| Brief Bio of the Speaker | Dr. Alexandre Tkatchenko currently leads a group composed of 12 researchers at the Fritz Haber Institute (FHI) of the Max Planck Society in Berlin, Germany. The main interest of our group is to push the limits of accuracy and efficiency of first-principles atomistic methods to enable predictive modeling of functional materials (pharmaceuticals, opto-electronics, energy materials).To achieve this goal, we combine concepts and ideas from quantum chemistry, density-functional theory, and statistical mechanics. Dr. Tkatchenko graduated in computer science, and did a PhD in physical chemistry at the Universidad Autonoma Metropolitana in Mexico City. He then moved to FHI as an Alexander von Humboldt Fellow in 2007, becoming a group leader in 2011. Dr. Tkatchenko has more than 60 publications, has given 35 invited talks at international conferences, and 33 invited seminars at universities and research centers across the world. |
Abstract | The ultimate goal of development in atomistic modeling is to enable simulations as accurate and as fast as possible. If we could systematically achieve chemical accuracy of 1 kcal/mol and time scale of seconds in atomistic simulations, this would in principle enable rational computational design of materials including pharmaceuticals, opto-electronic and energy materials. Our group has been addressing the issues of both the accuracy and the speed for first-principles atomistic simulations. I will present our recent developments in many-body van-der-Waals energy density functionals, which lead to accuracy better than 1 kcal/mol for molecular materials [1,2]. Applications will be shown for a wide variety of systems in biology, chemistry, and physics [3-5]. In terms of speed, we have been developing machine learning methods in which the problem of solving the molecular Schroedinger equation is mapped onto a nonlinear statistical regression problem of reduced complexity [6]. This talk will discuss the recent exciting developments in first-principles atomistic modeling, and the many challenges that remain to be addressed to enable truly predictive simulations of functional materials. [1] A. Tkatchenko and M. Scheffler, Phys. Rev. Lett.102, 073005 (2009). [2] A. Tkatchenko, R. A. DiStasio Jr., R. Car, and M. Scheffler, Phys. Rev. Lett. 108, 236402 (2012). [3] R. A. DiStasio Jr., O. A. von Lilienfeld, A. Tkatchenko, Proc. Natl. Acad. Sci. USA 109, 14791 (2012). [4] V. V. Gobre and A. Tkatchenko, Nature Communications 4:2341 (2013). [5] A. M. Reilly and A. Tkatchenko, J. Phys. Chem. Lett.4, 1028 (2013). [6] M. Rupp, A. Tkatchenko, K.-R. Mueller, O. A. von Lilienfeld, Phys. Rev. Lett. 108, 058301 (2012). |
