Vibration control requires high dynamic stiffness in mechanical structures for a reliable performance under extreme conditions. Dynamic stiffness composes the parameters of stiffness (K) and damping (η) that are usually in a trade-off relationship. This thesis study aims to break the trade-off relationship. After identifying the underlying mechanism of damping in composite materials and joint interfaces, this thesis studies the deposition technique and physical characteristics of nano-structured HDS (high dynamic stiffness) composite thick-layer coatings. The HDS composite were created by enlarging the internal grain boundary surface area through reduced grain size in nano scale (≤ 40 nm). The deposition process utilizes a PECVD (Plasma Enhanced Chemical Vapour Deposition) method combined with the HiPIMS (High Power Impulse Magnetron Sputtering) technology. The HDS composite exhibited significantly higher surface hardness and higher elastic modulus compared to Poly(methyl methacrylate) (PMMA), yet similar damping property. The HDS composites successfully realized vibration control of cutting tools while applied in their clamping interfaces. Compression preload at essential joint interfaces was found to play a major role in stability of cutting processes and a method was provided for characterizing joint interface properties directly on assembled structures. The detailed analysis of a build-up structure showed that the vibrational mode energy is shifted by varying the joint interface’s compression preload. In a build-up structure, the location shift of vibration mode’s strain energy affects the dynamic responses together with the stiffness and damping properties of joint interfaces. The thesis demonstrates that it is possible to achieve high stiffness and high damping simultaneously in materials and structures. Analysis of the vibrational strain energy distribution was found essential for the success of vibration control.

Ce travail s’inscrit dans le cadre du développement de nouvelles applications des matériaux III-Nitrure en général, et de l’alliage InGaN en particulier, pour la réalisation de cellules solaires à base de multi-jonction. Les nombreux avantages du matériau InGaN, à savoir son coefficient d’absorption élevé (105 cm−1), sa résistance thermique élevée, et sa tolérance aux radiations ainsi que sa bande interdite couvrant presque tout le spectre solaire (0.64 - 3.4eV), en font un sérieux candidat pour les dispositifs photovoltaïques. Ainsi une cellule solaire à quadruple jonctions permettrait l’obtention d’une efficacité au-delà de 50%. Cependant, les enjeux technologiques tels que la séparation de phase, le manque de substrat approprié donnant lieu à de forte densité de dislocations, et la difficulté de réalisation du dopage de type-p, sont considérés comme des obstacles pour atteindre les performances attendues. L’objectif de ce travail est d’étudier deux nouvelles approches qui peuvent résoudre les problèmes cités précédemment pour la réalisation de cellules solaires de haute efficacité à base d’InGaN. La première approche est dite approche "semibulk". Elle consiste à élaborer une structure multicouches InGaN/GaN épaisse avec une optimisation de l’épaisseur de chaque couche (InGaN et GaN), de façon que les couches de GaN soient suffisamment épaisses pour être efficaces, et assez mince pour permettre le transport des porteurs de charges par effet tunnel. Les couches InGaN quant à elles, doivent être assez épaisses et nombreuses afin d’absorber efficacement le rayonnement lumineux et suffisamment minces pour éviter la relaxation et l’apparition de dislocations. La deuxième approche consiste en la croissance de nanostructures InGaN qui autorise une incorporation d’indium élevée avec un matériau complètement relaxé et sans dislocation. La complète relaxation du matériau permet en outre de s’affranchir de l’effet piézoélectrique qui conduit à une chute du rendement. Nous avons pu démontrer que les cellules photovoltaïques à base d’In0.08Ga0.92N réalisées suivant l’approche "semibulk" présentent un pic de rendement quantique de 85%, ainsi qu’une efficacité de conversion en conditions AM 1.5G, presque trois fois plus élevée que l’état de l’art. Les premiers résultats obtenus sur les cellules photovoltaïques à base de nanostructures d’In0.08Ga0.62N sont très encourageants / The InGaN material system, with high absorption coefficient (105 cm−1) and a bandgap from 0.64 eV to 3.4 eV spanning the entire visual spectrum, make the development of all-InGaN multijunction solar cells with overall conversion efficiency larger than 50% theoretically possible. However, to reach this goal high-quality and thick InGaN layers with high indium concentration are required, which is not a trivial task. Studies of InGaN-based junctions with an indium mole fraction exceeding 0.3 are rare due to issues such as strong phase separation and relaxation of the layer due to lattice mismatch with the substrate which lead to InGaN layers with large dislocation density and indium-clustering. These material problems, significantly limit the performance of InGaN-based photovoltaic cells, and whatever the indium content, performance still remains far from the theoretical ones. The objective of this study is to investigate new approaches that may overcome the issues of phase separation and high dislocation density in InGaN materials with high indium concentration, for the realization of high efficiency InGaN based solar cells. Two novel approaches are proposed that may overcome the basic challenges involved in the InGaN hetero-junction solar cells. The first approach consists in the growth of a thick multi-layered InGaN/GaN absorber, called Semibulk. These GaN interlayers need to be thick enough to be effective and thin enough to allow carrier transport through tunneling. The InGaN layers need to be thick and numerous enough to absorb efficiently the incoming light beam, and thin enough to remain fully strained and without phase separation. The second approach consists in the growth of InGaN nano-structures to achieve high quality thick InGaN epitaxial layers with high indium concentration. It allows the elimination of the preexisting dislocations in the underlying template. It also allows strain relaxation of InGaN layers without any dislocations, leading to higher indium incorporation and reduced piezoelectric effect. The electro-optical characterization of semibulk In0.08Ga0.92N PV devices show a maximum external quantum efficiency (EQE) of 85%, which is the maximum EQE peak reported so far for an InGaN PIN heterojunction solar cell. The voltage dependence of the current density, under AM 1.5G solar spectrum for the semibulk In0.08Ga0.92N solar cells results in values of Jsc, Voc, fill factor (FF) and power conversion efficiency (PCE) as 0.57 mA/cm2, 1.04 V, 65% and 0.39% respectively. A comparison of the results to the literature show that the Jsc is four to five times of what has been reported for a bulk In0.08Ga0.92N PV structure. This value of Jsc lead to a PCE for the semibulk In0.08Ga0.92N-based PV cell which is at least three times higher than the PCE for the bulk In0.08Ga0.92N structure under AM 0 solar spectrum. For our second approach, high crystalline structural quality for InGaN nano-structures with 35% of indium concentration has been obtained. The electro-optical characterization for In0.09Ga0.91N nano-structure PV cells shows a significant enhancement in the performance of the devices. The PV devices result in a Jsc and Voc of 12 mA/cm2 and 1.89 V under concentrated light respectively

The work is focused on research and development of a suitable method for creating a selective emitter for the visible and near infrared region so that they are able to work optimally together with silicon photovoltaic cells in a thermophotovoltaic system. The aim of the work was to develop a new method of creating very fine structures outside the current standard, which will increase the emissivity of the base material to meet the needs of a selective emitter for the VID and NIR region. The methods available to us for the creation of structures were chosen, from which we eliminated all unsuitable ones and we introduced the optimal procedure and parameters for their creation for the selected method. In this work, we focused on both ceramic and metallic materials, whose heat resistance and selective properties are key to this work. Part of the development of the emitter structures was also the need for pretreatment of the substrate itself, where great emphasis was placed on the purity of materials and surface roughness. Since ceramic materials cannot achieve a surface roughness so low that the desired structures can be formed, these materials have been purposefully used primarily for the purpose of combining the base material with thin layers of other high temperature material. Their compatibility and suitability were verified in terms of adhesion and subsequent heat resistance. The main material for the formation of fine structures was purposefully chosen tungsten, for which we verified the influence of the formed structure on the emissivity as well as the thermal stability during long-term exposure to high temperatures. The work thus represents not only a new method of creating very fine structures, which are not normally formed in such subtlety, but also opens the way to new possibilities of combining more materials to achieve the required selectivity of the thermophotovoltaic emitter.

This semestral thesis focuses on the description of specific renewable resources in the form of thermophotovoltaic cells using selective radiators with micro/nano structures. This work deals with an introduction of renewable resources and specifically focuses on thermophotovoltaic. Thesis describes basic principles, but also influences affecting the proper functioning of these systems. It also focuses on selective radiators, which are created by mikro/nano structures, and factors that can affect their implementation or simulation. Part of the work are also examples of calculations of basic parameters of the structures, which will be used in the simulations. Next chapters are dealing with simulations which are analyzing thermal effects on termophotovoltaic system. Except the analysis itself there is also partial optimalization solving some of the negative thermal effects.

Die Kontrolle physikalischer Phänomene auf Oberflächen durch bestimmte Topographien ist eines der Ziele oberflächentechnischer Verfahren. Die Oberflächentopographie kann durch oberflächenmodifizierende Verfahren wie Direkte Laserinterferenzstrukturieren (DLIP) und das Direkte Laserschreiben (DLW) verändert werden. Dadurch können definierte und kontrollierte Mikro- und Nanostrukturen auf verschiedenen Materialien erzeugt werden. Darüber hinaus können spezifische Topographien entworfen und großflächig nachgebildet werden, welche die gleichen Oberflächeneigenschaft gewährleisten können. Diese Arbeit schlägt neue Ansätze zur Verbesserung der Mikro- und Nano-Oberflächenstrukturen vor, die durch DLIP auf Metalloberflächen erzeugt werden. DLIP Experimente werden in der Zweistrahlkonfiguration entweder mit infraroten Nano- oder Pikosekundenlasern durchgeführt. Damit werden die Möglichkeiten zur Verbesserung und Kontrolle von Oberflächeneigenschaften durch die Mikrofertigung mit Strukturperioden von 0,2 µm bis 7,2 µm erweitert. Anschließend wird die Homogenität der Oberflächentextur auf Basis der Pulsverteilung und der Laserparameter optimiert. Ein quantitatives Messschema der Homogenität, das auf etablierten Parametern wie mittlere Strukturhöhe, seiner Standardabweichung und Kurtosis basiert, wird vorgestellt. Darüber hinaus wird die Herstellung hierarchischer linien- und säulenartiger Mikrostrukturen mittels DLIP in Abhängigkeit von der Anzahl der Pulse und der Fluenz untersucht. Zusätzlich zu den Mikrostrukturen, die der Interferenzverteilung entsprechen, wurden gleichzeitig laserinduzierte periodische Oberflächenstrukturen (LIPSS) erzeugt, die zu hierarchischen Mikro- und Nanostrukturen führen. Überdies wird als weitere Technologie das DLW eingesetzt, um Mikrozellen im Bereich von 17 µm bis 50 µm zu generieren. Anschließen werden Mikro- und Nanostrukturen mittels DLIP auf den Mikrozellen hergestellt. Die finale Topographie besteht aus multiskaligen hierarchischen Mikro- und Nanostrukturen. Um den Durchsatz des DLIP-Verfahrens zu verbessern, wird ein Ablationsmodell entwickelt und mit experimentellen Daten verifiziert. Das Modell ermöglicht die Berechnung von Strukturtiefe in Abhängigkeit von optimalen Laserprozessparametern. Darüber hinaus wird die Benetzbarkeit auf den Mikrosäulen im Rahmen des Füllfaktors und der Kombination von hierarschischen und einskalen Strukturen ausgewertet. Dazu wird ein hydrophobes Lösungsmittel auf die hierarchischen Strukturen aufgetragen, um den Wasserkontaktwinkel auf bis zu 152 ° ± 2 ° und die Kontaktwinkelhysterese von 4 ° ± 2 ° zu erreichen. Mikrosäulen mit einer Periode von 5,20 µm werden auf einer Flugzeugtragfläche hergestellt. Auf diese Weise wird der mögliche Einfluss von Mikrostrukturen auf die Ermüdungseigenschaften untersucht. Schließlich werden Mikrosäulen mit ca. 40 % geringeren Reibungskoeffizienten als die Referenz in einem grenzflächengeschmierten Bereich getestet. Zusammenfassend kann ausgesagt werden, dass die durch DLIP erzeugten Mikrosäulen eine vielversprechende und gut realisierbare Struktur für die Oberflächenfunktionalisierung von Metallen darstellen.:Selbstständigkeitserklärung Abstract Kurzfassung Acknowledgments Symbols and abbreviations 1 Motivation 2 Theoretical background 2.1 Laser-matter interactions 2.2 Principle of interference 2.3 Wetting on solid surfaces 2.4 Introduction to friction 2.5 Introduction to fatigue 3 State of the art 3.1 Properties of natural surfaces 3.2 Texturing techniques for creating micro/nanoroughness 3.3 Surface microstructuring of metals using pulsed laser sources 3.3.1 Direct Laser Writing 3.3.2 Direct Laser Interference Patterning 3.3.3 Laser-Induce Periodic Surface Structures 3.3.4 Challenges for laser surface texturing methods 3.4 Surface properties affected by laser micro/nano texturing on metals 3.4.1 Impact of laser surface textures and chemistry on wettability 3.4.2 Control of the friction coefficient 3.4.3 Impact on fatigue performance 4 Materials and methods 4.1 Materials 4.2 Direct Laser Writing 4.3 Direct Laser Interference Patterning 4.4 Surface chemical treatment 4.5 Characterisation methods 4.5.1 Water Contact Angle 4.5.2 White Light Interferometry and Confocal Microscopy 4.5.3 Scanning Electron Microscopy and Energy Dispersive X-ray Spectroscopy 4.5.4 Raman Spectroscopy 4.5.5 X-ray Photoelectron Spectroscopy 4.5.6 Tribological test 4.5.7 Fatigue test 5 Results and discussion 5.1 Interference structuring of Ti6Al4V using nanosecond laser pulses 5.1.1 Strategy to fabricate homogeneous DLIP line-like structures 5.1.2 Development of topographical parameters for homogeneity quantification 5.1.3 Impact of process parameters on surface structure homogeneity 5.2 Interference structuring of stainless steel using picosecond laser pulses 5.2.1 Fabrication of hierarchical periodic micro/nanostructures 5.2.2 Control of nanostructure orientation 5.2.3 Fabrication of hierarchical pillar-like microstructures 5.2.4 Control of nanostructures on hierarchical periodic microstructures 5.3 Fabrication of multi-scale periodic structures by DLW and DLIP 5.3.1 Laser surface texturing of Ti6Al4V 5.3.2 Laser surface texturing of Al2024 5.4 Structuring of a large aircraft surface for a flight test 6. Development of an analytical ablation model for ps-DLIP 7. Surface properties of textured materials 7.1 Determination of wetting behaviour 7.1.1 Wetting transition on single and hierarchical microstructures 7.1.2 Surface chemistry influence on wetting 7.1.3 Wetting response after the chemical surface modification 7.2 Wetting on multi-scale periodic structures fabricated by DLW and DLIP 7.3 Tribological properties of laser treated surfaces 7.4 Influence of laser treated surfaces on fatigue 8. Conclusions and outlook References Curriculum Vitae List of publications

Lorsqu'un supraconducteur (S) est mis en contact électrique avec un métal normal (N), il transfert à ce dernier des propriétés supraconductrices sur une faible épaisseur : ce phénomène s'appelle "l'effet de proximité". L'objectif de cette thèse consiste à étudier la cohérence quantique dans N à travers ce phénomène grâce à un microscope à effet tunnel (STM) fonctionnant à 1.6 K et 50 mK. L'avantage de cette méthode expérimentale est d'être en mesure d'observer l'échantillon et de déterminer ses propriétés électroniques locales simultanément avec une grande résolution spatiale et énergétique. L'effet de proximité a déjà été observé expérimentalement dans le cas où les dimensions du métal normal et du supraconducteur peuvent être considérés comme infinies. En régime diffusif, la densité d'états locale présente un pseudo-gap dans N qui se referme à mesure que l'on s'éloigne de l'interface [S-N]. Cependant, si les dimensions de N sont plus petites que la longueur de cohérence de phase, un mini-gap relié à l'énergie de Thouless se développe, traduisant l'établissement des propriétés supraconductrices dans tout le volume du métal normal. En régime balistique, des conclusions similaires s'appliquent selon le caractère intégrable (pseudo-gap) ou chaotique (mini-gap) des structures métalliques en contact avec le supraconducteur. Afin d'observer expérimentalement ces effets, nous avons fabriqué par lithographie électronique des échantillons où des motifs de métal normal, de taille mésoscopique et de différentes géométries, sont présents à proximité d'un supraconducteur. Nous avons mis en évidence des pics très fins dans les spectres de la conductance tunnel différentielle locale. Nous pensons qu'ils révèlent la présence d'états résonnants d'Andreev, impliquant l'existence de trajectoires électroniques semi-classiques satisfaisant la relation de quantification de De Gennes-St James. Dans le cadre de cette thèse, nous avons aussi saisi l'opportunité d'étudier les propriétés électroniques des films supraconducteurs désordonnés de nitrure de titane. A mesure que l'épaisseur des films est réduite, la diminution du potentiel effectif d'attraction des électrons et les effets de localisation donnent lieu à une transition supraconducteur-isolant (T.S.I.), observée par des études de magnéto-transport. Grâce au STM, la détermination de la densité d'états locale sur ces mêmes films a permis de révéler la co-existence de domaines supraconducteurs et normaux ainsi qu'une évolution spatiale des mesures spectroscopiques très inhabituelle. Ce matériau présente donc des propriétés supraconductrices inhomogènes pouvant jouer un rôle important dans les mécanismes de la T.S.I.

The broad theme of the present research investigation is on the preparation and characterization of the ultra thin films. The emerging field of nano science and technology demands the realization of different materials in nanometer dimension and a comprehensive understanding of their novel properties. Especially, the properties of the semiconducting materials in the nano dimensions are quite different from their bulk phase. A phase transition from semimetalic to semiconducting nature occurs at a thickness < 5nm of Sb ultra thin films. These facts emphasize the need for preparing these materials as nano layers and studying their properties as a function their size. Among the various characterization methods available to study the structure and the interfaces, Raman spectroscopy has proved to be a useful technique. In addition to revealing the structural information, Raman spectroscopy can bring out the quantum size effects in the lattice vibrational spectra of lower dimensional solids, stress state of the film in the initial growth stages, chemical nature of materials etc. Raman spectroscopy studies on the quantum structure of Ge and Sb are limited. This is attributed to the two serious limitations of the conventional backscattering of Raman signal. 1. The back scattered Raman signal intensity from the ultra thin layer could be below the detection limit. 2. The lower penetration depth of the lasers could inhibit the information from the buried layers. These limitations could be overcome to a major extent by employing an optical interference technique called IERS. This is basically an anti-reflection structure consisting minimum of three layers. These three layers are essential for achieving the interference conditions. The thicknesses of each layer were calculated using a matrix method. IERS structure consists of 1. A reflecting layer at the bottom of the stack (Platinum or Aluminum) 2. The second layer which is grown above the reflecting layer is a transparent dielectric layer, which introduces the necessary phase shift and hence it is called phase layer.(SiO2 or CeO2) 3. The top ultra thin layer which is to be investigated (Ge or Sb), is grown over the dielectric film and it is the layer which absorbs the most of the incident exciting light and it is called the absorbing layer. In this trilayer structure the thickness of the phase layer and the absorbing layer are adjusted in such a way that the light reflected from the air-ultra thin layer interface and the dielectric-reflector interface are equal in amplitude but opposite in phase. This leads to the destructive interference and a perfect anti-reflection condition is achieved. This enhances the near surface local field and results in the enhanced Raman signal. Regarding the reflection layer, thermally evaporated Al films were used. But the surface studies revealed a large surface roughness of 2.7nm for area of 2 µm×2µm. Also Al is known to react with oxygen and formation of an oxide layer is favored. In an effort to overcome these problems, a platinum layer was chosen instead of Al as a reflecting layer. Dual ion beam sputter deposition was employed to prepare the platinum films and to study the surface property of the films prepared at different secondary ion current density. Thus the process parameters to get the Pt film with the required surface properties were optimized. To prepare the required phase layer, optical thin films of Ceria were used. The optical and structural property of ceria is found to be sensitive to the process parameters. Hence a new deposition technique for preparing the CeO2 thin films was adopted. This technique is called Dual ion beam Sputter Deposition (DIBSD). This technique involves, two ion sources (Kaufman type). One source is used to sputter the target, which is called the primary ion source and the other one is used to assist the growing film, which is called the secondary ion source. Both argon and oxygen were fed into the secondary ion source and oxygen ions in the mixture of the gases (Ar +O2) react with the growing film and the oxygen stoichiometry in the film is maintained. Also the secondary ion bombardment of the growing film helps in the densification and it leads to the increase in the refractive index of the ceria films. The films were found to grow with a preferential orientation along (111) direction. The optical properties of the films were studied by using the transmission spectra of the films from the spectrophotometer. Powder X-Ray diffraction, and Raman spectroscopy, were employed to study the structural properties. Atomic Force Microscopy was used to examine the surface topography and to estimate the surface statistics. A stress free ceria film with a high refractive index of 2.36 at 600nm was prepared for a secondary ion beam current density of 150µA/cm2 and a beam energy of 150 eV. Raman spectra and X-ray diffraction data of these films have revealed the formation of point defects in these films as a function of secondary ion current density. Germanium (Ge) ultra thin layers were prepared by using Ion Beam Sputter Deposition (IBSD) as this technique has a good control over the rate of deposition apart from various other advantages. The Ge ultra thin films were prepared on the multilayer stacks with Al or Pt as a reflecting layer. The germanium films were prepared for the various thicknesses ranging from 1-10 nm. These films were prepared on the multilayer stack of reflecting layer and phase layer. The films were prepared for the different substrate temperatures from 40 °C to 300 °C. The films thus prepared have been analyzed by Interference Enhanced Raman Spectroscopy (IERS) for the structural and quantum size effects, by RBS for the thickness and to study interface diffusion, and Atomic Force Microscopy (AFM) for the analysis of nano structure of the grown films and also for the surface statistics. The thickness of the Ge films was found to be same as that had been calculated from the rate of deposition of the films. The films showed increase in the grain sizes with increase in the thickness of the films. The nanostructure of the films from AFM images confirms this observation. IERS of the films shows the transition from the compressive to stress free nature of the film for the nominal thickness of 1 & 2 nm. The quantum size effects of the films show the asymmetric broadening and peak shift and these observations were studied using the spatial correlation model. The TEM studies on the samples with Pt as a reflecting layer show influence of the underlying layer of CeO2 by the formation Moiré fringes. Antimony (Sb) films were prepared for the different thicknesses (3-10nm) and at different substrate temperatures (40 °C - 200 °C) on the Pt/CeO2 multilayer stacks as the absorbing layer. IERS studies on the films were performed and the results are as follows. Sb films show crystallization with increase in thickness from 3nm to 4nm. The films show amorphous to crystalline transition for the substrate temperature of 200 °C. Quantum size effects on the samples due to the phonon confinement were analyzed by the spatial correlation model. The atomic force microscopic measurements for the nanostructural information on the samples showed that the grain sizes of the films increase with increase in the thickness. Also the surface morphology shows a definite change in the features for the transition of amorphous to crystallization phase. Chapter 1 introduces the importance of Ge and Sb in the present day technologies. The current state of research on these two materials has been discussed. The importance of ceria and Pt films has been highlighted in the context of IERS and for the applications elsewhere. The advantages and disadvantages of ion beam sputter deposition have been described. The importance of Raman spectroscopy as a characterization tool for the nano structures has been shown in this chapter along with an introduction on Raman spectroscopy. Also, the importance of the other complimentary characterization techniques has been discussed. Chapter 2 presents the experimental details used to deposit and characterize the thin films. Details of IBSD and DIBSD processes are given. The characterization pertaining to structural, surface, optical and compositional properties are dealt in detail. Method to compute the optical constants of a transparent film is also given. Chapter 3 presents the properties of reflecting layers. Structural, surface and the compositional (presence of Ar ion) properties of the DIBSD platinum thin films are presented. Chapter 4 presents the optical, structural and surface properties of DIBSD ceria thin films as a function of process parameters. Chapter 5 deals with the growth and Raman analysis of ultra thin Ge films with Al and Pt as reflecting layers. Chapter 6 deals with the growth and Raman analysis of ultra thin Sb films. Chapter 7 gives the summary of the thesis and the future scope of the work.

Mesoscopic spin systems can be designed and fabricated using modern nano-fabrication techniques. These systems can contain large numbers of patterned ferromagnetic elements, for which the shape will generally determine their effective mesospin dimensionality. The lateral arrangement of these mesospins can be further used to tune the interactions between them. With an appropriate choice of material, it is possible to define a temperature range where thermal fluctuations of these mesospins are experimentally accessible. To actively define this range, we use δ-doped Palladium, a three-layer system of Palladium—Iron—Palladium, for which the Curie-temperature scales with the Iron layer thickness. The patterned mesoscopic elements used in this work have a stadium-like shape that promotes a single magnetic domain state, thus making these islands behave as one-dimensional Ising-like mesospins that can be observed using magnetic imaging techniques. We investigate the impact on the magnetic order resulting from modifications of the square spin ice geometry. By adding, removing and merging elements in the square artificial spin ice architecture, energy-landscape variations can be realized. Firstly, an added interaction modifier is used to equilibrate the interactions between the mesospins at the vertex level, which can restore the degenerate ground state of the square spin ice model. Secondly, the removal of elements can lead to topologically frustrated spin systems, as not all building blocks can simultaneously be in their lowest energy state. Furthermore, the merging results in multiple element sizes in the mesospin system. As the magnetization reversal barrier is dependent on the element size, these mesospin systems have different energy barriers. The thermal ordering process in such a system differs from a single-size element system with its unique energy barrier. Using reciprocal space analysis tools like the magnetic spin structure factor we show that systems with multiple element sizes achieve a higher short-range order then their single-size element references. The magnetic order in mesoscopic spin systems could successfully be tailored by modifications of the lattice geometry.

Ce travail de thèse a pour objectif l'étude de la dynamique de l'aimantation dans des nano-objets à anisotropie magnétique perpendiculaire, soit métalliques Pt/Co(5 Å)/Pt soit semiconducteurs ferromagnétiques dilués GaMnAs. Le renversement de l'aimantation, sous l'action d'un champ magnétique dans des nano-plots ou sous l'action d'un courant polarisé en spin sur une paroi de domaines dans une piste, a été étudié par magnéto-optique Kerr polaire. La nano-structuration a été réalisée par un procédé classique dans le cas du GaMnAs et par un procédé élégant dans le cas de la couche ultramince de cobalt : l'irradiation aux ions hélium qui permet d'obtenir des nano-plots dans un environnement planaire paramagnétique. Dans ces deux systèmes, différents de par l'origine du ferromagnétisme et de l'anisotropie, la réduction des dimensions latérales joue un rôle important en impliquant une transition d'un renversement à plusieurs mécanismes à un renversement à un seul mécanisme. Si le mode de renversement de Néel-Brown permet d'expliquer les résultats obtenus pour les nano-plots de GaMnAs de diamètre 33 nm, il ne peut pas rendre compte des résultats expérimentaux obtenus sur les nano-disques de Pt/Co/Pt de diamètre 130 nm. L'étude du comportement magnétique individuel de ces nano-disques conduit à la mise en évidence d'une nucléation d'une gouttelette en périphérie, confortée par un modèle micromagnétique. Contrairement au cas des métaux, le déplacement de paroi induit par un courant polarisé en spin se révèle aisé dans une piste de GaMnAs. Les mesures réalisées à température effective constante ont montré la nécessité de considérer un mécanisme non-adiabatique de transfert de spin.

Im Rahmen dieser Arbeit wurde die Strukturierbarkeit ultradünner, funktioneller Polymerfilme anhand von Diazosulfonat-Terpolymer- und Aminoterpolymer-Schichten untersucht. Beide Polymersysteme enthalten eine photoaktive Gruppe in der Seitenkette, die sich durch gezielte UV-Bestrahlung verändern läßt. In den Diazosulfonat-Terpolymeren wird durch die Belichtung die Funktionalität zerstört, während bei den Aminoterpolymeren die Funktionalität durch die Belichtung erst freigelegt wird. Dafür wurden Strukturierungsmethoden für verschiedene Längenskalen erarbeitet und auf ihre Eignung geprüft. Der Nachweis der erfolgreichen Strukturierung wurde durch an die Längenskala angepaßte Methoden geführt und damit die erzeugten Strukturen sichtbar gemacht. Die Veränderungen im optischen Absorptionsverhalten konnten an makroskopischen Probenbereichen nachgewiesen werden. Sowohl der verwendete Aufbau für die Strukturierung (Belichtung) als auch die Detektion mit dem 2-Stahl-Spektrometer erwies sich als geeignet. Es konnte deutlich der Abbau der UV-Absorptionsbande der Diazosulfonat-Terpolymerfilme gezeigt und quantitativ untersucht werden. Dafür wurden Lichtdosen von etwa 0,35 ... 39 nJ/µm² eingebracht und deren Auswirkungen auf die Absorptionsänderung des Polymers direkt festgestellt. Diese Messungen zeigen, daß die eingebrachte Energie und nicht die Leistung (sofern diese unterhalb 2,5 mW liegt) entscheidend für die Modifikation der optischen Eigenschaften dieser Polymere ist. Anhand der Meßergebnisse konnte eine Abschätzung der Quantenausbeute durchgeführt werden, die für die Diazosulfonat-Terpolymerfilme einen Wert von (12 ± 6) % ergab. Auf der Mikrometer-Skala wurden unterschiedliche Ansätze verfolgt, um die optische Strukturierung nachzuweisen. Der Nachweis optischer Modifikationen der Diazosulfonat-Terpolymerfilme wurde nach Belichtung mit hohen Lichtdosen geführt, da er sich nur in diesem Energiebereich mit der erforderlichen Empfindlichkeit realisieren ließ. Für die Aminoterpolymerfilme wurden Strukturen durch Fluoreszenzmarkierung nachgewiesen, welche sich als sehr sensitiv herausstellte. Im Anschluß an die Belichtung konnten topographische Modifikationen mit dem AFM gemessen werden. Mit dem SNOM konnten diese Modifikationen bereits während der Belichtung direkt analysiert werden. Die getesteten Methoden der Raman-Spektroskopie und der Metallisierung mit anschließender Röntgen-Photoelektronenspektroskopie zeigten weder die benötigte Sensitivität noch Selektivität. Die untersuchten Polymersysteme können in Form ultradünner Filme auf unterschiedliche Substrate aufgebracht werden. In diesen Polymerfilmen wurden Strukturen von der Millimeter-Skala bis Nanometer-Skala erzeugt. Anhand von an die Größenskala angepaßten direkten und indirekten Nachweismethoden konnten Veränderungen der optischen, mechanischen und chemischen Eigenschaften der Polymere analysiert werden.