CONICET | Buscador de Institutos y Recursos Humanos

Sr2FeMoO6 is the double perovskite prototype of a family of compounds with potential important magnetic applications due to a significant decrease in its resistivity under a magnetic field. It undergoes a lattice phase transition at ~410 K becoming a ferrimagnetic metallic oxide at lower temperatures. In addition, ideal Fe, Mo octahedral ordering yields a half-metallic spin polarized, a half-metal character, due to a minority down-spin finite density of states at the Fermi level. The down-spin conduction band crossing EF is dominated by Fe3d-Mo4d t2g states while the up-spin band below EF is mostly due to the Fe 3d eg. This spin order depends decisively of octahedral order in the lattice. Thus, the half-metallic character is destroyed by lattice errors (antisite positions) in which Fe octahedra are replaced by the Mo counterpart. In this communication we comment on temperature dependent EXAFS measurements of the Fe-K (7112 eV), Mo-K (19999 eV) and W-LIII (10297 eV) edges of Sr2FeMo1-xWxO6 (0£ x £1) in the 28 K to 600 K range. We consider the Fe-Mo pure end compound with different degrees of antisite disorder. We also discuss on the nature of patches reducing carrier mobility and compare those measurements with earlier X-ray and neutron diffraction patterns [1] and far infrared spectra [2] reported by our group for the same samples. We find that the K-Fe broadens, relative to the those measured in FeO, Fe2O3 and Fe3O4, in agreement with a mixed valence state as for Fe 3+-Fe2+ for which we have already found a small polaron transport scenario in infrared transmission and reflectivity. The remarkable correspondence in the profiles K-Fe for x~0.5 suggests that for those replacements the Fe local structure remains the same though out the composition with the carrier path increasingly truncated as we increase the amount of W. Only for x>0.7 there is a significant change in the Fe edge profile indicating a majority of Fe2+ as consequence of W replacement. On the other hand W- LIII edge remain mostly unaltered as a function of concentration and temperature showing that a smooth valence change through out the composition, resulting from the replacement of Mo by W in the mixed compound, is not in agreement with our results. Rather, our data suggest that at least for x=0.8 an arrange in a cluster like formation in which W particularly remains with the insulating pure end valence state of W 6+. Our overall aim is to gain an understanding using long and short, local, range probes of spatially inhomogeneous systems with established complexity where giant responses might be consequence of small perturbations.[3] Our results may then relate to those found in other transition metal oxides where quenched disorder provide ground for complex behaviors