Application of Microprobe Analysis to the Reconstruction and Characterization of Dendritic Structures
- PDF / 959,313 Bytes
- 10 Pages / 593.972 x 792 pts Page_size
- 48 Downloads / 181 Views
the field of numerical simulation, such as computational fluid dynamics (CFD) or structural mechanics, the three-dimensional (3-D) geometric representation of the investigated structure is a crucial input parameter. Particularly, this is the case if complex-shaped solidification microstructures are investigated using numerical simulation, because microstructural features (e.g., number and connectivity of particles or dendrites) might affect the simulation results distinctly. Such features can only be determined exactly from 3-D microstructure representations.[1] For example, fluid-flow simulations through reconstructed dendritic networks indicated that the direction-dependent permeability is influenced strongly by the interfacial surface area, which varies with the fraction of solid.[2] If a certain structure has a well-defined geometry, then it is relatively straightforward to create its representation by using computer-aided design tools. In most engineering cases, this may apply. However, the evolving microstructures undergoing some growth phenomJOSEF DOMITNER, PhD Student, is with the Christian Doppler Laboratory for Multiphase Modeling of Metallurgical Processes, University of Leoben, A-8700 Leoben, Austria. Contact e-mail: josef. [email protected] ABDELLAH KHARICHA, Senior Scientist, MENGHUAI WU, Associate Professor, and ANDREAS LUDWIG, Professor, are with the Chair of Simulation and Modeling of Metallurgical Processes, University of Leoben. Manuscript submitted March 30, 2011. Article published online March 22, 2012 METALLURGICAL AND MATERIALS TRANSACTIONS A
ena can exhibit structures with complex shapes. Therefore, they cannot be described easily with simple geometric relations. To make such structures available for numerical simulations, they are captured using different analysis techniques. Commonly, computed Xray microtomography[3–6] or serial sectioning of the sample combined with micrograph stacking is used. The generated geometric information is then imported into the simulation software. An overview about first serial sectioning approaches carried out in the last century to study the shape of metal phases such as ferrite, cementite, or perlite is given by Kral et al.[7] The described approaches had the fundamental disadvantage that the manual sectioning and image acquisition procedures were extremely tedious. Nevertheless, in 1983 DeHoff already predicted that serial sectioning would become an important method to generate and to evaluate 3-D metallic microstructures by overcoming this awkward problem.[1] Li et al.[8] used the serial sectioning technique to obtain detailed 3-D microstructure images of an Al-alloy reinforced with Si-particles. Therefore, optical micrographs taken from a series of two-dimensional (2-D) sections were digitalized and then stacked together computationally. Prior to removing each section from the sample, fiducial pyramidal hardness indents of known proportions were made to enable the alignment of the sections and to measure the removed material thickness at each pass. Chawla et
Data Loading...
