Benchmarking density functional theory methods for modelling cationic metal-argon complexes

Noble gas chemistry is fascinating because noble gases can make formal chemical bonds with metal ions, despite their closed electronic configuration. Argon-metal ion complexes are particularly interesting since their bonding is halfway between dispersion and covalent interactions. Although many metal ion-noble gas complexes have been synthesized, there are still disagreeing theoretical descriptions about their bonding, which is not yet fully understood. Accurate experimental data are important as solid reference for theoretical methodologies, but such data are currently scarce for complexes of a metal ion with noble gas atoms. We measured infrared spectra of MArn+ (n = 3-5; M = Au, Ag, Pd) complexes and used these spectra as benchmark data for different theory levels within the density functional theory formalism. Several basis sets, exchange-correlation functionals, and the inclusion of dispersion corrections were considered. The agreement between the measured spectra and the calculations strongly depends on the applied level of theory. Functionals of a higher level of complexity do not consistently provide a better agreement with the experiment; this is particularly the case for the B3LYP hybrid functional that performs worse than the PBE GGA functional. On the other hand, the inclusion of dispersion corrections and the use of a large basis sets are crucial for a good description of the interaction between M+ and argon atoms.