A new procedure is suggested to improve genetic algorithms for the prediction of structures of nanoparticles. The strategy focuses on managing the creation of new individuals by evaluating the efficiency of operators (o1, o2,…,o13) in generating well-adapted offspring. This is done by increasing the creation rate of operators with better performance and decreasing that rate for the ones which poorly fulfill the task of creating favorable new generation. Additionally, several strategies (thirteen at this level of approach) from different optimization techniques were implemented on the actual genetic algorithm. Trials were performed on the general case studies of 26 and 55-atom clusters with binding energy governed by a Lennard-Jones empirical potential with all individuals being created by each of the particular thirteen operators tested. A 18-atom carbon cluster and some polynitrogen systems were also studied within REBO potential and quantum approaches, respectively. Results show that our management strategy could avoid bad operators, keeping the overall method performance with great confidence. Moreover, amongst the operators taken from the literature and tested herein, the genetic algorithm was faster when the generation of new individuals was carried out by the twist operator, even when compared to commonly used operators such as Deaven and Ho cut-and-splice crossover. Operators typically designed for basin-hopping methodology also performed well on the proposed genetic algorithm scheme.
Genetic algorithm is employed to survey an empirical potential energy surface for small NaxKy clusters with x + y ≤ 15, providing initial conditions for electronic structure methods. The minima of such empirical potential are assessed and corrected using high level ab initio methods such as CCSD(T), CR-CCSD(T)-L and MP2, and benchmark results are obtained for specific cases. The results are the first calculations for such small alloy clusters and may serve as a reference for further studies. The validity and choice of a proper functional and basis set for DFT calculations are then explored using the benchmark data, where it was found that the usual DFT approach may fail to provide the correct qualitative result for specific systems. The best general agreement to the benchmark calculations is achieved with def2-TZVPP basis set with SVWN5 functional, although the LANL2DZ basis set (with effective core potential) and SVWN5 functional provided the most cost-effective results.
Rate coefficients for the collisions were calculated employing quasiclassical trajectories and the first available set of potential energy surfaces for such excited nitrogen interactions. The details of the vibrational energy transfer are discussed, such as the contributions from reactive and non-reactive trajectories as well as the contribution of each electronic symmetry. The calculated state-to-state and state-to-all rate coefficients show that deactivation is far more probable than excitation, and multi-quanta deactivation play an important role.
12 September 2014 in Journal of Molecular Modeling