Heavy oil and tar sands are important hydrocarbon resources that are destined to play an increasingly important role in the oil supply of the world. A huge proportion of total world oil resources are in the form of these highly viscous fluids. The main recovery mechanism for these kinds of reservoirs is to somehow reduce their viscosity by the application of heat. In these extra heavy oil reservoirs, the reservoir has almost no injectivity, and therefore conventional steam flooding is hard to conduct. Steam Assisted Gravity Drainage (SAGD), however, reduces the viscosity of bitumen in place and the heated bitumen drains due to gravity forces towards the production well, where it is produced. Modeling and evaluating the production mechanisms in this process requires a thorough understanding of multi-phase flow parameters like relative permeability.
Relative permeability data depend on a number of different parameters among others temperature and fluid viscosity. Viscosities of the flowing fluids drop with temperature, which can affect the relative permeability data. There has been a long debate on the actual impact of temperature on the relative permeabilities. Although some authors have reported saturation range shifts and relative permeability curve variations by temperature, others have attributed these variations to artifacts inherent in the methods used and the systems tested. Viscous instabilities and fingering issues have been blamed for temperature dependencies reported, and some researchers have reported that relative permeability data changes due to oil/water viscosity ratio changes at different temperatures.
The variations in the experimental conditions have resulted in different and even contradictory results. There is specifically few experimental works conducted on Athabasca oil systems, and previously reported trends mainly apply to less viscous oils. This implies that the actual effect of temperature on flow behavior of fluids in the rock is case specific. Due to the contradictory reports and conclusions, which are due to variation in the systems being tested, it seemed necessary to conduct our own core flooding experiments, and investigate the curves of relative permeability. The objective was to obtain the imbibition relative permeability curves in an Athabasca oil type reservoir at different temperatures and oil viscosities, and figure out any possible trends of variations with temperature.
Before conducting the core flooding experiments, some fluid behavior experiments were done to figure out the properties of bitumen used in this study. These include fluid compositions, density, viscosity, molecular weight and oil/steam interfacial tension. These properties were further used in numerical simulation studies.
Core floodings were conducted on glass bead packs and sand packs saturated with heavy oils with varying viscosities. Displacement experiments with water were performed at different temperatures, and unsteady-state method of relative permeability measurement was conducted. The relative permeability data were determined by history matching the oil production data and pressure differential data in each experiment.
Results indicated a change in the water saturation range in the oil-water relative permeability curves. The shift was towards higher water saturations, meaning an increase in irreducible water saturation and a decrease in residual oil saturation. Regarding the shape of relative permeability data, no unique trend of either rising or falling with temperature was found for oil and water relative permeability curves. The viscous instabilities are believed to be present in the experiments.
As the same saturation range shift occurs by comparing the results at the same temperature level and by only changing the oil viscosity, this suggests that the temperature dependency of relative permeabilities can be attributed to the drop in oil to water viscosity ratio by temperature.
The variations of relative permeability data with temperature was therefore found to be more related to artifacts in the experimental procedures like viscous fingering, and fluid viscosity changes than fundamental flow properties.
Numerical simulations were accomplished on field scale SAGD and ES-SAGD (Expanding Solvent SAGD) operations testing the effect of relative permeability curves. Temperature dependent relative permeability data were tested and Oil production was found to be strongly dependant on the end point data. It is therefore suggested to use this option as a matching criterion when trying to history match SAGD field data.
Since the main experimental part of this study deals with temperature dependency of relative permeability data, the introduction of this thesis is totally devoted to introducing this concept and its measurement methods and a literature review on the works performed so far. The main thesis is composed of three main parts, the fluid behavior experiments on bitumen, one-dimensional flow studies and multi-dimensional flow part. The results of fluid behavior experiments are given in chapter 2. Chapters 3 and 4 are devoted to one-dimensional flow works and chapters 5 and 6 present the part of this thesis dealing with two and three-dimensional flow. It should, however, be mentioned that chapters 4 to 6 can be read independently, as the contents of these chapters are taken from previously published papers with some minor revisions.