Regensburg 2016 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
MM: Fachverband Metall- und Materialphysik
MM 18: Topical session: Integrated computational materials engineering for design of new materials IV
MM 18.2: Talk
Tuesday, March 8, 2016, 10:45–11:00, H39
Better (random) walking through chemistry: how not to get lost in vast configurational spaces — •Chiara Panosetti1, Konstantin Krautgasser1, Dennis Palagin2, Karsten Reuter1, and Reinhard Maurer3 — 1Technische Universität München — 2University of Oxford — 3Yale University
Ab initio structure prediction can systematically aid the computational discovery and rational design of new materials, as well as providing interpretative insights when atomistic details are difficult to resolve experimentally. However, global geometry optimization –the method of choice for finding chemically relevant (meta-)stable structures– is rarely applied to large-scale systems. The main challenge lies in the necessity of efficient ways to traverse configurational spaces in which the number of minima explodes with system size. We recently proposed an approach to global screening in chemically meaningful subspaces [1]. Following the fil rouge of the exploitation of chemically motivated trial moves, we now extend it to the study of organic molecules on surfaces –by suitably imposing partial constraints– and to complex interfaces with variable or unknown stoichiometry –by applying strategies to alleviate the strain of newly generated structures upon grand-canonical particle insertion. A selection of prototypical results will be presented (a retinoic acid analogue on Au(111) and metal-silicon clusters growth) to illustrate how all relevant portions of chemical space can be accessed with this approach, whereas “conventional” sampling often even struggles to produce sensible structures besides the starting geometry at all. [1] Nano Lett., doi: 10.1021/acs.nanolett.5b03388 (2015)