Challenges of modelling real nanoparticles: Ni@Pt electrocatalysts for the oxygen reduction reaction.

Theoretical/computational methods have been extensively applied to screen possible nano-structures attempting to maximize catalytic and stability properties for applications in electrochemical devices. This work shows that the method used to model core@shell structures is of fundamental importance in order to truly represent the physicochemical changes arising from the formation of a core-shell structure. We demonstrate that using a slab approach for modelling nanoparticles the oxygen adsorption energies are qualitatively well represented. Although this is a good descriptor for the catalytic activity, huge differences are found for the calculated surface stability between the results of a nano-cluster and those of a slab approach. Moreover, for the slab method depending on the geometric properties of the core and their similarity to the elements of the core or shell, contradictory effects are obtained. In order to determine the changes occurring as the number of layers and nano particle size are increased, clusters of Ni@Pt from 13 to 260 atoms were constructed and analyzed in terms of geometric parameters, oxygen adsorption, and dissolution potential shift. It is shown that the results of modelling the Ni@Pt nanoparticles with a cluster approach are in good agreement with experimental geometrical parameters, catalytic activity, and stability of a carefully prepared series of Ni@Pt nanostructures where the shell thickness is systematically changed. The maximum catalytic activity and stability are found for a monolayer of Pt whereas adding a second and third layer the behavior is almost the same than that in pure Pt nanoparticles.