Image
Venue MSG-024/025, Bernal Institute.
ABSTRACT
This talk will address challenging aspects of scientific computing and computational modelling in materials science. These are particularly the reproducibility and reusability of the models and data, and the usability and productivity of state-of-the-art tools, such as workflow management systems. An overview will be provided of two exemplary applications with some results: (1) multiscale modelling of charge transport in organic semiconductors and (2) virtual screening of catalysts for oxygen reduction and evolution reactions. Dr Kondov will propose generative and model-driven alternative approaches and recently developed tools to tackle the mentioned issues more effectively. For these two applications, he will show how to use the domain-specific language TextM to write models from scratch and to reuse models and data from previous simulation steps.
ABOUT THE PRESENTER
Ivan Kondov studied chemistry at the Sofia University where he obtained a master’s degree in theoretical chemistry and chemical physics in 1996. After a period of three years at the Bulgarian Academy of Sciences, he went to the Chemnitz University of Technology, Germany. There he received a PhD degree in theoretical physics in 2003 with a thesis on numerical schemes for solving quantum master equations describing charge and exciton transfer. As postdoc at the Technical University of Munich, he worked on simulating electron transfer processes at dye-semiconductor interfaces. In 2006 he moved to the Scientific Computing Center (SCC) at Karlsruhe Institute of Technology where since 2010, he has been the leader of the Simulation Laboratory NanoMicro, since 2015 the deputy head of the department Scientific Computing and Mathematics and since 2021 the spokesperson of SDL Materials Science. His current research includes multiscale modelling and high-throughput screening simulations using high performance computing and scientific workflows, modelling of tightly coupled multiscale systems, employing the concepts of model-driven engineering and service-oriented architecture in computational materials science.