Use of EBSD data in mesoscale numerical analyses
Experimentation, theory, and modeling have all played vital roles in defining what is known about microstructural evolution and the effects of microstructure on material properties. Recently, technology has become an enabling factor, allowing significant advances to be made on several fronts. Experimental evidence of crystallographic slip and the basic theory of crystal plasticity were established in the early 20th century, and the theory and models evolved incrementally over the next 60 years. During this time, modeling was primarily concerned with the average response of polycrystalline aggregates. While some detailed finite element modeling (FEM) with crystal plasticity constitutive relations was performed in the early 1980's, such simulations over taxed the capacity of the available computer hardware. Advances in computer capabilities led to a flurry of activity in finite element modeling in the next 10 years, thus increasing understanding of lattice orientation evolution and generating detailed predictions of spatial orientation distributions that could not be readily validated with existing experimental characterization methods. Significant advancements in material characterization, particularly automated electron backscatter diffraction (EBSD), have made it possible to conduct detailed validation studies of the FEM predictions. The data collected are extensive, and many questions about the evolution of microstructure and its role in determining mechanical properties can now be addressed. It is now possible to obtain a detailed map of lattice orientations on a fine size scale. This will allow detailed quantitative comparisons of experiments and newly emerging large scale continuum FEM simulations. This capability will facilitate model validation efforts aimed at predicting deformation induced structural features, such as shear bands and cell structures, as well as predictions of the effects of grain interactions. The insight gained from the coupling of EBSD and FEM studies will provide impetus for further development of microstructure models and theories of microstructure evolution. Early studies connecting EBSD data to detailed finite element models used manual measurements to define initial orientations for the simulations. In one study, manual measurements of the deformed structure were also obtained for comparison with the model predictions. More recent work has taken advantage of automated data collection on deformed specimens as a means of collecting detailed and spatially correlated data for FEM model validation. Although it will not be discussed here, EBSD data can also be incorporated in FEM analyses in a less direct manner that is suitable for simulations where the element size is much larger than the grain size. The purpose of such models is to account for the effects of evolving material anisotropy in macro-scale simulations. In these analyses, a polycrystal plasticity model (e.g., a Taylor model or a self-consistent model), or a yield surface constructed from a polycrystal plasticity model, is used to determine the constitutive response of each element. The initial orientations used in the polycrystal plasticity model can be obtained from EBSD analyses or by fitting distributions of discrete orientations to x-ray data. The use of EBSD data is advantageous in that it is easier to account for spatial gradients of orientation distribution within a part. Another area in which EBSD data is having a great impact is on recrystallization modeling. EBSD techniques can be used to collect data for quantitative microstructural analysis (Humphreys, 1998). This data can be used to infer growth kinetics of specific orientations, and this information can be synthesized into more accurate grain growth or recrystallization models (Vogel et al., 1996). A second role which EBSD techniques may play in recrystallization modeling is in determining initial structures for the models. A realistic starting structure is vital for evaluating the models, and attempts at predicting realistic structures with finite element simulations are not yet successful. As methodologies and equipment resolution continue to improve, it is possible that measured structures will serve as input for recrystallization models. Simulations have already been run using information obtained manually from a TEM.
- Research Organization:
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- W-7405-ENG-48
- OSTI ID:
- 15013398
- Report Number(s):
- UCRL-JC-137195-REV-1; TRN: US200604%%74
- Resource Relation:
- Conference: American Society of Metals Conference Minerals, Metals, Materials Society, St. Louis, MO, Oct 08 - Oct 12, 2000
- Country of Publication:
- United States
- Language:
- English
Similar Records
Multi-scale Modeling of Plasticity in Tantalum.
Modeling of friction-induced deformation and microstructures.