Discrepancy Between Crystal Plasticity Simulations and Far-Field High-Energy X-ray Diffraction Microscopy Measurements

A three-dimensional dataset of Inconel-625 generated as part of the Air Force Research Laboratory’s Additive Manufacturing Challenge Series is considered to interrogate modeling errors in a crystal-plasticity finite-element (CPFE) framework. Grain-average elastic strains from this framework are compared to measurements made with far-field high-energy X-ray diffraction microscopy (ff-HEDM) on a specimen loaded in tension at room temperature. The ff-HEDM measurements, taken when the specimen was held under load control, were made on the same microstructure considered in the CPFE framework. This one-to-one comparison enables a thorough investigation into modeling discrepancies, specifically those induced by improper boundary condition selection and inability to model stress relaxation events which are a likely characteristic of intermittent plasticity. It is shown that erroneously constraining the Poisson effect locally with kinematic boundary conditions induces over-estimates of off-loading-axis normal strains as far as 100 μ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu $$\end{document}m away from the over-constrained surface, which is roughly 14% of the overall height of the polycrystal in the loading direction. A second set of boundary conditions that relaxes the constraints on the previously over-constrained surface showed clear quantitative improvement in model predictions. Specifically, the errors in the normal components of strains in the off-loading-axes decreased. Further, it was observed that irrespective of which of the two boundary conditions was applied, the CPFE framework deviates considerably from measurements in the plastic regime compared to the elastic regime. This is mainly due to the increased frequency of strain drop (or stress relaxation) events occurring in the plastic regime, most likely caused by shearing of γ′\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma ^{\prime}$$\end{document} precipitates and other intermittent dislocation activity. The CPFE framework is unable to accommodate the stress relaxation in grains. Nevertheless, the framework is interrogated for its ability to predict such events by comparing predicted slip fields with measured grain-level relaxation events. It is demonstrated that the set of grains experiencing a strain drop (from ff-HEDM measurements) collectively had higher values of plastic strain (calculated from CPFE simulations) compared to the set of grains that did not experience a strain drop.