A Comparative Study on WAS, SWAS, and Solvent-Soak Scenarios Applied to Heavy-Oil Reservoirs Using Five-Spot Glass Micromodels S.A. Farzaneh, Heriot-Watt University; A.A. Dehghan and M.H. Ghazanfari, Sharif University of Technology; and R. Kharrat, Petroleum University of Technology Summary In this work, a series of solvent- and water-injection scenarios were conducted on horizontal five-spot glass micromodels that were saturated initially with heavy oil. Sandstone and limestone rock look-alike and network patterns with different pore structures were used in the experiments. The results show that the ultimate oil recovery of a water-alternating-solvent (WAS) scheme was greater than that of a simultaneously water-alternating-solvent (SWAS) scheme, and the efficiency of a solvent-soak scheme also offers a greater recovery. Likewise, the WAS scheme resulted in greater oil recovery when compared with continuous solvent injection (CSI), with the same amount of solvent consumption. Furthermore, some pore-scale phenomena, such as viscous finger- ing, diffusion of solvents into heavy oil, and localized entrapment of oil and solvent because of heterogeneity and/or water blockage, are also illustrated. The results of this work can be helpful for bet- ter understanding and verification of flow transport and pore-scale events during different solvent-based-injection scenarios in heavy- oil reservoirs. Introduction Because of the decline of conventional oil reserves, the exploita- tion of heavy oil is one of the primary interests to many oil com- panies (Butler and Mokrys 1991). The worldwide magnitude of these resources is on the order of 6 trillion bbl of oil in place (Das and Butler 1998). Currently, thermal processes are applied as enhanced-oil-recovery (EOR) processes to recover heavy oil by reducing the oil viscosity after heating the reservoir (Upreti et al. 2007). The viscosity of heavy oil can also be reduced by addition of solvents. The considerable energy losses inherent in the ther- mal processes may be avoided if a solvent is used instead of steam (Pooladi-Darvish et al. 1995). One of the emerging techniques for this purpose is the vapor extraction (VAPEX) process, in which a vaporized light-hydrocarbon solvent or a mixture of solvents is injected into the reservoir through a horizontal injection well and the diluted oil is drained by gravity and produced through a hori- zontal production well. However, the performance of the VAPEX process in problematic heavy-oil reservoirs such as those with lack of confinement of the solvent chamber, regions with large gas caps or bottomwater, and oil zones with high shale breaks is not satisfactory or economical (Scott 2002; Singhal et al. 1996, 1998). On the other hand, Spencer et al. (1969) found that the sol- vent has a highly effective ability to create the opening for water- flow paths. In addition, the initial solvent slug injected before water can improve the displacement efficiency through improve- ment in the mobility ratio. Therefore, the solvent flood and its dif- ferent injection schemes (e.g., WAS) appear to be the attractive alternative EOR technologies that can be justified economically for problematic heavy-oil reservoirs. Definitions and specifica- tions of heavy-oil problematic reservoirs are given elsewhere (Farouq Ali 1976). The large viscosity contrast between injected miscible solvent and reservoir fluids leads to unstable displacement. The simplest method for controlling this poor mobility ratio is to inject the sol- vent alternately with water. The water-alternating-gas (WAG) process, proposed by Caudle and Dyes (1958), has remained as the most widely practiced process to control displacements stably in oil fields (Christensen et al. 2001). The WAG process was aimed at improving gasflood conformance by simultaneously using the natural counteracting tendencies of the gas to rise and the water to descend. The combination of the higher microscopic- displacement efficiency of gas with the better volumetric-sweep efficiency of water helps significantly to increase the incremental oil recovery over a waterflood. The process of WAG can be clas- sified in different forms by the methods of fluid injection. The most common categorization is the difference between miscible and immiscible injection. Significant research efforts for optimiz- ing and increasing recovery from the WAG process have provided better understanding of the injectivity limitations and WAG-ratio optimizations (Christensen et al. 2001; Kulkarni and Rao 2005). Sohrabi et al. (2000; 2001) conducted a series of WAG tests using water-wet, oil-wet, and mixed-wet micromodels to demonstrate the difference in pore-scale flow mechanisms of the WAG process under different wettability conditions. They found that successive WAG cycles redistributed the fluids, creating fresh pathways for gas to enter the pores occupied by oil; hence, some of the oil that otherwise would not have been mobile under either gas or water injection alone was mobilized and produced. Their results showed that oil recovery was greater for WAG injection than for water or gas injection alone under any of the wetting conditions. It was also observed that WAG recovery was greater for strongly oil-wet or mixed-wet models than for strongly water-wet models. Fatemi et al. (2011) reported the results of a comprehensive series of cor- eflood experiments carried out in two different sandstone cores under natural water-wet and mixed-wet conditions. These included water injection, gas injection, and WAG injection, and the impact of wettability on the performance of these injection strategies was investigated. The results show that in both the water-wet and mixed-wet cores, the performance of WAG injec- tion was better than that of water injection or gas injection alone. The successful application of miscible and immiscible WAG processes in various field tests indicates that a high recovery fac- tor and economical solvent/oil ratios are achievable (Christensen et al. 2001; Kulkarni and Rao 2005). Nevertheless, there are no reports in the literature that investigate experimentally the effi- ciency of different injection schemes for the case of heavy-oil sys- tems, and there are still several technical challenges to be solved for these systems. The pore-scale events of heavy-oil recovery in the WAS, SWAS, and solvent-soak processes were not under- stood well enough to enable incorporating the pore-level physics of the process into mathematical models. Investigation of the physics of fluid distribution and flow behaviour in porous media for different injection scenarios of water and solvent at the pore Copyright V C 2012 Society of Petroleum Engineers This paper (SPE 158376) was accepted for presentation at the Canadian International Petroleum Conference, Calgary, 16–18 June 2009, and revised for publication. Original manuscript received for review 23 March 2009. Revised manuscript received for review 17 November 2011. Paper peer approved 10 January 2012. September 2012 Journal of Canadian Petroleum Technology 383