Cette thèse porte sur la manipulation du moment angulaire de la lumière à l'échelle sub-micronique. Le moment angulaire total de la lumière est composé d'une partie de spin, relié au degré de liberté de polarisation circulaire de la lumière, et d'une partie orbitale, relié au degré de libertés spatiaux de la lumière que sont sa direction de propagation (locale et globale) et sa distribution spatiale d'intensité. Le couplage spin-orbite existant entre ces deux contributions permet alors de manipuler les degrés de libertés spatiaux de la lumière par un simple contrôle de son état de polarisation circulaire. Dans cette thèse, nous avons étudié et exploité ce couplage à l'échelle sub-micronique dans deux nouveaux phénomènes que nous avons mis en évidence. Le premier met à profit ce couplage pour permettre d'exciter de manière unidirectionnelle des modes plasmoniques guidés. Une étude complète (numérique, expérimentale et analytique) de ce phénomène nouveau, basé sur un couplage entre le moment de spin du photon incident et le moment orbital extrinsèque des modes plasmoniques guidés dans la courbure d'un guide, est présentée. La deuxième étude présente une voie pour tirer parti du transfert de moment orbital de la lumière à un gaz d'électrons libres dans un métal afin de générer et contrôler le sens et la géométries de boucles de courants sub-microniques dans des structures métalliques. Ce contrôle permettrait la génération d'optomaimants nanométriques, entièrement contrôlés par la lumière, pouvant être modulés aux fréquences optiques. Ce travail a été soutenu par le LABEX Action. / This thesis focuses on the manipulation of the angular momentum of light at the nanoscale.The total angular momentum of light is composed of a spin component, connected to the polarization degree of freedom of light, and an orbital component, related to the spatial degrees of freedom of the light which are its propagation direction (local and global) and its intensity distribution. The spin-orbit coupling between these two contributions allows the control of the spatial degrees of freedom of light by a simple manipulation of its circular polarization state. In this thesis, we have studied and applied this coupling at the nanoscale anbd we have highlighted two new phenomenas. The first one takes part of this coupling to allows unidirectional excitation of plasmonic guided modes. A complete study (numerical, experimental and analytical) of this new phenomenon, based on a coupling between the spin of the incident photon and the extrinsic orbital momentum of the plasmonic guided modes within the curvature of a waveguide, is presented. The second study propose a way to benefit from the transfer of the angular momentum of light to the free electrons gas in a metal to generate and control the direction and the geometry of nanoscale current loops in metallic structures. this control would at optical frequencies. This work was supported by the LABEX Action.

Random number generation can be based on physical events with probabilistic character, or on algorithms that use complex or one-way functions, alternatively on both of these approaches. A magnetic vortex is a basic state of magnetization that forms in magnetic micro- and nanostructures of an appropriate shape, dimensions and material. Quantities of the magnetic vortex form randomly if ambient conditions are chosen eligibly. A concept of a true random number generator using a random switching of states of the magnetic vortex is presented in this thesis. This concept is realized and random numbers were experimentally generated and numbers were statistically analysed.

This thesis describes a portfolio of work aimed at the high field applications of superconductors and can be split into four main topics: The thermal stability of technical superconductors. This section investigates the effects of thermal perturbations on technical superconducting wire used in MRI scanner construction. The ultimate aim of this section is to predict how the architecture of the wire may affect its thermal stability. To this end a detailed finite element analysis model was constructed, verified by detailed experimental data, which could then be used to quickly and easily vary the wire’s parameters. Design of a high field pulsed electromagnetic coil for flux trapping in superconductors. This section details the design, construction and testing of a novel pulsed high field magnet. The design uses finite element analysis to predict the electromagnetic, thermal and structural properties of the coil. Explosive testing of high tensile fibres used in the construction of the high field coil. This section describes the refinement and use of a novel method for testing the mechanical properties of high tensile fibres in cylindrical geometries by using highly pressurized copper vessels. Pulsed field magnetization of bulk high temperature superconductors. This section discusses the process of magnetizing bulks of high temperature superconductors by using pulsed magnetic fields. It investigates how the trapped field varies with the magnitude and rise-time of the magnetizing field, sample temperature and time after magnetization.

<p>Materials with new or improved properties are crucial for technological development. To provide the foundation for future successful products, it is important to prepare and characterise new chemical compounds that could show unusual properties. The properties of magnetic materials are closely related to their crystal, magnetic and electronic structures. This thesis focuses on the novel synthesis and structural characterisation of a number of new ternary or <i>pseudo</i>-ternary silicides and germanides of manganese with iridium, cobalt or palladium. To provide a more complete picture of the complex magnetic properties, crystal and magnetic structure refinements by the Rietveld method of X-ray and neutron powder diffraction data are complemented by single-crystal X-ray diffraction, electron diffraction, magnetisation measurements and Reverse Monte Carlo simulations of magnetic short-range order. The experimental results are corroborated by first-principles electronic structure and total energy calculations. </p><p>A commensurate non-collinear antiferromagnetic structure is found for most compounds of the solid solution Mn₃Ir_1-yCo_ySi_1-xGe_x. The non-collinearity is a result of geometric frustration in a crystal structure with magnetic Mn atoms located on a three-dimensional network of triangles. The close structural similarity to the β-modification of elemental manganese, which does not order magnetically, inspired a closer theoretical comparison of the Mn₃Ir_1-yCo_ySi_1-xGe_x propertieswith β-Mn.</p><p>Magnetic frustration is also observed for Mn₄Ir_7-xMn_xGe₆, and is an important factor underlying the dramatic change from commensurate antiferromagnetic order to spin glass properties induced by a small variation in Mn concentration. Magnetic short-range order with dominant antiferromagnetic correlation is observed for Mn₈Pd₁₅Si₇, and results from a random distribution of Mn atoms in-between the geometrically frustrated magnetic moments on the Mn octahedra. </p><p>An incommensurate cycloidal magnetic structure, observed for IrMnSi, is stabilised by an electronic structure effect, which also accounts for the non-collinearity of the Mn₃IrSi type magnetic structure.</p>

Materials with new or improved properties are crucial for technological development. To provide the foundation for future successful products, it is important to prepare and characterise new chemical compounds that could show unusual properties. The properties of magnetic materials are closely related to their crystal, magnetic and electronic structures. This thesis focuses on the novel synthesis and structural characterisation of a number of new ternary or pseudo-ternary silicides and germanides of manganese with iridium, cobalt or palladium. To provide a more complete picture of the complex magnetic properties, crystal and magnetic structure refinements by the Rietveld method of X-ray and neutron powder diffraction data are complemented by single-crystal X-ray diffraction, electron diffraction, magnetisation measurements and Reverse Monte Carlo simulations of magnetic short-range order. The experimental results are corroborated by first-principles electronic structure and total energy calculations. A commensurate non-collinear antiferromagnetic structure is found for most compounds of the solid solution Mn3Ir1-yCoySi1-xGex. The non-collinearity is a result of geometric frustration in a crystal structure with magnetic Mn atoms located on a three-dimensional network of triangles. The close structural similarity to the β-modification of elemental manganese, which does not order magnetically, inspired a closer theoretical comparison of the Mn3Ir1-yCoySi1-xGex propertieswith β-Mn. Magnetic frustration is also observed for Mn4Ir7-xMnxGe6, and is an important factor underlying the dramatic change from commensurate antiferromagnetic order to spin glass properties induced by a small variation in Mn concentration. Magnetic short-range order with dominant antiferromagnetic correlation is observed for Mn8Pd15Si7, and results from a random distribution of Mn atoms in-between the geometrically frustrated magnetic moments on the Mn octahedra. An incommensurate cycloidal magnetic structure, observed for IrMnSi, is stabilised by an electronic structure effect, which also accounts for the non-collinearity of the Mn3IrSi type magnetic structure.

No description available.

Hoch manganhaltige Eisenbasislegierungen sind bei Raumtemperatur austenitisch und antiferromagnetisch (afm). Dabei besteht die Besonderheit, dass sich durch Legierung die afm Übergangstemperatur (Néeltemperatur) so einstellen lässt, dass sie nahe Raumtemperatur liegt. FeMn-Basislegierungen zeigen in Abhängigkeit von der Zusammensetzung Transformation Induced Plasticity (TRIP) und/oder Twinning Induced Plasticity (TWIP), d.h. die niedrige Stapelfehlerenergie dieser Legierungen führt zu verformungsinduzierter, metastabiler Phasenbildung (TRIP) bzw. zur Bildung von Verformungszwillingen (TWIP) und dadurch zu außerordentlich hoher Duktilität bei gleichzeitig hoher Verfestigung. Darüber hinaus haben FeMn-Basislegierungen einen ausgeprägten Magnetovolumeneffekt und magnetoelastischen Effekt durch magnetische Ordnung. Daher sind die untersuchten FeMnNiCr-Basislegierungen auch prototypisch für afm Élinvarlegierungen. Da Élinvar jedoch für invariable Elastizität steht, bedingt eine Anwendung als temperaturkompensierte Konstantmodullegierungen die Glättung der ausgeprägten magnetischen Anomalien, die industriell noch in keiner Anwendung realisiert wurde. Der Vorteil dies für eine Anwendung zu erreichen, läge in der Unempfindlichkeit feinmechanischer Bauelemente, gegenüber magnetischen Feldern, die bei den industriell verfügbaren ferromagnetischen Élinvarlegierungen nicht gewährleistet ist. Mit Bezug zu feinmechanischen Schwingsystemen spielen dabei neben der Einstellung der magnetoelastischen Eigenschaften die Prozessierbarkeit, Kaltumformbarkeit und Festigkeit sowie deren wechselseitige Beeinflussung eine große Rolle. Die vorliegende Arbeit befasst sich daher mit der Anwendbarkeit der untersuchten FeMnNiCr-Legierungen. Dabei wurden grundlegende Untersuchungen zur Plastizität durchgeführt, die die mechanische Zwillingsbildung in diesen Legierungen charakterisiert und ein Modell der mechanischen Zwillingsbildung bei kleinen plastischen Dehnungen vorschlägt, das eine Abschätzung der Stapelfehlerenergie erlaubt. Die Untersuchung des Antiferromagnetismus umgeformter Proben zeigt das Auftreten thermoremanenter Magnetisierung (TRM), deren Größe mit dem Umformgrad der untersuchten Proben skaliert. Sie wird den durch Umformdefekte erzeugten unkompensierten Momenten in der afm Spinstruktur zugeschrieben. Diese werden durch eine magnetische Feldkühlung magnetisiert und koppeln durch Austauschwechselwirkung an die umgebende antiferromagnetische Matrix unterhalb der Néeltemperatur. Das komplexe thermomagnetische Verhalten der unkompensierten Momente wird experimentell beschrieben und phänomenologisch gedeutet. Die Weiterentwicklung und Bewertung technischer, ausscheidbarer FeMnNiCrBe- und FeMnNiCr(Ti, Al)-Legierungen wird mit Bezug zu den grundlegenden Untersuchungen dargestellt. Es wird gezeigt, dass die neu entwickelten ausscheidbaren FeMnNiCr(Ti, Al)-Legierungen eine vielversprechende Ausgangsbasis darstellen, afm Élinvarlegierungen technisch umzusetzen. / High manganese iron-base alloys are austenitic and antiferromagnetic (afm) at room temperature. By further alloying it is possible to tune the afm transition temperature (Néel temperature) near room temperature. FeMn-base alloys show extraordinary strain hardening as well as ductility because of Transformation Induced Plasticity (TRIP) and/or Twinning Induced Plasticty (TWIP), i.e. in dependence on composition the generally low stacking fault energy in these alloys allows for the mechanically induced formation of metastable phases (TRIP) or deformation twinning (TWIP). Furthermore, magnetic order causes distinct magnetovolume and magnetoelastic effects in these afm FeMn-base alloys. The investigated FeMnNiCr-base alloys are therefore prototypic for afm Élinvar alloys. However, as Élinvar is meant for invariant elasticity, an application as temperature compensated alloy with constant elastic modulus requires the smoothing of the pronounced magnetic anomalies, that is not industrially available yet. The advantage of afm Élinvar alloys in precision mechanics applications, would be their impassiveness with respect to magnetic fields that is not achievable by their ferromagnetic counterparts. For precision components like mechanic oscillators not only the tuning of the magnetoelastic properties but also the processing, cold formability and mechanical properties as well as their interplay have strong influence. Therefore this work addresses the applicability of the studied FeMnNiCr alloys. Elementary investigations on plasticity characterise the occurrence of TWIP in these alloys and propose a modell for deformation twinning at low plastic strains that allows for an estimation of the stacking fault energy. The investigations on the antiferromagnetism of deformed samples show the appearance of thermoremanent magnetisation (TRM). Its magnitude scales with the degree of deformation. The TRM is therefore attributed to uncompensated moments in the afm spin structure due to deformation induced defects. These are magnetised by a magnetic field cooling and couple to the afm matrix by exchange interaction below the Néel temperature. The complex thermomagnetic behaviour of the uncompensated moments is experimentally described and phenomenologically explained. The further development and assessment of engineering-grade pecipitable FeMnNiCrBe and FeMnNiCr(Ti, Al) alloys is presented in relation to the aforementioned elementary investigations. It is shown that the newly developped precipitable FeMnNiCr(Ti, Al) alloys are good candidates for afm Élinvar alloys in application.

This work has been devoted to the study of phase transformations involving chemical ordering and magnetic properties evolution in bulk Fe-Pt alloys composed of nanometer-sized grains. A comprehensive study of phase transformations and ordering in Fe-Pt alloys is performed by a combination of in-situ neutron powder diffraction and thermal analysis. The dependence of ordering processes on the alloy composition and initial microstructure (homogeneous A1 phase or multilayer-type) is established. Through the use of mechanical alloying and subsequent heat treatment it has been possible to achieve the formation of chemically highly ordered L10 FePt and, in the case of the Fe-rich and Pt-rich compositions, L12 Fe3Pt and FePt3 phases, respectively. Whereas in Pt-rich alloys the decoupling effect of the FePt3 phase leads to coercivity improvement, in Fe-rich nanocomposites a peculiar nanometer scale multilayer structure gives rise to remanence enhancement due to large effects of exchange interactions between the crystallites of the phases. The structure, magnetic properties and magnetisation reversal processes of these alloys are investigated. Experimentally observed phenomena are understood on the basis of a simple two-particle interaction model. Neutron diffraction has also been used for the investigation of the magnetic structure of ordered and partially ordered nanocrystalline Fe-Pt alloys. It has been shown that the magnetic moment of Fe atoms in L10-type Fe Pt alloys is sensitive to the compositional order. The results are compared to density functional calculations.

Charakteristisch für Perlcellulose als Regenerat vom Typ Cellulose II sind sphärisch geformte, poröse Partikel mit einer hohen spezifischen Oberfläche und einer guten Bioverträglichkeit. Aufgrund ihrer Eigenschaften sind diese Cellulosemikropartikel besonders gut für medizinische Anwendungen geeignet. Im Mittelpunkt der Arbeit standen Herstellung, Charakterisierung und Modifizierung von Perlcellulosen mit Partikelgrößen von etwa 1 bis10 µm. Im Rahmen der Arbeit wurden zunächst sieben technische Cellulose-2,5-acetate mit vergleichbaren molekularen Eigenschaften auf ihre Eignung zur Herstellung von Perlcellulose nach dem in EP0750007 beschriebenen Acetatverfahren untersucht. Dabei erfolgte der Vergleich verschiedener Eigenschaften. Aus allen untersuchten Celluloseacetaten können Perlcellulosen synthetisiert werden. Als besonders geeignet erwies sich ein Produkt mit einer Molmasse von über 100.000, einem Verhältnis der Molmasse zur numerischen Molmasse von etwa 1,5 und einer guten Löslichkeit in Ethylacetat / Methanol (100:17,5). Die hergestellte Perlcellulose hat eine geringe Partikelgröße und eine relativ enge Größenverteilung. Damit erfüllt dieses Cellulose-2,5-acetat alle Anforderungen für die Synthese von Perlcellulose. Der entscheidende Verfahrensschritt zur Herstellung von Perlcellulose ist das Dispergieren der Emulsion mittels Inline-Ultraturrax. Die Partikelgrößenverteilung hängt im Wesentlichen von den Bedingungen während des Dispergierens ab. Im Rahmen der Arbeit gelang die reproduzierbare Herstellung von Cellulosemicrospheres mit einer Partikelgröße unter 5 µm. Für die Herstellung von Cellulosemikropartikeln mit definierten Eigenschaften ist neben den Synthesebedingungen auch die Charakterisierung der Perlcellulosen von entscheidender Bedeutung. Dafür wurden zunächst etablierte Verfahren verwendet (Partikelgrößenmessung, REM und Quecksilberporosimetrie). Parallel dazu erfolgte die Entwicklung bzw. Einführung neuer Methoden. Im Vordergrund stand die Untersuchung des Sedimentationsverhaltens der Perlcellulosen durch analytisches Zentrifugieren. Davon ausgehend konnte ein Verfahren zur Berechnung der Porosität aus dem Sedimentationsvolumen entwickelt werden. Zum Nachweis der kompletten Deacetylierung der Proben wurde die Ramanspektroskopie genutzt. Durch die Anwendung vorhandener und die Entwicklung neuer Methoden wird die genaue Einstellung von Eigenschaften der Perlcellulosen (z. B. Partikelgröße und deren Verteilung, Morphologie sowie Porosität) ermöglicht und deren Reproduzierbarkeit gewährleistet. Weitere Versuche hatten die Entwicklung von Endotoxinadsorbermaterial auf Basis von Perlcellulose und Polymyxin-B-sulfat (PMB) zum Ziel. Die Kopplung des PMB erfolgte meist nach Aktivierung der Proben mit Epichlorhydrin. Zunächst wurde die eingesetzte Epichlorhydrinmenge variiert, um das Optimum für die Aktivierung der Perlcellulosen zu finden. Weiterhin wurden unterschiedliche Mengen PMB angebunden und die Anbindung an nicht aktivierte Proben untersucht. Die Planung aller Versuche erfolgte jeweils nach Auswertung der an der Donau-Universität Krems durchgeführten Limulus- Amöbocyten- Lysat (LAL)-Tests. Mittels dieser Batchtests wurde die Wirksamkeit des Endotoxinadsorbermaterials sowohl im Vergleich zu unbehandeltem Blutplasma und als auch zu kommerziell erhältlichen Adsorbern auf Polystyrenbasis getestet. Endotoxinadsorber, die bei diesen Tests besonders gut bewertet wurden, konnten in einem up-scale- Versuch erstmals in größeren Mengen synthetisiert werden. Auch die direkte Herstellung von Endotoxinadsorbermaterial aus Perlcelluloseacetat konnte realisiert werden. Bei diesem neu entwickelten Verfahren erfolgen Deacetylierung und Aktivierung in einem Schritt. Damit kann die Herstellung vereinfacht werden. Zur Gewährleistung der Erstfehlersicherheit in extrakorporalen Blutreinigungssystemen sollen magnetisierte Perlcellulosepartikel als bioverträgliche Marker eingesetzt werden. Versuche zur Magnetisierung von Cellulosemikropartikeln während des Herstellungsprozesses zeigten, dass die Einbindung von Magnetit bei Erhalt der sphärischen Partikelstruktur prinzipiell auch auf diesem homogenen Syntheseweg möglich ist. / Bead cellulose is regenerated cellulose II characterized by spherically shaped, porous particles with a high specific surface and a good biocompatibility. Because of their properties these cellulose microspheres are especially suited for medical applications. The focus of this work was the synthesis, characterization and modification of bead cellulose with particle sizes between 1 to 10 µm. In the frame of this work seven technical cellulose-2.5-acetates were investigated with regard to their suitability for making bead cellulose according to the process described in EP0750007. These cellulose acetates have comparable molecular characteristics. Different properties were compared. Bead celluloses can be synthesized from all investigated cellulose acetates. A product with a molecular weight of more than 100,000 and with a ratio between molecular weight and numeric molecular weight of about 1.5 is special suited. This cellulose-2.5-acetate has a good solubility in ethyl acetate / methanol (100:17.5). The bead cellulose made from it has a low particle size and a relative narrow size distribution. Thus this cellulose acetate complies with the requirements for making bead cellulose. The most important process step for making bead cellulose is the dispersing of the emulsion using an inline-ultraturrax. The distribution of particle size depends mainly on the conditions during dispersing. A reproducible synthesis of cellulose microspheres with a particle size range below 5 µm was successfully achieved. In addition to determining conditions for manufacturing bead cellulose the characterization of the microspheres is essential to obtain bead cellulose with well defined properties. At first well-established methods of characterization were used (particle size measurement, SEM and mercury porosimetry). In parallel new methods were developed and implemented. The main focus was the investigation of sedimentation behaviour of bead cellulose using analytical centrifugation. Based on this knowledge of the sedimentation volume a new method to calculate the porosity was designed. Raman spectroscopy was used for detecting the complete deacetylation of the samples. By using well-established and newly developed methods properties of bead cellulose such as particle size and distribution, morphology and porosity can be accurately adjusted. In this way the reproducible synthesis of cellulose microspheres can be ensured. The aim of further experiments was to develop an endotoxin adsorber material based on a coupling of bead cellulose with Polymyxin B sulfate (PMB). The coupling with PMB was carried out after activation of the samples by using epichlorohydrin. At first the added epichlorohydrin amount was diversified in order to find the optimum for the activation of bead cellulose. Later the coupling of different amounts of PMB took place and the linking of PMB to non activated samples was investigated too. The planning of all experiments occurred after evaluation of Limulus amebocyte lysate (LAL) tests at Danube-University Krems. Using these batch tests the effectiveness of the endotoxin adsorber material was tested compared to untreated blood plasma as well as commercial available adsorbers based on polystyrene. Endotoxin adsorbers showing the best adsorption rate were then synthesized for the first time in larger quantities. Also the direct synthesis of endotoxin adsorber material based on bead cellulose acetate could be realized. Using this newly developed method, deacetylation and activation occur during the same step. This means manufacturing process can be simplified. Using magnetized bead cellulose as biocompatible marker particles is planned to achieve first fault safety in case of a membrane rupture during extracorporeal blood purification. Initial tests have shown that the magnetization of cellulose microspheres is possible during the manufacturing process. The incorporation of magnetite can be realized while keeping the spherical shape of the particles using this homogenous synthesis pathway.

This work has been devoted to the study of phase transformations involving chemical ordering and magnetic properties evolution in bulk Fe-Pt alloys composed of nanometer-sized grains. A comprehensive study of phase transformations and ordering in Fe-Pt alloys is performed by a combination of in-situ neutron powder diffraction and thermal analysis. The dependence of ordering processes on the alloy composition and initial microstructure (homogeneous A1 phase or multilayer-type) is established. Through the use of mechanical alloying and subsequent heat treatment it has been possible to achieve the formation of chemically highly ordered L10 FePt and, in the case of the Fe-rich and Pt-rich compositions, L12 Fe3Pt and FePt3 phases, respectively. Whereas in Pt-rich alloys the decoupling effect of the FePt3 phase leads to coercivity improvement, in Fe-rich nanocomposites a peculiar nanometer scale multilayer structure gives rise to remanence enhancement due to large effects of exchange interactions between the crystallites of the phases. The structure, magnetic properties and magnetisation reversal processes of these alloys are investigated. Experimentally observed phenomena are understood on the basis of a simple two-particle interaction model. Neutron diffraction has also been used for the investigation of the magnetic structure of ordered and partially ordered nanocrystalline Fe-Pt alloys. It has been shown that the magnetic moment of Fe atoms in L10-type Fe Pt alloys is sensitive to the compositional order. The results are compared to density functional calculations.