Steel Design from Fully Parameter-Free Ab Initio Computer Simulations

The high strength and formability of steels is based on a large number of competing mechanisms on the microscopic/atomistic scale. Among them are dislocation gliding, dynamic strain aging, mechanical twin formation and local martensitic phase transformations, for which stacking faults play a dominant role. Many of the underlying concepts are based on empirical and experimental data. For a deeper understanding, however, an atomistic simulation of those structural defects becomes more and more crucial. Recent advances in ab initio calculations have sparked a lot of interest in deriving this information from such completely parameter free methods. Employing ab initio methods allows exploring chemical trends, to deliver parameters for phenomenological models, and to identify new routes for the optimization of steel properties. A major challenge in applying these methods to the above questions is the inclusion of all relevant temperature effects on the desired properties. We have therefore developed a large range of computational tools to improve the capability and accuracy of first-principles methods in determining free energies. These combine electronic, vibrational, and magnetic effects in an integrated approach. Based on these simulation tools, we are able to successfully predict mechanical and thermodynamic properties of metals with hitherto not achievable accuracy.