(173e) On the Role of Viscosity Variation on the Onset of Natural Convection

Solvent assisted thermal recovery methods have shown promise in improving energy efficiency and reducing environmental footprint of steam based recovery processes. Natural convection may have an important effect during co-injection of solvent and steam into heavy oil reservoirs. It has potential to accelerate the oil recovery due to enhanced mass transfer between solvent and bitumen. In this study, the numerical simulation and linear stability analysis (LSA) for mass and heat transfer during the vertical growth of steam chamber in solvent and steam injection into heavy oil reservoirs are performed. The LSA is conducted based on transient concentration and temperature boundary layers in a porous medium saturated with heavy oil. The bottom surface is subject to a constant solvent concentration and a constant steam temperature. Galerkin method is used to represent the growth of perturbations and the obtained system of ODE equations is integrated using fourth-order Runge-Kutta method.

The obtained scaling relations show that the onset of instabilities is accelerated as the concentration and temperature dependency of heavy oil viscosity is enlarged. Moreover, it is observed that temperature dependency of viscosity has a greater effect on the onset of instabilities than its concentration dependency. However, the presence of solvents is crucial for occurrence of natural convection and they can lead to increased heavy oil production. The results of LSA are compared to the numerical simulation results and good agreement is obtained.

The proposed method provides a more realistic analysis of the problem since it considers transient nature for temperature and concentration fields and also includes the variation of heavy oil viscosity. The introduced scaling relations have a general form and they can be used to predict the onset and wavelength of instabilities under different conditions. Furthermore, the predicted wavelength of instabilities facilitates selection of the proper grid size in numerical simulation of solvent and steam recovery processes.