Copper on Ni(111): The electron states from submonolayer to several-monolayer coverages

Experimental and theoretical results are presented for Cu on an Ni(111) surface. Angle-integrated ultraviolet photoelectron spectroscopy (UPS) results ($h\nu =21.2\mathrm{}$ eV) are given for Ni(111) with increasing Cu coverages $\theta$ from 0.1 to 4 monolayers. The results at $\theta \ll 1$ refer to isolated Cu adatoms and those at $1<\mathrm{}\theta <4\mathrm{}$ to an ordered overlayer. The position and line shape of the Cu $3d$ peak are the same for $0.1<\mathrm{}\theta <1\mathrm{}$: This indicates that the main features of the Cu $3d$ band are determined by the interaction of Cu atoms with the substrate atoms. Evidence for such interaction is provided by the modifications of the substrate photoemission, which take place upon Cu deposition. At $\theta \simeq 4$ the line shape is more similar to that of bulk Cu. These results appear to be typical of the interaction of Cu atoms with a $d$-metal substrate: Comparison with those of a Cu monolayer deposited on an $\mathrm{sp}$ metal [Cu on Zn(0001)] shows that the Cu $3d$ peak is found at higher binding energy with respect to the Fermi energy. The monolayer case is discussed in detail in connection with theoretical results of a tight-binding calculation. The local density of states (LDS) calculated for the Cu overlayer, the Ni underlayers, and for the clean Ni(111) surface shows clearly the importance of the $d-d$ interaction between Cu and Ni atoms. The theory accounts for the Cu $3d$-peak position and the modification of the substrate density of states. Also, the measured Cu $3d$ line shape can be understood on the basis of the theory. The experimental and theoretical results are also discussed in connection with other available information on related topics.