Aluminum clusters graphene supported a DFT-based genetic algorithm study

Abstract

Genetic Algorithm (GA) in conjunction with Density Functional Theory (DFT) was implemented to optimize the geometric configurations of aluminum clusters (Al_n, n = 3–40). GA efficiently explored the configurational space by evolving randomly generated initial structures through selection, crossover, and mutation, yielding energetically metal clusters. Subsequently, DFT was employed to relax these structures, ensuring the precise determination of the most stable geometries. The results showed that the binding energy per aluminum atom increased progressively with cluster size, approaching values close to that of bulk aluminum for large clusters. Notably, clusters with fewer atoms (Al_n, n = 3–9) displayed diverse and irregular geometries, while larger clusters (Al_n, n = 10–40) exhibited defined and stable configurations. Additionally, the obtained aluminum clusters were supported on defected graphene surfaces, undergoing structural rearrangements while maintaining their energetic stability. These results highlight the influence of surface interactions on the minimum energy configurations. This study demonstrates the robust capabilities of the combined GA-DFT approach for predicting structural and energetic properties of metal clusters, offering valuable insights for applications in catalysis and materials science.