Dr. Swarnava Ghosh
National Center for Computational Sciences
Oak Ridge National Laboratory

Insights From the Electronic Structure for Multiscale Materials Modeling

ABSTRACT: The development of new materials with tailored properties is an important technological problem. The behavior of materials is predominantly due to multiscale effects. Therefore, it is important to accurately capture mechanisms at different length scales that give rise to novel macroscale properties. At the lower scales, first-principles (ab-initio) electronic structure methods, which are free from empirical inputs, have found immense success because of its high accuracy and can provide valuable insights into material behavior. However, there is tremendous computational cost associated with these simulations, and many of the widely used computational codes have limitations on the size and type of systems that can be studied. In this seminar, I will talk about real-space Density Functional Theory codes that can simulate large systems, and how we use information from lower scales for multiscale materials modeling. I will present applications in strain engineering of nanomaterials, tailoring the microstructure of lightweight Mg alloys for creep and spall resistance via thermo-mechanical processing. Next, I will talk about a linear scaling framework to accurately simulate defects in materials from first principles, free of ad-hoc parameter passing or artificial periodicity. Finally, I will present some recent results in defects in magnetic materials to engineer phase transitions.

BIOGRAPHY: Swarnava Ghosh is a Computational Scientist in the Advanced Computing for Chemistry and Materials Group in the Science Engagement Section at the National Center for Computational Sciences, Oak Ridge National Laboratory. Previously, he was a Postdoctoral Scholar in the same group and at the California Institute of Technology. He obtained his PhD and MS degrees from Georgia Institute of Technology and BS from Jadavpur University, India. His research interests include development of computational methods for first-principles calculation of materials and using these methods for multi-scale materials modeling.