遗传算法研究贵金属团簇的基态结构

在遗传算法的选择规则中加入配位数分析的方法 ,并应用基于紧束缚模型的二级矩近似势计算得到了含 38、5 5和 75个原子的铜和金团簇的基态结构 .运用共同近邻分析等方法比较了两种团簇基态原子排列方式的显著差异 ,并认为由于金原子间作用的短程特性使得含幻数原子数 (N =38、5 5 )的金团簇倾向于无序的内部排列方式

Studies on the Optimal Structures of Some Noble Metal Clusters by an Improved Genetic Algorithm

The authors have studied the ground state configurations of copper and gold clusters with 38, 55 and 75 atoms respectively by using Genetic Algorithm which has been modified by setting fitness function and selection rule of evolution. The interaction amony atoms is described by a Gupta-like semi-empirical potential based on the second-moment approximation of tight-binding scheme. The prominent differences between the optimal structures of copper and gold clusters have been illustrated by common-neighbor-analysis as well as the distribution of interatomic distance. The authors deemed that the short-range characteristic of coherent interaction have among gold atoms should answer for the disorder of atom arrangement in the ground state configurations (N=38, 55).