Abstract

Diffusion is one of the main mass transfer principles associated with a gas injection process, e.g., vapor extraction (VAPEX) and cyclic solvent injection (CSI), to reduce the viscosity of heavy oil by dissolution of light gas(es). A few efforts have been made to estimate the diffusivity of each gas component of a binary gas-mixture in heavy oil as a constant, but no attempts have been made to determine individual diffusivity of each gas component in the binary gas-mixture as concentration-dependent. In this study, the concentration-dependent diffusivities of CO2 and C3H8 in a CO2/C3H8 mixture, which preferentially diffuses in a Lloydminster heavy oil, are evaluated implicitly as a power function of gas concentration, pressure, and temperature through the oil viscosity. The coefficient and exponent of such a power function for each gas are determined once the deviations between the measured and calculated values for oil swelling factors and composition of each gas dissolved in heavy oil at the completion of a dynamic volume analysis (DVA) test are minimized by applying the 1D time-dependent finite element method (FEM). Compared with the constant diffusivities for the same test, considering concentration-dependency of the dissolved gases reproduces the experimentally measured oil swelling factors and compositions at the end of the test with a significantly enhanced accuracy. Then, such obtained concentration-dependent diffusivities are extended to reproduce the experimentally measured swelling factors of a pressure-decay test performed with another CO2/C3H8 gas-mixture and the same heavy oil. From such an evaluation, it can be inferred that the coefficient and exponent of the diffusivity correlation associated with a binary gas-mixture are functions of pressure, temperature, feed gas composition, and test duration, respectively. This is because the coefficient and exponent for one gas component may be affected by the presence of the other as the latter alters the composition of heavy oil in which the former diffuses.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call