An Experimental Study of Steam-Solvent Coinjection for Bitumen Recovery Using a Large-Scale Physical Model

Summary This paper presents a detailed investigation into compositional flow in solvent-assisted steam-assisted gravity drainage (SA-SAGD) with multicomponent hydrocarbon solvent “condensate”, which is readily available near thermal operation sites. Designing SA-SAGD with condensate requires understanding the complex interplay between phase behavior and fluid flow that affects bitumen production and solvent recovery. This research consisted of three stages. First, an SA-SAGD experiment was performed using a cylindrical physical model with an inner diameter of 0.45 m and a length of 1.2 m. Steam and synthetic condensate (2.7 mol%) were coinjected at 3500 kPa. Second, the experimental data were history-matched by using a numerical model, and the resulting model was analyzed for the detailed analysis of solvent flow behavior. Third, observations of compositional details in the SA-SAGD experiment were further investigated with a simulation case study using stochastic realizations of a 3-D heterogeneous reservoir. We present a novel way of analyzing how much of the injected solvent was used for in-situ bitumen dilution beyond the edge of an SA-SAGD chamber, which is referred to as “utilization factor” in this paper. Solvent utilization factor is related to, but different from solvent recovery factor because different components have different volatilities in the coinjected condensate. SA-SAGD experimental results showed that the produced condensate was heavier than the injected. Material balance analysis showed that the in-situ condensate composition for bitumen dilution was similar to the injected condensate composition, but the volatile components (C1 and C4) had lower recovery factors. The simulation case study for an upscaled 3-D heterogeneous reservoir confirmed that a large fraction of the volatile solvents in the condensate was used for in-situ bitumen dilution. The solvent recovery factor that has been used as an economic indicator for SA-SAGD does not represent how much of the injected solvent is used for in-situ bitumen dilution.