Mechanical Properties of St. Peter Sandstone A Comparison of Field and Laboratory Results

Abstract

The St. Peter sandstone is an arenaceous, ortho-quartzitic, sublittoral cratonic sheet sand of middle Ordovician age. It is remarkable in geographic extent, mineralogical composition, thickness and engineering properties. A vast majority of its extent is buried; only in the Upper Mississippi River valley is it readily exposed. The St. Peter sandstone behaves as a locked sand, a nearly cohesionless geo-material characterized by brittle behavior, a lack of interstitial cement, and high dilation rates at failure. When confined it is capable of supporting large loads with small deformations, even under saturated conditions, yet when confinement is removed it disintegrates and readily flows. Dry intact samples have a uniaxial strength of approximately 1 MPa and a Young’s Modulus of about 1 GPa. Triaxial tests conducted at confining pressures of less than 150 kPa yield an angle of internal friction of approximately 60 Because the St. Peter sandstone disintegrates so readily when unconfined, sampling the material is quite difficult. The St. Peter sandstone has been excavated as foundation material, and for tunnels and sewers, by the cities of Minneapolis and St. Paul for decades. Earliest testing was for those endeavors. As the interest in the design of underground spaces has grown, testing of the St. Peter sandstone has changed to meet that need but testing has been done predominantly in the laboratory. The purpose of this project was to evaluate the mechanical properties of St. Peter sandstone by comparing in situ tests with laboratory test results. Direct shear tests were conducted to evaluate strength-dilatancy behavior. Transmitted light and scanning electron microscopy were used to help explain the high friction angle of the material. At low confining pressures the St. Peter sandstone exhibits a friction angle of around 60 but with small cohesion, less than 100 kPa. The high angle of internal friction at failure may be due to locked sand particles or to post-depositional quartz overgrowths. Tests on pulverized, densely packed samples (with void ratios similar to intact samples) and loosely packed samples were conducted in the same fashion as the intact samples and yielded friction angles of approximately 45 and 35 respectively. Pressuremeter tests were performed in situ and the results were interpreted using elasto-plastic analysis. By properly considering system stiffness, a Young’s modulus of approximately 0.5 GPa was determined and a friction angle between 60 and 40 was estimated, depending on the assumed dilation angle.