As any other DFT code, the geometry optimization in WIEN2k finds the local minimum that is 'nearest' to the cell you used as a start. It's very stable and efficient in doing so. If you want to find the global minimum - i.e. a crystal that can be entirely different from the structure you start from - you need to couple your DFT code to a code that is meant for global structure search. Examples are AIRSS (random structure search), USPEX (evolutionary algorithm) or CALYPSO (particle swarm algorithm).
Stefaan Van: Wien <wien-boun...@zeus.theochem.tuwien.ac.at> Namens ??? Verzonden: zaterdag 8 september 2018 10:35 Aan: wien <wien@zeus.theochem.tuwien.ac.at> Onderwerp: [Wien] Whether structure optimization can achieve global minimization? I browsed almost all of the mailing lists, but I didn't find this topic, I would like to inquire about the structure of the optimization of the global minimum. It is difficult to find the global minimum in the high dimensional potential energy surface. It requires us to traverse the potential energy surface, eliminate many local minimums, and finally find the global minimum. Some algorithms for searching for global minimization include genetic evolution algorithm, random searching, simulated annealing and so on. My question is whether Wien2k can achieve global minimization, and if so, how do I need to do that? Any comment(s) would be highly appreciated.
_______________________________________________ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
