heavy components. The present experimental data have been compared with results calculated with two predictive models, PPR78 and PR2SRK, based on the Peng–Robinson 78 (PR78) and Soave–Redlich–Kwong (SRK) equations of state with group-contribution formulae for the binary interaction parameters. Careful attention was paid to the critical constants and acentric factor of high molar-mass components. Since the mixture also contained several light substances with critical temperatures below some or all experimental temperatures, we investigated the use of the Boston–Mathias modification of the PR78 and SRK equations. The results showed that these models can predict with reasonable accuracy the vapor–liquid equilibria of systems containing CO2 and complex hydrocarbon mixtures without the need to regress multiple binary parameters against experimental data.