Premium
Structure Prediction of Inorganic Nanoparticles with Predefined Architecture using a Genetic Algorithm
Author(s) -
Woodley Scott M.,
Sokol Alexey A.,
Catlow C. Richard A.
Publication year - 2004
Publication title -
zeitschrift für anorganische und allgemeine chemie
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.354
H-Index - 66
eISSN - 1521-3749
pISSN - 0044-2313
DOI - 10.1002/zaac.200400338
Subject(s) - algorithm , computer science , genetic algorithm , semiconductor , architecture , stage (stratigraphy) , ranging , space (punctuation) , phase space , biological system , materials science , physics , optoelectronics , biology , machine learning , art , paleontology , thermodynamics , visual arts , operating system , telecommunications
A multistage scheme has been developed for the structure prediction of viable clusters, where a genetic algorithm is implemented within the first stage to generate approximate structures that can be refined via direct energy minimisation techniques in a later stage. Inclusion and exclusion zones, regions of space where atoms are allowed and forbidden, respectively, are used to reduce the search space and to steer the predicted structures so that they have a predefined architecture, e.g. bubble‐like as oppose to dense phase‐like. In a final stage, electronic, rather than atomistic, calculations were performed. The method has successfully generated sets of stable II‐VI semiconductor clusters (MX) n , with n ranging from 1 to 15. This method allows one to predict different types of morphology that may not necessarily contain the global minimum structure, but are of particular importance for small clusters in nature and the laboratory.