Computational property prediction using the MedeA software platform

I&EC 99

Alexander Mavromaras, amavromaras@materialsdesign.com, Walter Wolf, WWolf@materialsdesign.com, Mikael Christensen, mchristensen@materialsdesign.com, Erich Wimmer, ewimmer@materialsdesign.com, and P. W. Saxe, psaxe@materialsdesign.com. Materials Design Inc, PO Box 2000, Angelfire, NM 87710
Controlling and designing materials down to the sub-micron and nanometer scale requires an “atomistic” understanding of matter, in particular of structure-property relationships and thermodynamics. As highly accurate measurements are increasingly difficult and cost intensive to perform, computational methods have become a widely accepted R&D technology component. Examples are modeling-based design of experiment, computational materials prescreening and characterization, structural analysis and visualization and increasingly the prediction of mechanical and thermodynamical properties. In this context, Materials Design has created the software platform MedeA, which integrates leading computational methods like the Vienna Ab-Initio-Simulation Package (VASP), the Monte-Carlo statistical mechanics program GIBBS, and experimental databases such as the Inorganic Crystal Structure Database (ICSD) with fully automated procedures to compute phonon dispersions and thermo-mechanical properties. MedeA facilitates the mining and interpretation of experimental data and allows for fast prediction of structural, mechanical, thermodynamical, and thermochemical data, which are often hard to extract from experiment. MedeA's advanced composition fingerprinting and structure building tools simplify the setup of complex atomic arrangements like point defects, surfaces, heterogeneous interfaces and grain boundaries. The software's powerful integration of structure building, visualization, computational task automation and analysis will be illustrated by specific application examples including diffusion in metals, alloy design and metal/ceramic surfaces and interfaces.

Keywords: impurity metal-host interactions, diffusion, point defects, grain boundary stability, interfaces, computational property prediction, mechanical and thermodynamical properties, first-principles methods, experimental databases