David Cereceda

Unraveling the temperature dependence of the yield strength in single-crystal tungsten using atomistically-informed crystal plasticity calculations

Abstract We use a physically-based crystal plasticity model to predict the yield strength of body-centered cubic (bcc) tungsten single crystals subjected to uniaxial loading. Our model captures the thermally-activated character of screw dislocation motion and full non-Schmid effects, both of which are known to play critical roles in bcc plasticity. The model uses atomistic calculations as the sole source of constitutive information, with no parameter fitting of any kind to experimental data. Our results are in excellent agreement with experimental measurements of the yield stress as a function of temperature for a number of loading orientations. The validated methodology is employed to calculate the temperature and strain-rate dependence of the yield strength for 231 crystallographic orientations within the standard stereographic triangle. We extract the strain-rate sensitivity of W crystals at different temperatures, and finish with the calculation of yield surfaces under biaxial loading conditions that can be used to define effective yield criteria for engineering design models.

Modeling dynamic fragmentation of concrete under high strain-rate loading

Extreme high-rate loading conditions in structural materials trigger a complex process of fragmentation involving probabilistic, energetic and mechanical aspects. In this work we discuss a one-dimensional model based on [1] that captures the physics of dynamic fracture and fragmentation in concrete at strain rates from 103 to 105 /s, with particular interest in the higher strain rate values. In particular, the model considers a one-dimensional bar under a uniform tensile initial strain rate, with a stochastically varying strength. Initial results for the relationship between average fragment size and strain rate show good agreement with shock tube experiments on concrete panels. However, the predicted distribution of fragment size exhibits a smaller variance than that observed in the experiments. Future work will evaluate this difference in the results, which could be the result of the one-dimensionality of the model, heterogeneity of strain rate in the shock tube tests, experimental measurement errors, or a combination of all of these. Further investigations to extend the present model to other brittle materials like glass and concrete are also currently under development.