We propose a hypothesis and a numerical model for the formation of branching plagioclase textures visible at both macroscopic (∼cm to ∼m) and microscopic scale within melagabbro of the Isle of Rum, Scotland, based on macroscopic, microscopic observations and relevant geological history. The plagioclase crystals are typically linked as twins and form meshes of planar stellate structures (m-scale) with a large range in geometrical organization from patchy to radiating. Evidence of macroscopic crystal aggregation and alignment is attributed to interfacial free energy minimization at the microscopic scale during growth. Accordingly, a binary immiscible Lattice Boltzmann model was developed to simulate diffusion of simplified plagioclase in the melt phase. Isothermal phase transitions modelled via first order chemical reactions are subsequently coupled with stochastic dynamics at the crystal growth front to simulate energy minimization processes including twinning during crystallization in an igneous environment. The solid phase and the liquid phase are coupled with a temporal flexibility that sets the overall ratio between the rate of diffusion and chemical enrichment in the liquid state and the rate of crystallization. The parameter space of the model is explored extensively, followed by a reasonable transcription of physical parameters and an estimation of other parameters to construct realistic simulation scenarios yielding synthetic plagioclase stellates. The results are presented, analyzed and discussed. They appear to be in reasonable qualitative agreement with observations, and several aspects of the natural stellates such as the stellate spacing and long continuous stretches of plagioclase with epitaxial junctions seem to be in reasonable quantitative agreement with observations.

Les matériaux piézoélectriques, comme le titanate-zirconate de plomb Pb(ZrxTi1-x)O3 (PZT), l’oxyde de zinc ZnO, ainsi que la solution solide de Pb(Mg1/3Nb2/3)O3-PbTiO3 (PMN-PT), sont actuellement l’objet d’études de plus en plus nombreuses à cause de leurs applications innovantes dans les systèmes micro-électromécaniques (MEMS). Afin de les intégrer sur substrat de silicium, certaines précautions doivent être prises en compte concernant par exemple des couches tampon, les électrodes inférieures. Dans cette thèse, des films piézoélectriques (PZT et PMN-PT) ont été épitaxiés avec succès sous forme de monocristaux sur silicium et SOI (silicon-on-insulator) par procédé sol-gel. En effet, des études récentes ont montré que les films piézoélectriques monocristallins semblent posséder des propriétés supérieures à celles des films polycristallins, permettant ainsi une augmentation de la performance des dispositifs MEMS. Le premier objectif de cette thèse était de réaliser l'épitaxie de film monocristallin de matériaux piézoélectriques sur silicium. L'utilisation d’une couche tampon d'oxyde de gadolinium (Gd2O3) ou de titanate de strontium (SrTiO3 ou STO) déposés par la technique d’épitaxie par jets moléculaires (EJM) a été explorée en détail pour favoriser l’épitaxie du PZT et PMN-PT sur silicium. Sur le système Gd2O3/Si(111), l’étude par diffraction des rayons X (XRD) de la croissance du film PZT montre que le film est polyphasé avec la présence de la phase parasite pyrochlore non ferroélectrique. Cependant, le film PZT déposé sur le système STO/Si(001) est parfaitement épitaxié sous forme d’un film monocristallin. Afin de mesurer ses propriétés électriques, une couche de ruthenate de strontium conducteur SrRuO3 (SRO) déposée par ablation laser pulsé (PLD) a été utilisée comme l'électrode inférieure à cause de son excellente conductibilité et de sa structure cristalline pérovskite similaire à celle du PZT. Les caractérisations électriques sur des condensateurs Ru/PZT/SRO démontrent de très bonnes propriétés ferroélectriques avec présence de cycles d'hystérésis. Par ailleurs, le matériau relaxeur PMN-PT a aussi été épitaxié sur STO/Si comme l’a confirmé la diffraction des rayons X ainsi que la microscopie électronique en transmission (TEM). Ce film monocristallin est de la phase de perovskite sans présence de pyrochlore. En outre, une étude en transmission du rayonnement infrarouge au synchrotron a prouvé une transition de phase diffuse sur une large gamme de température, comme attendue dans le cas d’un relaxeur. L'autre intérêt d'avoir des films PZT monocristallins déposés sur silicium et SOI est de pouvoir utiliser les méthodes de structuration du silicium bien standardisées maintenant pour fabriquer les dispositifs MEMS. La mise au point d’un procédé de micro-structuration en salle blanche a permis de réaliser des cantilevers et des membranes afin de caractériser mécaniquement les couches piézoélectriques. Des déplacements par l'application d'une tension électrique ont ainsi pu être détectés par interférométrie. Finalement, cette caractérisation par interférométrie a été combinée avec une modélisation basée sur la méthode des éléments finis. Dans le futur, il sera nécessaire d’optimiser le procédé de microfabrication du dispositif MEMS afin d’en améliorer les performances électromécaniques. Enfin, des caractérisations au niveau du dispositif MEMS lui-même devront être développées en vue de leur utilisation dans de futures applications. / Recently, piezoelectric materials, like lead titanate zirconate Pb(ZrxTi1-x)O3 (PZT), zinc oxide ZnO, and the solid solution Pb(Mg1/3Nb2/3)O3-PbTiO3 (PMN-PT), increasingly receive intensive studies because of their innovative applications in the microelectromechanical systems (MEMS). In order to integrate them on silicon substrate, several preliminaries must be taken into considerations, e.g. buffer layer, bottom electrode. In this thesis, piezoelectric films (PZT and PMN-PT) have been successfully epitaxially grown on silicon and SOI (silicon-on-insulator) in the form of single crystal by sol-gel process. In fact, recent studies show that single crystalline films seem to possess the superior properties than that of polycrystalline films, leading to an increase of the performance of MEMS devices. The first objective of this thesis was to realize the epitaxial growth of single crystalline film of piezoelectric materials on silicon. The use of a buffer layer of gadolinium oxide(Gd2O3) or strontium titanate (SrTiO3 or STO) deposited by molecular beam epitaxy (MBE) has been studied in detail to integrate epitaxial PZT and PMN-PT films on silicon. For Gd2O3/Si(111) system, the study of X-ray diffraction (XRD) on the growth of PZT film shows that the film is polycrystalline with coexistence of the nonferroelectric parasite phase, i.e. pyrochlore phase. On the other hand, the PZT film deposited on STO/Si(001) substrate is successfully epitaxially grown in the form of single crystalline film. In order to measure the electrical properties, a layer of strontium ruthenate (SrRuO3 or SRO) deposited by pulsed laser deposition (PLD) has been employed for bottom electrode due to its excellent conductivity and perovskite crystalline structure similar to that of PZT. The electrical characterization on Ru/PZT/SRO capacitors demonstrates good ferroelectric properties with the presence of hysteresis loop. Besides, the relaxor ferroelectric PMN-PT has been also epitaxially grown on STO/Si and confirmed by XRD and transmission electrical microscopy (TEM). This single crystalline film has the perovskite phase without the appearance of pyrochlore. Moreover, the study of infrared transmission using synchrotron radiation has proven a diffused phase transition over a large range of temperature, indicating a typical relaxor ferroelectric material. The other interesting in the single crystalline PZT films deposited on silicon and SOI is to employ them in the application of MEMS devices, where the standard silicon techniques are used. The microfabrication process performed in the cleanroom has permitted to realize cantilevers and membranes in order to mechanically characterize the piezoelectric layers. Mechanical deflection under the application of an electric voltage could be detected by interferometry. Eventually, this characterization by interferometry has been studied using the modeling based on finite element method and analytic method. In the future, it will be necessary to optimize the microfabrication process of MEMS devices based on single crystalline piezoelectric films in order to ameliorate the electromechanical performance. Finally, the characterizations at MEMS device level must be developed for their utilization in the future applications.

Cette thèse de doctorat traite de la mise en œuvre du procédé de soulèvement épitaxial (ou ELO pour epitaxial lift-off) à partir d'un substrat d'InP permettant le détachement des couches actives et le recyclage du substrat afin de rendre économiquement compétitive la fabrication de cellules solaires multi-jonctions pour le photovoltaïque concentré. Ce procédé, qui consiste à sous-graver sélectivement une couche sacrificielle comprise entre le substrat et les couches actives, est bien connu et maîtrisé sur un substrat de GaAs avec l'utilisation d'une couche sacrificielle d'AlAs d'épaisseur voisine de 5 nm, ce qui n'est pas possible sur un substrat d'InP en raison du fort désaccord de maille cristalline existant entre l'AlAs et l'InP. Pour l'adapter à un substrat d'InP, le développement d'une couche sacrificielle spécifique basée sur un super-réseau AlAs/InAlAs a été réalisé, ce qui permet de contourner les problématiques liées au désaccord de maille et à la croissance de matériaux contraints. Après optimisation des conditions de croissance de ce super-réseau, les épaisseurs atteintes et donc les vitesses de sous-gravure obtenues en utilisant ce type de couche sacrificielle ont satisfait aux exigences du procédé ELO. Ensuite, le report et le soulèvement de structures actives de cellules solaires InGaAs en couches minces cristallines ont été développés. Les cellules solaires ainsi fabriquées ont montré des performances semblables à celles réalisées par épitaxie standard sur un substrat d'InP, voire meilleures sous concentration en raison d'effets de confinement optique. Finalement, le recyclage du substrat d'InP réalisé avec un procédé utilisant seulement deux étapes de nettoyage par voies chimiques humides, a permis de produire des surfaces d'InP de qualité suffisante pour réaliser une reprise d'épitaxie satisfaisante. / This PhD thesis deals with the implementation of the epitaxial lift-off (ELO) process from an InP substrate allowing the detachment of active layers and the substrate recycling. The final target is to realize multi-junction solar cells in an economically competitive way for concentrated photovoltaic. The ELO process consists in the under-etching of a sacrificial layer inserted between the substrate and the active layers. It is well known and mastered on a GaAs substrate with the use of a sacrificial layer of AlAs with a thickness of about 5 nm. Such a layer is not usable on an InP substrate due to the high lattice mismatch between AlAs and InP. In order to adapt the ELO process to an InP substrate, this work aimed to develop a specific sacrificial layer based on an AlAs/InAlAs superlattice. Thus, it is possible to circumvent problems related to the lattice mismatch and to the strained layer growth. After optimization of growth conditions of this superlattice, using this type of sacrificial layer, we achieve a sufficient thickness and therefore a sufficient under-etching rate in order to meet the requirements of the ELO process. Then, the transfer and lift-off of thin crystalline film based InGaAs solar cells have been developed. This kind of solar cells showed performances similar to those obtained with a standard epitaxial growth on an InP substrate, or even better under concentration due to optical confinement effects. Finally, the recycling of the InP substrate carried out by a process using only two wet chemical cleaning steps made it possible to produce InP surfaces of sufficient quality to achieve a promising second epitaxial growth.

L’intégration monolithique hétérogène de composants III-N sur silicium (Si) offre de nombreuses possibilités en termes d’applications. Cependant, gérer l’hétéroépitaxie de matériaux à paramètres de maille et coefficients de dilatation très différents, tout en évitant les contaminations, et concilier des températures optimales de procédé parfois très éloignées requière inévitablement certains compromis. Dans ce contexte, nous avons cherché à intégrer des transistors à haute mobilité électronique (HEMT) à base de nitrure de Gallium (GaN) sur substrat Si par épitaxie sous jets moléculaires (EJM) en vue de réaliser des circuits monolithiques GaN sur CMOS Si. / The monolithic integration of heterogeneous devices and materials such as III-N compounds with silicon (Si) CMOS technology paves the way for new circuits applications and capabilities for both technologies. However, the heteroepitaxy of such materials on Si can be challenging due to very different lattice parameters and thermal expansion coefficients. In addition, contamination issues and thermal budget constraints on CMOS technology may prevent the use of standard process parameters and require various manufacturing trade-offs. In this context, we have investigated the integration of GaN-based high electron mobility transistors (HEMTs) on Si substrates in view of the monolithic integration of GaN on CMOS circuits.

Die Bildung von CoSi$_2$-Schichten mittels TIME-Verfahren (TIME: Ti-Interlayer Mediated Epitaxy) wurde untersucht. Dabei wurde die Ti-Zwischenschicht durch eine Hf-Zwischenschicht ersetzt. Der Einfluss der Prozessparameter (Tempertemperatur, Temperzeit, Aufheizrate und Ausgangsschichtdicken) und des Metalls (Hf, Ti, Zr) der Zwischenschicht auf die Reaktion der metallischen Doppelschichten mit Si(100) wurde ermittelt. Zur Charakterisierung der Proben wurden RBS-, TEM-, XRD- und AES-Untersuchungen durchgefuehrt. Die Ausbildung eines Mehrschichtsystems nach der Temperung der Doppelschichten in Abhaengigkeit der Prozessparameter wird dargestellt. Es wird gezeigt, dass die Prozessparameter die Temperatur bestimmen, bei der die CoSi$_2$-Keimbildung stattfindet. Anhand dieser Untersuchungen wird nachgewiesen, dass sich mit Erhoehung der CoSi$_2$-Keimbildungstemperatur die epitaktische Qualitaet der gebildeten CoSi$_2$-Schicht verbessert. Die Erklaerung des Reaktionsablaufs der metallischen Doppelschichten mit Si(100) erfolgt anhand eines aufgestellten Reaktionsschemas. Zur Entfernung der Deckschicht wurden verschiedene Aetzverfahren angewandt und deren Wirkung verglichen. / The formation of a CoSi$_2$ layer by solid phase reaction of metallic bilayer with Si (TIME: Ti-Interlayer Mediated Epitaxy) was investigated. In this work the Ti was replaced by Hf. The influence of the annealing temperature, the annealing time, the heating rate and the thicknesses of the metallic layers on the reaction of the bilayer with Si was determined. The samples were characterised by Rutherford-backscattering (RBS), Transmission-Electron-Microscopy (TEM), X-ray-Diffraction and Auger-Electron-Spectroscopy (AES) studies. During the annealing of the samples a system of layers is formed. It was shown, that the annealing parameters and the thicknesses of the layer determine the temperature, on which the nucleation of CoSi$_2$ occurs. A decrease of this nucleation temperature leads to an improvement of the quality of the epitaxial CoSi$_2$ layer. A model of reaction is presented, which explains the reaction of the metallic bilayer with Si. The removal of the top layer by several etching procedures was investigated and the results were compared.

The term "topological insulator" (TI) represents a novel class of compounds which are insulating in the bulk, but simultaneously and unavoidably have a metallic surface. The reason for this is the non-trivial band topology, arising from particular band inversions and the spin-orbit interaction, of the bulk. These topologically protected metallic surface states are characterized by massless Dirac dispersion and locked helical spin polarization, leading to forbidden back-scattering with robustness against disorder. Based on the extraordinary features of the topological insulators an abundance of new phenomena and many exciting experiments have been proposed by theoreticians, but still await their experimental verification, not to mention their implementation into applications, e.g. the creation of Majorana fermions, advanced spintronics, or the realization of quantum computers. In this perspective, the 3D TIs Bi2Te3 and Bi2Se3 gained a lot of interest due to their relatively simple electronic band structure, having only a single Dirac cone at the surface. Furthermore, they exhibit an appreciable bulk band gap of up to ~ 0.3 eV, making room temperature applications feasible. Yet, the execution of these proposals remains an enormous experimental challenge. The main obstacle, which thus far hampered the electrical characterization of topological surface states via transport experiments, is the residual extrinsic conductivity arising from the presence of defects and impurities in their bulk, as well as the contamination of the surface due to exposure to air. This thesis is part of the actual effort in improving sample quality to achieve bulk-insulating Bi2Te3 films and study of their electrical properties under controlled conditions. Furthermore, appropriate capping materials preserving the electronic features under ambient atmosphere shall be identified to facilitate more sophisticated ex-situ experiments. Bi2Te3 thin films were fabricated by molecular beam epitaxy (MBE). It could be shown that, by optimizing the growth conditions, it is indeed possible to obtain consistently bulk-insulating and single-domain TI films. Hereby, the key factor is to supply the elements with a Te/Bi ratio of ~8, while achieving a full distillation of the Te, and the usage of substrates with negligible lattice mismatch. The optimal MBE conditions for Bi2Te3 were found in a two-step growth procedure at substrate temperatures of 220°C and 250°C, respectively, and a Bi flux rate of 1 Å/min. Subsequently, the structural characterization by high- and low-energy electron diffraction, photoelectron spectroscopy, and, in particular, the temperature-dependent conductivity measurements were entirely done inside the same ultra-high vacuum (UHV) system, ensuring a reliable record of the intrinsic properties of the topological surface states. Bi2Te3 films with thicknesses ranging from 10 to 50 quintuple layers (QL; 1QL~1 nm) were fabricated to examine, whether the conductivity is solely arising from the surface states. Angle resolved photoemission spectroscopy (ARPES) demonstrates that the chemical potential for all these samples is located well within the bulk band gap, and is only intersected by the topological surface states, displaying the characteristic linear dispersion. A metallic-like temperature dependency of the sheet resistance is observed from the in-situ transport experiments. Upon going from 10 to 50QL the sheet resistance displays a variation by a factor 1.3 at 14K and of 1.5 at room temperature, evidencing that the conductivity is indeed dominated by the surface. Low charge carrier concentrations in the range of 2–4*10^12 cm^−2 with high mobility values up to 4600 cm2/Vs could be achieved. Furthermore, the degradation effect of air exposure on the conductance of the Bi2Te3 films was quantified, emphasizing the necessity to protect the surface from ambient conditions. Since the films behave inert to pure oxygen, water/moisture is the most probable source of degeneration. Moreover, epitaxially grown elemental tellurium was identified as a suitable capping material preserving the properties of the intrinsically insulating Bi2Te3 films and protecting from alterations during air exposure, facilitating well-defined and reliable ex-situ experiments. These findings serve as an ideal platform for further investigations and open the way to prepare devices that can exploit the intrinsic features of the topological surface states.:Abstract Kurzfassung Acronyms List of Symbols Introduction 1 Topological insulators 1.1 Basic theory of topological insulators 1.2 3D topological insulator materials: bismuth chalcogenides 2 Experimental techniques 2.1 General layout of the UHV-system 2.2 Molecular beam epitaxy 2.3 Structural and spectroscopic characterization 2.3.1 RHEED and LEED 2.3.2 Photoelectron spectroscopy 2.3.3 Ex situ x-ray diffraction 2.4 In situ electrical resistance measurements 2.4.1 In situ transport setup 2.4.2 Measurement equipment and operation modes 2.5 Substrates and sample holders 3 MBE growth and structural characterization of Bi2Te3 thin films 3.1 Bi2Te3 growth optimization and in situ structural characterization 3.1.1 1-step growth on Al2O3 (0001) 3.1.2 2-step growth on Al2O3 (0001) 3.1.3 2-step growth on BaF2 (111) 3.2 Ex situ structural characterization 4 In situ spectroscopy and transport properties of Bi2Te3 thin films 4.1 In situ spectroscopy of Bi2Te3 thin films 4.1.1 XPS 4.1.2 ARPES 4.2 Combined ARPES and in situ electrical resistance measurements of bulk-insulating Bi2Te3 thin films 4.2.1 Quality of the in situ electrical sample contacts 4.2.2 Verification of the intrinsic conduction through topological surface states of bulk-insulating Bi2Te3 thin films 5 Effect of surface contaminants on the TI properties 5.1 Effect of air exposure on the electrical conductivity of Bi2Te3 surfaces 5.2 Determination of the contaminants causing degradation of the TI properties 5.3 Long-time resistance behavior of a Bi2Te3 film exposed to minimal traces of contaminants 6 Protective capping of bulk-insulating Bi2Te3 thin films 6.1 Capping with BaF2 6.1.1 MBE growth and structure of BaF2 on Bi2Te3 thin films 6.1.2 Electron spectroscopy and electrical transport properties of BaF2 capped Bi2Te3 6.2 Capping with tellurium 6.2.1 MBE growth and structure of Te on Bi2Te3 thin films 6.2.2 Photoelectron spectroscopy and electrical transport properties of Te capped Bi2Te3 6.2.3 De-capping of Te 6.2.4 Efficiency of Te capping against air exposure 7 Conclusion and outlook Bibliography Versicherung Curriculum vitae Veröffentlichungen / Der Begriff "Topologischer Isolator" (TI) beschreibt eine neuartige Klasse von Verbindungen deren Inneres (engl. Bulk) isolierend ist, dieses Innere aber gleichzeitig und zwangsläufig eine metallisch leitende Oberfläche aufweist. Dies ist begründet in der nicht-trivialen Topologie dieser Materialien, welche durch eine spezielle Invertierung einzelner Bänder in der Bandstruktur und der Spin-Bahn-Kopplung im Materialinneren hervorgerufen ist. Diese topologisch geschützten, metallischen Oberflächenzustände sind gekennzeichnet durch eine masselose Dirac Dispersionsrelation und gekoppelte Helizität der Spinpolarisation, welche die Rückstreuung der Ladungsträger verbietet und somit zur Stabilisierung der Zustände gegenüber Störungen beiträgt. Auf Grundlage dieser außergewöhnlichen Merkmale haben Theoretiker eine Fülle neuer Phänomene und spannender Experimente vorhergesagt. Deren experimentelle Überprüfung steht jedoch noch aus, geschweige denn deren Umsetzung in Anwendungen, wie zum Beispiel die Erzeugung von Majorana Teilchen, fortgeschrittene Spintronik, oder die Realisierung von Quantencomputern. Aufgrund ihrer relativ einfachen Bandstruktur, welche nur einen Dirac-Kegel an der Oberfläche aufweist, haben die 3D TI Bi2Te3 und Bi2Se3 in den letzten Jahren großes Interesse erlangt. Weiterhin besitzen diese Materialien eine merkliche Bandlücke von bis zu ~0,3 eV, welche sogar Anwendungen bei Raumtemperatur ermöglichen könnten. Dennoch ist deren experimentelle Umsetzung nachwievor eine enorme Herausforderung. Das Haupthindernis, welches bis jetzt insbesondere die elektrische Charakterisierung the topologischen Oberflächenzustände behindert hat, ist die zusätzliche Leitfähigkeit des Materialinneren, welche durch Kristalldefekte und Beimischungen, sowie die Verunreinigung der Probenoberfläche durch Luftexposition bedingt wird. Die vorliegende Arbeit liefert einen Beitrag zu aktuellen den Anstrengungen in der Verbesserung der Probenqualität der TI um die Leitfähigkeit des Materialinneren zu unterdrücken, sowie die anschließende Untersuchung der elektrischen Eigenschaften unter kontrollierten Bedingungen durchzuführen. Weiterhin sollen geeignete Deckschichten identifiziert werden, welche die besonderen elektronischen Merkmale der TI nicht beeinflussen sowie diese gegen äußere Einflüsse schützen, und somit die Durchführung anspruchsvoller ex situ Experimente ermöglichen können. Die untersuchten Bi2Te3 Schichten wurden mittels Molekularstrahlepitaxie (MBE) hergestellt. Es konnte gezeigt werden, dass es allein durch Optimierung der Wachstumsbedingungen möglich ist Proben herzustellen, die gleichbleibend isolierende Eigenschaften des TI Inneren aufweisen und Eindomänen-Ausrichtung besitzen. Die zentralen Faktoren sind hierbei die Aufrechterhaltung eines Flussratenverhältnisses von Te/Bi ~8 der einzelnen Elemente, sowie die Wahl einer ausreichend hohen Substrattemperatur, um ein vollständiges Abdampfen (Destillation) des überschüssigen Tellur zu erreichen. Weiterhin müssen Substrate mit gut angepassten Gitterparametern verwendet werden, welches bei BaF2 (111) gegeben ist. Optimales MBE Wachstum konnte durch ein Zwei-Stufen Prozess bei Substrattemperaturen von 220°C und 250°C und einer Bi-Verdampfungsrate von 1 Å/min erreicht werden. Die nachfolgende Charakterisierung der strukturellen Eigenschaften, Photoelektronenspektroskopie, sowie temperaturabhängige Leitfähigkeitsmessungen wurden alle in einem zusammenhängenden Ultrahochvakuum-System durchgeführt. Auf diese Weise wird eine zuverlässige Erfassung der intrinsischen Eigenschaften der TI sichergestellt. Zur Überprüfung, ob die Leitfähigkeit der Proben tatsächlich nur durch die Oberflächenzustände hervorgerufen wird, wurden Filme mit Schichtdicken im Bereich von 10 bis 50 Quintupel-Lagen (QL; 1QL~ 1 nm) hergestellt und charakterisiert. Winkelaufgelöste Photoelektronenspektroskopie (ARPES) belegt, dass das chemische Potential (Fermi-Niveau) in allen Proben innerhalb der Bandlücke der Bandstruktur des Materialinneren liegt und nur von den topologisch geschützten Oberflächenzuständen gekreuzt wird, welche die charakteristische lineare Dirac Dispersionsrelation aufweisen. Die temperaturabhängigen Widerstandsmessungen zeigen ein metallisches Verhalten aller Proben. Bei der Variation der Schichtdicke von 10 zu 50QL wird eine Streuung des Flächenwiderstandes vom Faktor 1,3 bei 14K und 1,5 bei Raumtemperatur beobachtet. Dies beweist, dass die gemessene Leitfähigkeit vorrangig durch die topologisch geschützten Oberflächenzustände hervorgerufen wird. Eine geringe Oberflächenladungsträgerkonzentration im Bereich von 2–4*10^12 cm^−2 und hohe Mobilitätswerte von bis zu 4600 cm2/Vs wurden erreicht. Weiterhin wurden die negativen Auswirkungen auf die Eigenschaften der TI durch Luftexposition quantifiziert, welches die Notwendigkeit belegt, die Oberfläche der TI vor Umgebungseinflüssen zu schützen. Die Proben verhalten sich inert gegenüber reinem Sauerstoff, daher ist Wasser aus der Luftfeuchte höchstwahrscheinlich der Hauptgrund für die beobachtbare Verschlechterung. Darüber hinaus konnte epitaktisch gewachsenes Tellur als geeignete Deckschicht ausfindig gemacht werden, welches die Eigenschaften der Bi2Te3 Filme nicht beeinflusst, sowie gegen Veränderungen durch Luftexposition schützt. Die gewonnenen Erkenntnisse stellen eine ideale Grundlage für weiterführende Untersuchungen dar und ebnen den Weg zur Entwicklung von Bauelementen welche die spezifischen Besonderheiten der topologischen Oberflächenzustände.:Abstract Kurzfassung Acronyms List of Symbols Introduction 1 Topological insulators 1.1 Basic theory of topological insulators 1.2 3D topological insulator materials: bismuth chalcogenides 2 Experimental techniques 2.1 General layout of the UHV-system 2.2 Molecular beam epitaxy 2.3 Structural and spectroscopic characterization 2.3.1 RHEED and LEED 2.3.2 Photoelectron spectroscopy 2.3.3 Ex situ x-ray diffraction 2.4 In situ electrical resistance measurements 2.4.1 In situ transport setup 2.4.2 Measurement equipment and operation modes 2.5 Substrates and sample holders 3 MBE growth and structural characterization of Bi2Te3 thin films 3.1 Bi2Te3 growth optimization and in situ structural characterization 3.1.1 1-step growth on Al2O3 (0001) 3.1.2 2-step growth on Al2O3 (0001) 3.1.3 2-step growth on BaF2 (111) 3.2 Ex situ structural characterization 4 In situ spectroscopy and transport properties of Bi2Te3 thin films 4.1 In situ spectroscopy of Bi2Te3 thin films 4.1.1 XPS 4.1.2 ARPES 4.2 Combined ARPES and in situ electrical resistance measurements of bulk-insulating Bi2Te3 thin films 4.2.1 Quality of the in situ electrical sample contacts 4.2.2 Verification of the intrinsic conduction through topological surface states of bulk-insulating Bi2Te3 thin films 5 Effect of surface contaminants on the TI properties 5.1 Effect of air exposure on the electrical conductivity of Bi2Te3 surfaces 5.2 Determination of the contaminants causing degradation of the TI properties 5.3 Long-time resistance behavior of a Bi2Te3 film exposed to minimal traces of contaminants 6 Protective capping of bulk-insulating Bi2Te3 thin films 6.1 Capping with BaF2 6.1.1 MBE growth and structure of BaF2 on Bi2Te3 thin films 6.1.2 Electron spectroscopy and electrical transport properties of BaF2 capped Bi2Te3 6.2 Capping with tellurium 6.2.1 MBE growth and structure of Te on Bi2Te3 thin films 6.2.2 Photoelectron spectroscopy and electrical transport properties of Te capped Bi2Te3 6.2.3 De-capping of Te 6.2.4 Efficiency of Te capping against air exposure 7 Conclusion and outlook Bibliography Versicherung Curriculum vitae Veröffentlichungen

This report collects selected outstanding scientific and technological results obtained within the frame of the European project "FLASiC" (Flash LAmp Supported Deposition of 3C-SiC) but also other work performed in adjacent fields. Goal of the project was the production of large-area epitaxial 3C-SiC layers grown on Si, where in an early stage of SiC deposition the SiC/Si interface is rigorously improved by energetic electromagnetic radiation from purpose-built flash lamp equipment developed at Forschungszentrum Rossendorf. Background of this work is the challenging task for areas like microelectronics, biotechnology, or biomedicine to meet the growing demands for high-quality electronic sensors to work at high temperatures and under extreme environmental conditions. First results in continuation of the project work – for example, the deposition of the topical semiconductor material zinc oxide (ZnO) on epitaxial 3C-SiC/Si layers – are reported too.

Manufacturing of electronics devices, continuously decreasing in size, commonly requires the vapor phase deposition of materials into small structures on a wafer, often at a nanometer scale. In this thesis the goal is to simulate vapor-phase deposition processes at a scale where fully atomistic simulations using Molecular Dynamics are no longer feasible. This is achieved by combing two methods, one simulating the gas flow and deposition processes and another method simulating the changing surface. A Particle Monte Carlo method, specifically designed for free molecular flow, the typical flow regime at this length scale, is used. The simulation of growing surfaces uses the Level Set Method. Combining these two methods requires some additional coupling steps presented in this work. With the coupled model, different deposition processes are simulated within trenches to observe how well these processes perform for achieving a uniform deposition, as well as evaluating different process conditions.:Table of Contents List of Figures List of Tables List of Abbreviations List of Symbols 1 Introduction 2 Basics 2.1 Surface deposition processes 2.1.1 Chemical Vapor Deposition 2.1.2 Atomic Layer Deposition 2.1.3 Physical Vapor Deposition 2.2 Simulation approaches for surface depositions 2.2.1 Modeling chemical reactions on a surface 2.2.2 Interaction tables for PVD 2.3 Flow regimes 2.4 Molecular Dynamics 2.5 Particle Monte Carlo 2.6 Marker Particle Method 2.7 Level Set Method 2.7.1 Re-initialization of the signed distance function 2.7.2 Extension Velocities 2.7.3 Fast Marching Method 2.7.4 Upwind scheme 2.7.5 Curvature 2.8 Marching-Squares/Cubes Algorithm 3 Methods and Implementation 3.1 Software 3.1.1 External libraries 3.1.2 Geosect 3.2 Initialization of the signed distance field 3.3 Coupling between particle simulations and Level Set 3.3.1 The simulation cycle 3.3.2 Conversion from a grid to a discrete mesh 3.3.3 Extension of growth rates from a mesh to a grid 3.4 Integrating the Level Set Equation 3.4.1 Splitting the number of particles between different steps 3.4.2 Re-initializing the signed distance function 3.4.3 Handling surface coverage 3.4.4 The full update of the surface 3.5 Curvature dependent reflow 3.6 Level Set for radial symmetry 4 Verification 4.1 Testing different integration schemes 4.1.1 Growth of a circle in a linear velocity field 4.1.2 PVD in trenches 4.2 Mass preservation during curvature dependent reflow 4.3 Comparisons between 2D, radial 2D and 3D 4.3.1 Comparing 2D and 3D 4.3.2 Comparing radial 2D and 3D 5 Process Simulations 5.1 Resputter process using a PVD 5.1.1 Simulations and their parameters 5.1.2 Surfaces after the deposition step 5.1.3 Surface growth in the resputter step 5.1.4 Conditions for improved layer thickness 5.2 CVD with an effective sticking coefficient 5.3 Incomplete ALD cycles 5.4 Deposition onto a complex 3D shape 6 Conclusion Bibliography Acknowledgment Statement of authorship

Sources of single and polarization-entangled photons are an essential component in a variety of potential quantum information applications. Suitable emitters need to generate photons deterministically and at fast repetition rates, with highest degrees of single-photon purity, entanglement and indistinguishability. Semiconductor quantum dots are among the leading candidates for this task, offering entangled-photon pair emission on-demand, challenging current state-of-the-art sources based on the probabilistic spontaneous parametric down-conversion (SPDC). Unfortunately, their susceptibility to perturbations from the solid-state environment significantly affects the photon coherence and entanglement degree. Furthermore, most quantum dot types suffer from poor wavelength control and emitter yield, due to a random growth process. This thesis investigates the emerging family of GaAs/AlGaAs quantum dots obtained by in-situ Al droplet etching and nanohole infilling. Particular focus is laid on the interplay of growth parameters, quantum dot morphology and optical properties. An unprecedented emission wavelength control with distributions as narrow as ± 1 nm is achieved, using four independent growth parameters: The GaAs infilling amount, the deposition sequence, the migration time and the Al concentration in the barrier material. This enables the generation of large emitter ensembles tailored to match the optical transitions of rubidium, a leading quantum memory candidate. The photon coherence is enhanced by an optimized As flux during the growth process using the GaAs surface reconstruction. With these improvements, we demonstrate for the first time two-photon interference from separate, frequency-stabilized quantum dots using a rubidium-based Faraday filter as frequency reference. Two-photon resonant excitation of the biexciton state is employed for the coherent and deterministic generation of photon pairs with negligible multi-photon emission probability. The GaAs/AlGaAs quantum dots exhibit a very small average fine structure of (4.8 ±2.4) µeV and short average radiative lifetimes of 200 ps, enabling entanglement fidelities up to F = 0.94, which are among the highest reported for any entangled-photon source to date. Furthermore, almost all fabricated emitters on a single wafer exhibit fidelities beyond the classical limit - without any post-growth tuning. By embedding the quantum dots into a broadband-optical antenna we enhance the photon collection efficiency significantly without impairing the high degrees of entanglement. Thus, for the first time, quantum dots are able to compete with SPDC sources, paving the way towards the realization of a semiconductor-based quantum repeater - among many other key enabling quantum photonic elements.:Contents List of Figures ix List of Tables xiii 1 Introduction 1 1.1 Researchmotivation ...................1 1.1.1 Structure of this thesis ................. 3 1.2 Applications based on entangled photons ............. 4 1.2.1 Quantum bits ...................4 1.2.2 Quantum key distribution ................ 5 1.2.3 Qubit teleportation .................. 7 1.2.4 Teleportation of entanglement ..............9 1.2.5 The photonic quantumrepeater .............. 10 1.3 Generation of entangled photons ...............12 1.3.1 The ideal entangled-photon source ............. 12 1.3.2 Non-deterministic photon sources ............. 13 1.3.3 Deterministic photon sources ..............14 2 Fundamentals 17 2.1 Semiconductor quantumdots ................17 2.1.1 Introduction to semiconductor quantum dots .......... 17 2.1.2 Formation of confined excitonic states ............ 19 2.1.3 Energy hierarchy of excitonic states ............. 21 2.2 Entangled photons from semiconductor quantumdots ......... 22 2.2.1 The concept of entanglement ............... 22 2.2.2 Polarization-entangled photon pairs fromthe biexciton radiative decay .. 23 2.2.3 Origin and impact of the exciton fine structure splitting ....... 25 2.2.4 Impact of spin-scattering, dephasing and background photons on the degree of entanglement ..................29 2.3 Quantum dot entangled-photon sources - State of the art ........32 2.4 Exciton radiative lifetime .................. 34 2.4.1 The concept of radiative lifetime .............. 34 2.4.2 Measurement of the radiative lifetime ............35 2.5 Single-photon purity ...................37 2.5.1 Photon number distributions ............... 37 2.5.2 Second-order correlation function .............38 2.5.3 Measurement of the second-order correlation function ....... 40 2.6 Measurement of entanglement ................42 2.6.1 Quantum state tomography ...............43 2.7 Photon coherence and spectral linewidth .............46 2.7.1 The concept of coherence ................ 46 2.7.2 First-order coherence ................. 46 2.7.3 Relation between coherence and spectral linewidth ........ 49 2.7.4 homogeneous vs. inhomogeneous broadening in single quantumdots ..50 2.8 Photon indistinguishability .................51 2.8.1 Hong-Ou-Mandel interference ..............51 2.8.2 Hong-Ou-Mandel interference between photons fromseparate sources .. 52 2.8.3 The Bell state measurement with linear optics .......... 53 3 Experimentalmethods 55 3.1 The GaAs and AlAs material system ............... 55 3.2 Molecular beam epitaxy ..................56 3.2.1 The Concept of molecular beam epitaxy ...........56 3.2.2 Layout and components of the III-V Omicron MBE ........58 3.2.3 Growth parameters .................. 59 3.2.4 Reflection high-energy electron diffraction (RHEED) ........ 60 3.2.5 Growth rate determination using RHEED oscillations .......61 3.3 Optical setups .....................63 4 Results 67 4.1 Growth of GaAs/AlGaAs quantum dots by in-situ Al droplet etching .....68 4.1.1 Motivation for the study of GaAs / AlGaAs quantum dots ......68 4.1.2 GaAs / AlGaAs quantum dot growth process ..........69 4.1.3 Interplay between growth parameters, quantumdot morphology and optical properties ................. 71 4.1.4 Nanohole morphology and quantumdot formation ........ 73 4.1.5 Optical characterization ................75 4.1.6 Deterministic wavelength control .............77 4.1.7 Photon coherence and radiative lifetime ...........84 4.1.8 Decoherence processes in semiconductor quantum dots ......86 4.1.9 Chamber conditioning and growth process optimization ......87 4.1.10 Arsenic flux calibration using the GaAs surface reconstruction ..... 88 4.1.11 Enhanced photon coherence after growth process adjustments ....92 4.2 Two-photon interference from frequency-stabilized GaAs/AlGaAs quantum dots .................94 4.2.1 Frequency tuning of semiconductor quantumdots ........95 4.2.2 Experimental setup .................. 95 4.2.3 Optical characterization of the separate GaAs/AlGaAs quantum dots ... 98 4.2.4 Faraday anomalous dispersion optical filter and frequency feedback ... 99 4.2.5 Two-photon interference between remote, frequency-stabilized quantum dots 100 4.3 Solid-state ensemble of highly entangled photon sources at rubidiumatomic transitions ........................102 4.3.1 Fine-structure splitting ................103 4.3.2 Resonant excitation of the biexciton state ...........105 4.3.3 Single photon purity and radiative lifetime ........... 107 4.3.4 Radiative lifetime of GaAs/AlGaAs quantumdots - comparison to other quantumdot types ...................108 4.3.5 Degree of entanglement ................109 4.3.6 Highly-efficient extraction of the obtained entangled photons ..... 116 5 Conclusions 119 5.1 Summary ....................... 119 5.2 Discussion and outlook ..................122 Bibliography 127 Publications and scientific presentations 150 Acknowledgments 154 Selbstständigkeitserklärung 157 Curriculum vitae 157

In dieser Arbeit werden ferroelektrische Phasenübergänge in verspannten (K,Na)NbO3-Schichten erstmals mit Hilfe temperaturabhängiger dreidimensionaler Röntgenbeugung untersucht. Der Fokus liegt auf stark anisotrop verspannten Dünnschichten, die bei Raumtemperatur ein geordnetes Fischgräten-Domänenmuster mit einer periodischen Anordnung von monoklinen a1a2/MC-Phasen aufweisen. Bei Erhöhung der Temperatur durchlaufen die (K,Na)NbO3-Dünnschichten einen ferroelektrischen Phasenübergang in die orthorhombische Hochtemperaturphase, die sich durch regelmäßige, alternierenden a1/a2-Streifendomänen mit ausschließlich lateraler Polarisation auszeichnet. In-plane Röntgenmessungen zeigen, dass die Filmeinheitszellen eine kleine Verzerrung in der Ebene erfahren, was zur Bildung von vier verschiedenen Einheitszellvarianten und damit vier verschiedenen (Super-)Domänenvarianten führt. Durch den Vergleich von Röntgenbeugungsmessungen verschiedener Bragg-Reflexe an Filmen mit unterschiedlicher Schichtdicke ist es möglich, die spezifischen Beugungsmerkmale zu unterscheiden und sie den einzelnen Phasen zuzuordnen. Mit Hilfe von in situ temperaturabhängiger Röntgenbeugung ist es daher möglich, die Details des Phasenübergangs vom Fischgräten in das Streifen-Domänenmuster aufzudecken. Es zeigt sich, dass dieser sich über einen großen Temperaturbereich erstreckt und in mehreren Schritten vollzieht. Die Beobachtung von Phasenkoexistenz innerhalb des Übergangs und einer thermischen Hysterese in der Phasenübergangstemperatur lassen auf einen Phasenübergang erster Art schließen. Zudem hängt die Phasenübergangstemperatur stark von der Kaliumkonzentration x in der KxNa1-xNbO3-Dünnschicht ab und kann durch eine Änderung von x=0,95 (stärker kompressiv verspannt) auf x=0,8 (stärker tensil verspannt) um etwa 60 K erhöht werden. Darüber hinaus ist dies die erste Studie, die experimentell beobachtete dreidimensionale Domänenanordnungen direkt mit Berechnungen aus Phasenfeldsimulationen vergleicht. / In this work, ferroelectric phase transitions in strained (K,Na)NbO3 films are studied for the first time using in situ temperature-dependent three-dimensional X-ray diffraction. The focus lies on strongly anisotropically strained thin films, which exhibit a well-ordered herringbone domain pattern with a periodic arrangement of monoclinic a1a2/MC phases at room temperature. Upon increasing temperatures, the (K,Na)NbO3 thin films undergo a ferroelectric phase transition to the orthorhombic high-temperature phase, which is characterized by a regular pattern of alternating a1/a2 stripe domains with pure lateral polarization. In-plane X-ray measurements show that the film unit cells undergo a small in-plane distortion, leading to the formation of four different unit cell variants and thus four different (super)domain variants. By comparing X-ray diffraction measurements of different Bragg reflections of films with different film thicknesses, it is possible to distinguish the specific diffraction features and assign them to the individual phases observed at the different temperatures. Using in situ temperature-dependent X-ray diffraction, it is therefore possible to reveal the details of the phase transition from the a1a2/MC herringbone to the a1/a2 stripe domain pattern. It is shown to extend over a wide temperature range and to occur in several steps. The observation of phase coexistence within the transition and a thermal hysteresis in the phase transition temperature suggests a first-order type phase transition. Moreover, the phase transition temperature strongly depends on the molar concentration of potassium x in the KxNa1-xNbO3 thin film and can be increased by about 60 K by changing x=0.95 (more compressively strained) to x=0.8 (more tensile strained). Furthermore, this is the first study to directly compare experimentally observed three-dimensional domain arrangements with calculations from phase field simulations.