Statistically representative three-dimensional microstructures based on orthogonal observation sections
- PDF / 2,438,878 Bytes
- 11 Pages / 607.68 x 788.88 pts Page_size
- 68 Downloads / 142 Views
12/29/04
9:36 PM
Page 1969
Statistically Representative Three-Dimensional Microstructures Based on Orthogonal Observation Sections DAVID M. SAYLOR, JOSEPH FRIDY, BASSEM S. EL-DASHER, KEE-YOUNG JUNG, and ANTHONY D. ROLLETT Techniques are described that have been used to create a statistically representative three-dimensional model microstructure for input into computer simulations using the geometric and crystallographic observations from two orthogonal sections through an aluminum polycrystal. Orientation maps collected on the observation planes are used to characterize the sizes, shapes, and orientations of grains. Using a voxel-based tessellation technique, a microstructure is generated with grains whose size and shape are constructed to conform to those measured experimentally. Orientations are then overlaid on the grain structure such that distribution of grain orientations and the nearest-neighbor relationships, specified by the distribution of relative misorientations across grain boundaries, match the experimentally measured distributions. The techniques are applicable to polycrystalline materials with sufficiently compact grain shapes and can also be used to controllably generate a wide variety of hypothetical microstructures for initial states in computer simulations.
I. INTRODUCTION
COMPUTER models have become vital and cost-effective tools for predicting the lifetime and performance of materials. By conducting virtual experiments, these models have provided us with a facile means to examine material response to a wide variety of experimental conditions. However, it is important to consider that the behavior and performance of polycrystalline materials depend not only on the intrinsic properties of the material, but also on the microstructure. The microstructure of a fully dense, single-phase polycrystal can be described by a distribution of grain sizes, shapes, and orientations. In order for the models to have predictive power, it is imperative to have starting configurations that are representative of real microstructures. A direct method to accomplish this is to experimentally characterize microstructures using an automated electron backscattered diffraction (EBSD) technique, which will provide a spatially resolved orientation map of a sample surface.[1] This approach has been utilized in several recent studies (e.g., References 2 through 6); however, it is limited to two-dimensional structures. The extension of the direct-observation approach to three-dimensional structures would require the very difficult and laborious task of serial sectioning and volume reconstruction. This approach has been adopted by a few groups using synchrotron radiation,[7,8] but it will take many years for three-dimensional diffraction techniques to become generally available, if it ever happens. An DAVID M. SAYLOR, Postdoctoral Research Associate, is with MSEL, National Institute of Standards and Technology, Gaithersburg, MD 20899. Contact e-mail: [email protected] JOSEPH FRIDY, Senior Staff Computer Scientist, is with
Data Loading...
