GIGA: a versatile genetic algorithm for free and supported clusters and nanoparticles in the presence of ligands

Abstract

We present a versatile parallelised genetic algorithm, which is able to perform global optimisation from first principles for pure and mixed free clusters in the gas phase, supported on surfaces or in the presence of one or several atomic or molecular species (ligands or adsorbates). The genetic algorithm is coupled to different quantum chemical software packages in order to permit a large variety of methods for the global optimisation. The genetic algorithm is also capable of optimising different electronic spin multiplicities explicitly, which allows global optimisation on several potential energy hypersurfaces in parallel. We employ the genetic algorithm to study ligand-passivated clusters [Cd₃Se₃(H₂S)₃]⁺ and to investigate adsorption of [Pt₆(H₂O)₂]⁺ supported on graphene. The explicit consideration of the electronic spin multiplicity during global optimisation is investigated for nanoalloy clusters Pt₄V₂.