Comment on "Fano Resonance for Anderson Impurity Systems"

Abstract

In a recent Letter, Luo et al. (Phys. Rev. Lett. 92, 256602 (2004)) analyze the Fano line shapes obtained from scanning tunneling spectroscopy (STS) of transition metal impurities on a simple metal surface, in particular of the Ti/Au(111) and Ti/Ag(100) systems. As the key point of their analysis, they claim that there is not only a Fano interference effect between the impurity d-orbital and the conduction electron continuum, as derived in Ujsaghy et al. (Phys. Rev. Lett. 85, 2557 (2000)), but that the Kondo resonance in the d-electron spectral density has by itself a second Fano line shape, leading to the experimentally observed spectra. In the present note we point out that this analysis is conceptually incorrect. Therefore, the quantitative agreement of the fitted theoretical spectra with the experimental results is meaningless.