Determination of Multiphase Boundaries and Swelling Factors of Solvent(s)–CO₂–Heavy Oil Systems at High Pressures and Elevated Temperatures

Abstract

A generalized methodology has been proposed and successfully applied to determine multiphase boundaries as well as swelling factors of solvent(s)−CO 2 −heavy oil systems at high pressures and elevated temperatures. Experimentally, twoand three-phase boundaries and swelling factors have been respectively measured by conducting PVT tests in the temperature range of 280.45 to 396.15 K. Theoretically, the Peng−Robinson equation of state (PR EOS) combined with the modified alpha function has been applied to describe phase behavior of the solvent(s)−CO 2 −heavy oil systems. More specifically, an exponential distribution function is used to split a heavy oil sample, whereas the logarithm-type lumping method is employed to group single carbon numbers (SCNs) into multiple carbon numbers (MCNs). The exponents associated with two binary interaction parameter (BIP) correlations are respectively tuned for the alkane solvent-pseudocomponent pair and CO 2 -pseudocomponent pair to match the measured saturation pressures. It is found that six pseudocomponents combined with the BIP correlation as a function of critical volume is sufficient to predict saturation pressure with an absolute average relative deviation (AARD) of 5.07%. In addition, the PR EOS model associated with the selected parameters is applied to predict three-phase boundaries for a C 3 H 8 −CO 2 −heavy oil mixture and a n-C 4 H 10 −CO 2 −heavy oil mixture yielding an overall AARD of 4.58%. As for swelling factors, the Peneloux et al. method provides the minimum AARD of 2.09% in comparison with 4.00% from the Jhaveri et al. method and 2.41% from the Twu et al. method, respectively.

Direct perturbation of the Peng–Robinson attraction and covolume parameters for reservoir fluid characterization

“…Fig. 13 compares the predicted phase boundaries with the data points given in Li et al (2013b). Although the lower pressure boundary is overestimated, the higher pressure boundary is predicted quite accurately with no further adjustment of the fluid model from the direct PnA algorithm.…”

“…If the BIPs for CO 2 with the pseudo components are altered from 0.10 to 0.15, the AAD is reduced to 7.6% for Feed #4 and to 0.4% for Feed #5. Li et al (2013b) presented P-T conditions for three-phase behavior for the mixture of 83.2% CO 2 , 11.8% n-C 4 , and 34.3% heavy oil 4 on the molar basis (Feed #15 in Li et al, 2013b). Fig.…”

“…13. The three-phase boundaries measured and predicted for a mixture of heavy oil 4 (Table 5) with CO 2 and n-C 4 (Feed #15 of Li et al, 2013b). Table 3.…”

Phase Behaviour and Viscosity Reduction of CO2-Heavy Oil Systems at High Pressures and Elevated Temperatures

“…It was reported that the carbon number of reservoir fluids exhibits an approximate linear relationship with the logarithm of its corresponding mole fraction (Pedersen et al, 1983). Such relationship can be found when the carbon number is higher than a certain number, e.g., C 6 for gas condensate (Pedersen et al, 1984) and C 14 for heavy oil (Li et al, 2013b). It is correlated as the following equation (Pedersen and Christensen, 2006),…”

“…Lumping Previous work has grouped SCNs into two to eight pseudocomponents, where it is found that six pseudocomponents are sufficient to characterize Lloydminster heavy oil with a good accuracy (Li et al, 2013b). Therefore, the SCNs are lumped into six pseudocomponents by using the following mixing rule (Hong, 1982),…”