Sample disturbance alters the constitutive behaviour of oil-saturated sands. This research investigated the possibility of using an oil-free material to determine the constitutive behaviour of an oil-saturated material, and thereby avoid the problem of sample disturbance. St. Peter sandstone was chosen as the oil-free analog for oil-saturated McMurray sand based on their similar grain shapes, gradations, densities and mineralogies.

A high pressure triaxial testing facility was designed and built to operate under confining stresses up to 35 MPa. A standard Ottawa sand, Ottawa 360/0, was used to show that the facility functions properly.

St. Peter sandstone, St. Peter 220/0.34, was triaxially tested under drained conditions over the confining stress range of 1.5-13.5 MPa.

The behaviour of the Ottawa sand, St. Peter sandstone and McMurray oil sand were analysed using three methods: the Mohr-Coulomb method, Bolton's Relative Dilation Index method and the state parameter concept. The state parameters ($\Psi$) for the oil-free sandstone, which were determined from the author's isotropically consolidated drained triaxial tests, were compared to those for the oil-saturated material, which were determined from isotropically consolidated drained test data obtained from the literature. The author determined the state parameters from an estimated steady state line and the knowledge that peak dilation rate strongly correlates to the state parameter.

The analyses indicate that an oil-free St. Peter sandstone behaves similarly to an oil-saturated McMurray sands with regards to peak dilation rate, strength and stiffness under elevated to high confining stresses.