Skip over navigation

 

Experimental Validation of Crystal Plasticity Models

Jay Carroll (Sandia National Laboratories), Hojun Lim (Sandia National Laboratories), Brad Boyce (Sandia National Laboratories), Corbett Battaile (Sandia National Laboratories), Christopher Weinberger (Sandia National Laboratories)

Plasticity at Different Length Scales

Mon 9:00 - 10:30

CIT 219

Modeling deformation behavior of metals at the grain scale is challenging due the complicated behavior of a heterogeneous grain structure with varying grain shapes, orientations, and interaction effects. Although relatively simple stress projection factors, such as the Schmid and Taylor factors, have been used with some success, these only consider each grain in isolation, ignoring the contribution of grain neighborhoods on stress and strain heterogeneity. Crystal plasticity finite element models (CP-FEM) have the capability to employ stress projection factors in a framework that considers grain interactions. However, before these models can be used with any confidence, they must be validated against experimental measurements that are directly comparable to model predictions. Most validation comparisons have thus far been limited to qualitative visual comparisons between model predictions and experimental measurements of strain fields or grain rotation fields. In this work, we will provide a review of previous validation attempts of crystal plasticity models and introduce our work toward a method for quantitative comparison of CP-FEM models with validation experiments. BCC tantalum tensile specimens were annealed to mitigate the effects of unknown subsurface microstructure, resulting in a pseudo-two-dimensional grain structure with grain sizes on the order of millimeters. The grain structure of these oligocrystals (less than 20 grains in the gage section) was characterized using electron backscatter diffraction (EBSD). Microstructure measurements were combined with deformation measurements from digital image correlation (DIC) and optical profilometry to obtain a data set that can be directly modeled and used for quantitative validation on a pointwise basis. We find reasonable agreement between the CP-FEM model and experimental measurements in most grains. Possible causes of localized disagreement, and suggestions for improvement, will also be discussed.