SGER: Realistic 3-D Microstructure Simulations for Microstrucutral Science: Implications for Microstructure-Based Materials Design

Georgia Tech Research Corporation, Atlanta GA

Investigators

Abstract

TECHNICAL: Representation, modeling, and simulations of microstructures at relevant length scales are of central importance for understanding fundamental relationships between processing, structure, and properties of materials as well as for development of microstructure-based materials design and development methodologies. Unfortunately, a significant roadblock in the development of predictive quantitative materials science and efficient microstructure-based materials design methodologies is our inability to mathematically represent and numerically simulate the geometry of realistic multi-phase multi-scale three-dimensional (3-D) stochastic material microstructures and incorporate them in the models and simulations of material properties and performance. Accordingly, the objective of this research is to develop the required methodology for simulations of realistic multi-phase multi-scale stochastic three-dimensional microstructures and to implement them in the computational models of materials behavior. The methodology will be developed through its applications to (i) powder metallurgy processed and extruded 3-D microstructures of discontinuously reinforced Al-alloy (DRA) composites containing spatially clustered SiC particles and pores of complex shapes, (ii) multi-phase multi-scale microstructures of cast Mg-alloys that are of current interest for automotive applications, and (iii) microstructures of solid oxide fuel cell (SOFC) cathodes containing three phases. The research is of high-risk nature because numerous conceptual and computational issues need to be successfully addressed to develop the required methodology. On the other hand, if successful, the modeling and simulation methodology will be transformative of the way materials scientists study processing-microstructure-properties relationships, and will lead to a new paradigm for efficient microstructure-based materials design on the basis of few experiments and large simulated properties data sets obtained from virtual experiments. NON-TECHNICAL: The quantitative and predictive microstructure-based materials design approach would be of high interest not only to the Metals community, but would also benefit other branches of Materials Science. The correlations among the simulation parameters and the process conditions can be used to identify a range of process conditions that is likely to yield the material with the required microstructure that has a desired combination of properties. The high-resolution large-volume multi-phase multi-scale simulated material microstructures generated in this research will be made available to the materials research community via an FTP site or website. In addition, the work would support two graduate students.