Investigations of the electronic, magnetic and crystalline structure of perovskite oxides and an oxide-oxide interface

The mineral perovskite CaTiO₃ lends its name to the class of compounds with composition ABX₃, which have the same type of crystal structure known as the perovskite structure. Here, A and B are cations while X is typically a halogen or oxygen anion. The bigger cation A and the X anions form a cubic close packing AX₃ with the smaller B cation occupying one quarter of the octahedral sites. The underlying work deals with three classes of perovskite oxides, the ''titanates'', the ''cuprates'' and the ''manganites'', where the central B cations are Ti, Cu and Mn respectively, each class with very characteristic properties. Strontium titanate, STO, like so many oxides is an insulator, yet paraelectric and diamagnetic. It gained special interest as a commercially available, high quality substrate for the growth of hightemperature superconductors and other oxide thin films. The huge class of perovskite cuprates is most famous for members like YBCO, the well-known high-temperature superconductor. Finally, the doped, mixed-valent perovskite manganites LXMO (here X = Sr, Ca, Ce) have to be mentioned. They appeared on the screen of experimental and theoretical physicists in the 1950s when ferromagnetism and magnetoresistance were discovered in these compounds, leading to the theory of double exchange. In the mid-1990s they further increased their importance, following the description of the Jahn-Teller polaron and the discovery of the colossal magnetoresistance phenomenom. They may be derived from an insulator, LaMnO₃, but the doping allows the control of electric and magnetic properties over a wide range, i.e. from insulating behavior to metallic conductivity or from diamagnet to ferromagnet. Further notable effects are found when electric or magnetic fields are applied, including spin polarization and the related effect of colossal magnetoresistance. These substances may be approximated as consisting of manganese atoms and their surrounding oxygen octahedra while the rather passive lanthanum (or strontium in STO) mainly donates it's three electrons to MnO₆ (or TiO₆). These corner-shared units are not only the building blocks of a tightly bound 3D network; they are also responsible for the transport and magnetic properties. It is thus easy to see, how perturbations like doping, strain, cation size mismatch, anion vacancies or the Jahn-Teller effect can affect the properties of these samples. Another way to look at the perovskites is to think of a stack of alternating layers. Both concepts are used in this work, whenever they fit. In SrTiO₃ for example, charge neutral [Sr²⁺O^2-] and [Ti⁴⁺O₂^2-] layers follow upon each other when viewed along the [001] direction. In this picture LaMnO₃ consists of an array of positively charged [La³⁺O^2-] and negatively charged [Mn³⁺O₂^2-] layers. When these two materials meet at an interface (even with vacuum), a polar discontinuity is created which must lead to some form of structural or electronic reconstruction. For an interface between for instance LaAlO₃ and SrTiO₃ it is important to know which layers meet there, since the SrO-AlO₂ interface is insulating while the TiO₂-LaO interface is conducting. The first set of experiments (chapter 5) described in the underlying work deals with the termination of STO and the changes found on the surface for different preparation conditions. The knowledge and control of the exact termination of the substrate is of importance for the overlayers. Atomic control of the arrangement at the surface is the key for obtaining high-quality overlayers with the desired properties. The second and third part of the experiments was performed on doped manganites, overlayers on STO. The electronic, magnetic and crystal structure is in detail examined for the Cerium doped lanthanum manganite LCeMO in chapter 6. With this knowledge in mind, the electronic and crystalline structures of Strontium doped LSMO, Calcium doped LCMO and undoped LMO are discussed in chapter 7. The focus lies on a multi-technique approach to correlate changes of one parameter to its effects on others. The characterization of a YBCO / LCMO heterostructure on STO concludes this work (chapter 8). Again it is the electronic, magnetic and crystalline structure at the interface that proves crucial for the properties of this unusual system which artificially combines a superconductor (SC) and a ferromagnet (FM). This structure makes two antagonistic effects meet, the tendency of a superconductor to pair two electrons with different spin into Cooper pairs and the tendency of a ferromagnet to align all spin moments. This fascinating competition shows the wide range of properties that the perovskite oxides offer when they are used accordingly and combined usefully. The above mentioned polar discontinuity at the interface between a manganite and a cuprate is examined. The thesis is completed by a summary and an outlook.