Global geometry optimization of atomic and molecular clusters by genetic algorithms

摘要

One of the numerous NP-hard global optimization problems in theoretical chemistry is that of finding the global minimum energy structure of an atomic or molecular cluster. This paper describes our method of solving this problem with specialized genetic algorithms, both on the level of empirical model potentials and in combination with extremely expensive calculations from first principles. Discussing several current application examples, we show that our approach is a valuable new tool in hot topics of cluster chemistry.