No abstract

To obtain a well-determined constitutive database for the Alaskan frozen soil at confining pressures up to 100 MPa and temperatures down to -25°C, a series of laboratory tests was conducted using a unique high-pressure, low-temperature apparatus and the split Hopkinson pressure bar (SHPB). Quasi-static compression tests and indirect tension (or Brazilian) tests, are required to constrain the variabilities of material properties of frozen soil. The SHPB tests are required to obtain dynamic compression properties and the strain rate dependency of the frozen soil. The results from laboratory material testing showed that Alaskan frozen soil exhibits pressure and temperature dependence, rate sensitivity, anisotropy, brittle and ductile behavior, volumetric compaction, and dilation. The rate-sensitive and anisotropic version of a plasticity model, being developed by Fossum and Fredrich (2000), was able to represent the deformation behavior of such a complex material very well. This model includes high strain-rate sensitivity and anisotropy in both the elastic and plastic regimes. The model is defined through a continuous yield and loading surface for unified dilation and compaction phenomena. It is envisioned that this model will be used to predict the deformation and failure of frozen soil under the dynamic loading conditions resulting from projectile penetration into frozen soil targets.

This report investigates the validity of several key assumptions in classical plasticity theory regarding material response to changes in the loading direction. Three metals, two rock types, and one ceramic were subjected to non-standard loading directions, and the resulting strain response increments were displayed in Gudehus diagrams to illustrate the approximation error of classical plasticity theories. A rigorous mathematical framework for fitting classical theories to the data, thus quantifying the error, is provided. Further data analysis techniques are presented that allow testing for the effect of changes in loading direction without having to use a new sample and for inferring the yield normal and flow directions without having to measure the yield surface. Though the data are inconclusive, there is indication that classical, incrementally linear, plasticity theory may be inadequate over a certain range of loading directions. This range of loading directions also coincides with loading directions that are known to produce a physically inadmissible instability for any nonassociative plasticity model.

Management jointly sponsored a program to evaluate elastomeric O-ring seal materials for radioactive material shipping containers. Thk report presents the results of low-and high-. temperature tests conducted on 27 common elastomeric compounds. ACKNOWLEDGMENTS I thank the many Sandia National Laboratories Transportation Systems and Development Department members who have contributed to this work, including

No abstract