Finding stable minima using a nudged-elastic-band-based optimization scheme

Optimization is essential in many scientific and economical areas, but it is often too complex to be tackled by simple straightforward calculations or by trial and error. Two well-known methods to find low-lying minima in such complex systems are simulated annealing and the genetic algorithm. In these methods artificial fluctuations control the probability of the system to overcome a local minimum having a certain depth. Here we present a complementary scheme that is based on the nudged-elastic-band method ordinarily used to find saddle points and we apply the scheme to find the most stable isomers of the phosphorus P(4), P(8) molecules and the corresponding molecules of As(n), Sb(n), and Bi(n) (n = 4,8) in the framework of the density functional theory. In the case of n = 8 we have found stable and metastable configurations, some of which are new and have similar energies. As a by-product we obtained an upper bound for the energy barriers between these configurations.