X-ray Raman spectroscopy on iridate perovskites

In this thesis a X-ray Raman scattering study of the electronic structure of the first two compounds of the Ruddlesden-Popper series Sr_(n+1)Ir_(n)O_(3n+1) (n = 1, 2) of iridates is presented. The measurements were performed at beam line ID20 at the European Synchrotron Radiation Facility, Grenoble. 5d transition metal oxides, iridates in particular, have recently been intensively explored as they display new fascinating phenomena, arising from the strong spin-orbit coupling to which they are subjected. Indeed, a simple Hubbard model, applied with great success to 3d transition metal oxides, would predict a metallic state for these 5d compounds, in view of the larger bandwidth and smaller electron correlation in the 5d orbitals; instead, some iridates, among which the samples studied, are insulators. The opening of a gap is due to the strong spin-orbit coupling which enhances the effect of correlation, narrows the effective bandwidth and isolates the so-called J_(eff) = 1/2 ground state. This peculiar ground state is strictly achieved only if the energies at play, most especially the cubic and tetragonal components of the crystal field splitting and the spin-orbit coupling, follow a precise hierarchy. The aim of this work is to determine the cubic crystal field splitting of the Ir 5d states in Sr2IrO4 and Sr3Ir2O7 by X-ray Raman scattering, a bulk sensitive and self-absorption free probe. Indeed, spin-orbit coupling strength and tetragonal crystal field splitting have already been experimentally determined by other authors. By focusing our attention on the O K edge and exploiting the orientation dependence of the spectra, we were able to assign features in the 528-535 eV energy loss range to specific transitions involving the Ir 5d orbitals. This has allowed us to extract values for the cubic crystal field splitting: 3.8 ± 0.82 eV in Sr2IrO4 and 3.55 ± 0.13 eV in Sr3Ir2O7. Furthermore, we found values for the tetragonal crystal field splitting acting on the e_g states: 1.6 ± 0.82 eV in Sr2IrO4 and 1.9 ± 0.13 in Sr3Ir2O7. This work is the first direct experimental determination of the cubic crystal field splitting in the two iridates. A complete electronic structure of the two compounds is finally achieved and the implicit theoretical assumptions, which are essential for the establishment of the J_(eff) = 1/2 ground state, are confirmed. Furthermore, this is one of the first X-ray Raman scattering studies at the O K edge of transition metal oxides in general: we demonstrate that this spectroscopic technique can be used to obtain a detailed picture of the electronic transitions in these materials and, more generally, our work paves the way for similar detailed studies of correlated electron systems such as the high temperature cuprate superconductors or nichelates. In particular, X-ray Raman spectroscopy is suitable for measurements in extreme environments, such as high pressure, and therefore it is a valid substitute of other techniques (e. g. soft X-ray absorption spectroscopy) which cannot be accomplished in such conditions.