Development of numerical simulator for predicting oil recovery process of low-salinity water flooding

The present study was aimed to develop a numerical simulator for predicting oil recovery process during low-salinity water flooding (LSWF). While various mechanisms have been proposed in the past decades, this study focused primarily on the cation exchange. The authors' previous core flooding experiments with sandstone suggested the possibility of cation exchange as the predominant phenomenon contributing to the additional oil recovery; divalent cations, which form bridges between the negatively-charged polar-oil components and clay minerals, are increasingly replaced by H⁺ as salinity decreases, leading to the detachment of the oil from the rock surface. Moreover, rock wettability changes from oil-wet to water-wet owing to the ion exchange. Furthermore, the increase in differential pressure has been observed during the past core flooding experiments suggesting wettability alteration and/or the blockage of flow path. On the contrary, in carbonate rocks on which crude oil is directly attached on the positively charged rock surface, SO₄^2- ions in LSW promote oil recovery by weakening the ionic bonds between the crude oil and the rock surface. Therefore, we developed the pseudo multi-compositional simulator, which can deal with 1-dimensional, 3-phase (oil, water and solid) and 19-component (non-polar oil, polar oil, sand, H₂O, NaCl, CaCl₂ , MgCl₂ , H₂CO₃ , CaCO₃ , MgSO₄ , H⁺ , Na⁺ , Ca²⁺ , Mg²⁺ , OH^- , SO₄^2- , Cl^-, HCO₃^-, CO₃^2-) problems, aiming at more accurate reproduction of the oil recovery process during LSWF in both sandstone and carbonate reservoirs. The reactions of the pseudo multi-components were calculated taking into account ionization, adsorption/desorption of ions, diffusion of ions, and fine migration. The functions of the simulator were verified by comparing with analytical results. Thereafter, the results of past core flooding experiments were successfully reproduced using this simulator, demonstrating the importance of the above-mentioned mechanisms in LSWF.