Unusual electron-doping effects inSr2−xLaxFeMoO6

Abstract

We have investigated the electronic structure of electron-doped Sr2−xLaxFeMoO6 (x=0.0 and 0.2) by photoemission spectroscopy and band-structure calculations within the local-density approximation+U (LDA+U ) scheme. A characteristic double-peak feature near the Fermi level (EF) has been observed in the valence-band photoemission spectra of both x=0.0 and 0.2 samples. A photon-energy dependence of the spectra in the Mo 4d Cooper minimum region compared with the band-structure calculations has shown that the first peak crossing EF consists of the (Fe+Mo) t 2g↓ states (feature A) and the second peak well below EF is dominated by the Fe e g↑ states (feature B). Upon La substitution, the feature A moves away from EF by ∼50 meV which is smaller than the prediction of our band theory, 112 meV. In addition, an intensity enhancement of both A and B has been observed, although B is not crossing EF. Those two facts are apparently incompatible with the simple rigid-band shift due to electron doping. We point out that such phenomena can be understood in terms of the strong Hund's rule energy stabilization in the 3d 5 configuration at the Fe sites in this compound. From an observed band-narrowing, we have also deduced a mass enhancement of ∼2.5 with respect to the band theory, in good agreement with a specific heat measurement.

La-induced changes in the magnetic and electronic properties ofSr2−xLaxFeMoO6

“…The small decrease in S D0 observed for x = 0.5 is consistent with electron doping and an increasing Fermi energy based on band structure calculations and assuming a rigid band shift. 11 In contrast, the drag component is a very sensitive function of the balance between normal and umklapp scattering events, and that balance is strongly influenced by changes to the Fermi surface. 22,26 The zero-field resistivity has been measured for all the samples and shows a small increase in the resistivity with decreasing temperature.…”

“…This structure would need to remain that close even as the doping is changed by the introduction of as much as 0.5 La per formula unit, which appears to be highly unrealistic and not supported by band structure calculations. 11 In fact, band structure calculations show an increasing density of states for up to ϳ300 meV below and above the Fermi energy in the pure compound. Thus, assuming that the entire temperature dependence of S͑T͒ is due only to the proximity to a sharp peak in the DOS is untenable if it is to model the persistence of the systematic pattern over the large doping range in the present studies.…”

“…The situation is possibly more complicated, as we have mentioned earlier. For example, it has been suggested that Fe should remain in the 3+ state 11 and there is no welldefined Mo moment of 1 B per formula unit, since for the pure compound the electron in the originally down-spin Mo͑t 2g ͒ orbital is itinerant. 10 We find that the decrease in the high-field magnetization with increasing La doping can be primarily attributed to ASD, which is apparent in the Fig.…”

“…11 However, Mössbauer data have been interpreted in terms of a mixed Fe 2+ /Fe 3+ state in the pure compound. 12-14 A similar conclusion was reached from x-ray magnetic circular dichroism measurements.…”

Magnetism inSr2CrMoO6: A combinedab initioand model study

“…Many of these compounds show half-metallic behavior, which is of importance for spintronics and related technological applications. [7][8][9][10][11][12][13][14] The high magnetic transition temperatures (T c ) exhibited by compounds like Sr 2 FeMoO 6 , open up the possibility of room temperature application.…”