US Army Corps of Engineers: Engineer Research and Development Center (Knowledge Core) · 1987 · 37 citations · 0 references
Open access
Natural SciencesMaterials EngineeringMaterials ScienceGeotechnical EngineeringMechanical Behavior Of MaterialsSoil MechanicsCreep BehaviorCreep DeformationDeformation MechanismsDuctile DeformationDuctilityStress-strain BehaviorBrittle Fracture BehaviorBrittle Fracture TheoryCompressive StrengthCreep Under Strain RateDislocation CreepDuctile FailureLong-term Creep
Uniaxial constant-stress and constant-strain-rate compression tests were conducted on more than 200 remolded, saturated, frozen specimens of Fairbanks silt under various conditions. A series of curves of stress vs strain rate for various temperatures for strain rates shows a close strength correspondence between the constant-stress and constant-strain-rate tests. All of these complete stress vs strain rate curves could not be described by a single power law or exponential equation, indicating that different deformation mechanisms are dominant within different ranges of strain rate. Two critical strain rates for distinguishing between the different deformation mechanisms were observed to be near 10 to the power -3 and 10 to the power -6 s to the power -1 for the medium-dense frozen Fairbanks silt. The former indicates the transition from ductile failure to moderate brittle fracture as strain rate increases, while the latter indicates the transition from dislocation creep to glide creep (by the authors' definition). Based on the changes in flow law, two fundamental creeps were classified: short-term creep, which is governed by glide creep, and long-term creep, which is governed by dislocation creep. The general formula for the failure criterion of frozen silt is given. The failure strain was not sensitive to temperature and strain rate over a certain range of strain rates, but it was very sensitive to density. Assur's creep model (1980) for ice was used to fit the creep data in this study. It works well for short-term creep but does not fit as well for long-term creep. The rate process theory was applied to the creep data. A very high value of experimental activation energy was obtained for lower stresses, and a very high value of apparent activation energy was observed for higher temperatures. The peak compressive strength was very sensitive to temperature and strain rate but relatively insensitive to density. While the initial tangent modulus is not sensitive to strain rate, it increases with decreasing temperature and density.