Bulk screening in core-level photoemission from Mott-Hubbard and charge-transfer systems

Abstract

We report bulk-sensitive hard X-ray (hν = 5.95 keV) core-level photoemission spectroscopy (PES) of single crystal V1.98Cr0.02O3 and the high-Tc cuprate Bi2Sr2CaCu2O 8+δ (Bi2212). V1.98Cr0.02O3 exhibits low binding energy "satellites" to the V 2p "main lines" in the metallic phase, which are suppressed in the antiferromagnetic insulator phase. In contrast, the Cu 2p spectra of Bi2212 do not show temperature dependent features, but a comparison with soft X-ray PES indicates a large increase in the 2p 5 3d 9 "satellites" or 3d 9 weight in the bulk. Cluster model calculations, including full multiplet structure and a screening channel derived from the coherent band at the Fermi energy, give very satisfactory agreement with experiments.PACS numbers: 71.30.+h, 74.72.Hs, 78.20.Bh, Core-level photoemission spectroscopy (PES) has played a very important role in our understanding of the electronic structure of correlated transition metal (TM) and rare-earth compounds.[1] The appearance of strong satellite structure accompanying the main peaks in correlated systems is well known and systematic variations in the position and intensities of these satellites provide us important clues to their electronic structure.[2, 3, 4] The inter-atomic configuration-interaction approach, using a cluster model or Anderson impurity model , gives a quantitative interpretation for satellite intensities and positions, leading to an accurate description of the ground state and excitation spectrum. [2,3,4] In this approach, the physics of TM compounds can be described in terms of a few parameters, namely, the d-d Coulomb repulsion energy U , the charge-transfer energy ∆, the ligand p-TM d hybridization energy V , and the core-hole-d electron Coulomb attraction energy U dc . Zaanen, Sawatzky and Allen[5] proposed a classification scheme for TM compounds which soon evolved into a paradigm. In this scheme, the band gaps of late TM compounds are socalled charge-transfer (CT) type with U > ∆. NiO and CuO are typical examples of CT insulators while the high-T c cuprates are CT insulators driven metallic by doping. In contrast, the early TM compounds, with U < ∆ are Mott-Hubbard (MH) systems. V 2 O 3 , with its alloys, plays the role of a classic MH system displaying a correlation induced metal-insulator transition. [6,7,8] While the old picture of the MH metal-insulator transition involved a complete collapse or a coalescence of the lower and upper MH bands into a single band in the metal phase, photoemission studies showed the formation of a well-defined coherent band at the Fermi level in the presence of remnant MH bands for a series of correlated oxides [9] and very recently, also for V 2 O 3 .[10] The experimental results are in excellent agreement with calculations using dynamic mean-field theory (DMFT). [10,11] In spite of these successes of PES, the surface sensitivity of PES has often led to controversies regarding surface versus bulk electronic structure, and hence, hard Xray (HX)-PES is very important and promising. [12,13] With the develop...

Nature of electron correlation and hybridization in NixCu1−xMnSb Heusler alloys

“…7,23 Thus, the C1 and C2 satellites may be related to charge transfer between Mn 3d and metallic conduction bands at the Fermi level. Similar temperature dependent satellite structures have also been observed in Mn 2p spectra of manganites 14 as well as 2p spectra of other TMs 24 and ruthenates 25 and have been attributed to charge transfer between valence 3d states and the conduction band. Hence, the C1 and C2 satellite intensities can be viewed as a clear signature of strength of hybridization between Mn 3d and the conduction band.…”

“…As a result there will be enhanced core-hole repulsion between the 2p 5 core and 3d valence states, and consequently the CT peak will appear on the lower binding energy side, as is the case for C1. 24 On the other hand, when electrons are transferred from the 3d band, the number of unpaired electron reduces and the CT peak will appear on the higher binding energy side, as observed for C2. The broader spectral width of C2 compared to C1 may be attributed to additional charge transfer from Mn 3d to neighboring atoms.…”

“…Appearance of C1 can be attributed to electron transfer from conduction to 3d band, while C2 to electron transfer from 3d to conduction band. 14,24 This can be understood as follows : the photoemission of a Mn 2p electron leads to a 2p 5 final state with a core-hole. Now for a charge transfer (CT) peak for which there is an electron transfer into 3d states, there are more unpaired 3d electrons.…”

“…25 This is because charge transfer leads to change in electron binding energy according to the strength of electron-electron and electron-hole coulomb potentials. 24,25 On the other hand these potentials determine the strength of electron correlation. Hence, a larger relative energy shift of the CT satellite indicates stronger electron correlation.…”

Coherent and Incoherent Excitations of Electron-DopedSrTiO3

“…In fact, there is a recent trend to understand the valence-band spectra in terms of screening orbitals [35,36,37], a concept originally developed to understand the core-level spectra [38]. The coherent state, on the other hand, is understood as a final state which is non-locally screened by a coherent band having mainly-d character [37,39]. These assignments are similar to those adopted for LaTiO 3 [36] and (Ca,Sr)VO 3 [37], although STO:Nb is a non-integer filling system.…”