Photoemission and x-ray absorption spectroscopy studies on cubic pyrochlore ruthenatesBi2−xYx

Abstract

Electronic structures of ternary pyrochlore ruthenates Bi 2Ϫx Y x Ru 2 O 7 (xϭ0.0, 0.4, 1.0, 1.6, and 2.0͒ were investigated by using photoemission spectroscopy ͑PES͒ and x-ray absorption spectroscopy. The oxygen 2p band and the unoccupied conduction bands can be explained by the band calculation results, while the obtained Ru 4d partial PES spectral weight turns out to clearly show a departure from the band-structure calculation results, especially for high Y concentrations. These facts indicate that the metal-insulator transition around the composition xϳ1.2 is driven by the electron correlation effect among the occupied Ru 4d states.

Electronic structure of epitaxial SrRuO3 films studied by resonant photoemission

“…3). For both spectra, parameters coincide within the experimental error, and our obtained value of the resonance energy E res ¼53 71 eV agrees with the published data (Park et al, 2004a(Park et al, , 2004bFujioka et al, 1997). The relatively large linewidth (wE9-10 eV) of resonance is consistent with the expectation that the 4d orbital is quite extended and bonding with the O 2p orbitals is strong (Park et al, 2004b).…”

“…However, the previous attempt (Fujioka et al, 1997) in this direction was not quite successful partly because the resonance is not very strong compared with the cross-section (CS) variation of the O 2p level. Another method based on the Cooper minimum (CM) phenomenon of Ru 4d photoionization CS (Fano and Cooper, 1968) was used to deduce the Ru PSW in SrRuO 3 (Park et al, 2004b) and Bi 2 À x Y x Ru 2 O 7 (Park et al, 2004a). In spite of the fact that the main features of so obtained PSWs agree with the theoretical band-structure computations, the apparent disagreement between the PSWs and calculations calls for an explanation.…”

“…As it was discussed in the introduction, this disagreement is not surprising; however, it shows that the assumption about the mainly O 2p character of the VB spectrum at CM (Park et al, 2004a(Park et al, , 2004bFujioka et al, 1997) should be treated with caution. From the other hand, the similarity in the energy dependence of the contamination peak F and SrRuO 3 VB peaks with dominating O 2p states (C-E) confirms that the first one can be successfully used for the essential flattening of O 2p spectral weight (Park et al, 2004a(Park et al, , 2004b, at least at hn Z70 eV.…”

“…In principle, resonant photoelectron spectroscopy may be utilized to deduce a partial spectral weight (PSW) of elements in the valence band (VB) by making use of the Ru 4p-4d resonance, which was reported to occur around 52-54 eV (Park et al, 2004a(Park et al, , 2004bFujioka et al, 1997). However, the previous attempt (Fujioka et al, 1997) in this direction was not quite successful partly because the resonance is not very strong compared with the cross-section (CS) variation of the O 2p level.…”

“…A possible way to do it is to utilize the similarity in the photon energy dependence of the CSs for 2p states in oxygen and (O 2p-C 2s) orbitals in CO (Wilhelmy et al, 1994;Plummer et al, 1977;Yeh and Lindau, 1985). Such normalization to the intensity of adventitious contamination peaks around $ 10 eV binding energy (BE) from carbon monoxide in principle should provide almost flat O 2p spectral weight in this region (Park et al, 2004a(Park et al, , 2004b. However, the possible enhancement of O 2p CS due to the hybridization with Ru 4d orbitals in ruthenium compounds may destroy this balance, resulting in considerable energy dependence of normalized O 2p PSWs.…”