Validity of crystal plasticity models near grain boundaries: a contribution of elastic strain measurements at the micron scale

TL;DR

This study combines synchrotron Laue microdiffraction and Digital Image Correlation to evaluate the accuracy of crystal plasticity models in predicting elastic strains near grain boundaries in a stainless steel bicrystal, revealing overestimations and the need for model improvements.

Contribution

It provides experimental validation of crystal plasticity models at micron scale near grain boundaries, highlighting discrepancies in elastic strain predictions.

Findings

Total strains are well predicted by models.

Elastic strains are overestimated near grain boundaries.

Models require refinement for accurate stress prediction.

Abstract

Synchrotron Laue microdiffraction and Digital Image Correlation measurements were coupled to track the elastic strain field (or stress field) and the total strain field near a general grain boundary in a bent bicrystal. A 316L stainless steel bicrystal was deformed in situ into the elasto-plastic regime with a four-point bending setup. The test was then simulated using finite elements with a crystal plasticity model comprising internal variables (dislocation densities on discrete slip systems). The predictions of the model have been compared with both the total strain field and the elastic strain field obtained experimentally. While activated slip systems and total strains are reasonably well predicted, elastic strains appear overestimated next to the grain boundary. This suggests that conventional crystal plasticity models need improvement to correctly model stresses at grain boundaries.