CAD-based reconstruction of 3D polycrystalline alloy microstructures from FIB generated serial sections
Introduction
Advanced metallic materials used in many industrial applications have complex multi-colony, multi-phase polycrystalline aggregates in their microstructure as shown in Fig. 1. The mechanical behavior and fatigue failure response are intricately governed by microstructural features that include morphological and crystallographic characteristics, e.g. shape, size and location of phases in the colony structure, relative colony size and locations, crystal orientations and misorientations, grain boundary geometry etc. Detailed micromechanical computational models are being used to understand deformation and damage mechanisms and throw light on the stochastic nature of failure and fatigue phenomena of these materials [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]. While, the computational models of polycrystalline materials implementing crystal plasticity models are making great strides in predicting the stress–strain behavior with reasonable accuracy, ductility and fatigue failure predictions with high fidelity are still far from mature. Morphological and crystallographic heterogeneities in the microstructures result in strong anisotropy and localized non-homogeneous deformation, which impose severe challenges to these computational models. Experimental studies [11] suggest that the growth of crystallographic microslip bands along active slip systems of plastic flow causes localized instability due to compatibility requirements between interacting grains. They continue to grow across grain boundaries due to grain structure instability and eventually manifest as macroscopic shear bands. The interaction of microscopic shear bands with transverse grain boundaries also leads to grain boundary microcracking, which grows in size and merge to cause fracture.
It is important for computational models to capture the 3D geometric and crystallographic details of grain morphology, as well as their distribution in the polycrystalline aggregate for robust prediction of their properties. An automated approach of characterizing 3D microstructure using a dual beam focused ion beam (FIB)–SEM system has been recently developed [26] to acquire 3D orientation data of a succession of sections in the material microstructure. Using a FIB column in the microscope, highly localized micromachining and ion imaging is conducted. Following this, high resolution electron back-scatter diffraction (EBSD) images are acquired by a SEM column for grain orientations. These experimental advancements have made it possible to seamlessly reconstruct high fidelity 3D grain and subgrain microstructures of polycrystalline materials. The high fidelity 3D microregions can then be discretized and analyzed by computational methods like the finite element analysis methods for an accurate and reliable prediction of material properties. This paper develops a CAD-based method of creating 3D grain structures through post-processing of the FIB–SEM generated OIM data. A seamless reconstruction process will have certain characteristics and features that are summarized below.
- •
Input/output data: The input to the process is crystallographic orientation data of a metallic specimen in a 3D square grid. The output is a collection of solid bodies, with each body representing an individual grain. A requirement is that there be no overlap or gaps between them.
- •
Data uncertainty: Uncertainty corresponding to un-indexed points, incorrectly indexed points, misalignment, scatter marks etc. are to be expected in the experimentally acquired data, and has to be effectively dealt with.
- •
Process automation: The entire process from reading of the experimentally generated orientation maps to the creation of grain geometries in the aggregate should be automated, such that no, or minimum, additional user input is required.
- •
Robustness: The grain ensemble reconstruction procedure should be adequately robust to deal with different material microstructures, for which the sectional data is available. This requires identification of unstable operations and their removal.
- •
Requirements for finite element mesh: Typically crystal plasticity simulations of polycrystalline microstructures require prohibitively high computations, especially for models that represent microstructural details. It is therefore desirable to generate the microstructural details, keeping in mind both accuracy and efficiency considerations. Optimal representation with respect to the number of nodes and elements in the finite element mesh should be generated to retain both accuracy and efficiency of the eventual computational analysis.
This paper addresses the development of a seamless methodology for simulating polycrystalline metal microstructures from FIB–SEM generated serial sections using primitives used in CAD methods. A unique strength of this method is that it is entirely possible to monitor and control the resolution of the simulated microstructure for accuracy and efficiency needed for materials modeling. A commercial CAD package Unigraphics NX3 [31] (henceforth referred to as NX3) is used to perform all operations in the polycrystalline microstructure reconstruction. NX3 allows direct access to most of its geometric modeling and manipulation facilities through Open C API interface. A special module has been developed through this interface to reconstruct microstructure without any user intervention. Section 2 reviews some of the related work in this general area. Section 3.1 discusses steps for data collection and cleanup procedures, while Sections 3.2 Domain construction for individual grains, 3.3 Compatibility requirements in a polycrystalline aggregate, 3.4 Defeaturing spurious artifacts of the reconstructed grains describes the reconstruction process. Finally validation of this method with respect to microstructural characteristics is discussed in Section 4.
Section snippets
Brief review of microstructure reconstruction methods
Since polycrystalline deformation is predominantly 3D in nature, it is essential that the microstructural models be developed with detailed 3D information. Techniques based on ultrasonics or its variants, such as acoustic microscopy or laser ultrasonics [12], [13] rely on good reflection properties and have limited application in metals. While X-ray-based computed tomography [14], [15] methods are widely used in 3D solid model generation, they are generally deficient in achieving the resolution
Reconstructing the 3D polycrystalline microstructures
Fig. 2 is a flowchart of the sequence of operations that are necessary to construct the 3D microstructure from experimentally acquired section data. Each of the steps is discussed in this section.
Numerical results and algorithm validation
The CAD-based methodology described in Section 3 is now tested for reconstructing the microstructure of a fine-grained polycrystalline nickel-base superalloy, IN100. Fig. 17 (a) shows a section of the microstructure in a voxel-based representation using the experimental data after alignment. Fig. 17(b) shows a 3D representation of the reconstructed grain microstructure with the colors corresponding to a given orientation. Visual comparison of the two figures shows good agreement between the
Conclusion
This paper is aimed at the development of a robust and comprehensive CAD-based methodology for simulating 3D microstructures of polycrystalline metals using crystallographic input data on sections created by a focused ion beam (FIB)–scanning electron microscopy (SEM) system. The 3D dual-beam FIB–SEM is very effective for serial-sectioning micron/submicron scale metallic specimens and subsequently obtaining crystallographic orientation maps for each section. Orientation maps from each of these
Acknowledgements
The authors from OSU acknowledge the support of Air Force Office of Scientific with Grant # FA9550-05-1-0067 (Program Director: Dr. James Tiley) and the support of Office of Naval Research with Grant # N00014-05-1-0504 (Program Director: Dr. Julie Christodolou). This support is gratefully acknowledged. Computer support by the Ohio Supercomputer Center through grant # PAS813-2 is also acknowledged.
References (32)
- et al.
Modeling cyclic ratcheting based fatigue life of HSLA steels using crystal plasticity FEM simulations and experiments
Int J Fatigue
(2006) - et al.
Crystallographic texture evolution in bulk deformation processing of fcc metals
J Mech Phys Solids
(1992) - et al.
Elasto-viscoplastic constitutive equations for polycrystalline fcc materials at low homologous temperatures
J Mech Phys Solids
(2002) - et al.
On the mechanical behavior of AA 7075-T6 during cyclic loading
Int J Fatigue
(2003) - et al.
Polycrystal orientation distribution effects on microslip in high cycle fatigue
Int J Fatigue
(2003) - et al.
Damage assessment in an Al/SiC composite during monotonic tensile tests using synchrotron X-ray microtomography
Mater Sci Engng A
(1997) - et al.
Image-based modeling of the response of experimental 3D microstructures to mechanical loading
Scripta Mater
(2006) - et al.
Arbitrary topology shape reconstruction from planar cross sections
Graphical Models Image Process
(1996) - et al.
3D reconstruction and characterization of polycrystalline microstructure using a FIB–SEM system
Material Characterization
(2006) - et al.
3D polycrystalline microstructure reconstruction from FIB generated serial sections for FE Analysis
Comput Mater Sci
(2007)
Nonuniform deformations in polycrystals and aspects of the validity of the Taylor model
J Mech Phys Solids
A numerical study of localized deformation in bicrystals
Mech Mater
Crystal plasticity based FE model for understanding microstructural effects on creep and dwell fatigue in Ti-6242
ASME J Engng Mater Tech
Crystal plasticity modeling of deformation and creep in polycrystalline Ti-6242
Metal Mater Trans A
An accelerated methodology for the evaluation of critical properties of polyphase alloys
Metallurgical and Mater Trans A
Microscopic versus macroscopic aspect of shear band deformation
Acta Metall Mater
Cited by (56)
Automated identification of deformation twin systems in Mg WE43 from SEM DIC
2020, Materials CharacterizationCitation Excerpt :The efficient collection, processing, and organization of large multi-modal experimental datasets is needed to define representative material testbeds for the development and application of these new data mining approaches, in addition to its significance for studies of specific materials systems. Therefore, there are also ongoing efforts to collect and reconstruct large scale, high resolution, multi-modal datasets in a high-throughput manner [5–11]. Magnesium (Mg) is a promising material for lightweight structural applications, including in the automotive, aerospace, electronics, and biomedical device industries [12–15].
Anisotropic polycrystal plasticity due to microstructural heterogeneity: A multi-scale experimental and numerical study on additively manufactured metallic materials
2020, Acta MaterialiaCitation Excerpt :The grain morphology (size, shape and shape-axes orientation) distribution of the as-built material was characterized by EBSD of relatively large areas on two orthogonal planes: parallel to the BD (designated by ∥BD) and orthogonal to the BD (designated by ⊥BD). The alternative approach for a more precise 3D characterization of the microstructure would be 3D-EBSD using a focused ion beam (FIB) automated serial sectioning technique [97–101]. Moreover, dislocation density and twin fraction in the as-built/undeformed material as well as their evolution with respect to strain using differently strained material states were also estimated by means of EBSD orientation maps on ∥BD and ⊥BD sections.
Automated meshing of electron backscatter diffraction data and application to finite element micromagnetics
2019, Journal of Magnetism and Magnetic MaterialsInitializing type-2 residual stresses in crystal plasticity finite element simulations utilizing high-energy diffraction microscopy data
2018, Materials Science and Engineering: AIntegrated approach for geometric modeling and interactive visual analysis of grain structures
2018, CAD Computer Aided DesignInvestigation of plastic strain rate under strain path changes in dual-phase steel using microstructure-based modeling
2017, International Journal of Plasticity