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

DSpace Repositorium (Manakin basiert)


Dateien:

Zitierfähiger Link (URI): http://nbn-resolving.de/urn:nbn:de:bsz:21-opus-68128
http://hdl.handle.net/10900/49873
Dokumentart: Dissertation
Erscheinungsdatum: 2012
Sprache: Englisch
Fakultät: 7 Mathematisch-Naturwissenschaftliche Fakultät
Fachbereich: Chemie
Gutachter: Chassé, Thomas (Prof. Dr.)
Tag der mündl. Prüfung: 2013-02-14
DDC-Klassifikation: 540 - Chemie
Schlagworte: Röntgen-Photoelektronenspektroskopie , Magnetischer Röntgenzirkulardichroismus , Perowskit
Freie Schlagwörter: Grenzfläche , Oberfläche , Manganate
Interface , Surface , Manganites
Lizenz: http://tobias-lib.uni-tuebingen.de/doku/lic_mit_pod.php?la=de http://tobias-lib.uni-tuebingen.de/doku/lic_mit_pod.php?la=en
Gedruckte Kopie bestellen: Print-on-Demand
Zur Langanzeige

Inhaltszusammenfassung:

Die vorliegende Arbeit beschäftigt sich mit der Untersuchung der elektronischen, magnetischen und kristallinen Struktur von Perowskit-Oxiden, Strontiumtitanat STO, dotierte Lanthanmanganate LXMO (X = Ca, Sr, Ce) und Yttriumbariumcuprat YBCO. Aufgrund der elektronischen Korrelation in diesen Materialien wirken sich Einflüsse wie Verspannungen und Schichtdicke, Größe und Ladung von Dotieratomen oder der Sauerstoffgehalt von dünnen Filmen stark auf die physikalischen Eigenschaften aus. Diese Struktur-Wirkungsbeziehungen wurden mit Hilfe von röntgenbasierten Methoden wie Röntgenabsorptionsspektroskopie (x-ray absorption spectroscopy, XAS), Röntgenphotoelektronenspektroskopie (x-ray photoelectron spectroscopy, XPS) und Röntgenphotoelektronenbeugung (x-ray photoelectron diffraction, XPD) untersucht. Die Untersuchung der Oberflächenterminierung von STO durch XPD in Kapitel 5 eröffnet die Arbeit. Es konnte durch die Kombination von Experiment und Theorie gezeigt werden, dass, in Abhängigkeit von den Präparationsbedingungen, sowohl SrO- als auch TiO2-terminiertes STO hergestellt werden kann. Die außerordentlich gute Übereinstimmung zwischen gemessenen und simulierten Daten erlaubt die Unterscheidung mit Hilfe von XPD und lässt darauf schließen, dass chemisch geätzte Proben nicht zwangsläufig vollständig TiO2-terminiert sind. Die besetzte (XPS) und unbesetzte (XAS) Zustandsdichte von LCeMO wird in Kapitel 5 ebenso untersucht wie die magnetische Struktur und mit Transporteigenschaften verknüpft. Nur durch diesen multidisziplinären Ansatz lassen sich die physikalischen Eigenschaften mit der Art der Dotierung korrelieren. Es ergibt sich ein konsistentes Bild dieses formal elektronendotierten Materials, in dem sich die gewünschten Valenzen, Mn2+, Mn3+ und Ce4+, und die gewünschten Eigenschaften, CMR-Effekt und Metall-Isolator-Übergang, nicht in Einklang bringen lassen. Die formal korrekten Valenzen und eine Elektronendotierung führen zu einem rein halbleitenden Verhalten der Probe, während die erwünschten Eigenschaften nur in Proben gefunden werden die löcherdotiert sind. Der Widerspruch zwischen der Beobachtung von tetravalentem Cer und Löcherdotierung lässt sich durch Selbstdotierung erklären: Dabei sorgen Kationenvakanzen durch CeO2-Ausscheidungen zu einer gemischten Mn3+, Mn4+ Valenz trotz der Detektion von Ce4+ (und Ce3+). Diese Kationenvakanzen werden auch in Kapitel 7 noch einmal thematisiert, in dem die XPD-Daten von undotiertem Lanthanmanganat LMO mit dotierten Manganaten LXMO verglichen werden. Es zeigen sich, neben starken Hinweisen für eine MnO2- Terminierung aller untersuchten Proben, nur geringe Unterschiede in den Beugungsdaten von Mangan und Sauerstoff der verschiedenen Materialien. Das Grundgerüst der Manganate, eckenverknüpfte MnO6-Oktaeder, scheint gegen chemischen Druck relativ stabil zu sein. Auf der anderen Seite werden Unterschiede in den Beugungsmustern der Dotieratome zwischen diesen Oktaedern gefunden. Man findet eine kubische Umgebung für Ca und Sr in LCMO und LSMO während Cer in LCeMO ein diffuses Beugungsbild ohne Struktur liefert. Zusammen mit den Ergebnissen aus Kapitel 5 erhärtet sich damit der Verdacht, dass sich das Dotieratom Ce nicht ausschließlich zwischen den Oktaedern befindet sondern Kationenvakanzen durch Phasenseparation bildet. Das achte Kapitel präsentiert Ergebnisse von Doppellagen bestehend aus dem Ferromagneten LCMO und dem Hochtemperatursupraleiter YBCO in direktem Kontakt. Erneut wird die elektronische und magnetische Struktur durch die Kombination von Transporteigenschaften und spektroskopischen Ergebnissen studiert. Während die magnetischen Eigenschaften dieser Grenzfläche zwischen den beiden widerstreitenden Phänomenen, Ferromagnetismus und Cooper-Paar-Bildung, im Einklang mit anderen Veröffentlichungen ist, ergeben sich Diskrepanzen bei der orbitalen Besetzung zu früheren Ergebnissen. Eine alternative Erklärung wird in Form einer Entkopplung zwischen Magnetismus und Supraleitung durch eine isolierende Schicht nahe der Grenzfläche gegeben. Die bisher diskutierten Mechanismen konnten nicht reproduziert werden, da die atomare Struktur der Grenzflächen nicht direkt vergleichbar ist. Es zeigt sich, dass auch hier der Einfluss der kristallinen Struktur in Zukunft genauer untersucht werden muss. Zusammenfassend lässt sich sagen, dass durch multidisziplinäre Ansätze in jedem experimentellen Kapitel der vorliegenden Arbeit interessante Erkenntnisse über die Struktur-Wirkungsbeziehungen von elektronischer, magnetischer und kristalliner Struktur in den Perowskit-Oxiden erzielt werden konnten. Es zeigt sich, dass die komplexe Natur der hier untersuchten Materialien sowohl viele interessante Fragen aufwirft als auch spannende Möglichkeiten bietet. Der Weg zu ihrem Verständnis ist geebnet, der Weg bis zur Kontrolle der korrelierten Elektronen ist noch weit.

Abstract:

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 [Sr(2+)O(2-)] and [Ti(4+)O2(2-)] layers follow upon each other when viewed along the [001] direction. In this picture LaMnO3 consists of an array of positively charged [La(3+)O(2-)] and negatively charged [Mn(3+)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 (chapter5) 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.

Das Dokument erscheint in: