In a recent Letter [Phys. Rev. Lett. 107, 187203 (2011)], Fujii et al. reported Mn 2p photoelectron emission spectra for (Ga,Mn)As recorded using hard x-rays. Due to the enhanced bulk sensitivity, hard-x-ray spectra reveal an extra low-binding-energy peak, which is absent in surface-sensitive spectra recorded using soft x-rays. Based on Anderson-impurity-model calculations, Fujii et al. assigned the low-binding-energy peak to a cd6L2 final state, and related the variations in its intensity to variations in the As 4p-Mn 3d hybridization strength V. We show here that the definition of the charge-transfer energy considered by Fujii et al. is different from that considered in the Zaanen-Sawatzky-Allen diagram. We note that the Anderson impurity model is insufficient to describe low-binding-energy peaks in hard-x-ray core-level photoemission for transition-metal compounds on the verge of a metal-insulator transition. We propose a more plausible origin for the (Ga,Mn)As low-binding-energy peak, related to the nature of its metal-insulator transition.
Download