Physico-Chemical Insights into Gas-Phase and Oxide-Supported Sub-Nanometre AuCu Clusters

Abstract

Catalysis by AuCu nanoclusters is a promising scientific field. However, our fundamental understanding of the underlying mechanisms of mixing in AuCu clusters at the sub-nanometre scale and their physico-chemical properties in both the gas-phase and on oxide supports is limited. We have identified the global minima of gas-phase and MgO(100)-supported AuCu clusters with 3–10 atoms using the Mexican Enhanced Genetic Algorithm coupled with density functional theory. Au and Cu adatoms and supported dimers have been also simulated at the same level of theory. The most stable composition, as calculated from mixing and binding energies, is obtained when the Cu proportion is close to 50%. The structures of the most stable free AuCu clusters exhibit Cu-core/Au-shell segregation. On the MgO surface however, there is a preference for Cu atoms to lie at the cluster-substrate interface. Due to the interplay between the number of interfacial Cu atoms and surface-induced cluster rearrangement, on the MgO surface 3D structures become more stable than 2D structures. The O-site of MgO surface is found to be the most favourable adsorption site for both metals. All dimers favour vertical (V) configurations on the surface and their adsorption energies are in the order: AuCu < CuCu < AuAu < AuCu (where the underlined atom is bound to the O-site). For both adatoms and AuCu dimers, adsorption via Cu is more favourable than Au-adsorbed configurations, but, this disagrees with the ordering for the pure dimers due to a combination of electron transfer and the metal-on-top effect. Binding energy (and second difference) and HOMO-LUMO gap calculations show that even-atom (even-electron) clusters are more stable than the neighbouring odd-atom (odd- electron) clusters, which is expected for closed- and open-shell systems. Supporting AuCu clusters on the MgO(100) surface decreases the charge transfer between Au and Cu atoms calculated in free clusters. The results of this study may serve as a foundation for designing better AuCu catalysts.

Ultralow reaction barriers for CO oxidation in Cu–Au nanoclusters

“…1. Our results reproduced previous calculations for smaller clusters 53–56 or improved on them for Cu 2 Au 8 , Cu 8 Au 2 , and Cu 3 Au 7 clusters. The ground-state structures of Cu n Au m clusters with 11 ≤ n + m ≤ 15 were found for the first time.…”

Carbon vacancy-assisted stabilization of individual Cu5 clusters on graphene. Insights from ab initio molecular dynamics

“…Analyzing the case of the Cu 10 dimer in Table 2, it can be seen that all considered methods consistently predict a D 2dsymmetry structure as the energetically favored configuration, in agreement with previous ab initio work 48 and DFT studies. [80][81][82][83] Panel c of Fig. 1 shows that this Cu 10 geometry is composed of two trigonal bipyramidal 3D-Cu 5 structures rotated by ca.…”

Structure prediction in low dimensions: concepts, issues and examples

“…In recent years, global searches for the optimal structure of small and medium-sized metal clusters placed on insulator substrates have also been performed, using genetic/evolutionary algorithms. Examples are AuCu clusters on oxides [91], PdAu and PdPt on MgO [92,93] or Cu clusters on ZnO [94].…”

“…In recent years, global searches for the optimal structure of small and medium-sized metal clusters placed on insulator substrates have also been performed, using genetic/evolutionary algorithms. Examples are AuCu clusters on oxides [ 91 ], PdAu and PdPt on MgO [ 92 , 93 ] or Cu clusters on ZnO [ 94 ].

Figure 6 Optimal position of n CO molecules on a Cu 6 cluster ( n = 1–6); Cu, O and C are drawn as gold, grey and red spheres, respectively [ 89 ].

…”