Experimental and Theoretical Investigation of Oil and Gas Trapping Under Two- and Three-Phase Flow Including Water Alternating Gas (WAG) Injection

Abstract

Abstract Relative permeability, hysteresis effects, and trapped phase saturations are key parameters for reliable simulation of processes involved in oil recovery including WAG injection. Although hydrocarbon gas and CO2, which are widely used in WAG schemes, are likely to be injected at very low IFT (near-miscible) conditions into reservoir rocks with mixed wettability, current models are based on high IFT two-phase flow conditions and water-wet systems. In this paper, we investigate the characteristic properties of rock/fluids systems (wettability, immobile water saturation, permeability, saturation history and oil/gas IFT) that influence the entrapment of gas and oil in petroleum reservoirs. We report a series of core flood experiments which were performed in water-wet and mixed-wet rocks in order to measure the trapped oil and gas saturation as a function of their initial saturations. The experiments include both two-phase (oil/water, oil/gas and water/gas systems) and three-phase flow (WAG injection). Fluids saturations were calculated using material balance as well as x-ray scans of the cores. In both water-wet and mixed-wet cores, trapped gas saturation obtained for two-phase water-gas systems (Sgtw) are higher than those obtained for two-phase gas-oil systems (Sgto) under low gas-oil IFT. The differences of measured trapped gas saturations during three-phase and two-phase water/gas systems, especially for small Sgi values, were not significant. Both three-phase trapped gas and trapped oil saturations were larger in water-wet systems compared to what was obtained in mixed-wet systems. Measured three-phase trapped oil and gas saturations for lower permeability rock (65mD) were larger than those of the 1000 mD core sample. We also investigated the effect of trapped gas saturation (Sgt) on the amount of residual oil saturation at the end of water injection periods of WAG (Sorw). It is found that the Sorw increases linearly by decreasing the Sgt, and approaches to the two-phase Sorw (after primary waterflooding) at its limit where Sgt = 0. Sorw vs. Sgt curve of the water-wet system lies above that of the mixed-wet system. This means that the amount of oil that is trapped by water in the presence of gas increases as the porous medium becomes more water-wet. From the different parameters studied here, injection scenario, permeability and oil/gas IFT have the largest effect on the slope of the Sorw vs. Sgt curve. For both water-wet and mixed-wet systems it was found that total trapped hydrocarbon saturations (for different water injections of WAG injection) remained close to the residual oil saturation at the end of primary waterflooding. The above experimental results are discussed and explained based on our understanding of pore-scale and core-scale displacement mechanisms of multiphase flow and cyclic injections (especially WAG injection) in porous media. Using our experimental results, we demonstrate that although some previously developed empirical trap models are able to capture the trends of trapped gas and trapped oil saturations for two-phase systems, but the observed trends in three-phase (especially for mixed-wet system) cannot be captured using available models. This further emphasises the need for developing more reliable models for fluid displacements in three-phase flow regime.