Tortuosity factor exerts a major influence in fluid flow in porous medium. However, this factor is very difficult to quantify from experimental measurements particularly in deformable medium. It is commonly treated as an empirical or curve-fitting parameter in determination of permeability. This study attempts to use electrical resistivity measurement method as an analytical tool to study how changes in grain fabric affect changes in tortuosity. The first phase of the study concentrates on development of a framework on the flow tortuosity behaviour observed in idealized granular assemblies of regular packings with uniform spheres subjected to various types of grain fabric alteration, such as matrix dilation and grain rearrangement (rolling and overriding). Then, the framework is extended to interpret data measured in real random systems of intact and sheared oil sand specimens. It was found that the locked structure in intact oil sand specimens produce relatively high tortuosity factors as compared to those in the idealized packings. Shear deformation disrupts this intrinsic structure reducing the tortuosity factor to about one-fifth of the initial value. This peculiar behaviour implies that the permeability of oil sand is very sensitive to shear deformation or grain fabric alteration.


The Athabasca oil sand deposit in northern Alberta, Canada, contains 200 billion m3 of bitumen in place (ERCB 1991). Oil sand is a very dense granular material with interlocked fabric (Dusseault and Morgenstern 1979; Wong 1999). The main mineral composition is dominated by quartz. Its in situ porosity varies from 32 to 35% whereas its maximum porosity can be as high as 45 to 47%. Viscous bitumen filling the intergranular spaces does not affect the geotechnical properties of the material. Steam stimulation is one of the viable methods to extract bitumen from the oil sand ore. Large volume of steam of high pressures (up to 10 MPa) and elevated temperatures (up to 300 °C) are injected into oil sand reservoirs for extraction. Because oil sands are dense uncemented sands, the resulting deformation induced during the recovery processes could cause significant changes in its hydraulic properties. The understanding of its hydraulic properties due to shear deformation is important for design of the steam stimulation recovery processes.

Wong et al. (1991) conducted pulse-decay tests to determine changes in absolute and effective permeabilities (to water) of oil sand under conventional triaxial compression. Touhidi-Baghini (1998) used the constant flow rate method to quantify changes in absolute permeability of locked sands specimens under triaxial compressions along various stress paths. They observed that volumetric dilation in triaxial compression could be as high as up to 10% and the permeability could be increased by 5–6 times even though localized multiple shear bands were developed in the specimens. In these conventional permeability tests, the measured hydraulic properties are average responses of the test specimens subject to test boundary conditions, and are significantly influenced by the formation of localized shear bands. They are no longer unique, and become dependent of specimen dimension and test conditions.

This content is only available via PDF.
You can access this article if you purchase or spend a download.