Recovering heavy or viscous oils is always a challenge. If the reservoirs are deep, mining is not possible. SAGD, a very well known in-situ recovery steam process is a very efficient method. By injecting steam in the reservoir, the oil is heated, decreasing its viscosity, modifying wettability and thus making it flow more easily. Relative permeability equations are key to be able to modelize the SAGD process and to determine the best optimization parameters (for a better energy efficiency and potentially reduce costs linked to this thermal EOR process). There are very few laboratory experiments (Quettier, 1988; Schembre, 2005; Nakornthap, 1986; Maini, 1985; Kovscek, 2014) in the literature about relative permeability changes with temperature.
This paper describes the results of the 5 laboratory experiments performed on porous medium (unconsolidated sand) with extra-heavy oil at reservoir conditions by injecting hot water or steam. Experiments are made in reconstituted cores of 50cm long and 5 cm of diameter, with the sand and the oil from a specific field. The isothermal hot water injection experiments have been done at 36 bars, and 60°C, 120°C, 180°C and 240°C. Steam injection experiment has been done at 36bars, 245°C in a core initially at 36bars, 60°C. This report explains the different steps of the experiments, the conditioning and the procedure. Results are analyzed and interpreted to give end points data and identify the production mechanisms for each coreflood. Experimental parameters have been acquired accurately and data check has been done carefully.
Swi values increase with temperature; Sor has a lower value for 180°C than for other temperatures. An optimum of production in liquid waterflooding has been observed for this experiment (+10% compared to isothermal experiment at 120°C and 240°C), and a recovery after 20 PV comparable to the one of the steam injection experiment after 3 PV. This kind of behavior has already been published in literature and is often linked to change of wettablity with temperature (Maini, 1985; Kovscek, 2014; Jadhunandan, 1995). To confirm this behavior, the same experiment could be reproduced.
Steam injection shows the best efficiency, with a fast recovery: 77% in less than 0.7 PV injected and a final recovery of 80%.
These experiments allow a better understanding of hot-water and steam displacement mechanisms and provide Swi, Sor, Kw, Ko, RF changes with temperature. They can be used to improve the modeling of steam based processes as SAGD.