Les ferrofluides sont des suspensions colloïdales de nanoparticules magnétiques dispersées dans un liquide porteur. La possibilité de moduler les propriétés des ferrofluides in situ en appliquant un champ magnétique externe leur procure un fort potentiel d’étude, à la fois d’un point de vue fondamental ou pour des applications industrielles variées. En particulier, les nanospinels de ferrite ferrimagnétiques MFe2O4 (M = Fe2+, Co2+, Mn2+…) sont largement étudiés pour leurs propriétés électriques et magnétiques. Plus spécifiquement, une forte énergie d’anisotropie de ces matériaux à l’échelle nanométrique est requise pour des applications dans le stockage de l’information ou l’hyperthermie pour lesquels ils sont considérés. Une connaissance fine des mécanismes régissant ces propriétés d’anisotropies magnétiques est ainsi primordiale pour la création de nouveaux objets aux propriétés magnétiques contrôlées à l’échelle nanométrique. L’originalité de notre approche consiste à utiliser une technique fine du magnétisme, le dichroïsme magnétique circulaire des rayons X (XMCD) à l’étude des anisotropies et couplages magnétiques des nanospinels composants les ferrofluides. Au cours de cette thèse, nous nous sommes intéressés à différentes stratégies possibles pour induire une forte énergie d’anisotropie aux nanospinels de ferrite par l’utilisation de cobalt. Des nanoparticules de tailles et compositions variées ont été obtenues par différentes voies de synthèse, et nous démontrons que l’anisotropie magnétique de ces systèmes est fortement gouvernée par la symétrie de site du Co2+ en structure spinel qui peut être directement corrélé au processus de synthèse utilisé. Nous nous sommes aussi intéressés à l’ordre et au couplage magnétique de ferrite spinels structurés en coeur-coquille, dont le cœur et la coquille sont réalisés à partir de matériaux aux propriétés magnétiques intrinsèques différentes. Nous montrons ainsi que pour des nanospinels MnFe2O4@CoFe2O4, la très fine coquille formée de CoFe2O4 impose une forte anisotropie magnétique au cœur doux de MnFe2O4. Enfin, nous nous sommes intéressés à une troisième classe de ferrofluide à base de nanospinels, les ferrofluides binaires, constitué d’un mélange physique de ferrofluides aux propriétés magnétiques intrinsèques différentes. Pour de tels systèmes, il est essentiel de préserver le liquide porteur du ferrofluide pour ne pas dénaturer les interactions entre particules existantes. L’un des objectifs de cette thèse fut donc d’étendre la technique du XMCD à l’étude d’échantillons de ferrofluides in situ, dans leur phase liquide ou gelée. Nous avons débuté la conception d’une cellule liquide compatible avec les rayons X mous et un environnement ultra-vide sur la ligne de lumière DEIMOS (SOLEIL) qui est toujours en développement... / Ferrofluids are colloidal suspensions of magnetic nanoparticles dispersed in a carrier liquid. The intimate interaction between the magnetic nanoparticles and the liquid provides a unique system, from both fundamental and industrial application point of views, whose flow and properties can be precisely controlled using an external magnetic field. Magnetic nanoparticles of spinel ferrites MFe2O4 (M = Fe2+, Co2+, Mn2+…) are of particular scientific interest and have been extensively studied for their electrical and magnetic properties. Spinel ferrites find potential applications, notably in storage devices, for computers, or hyperthermia, for cancer treatment, where high magnetic anisotropy energies are required at the nanoscale. However, deeper knowledges of the fine mechanisms playing a significant role on the magnetic anisotropies existing in the nanospinels are necessary to help the creation of rationalized materials with controlled magnetic anisotropies for the requirement of the system. In this thesis, we have used X-ray Magnetic Circular Dichroism (XMCD) as an original approach for probing the magnetic anisotropies and magnetic couplings of nanospinels obtained in ferrofluids. The nanoparticles are iron bearing spinels for which cobalt ions have been introduced in the spinel structure of the nanoparticles as a true makers of magnetic anisotropy. First, magnetic nanospinels have been synthesized by tuning their size and composition and using different synthesis processes. XMCD investigations revealed that the coercive field of the nanospinels is governed by the concentration of Co2+ ions sitting in octahedral sites of the spinel structure, and this can be directly linked to some synthesis parameters. Then, we have investigated core@shell nanoparticles, which can be synthesized with an appropriate choice of magnetic anisotropies for the core and the shell in order to tailor optimal magnetic properties. In the case of MnFe2O4@CoFe2O4, our findings reveal that the very thin CoFe2O4 shell imposes a strong magnetic anisotropy to the otherwise very soft MnFe2O4 core. The other class of ferrofluids that has been investigated during this thesis are binary ferrofluids that are constituted of two different types of magnetic nanoparticles. For such systems, the carrier liquid must be preserved to understand the magnetic interactions in the ferrofluid as they are. Another motivation of this thesis was thus to extend XMCD to the in situ investigation of the nanospinels dispersed in ferrofluids. We have been started a liquid cell development in the DEIMOS beamline at SOLEIL. The setup is still in progress and is aimed at being compatible with soft X-Rays short penetration depth and ultra-high vacuum environment. Hard X-ray photon-in/photon-out spectroscopy coupled to XMCD (1s2p RIXS-MCD) can be a very valuable alternative to soft X-ray XMCD at K-edge of 3d elements when liquid cell sample environment is required. The instrumental development of a liquid cell used with 1s2p RIXS-MCD spectroscopy allowed us to investigate the nanoparticles directly in the ferrofluids revealing interparticles magnetic couplings in binary ferrofluids.

The capability to probe the dispersion of elementary spin, charge, orbital, and lattice excitations has positioned resonant inelastic x-ray scattering (RIXS) at the forefront of photon science. In this work, we will investigate how RIXS can contribute to a deeper understanding of the orbital properties and of the pairing mechanism in unconventional high-temperature superconductors. In particular, we will show how direct RIXS spectra of magnetic excitations can reveal long-range orbital correlations in transition metal compounds, by discriminating different kind of orbital order in magnetic and antiferromagnetic systems. Moreover, we will show how RIXS spectra of quasiparticle excitations in superconductors can measure the superconducting gap magnitude, and reveal the presence of nodal points and phase differences of the superconducting order parameter on the Fermi surface. This can reveal the properties of the underlying pairing mechanism in unconventional superconductors, in particular cuprates and iron pnictides, discriminating between different superconducting order parameter symmetries, such as s, d (singlet pairing) and p wave (triplet pairing).

L’utilisation de sulfure de lithium (Li2S), homologue lithié du soufre, en tant que matériau actif d’électrode positive permet le développement de systèmes à haute densité d’énergie plus sûrs, sans lithium métallique en tant qu’électrode négative. Appelés accumulateurs Lithium-ion/Soufre, leur mécanisme de fonctionnement, complexe et mal compris, reste semblable à celui des accumulateurs Li/S à l’exception de la 1ère charge. Cette 1ère charge présente une forte polarisation et très peu reproductible. L’objectif principal de ces travaux de thèse a donc été dédié à la compréhension des mécanismes mis en jeu lors de la charge initiale d’un accumulateur Li-ion/Soufre et en particulier l’étude de l’influence de la taille des cristallites du matériau pristine Li2S sur le mécanisme électrochimique. Il a été mis en évidence, dans un premier temps, la forte réactivité du Li2S nanométrique avec le liant polymère conventionnel PVdF (Polyfluorure de Vinylidène) conduisant au développement d’une nouvelle formulation d’électrode à base de PEO (Polyoxyde d’éthylène). Les résultats électrochimiques ont montré que la diminution de la taille des cristallites couplée à l’augmentation de la surface BET permet d’abaisser drastiquement la polarisation de la 1ère charge. Des caractérisations operando, de Diffraction des Rayons X (DRX) et Diffusion Inélastique Résonante des rayons X (RIXS), ont permis d’interpréter le rôle de la taille des cristallites et de la surface BET. Les résultats DRX ont montré une coexistence du Li2S et du β-S8 tout au long de la charge du Li2S micrométrique, tandis qu’aucun intermédiaire polysulfure soluble n’a été détecté par RIXS. Ces résultats sous-entendent donc une possible conversion solide/solide directe (Li2S micrométrique-->S8). A l’inverse, un comportement classique (Li2S-->Polysulfure solubles-->S8) a été observé à partir d’un matériau Li2S nanométrique avec l’existence successive des deux phases solides et la présence de polysulfures en solution. / Using Li2S instead of S8 as active material allows metallic lithium free batteries, also called Lithium-ion/Sulfur batteries, to be developed and safer systems with high energy density to be designed. The main difference between S8 and Li2S-based systems lies in the first charge. Indeed, during this first charge, a high polarization occurs with lack of reproducibility. Then, the main goal of this work is to focus on the analysis and understanding of the Li2S particle size impact on the electrochemical mechanism during the first charge of a Li-ion/Sulfur battery. Three Li2S types have been studied in this work: two nanometric Li2S and a micrometric one. Firstly, classical PVdF (polyvinylidenefluoride) binder was demonstrated to be highly reactive with nanometric Li2S leading to a new formulation based on PEO (polyethylene oxide) to be developed. Electrochemical investigations confirmed that starting with Li2S nanoparticles can effectively suppress the overall charge polarization. To go deeper, operando characterizations such as X-Ray Diffraction (XRD) and Resonant Inelastic X-ray Scattering (RIXS) have been carried out in order to correlate the particle size and the BET surface area effects. XRD results show that Li2S and β-sulfur phases coexist almost all along the first charge when starting with micrometric Li2S, while no polysulfides are detected by RIXS analysis. Therefore, a solid/solid (micrometric Li2S-->S8) reaction is suggested when using micrometric Li2S. On the opposite, when starting with nanometric Li2S particles, a very classical behavior (Li2S-->Polysulfides in solution-->S8) is obtained with the successive existence of the two solid phases with polysulfides in solution.

In the last decade, a large number of studies have been devoted to the peculiarities of correlated physics found in the quasi-two-dimensional square lattice iridium oxides. It was shown that this 5d family of transition metal oxides has strong structural and electronic similarities to the famous 3d family of copper oxides. Moreover, a delicate interplay of on-site spin-orbit coupling, Coulomb repulsion and crystalline electric field interactions is expected to drive various exotic quantum states. Many theoretical proposals were made in the last decade including the prediction of possible superconductivity in square-lattice iridates emerging as a sister system to high-Tc cuprates, which however met only limited experimental confirmation. One can, therefore, raise a general question: To what extent is the low-energy physics of the quasi-two-dimensional square-lattice iridium oxides different from other transition metal oxides including cuprates? In this thesis we investigate some of the effects which are usually neglected in studies on iridates, focusing on quasi-two-dimensional square-lattice iridates such as Sr2IrO4 or Ba2IrO4. In particular, we discuss the role of the electron-phonon coupling in the form of Jahn-Teller interaction, electron-hole asymmetry introduced by the strong correlations and some effects of coupling scheme chosen to calculate multiplet structure for materials with strong on-site spin-orbit coupling. Thus, firstly, we study the role of phonons, which is almost always neglected in Sr2IrO4, and discuss the manifestation of Jahn-Teller effect in the recent data obtained on Sr2IrO4 with the help of resonant inelastic x-ray scattering. When strong spin-orbit coupling removes orbital degeneracy, it would at the same time appear to render the Jahn-Teller mechanism ineffective. We show that, while the Jahn-Teller effect does indeed not affect the antiferromagnetically ordered ground state, it leads to distinctive signatures in the spin-orbit exciton. Second, we focus on charge excitations and determine the motion of a charge (hole or electron) added to the Mott insulating, antiferromagnetic ground-state of square-lattice iridates. We show that correlation effects, calculated within the self-consistent Born approximation, render the hole and electron case very different. An added electron forms a spin-polaron, which closely resembles the well-known cuprates, but the situation of a removed electron is far more complex. Many-body configurations form that can be either singlets and triplets, which strongly affects the hole motion. This not only has important ramifications for the interpretation of angle-resolved photoemission spectroscopy and inverse photoemission spectroscopy experiments of square lattice iridates, but also demonstrates that the correlation physics in electron- and hole-doped iridates is fundamentally different. We then discuss the application of this model to the calculation of scanning tunneling spectroscopy data. We show that using scanning tunneling spectroscopy one can directly probe the quasiparticle excitations in Sr2IrO4: ladder spectrum on the positive bias side and multiplet structure of the polaron on the negative bias side. We discuss in detail the ladder spectrum and show its relevance for Sr2IrO4 which is in general described by more complicated extended t-J -like model. Theoretical calculation reveals that on the negative bias side the internal degree of freedom of the charge excitation introduces strong dispersive hopping channels encaving ladder-like features. Finally, we discuss how the choice of the coupling scheme to calculate multiplet structure can affect the theoretical calculation of angle-resolved photoemission spectroscopy and scanning tunnelling spectroscopy spectral functions.

This thesis covers different problems that arise due to crystal and pseudospin anisotropy present in 3d and 5d transition metal oxides. We demonstrate that the methods of computational quantum chemistry can be fruitfully used for quantitative studies of such problems. In Chapter 2, Chapter 3, and Chapter 7 we show that it is possible to reliably calculate local multiplet splittings fully ab initio, and therefore help to assign peaks in experimental spectra to corresponding electronic states. In a situation of large number of peaks due to low local symmetry such assignment using semi-empirical methods can be very tedious and non-unique. Moreover, in Chapter 4 we present a computational scheme for calculating intensities as observed in the resonant inelastic X-ray scattering and X-ray absorption experiments. In our scheme highly-excited core-hole states are calculated explicitly taking into account corresponding orbital relaxation and electron polarization. Computed Cu L-edge spectra for the Li2CuO2 compound reproduce all features present in experiment. Unbiased ab initio calculations allow us to unravel a delicate interplay between the distortion of the local ligand cage around the transition metal ions and the anisotropic electrostatic interactions due to second and farther coordination shells. As shown in Chapter 5 and Chapter 6 this interplay can lead to the counter intuitive multiplet structure, single-ion anisotropy, and magnetic g factors. The effect is quite general and may occur in compounds with large difference between charges of metal ions that form anisotropic environment around the transition metal, like Ir 4+ in plane versus Sr 2+ out of plane in the case of Sr2IrO4. An important aspect of the presented study is the mapping of the quantum chemistry results onto simpler physical models, namely extended Heisenberg model, providing an ab initio parametrization. In Chapter 5 we employ the effective Hamiltonian technique for extracting parameters of the anisotropic Heisenberg model with single-ion anisotropy in the case of quenched orbital moment and second-order spin-orbit coupling. Calculated strong easy-axis anisotropy of the same order of magnitude as the symmetric exchange is consistent with experimentally-observer all-in/all-out magnetic order. In Chapter 6 we introduce new flavour of the mapping procedure applicable to systems with first-order spin-orbit coupling, such as 5d 5 iridates based on analysis of the wavefunction and interaction with magnetic field. In Chapter 6 and Chapter 7 we use this new procedure to obtain parameters of the pseudospin anisotropic Heisenberg model. We find large antisymmetric exchange leading to the canted antiferromagnetic state in Sr2IrO4 and nearly ideal one-dimensional Heisenberg behaviour of the CaIrO3, both agree very well with experimental findings.

The capability to probe the dispersion of elementary spin, charge, orbital, and lattice excitations has positioned resonant inelastic x-ray scattering (RIXS) at the forefront of photon science. In this work, we will investigate how RIXS can contribute to a deeper understanding of the orbital properties and of the pairing mechanism in unconventional high-temperature superconductors. In particular, we will show how direct RIXS spectra of magnetic excitations can reveal long-range orbital correlations in transition metal compounds, by discriminating different kind of orbital order in magnetic and antiferromagnetic systems. Moreover, we will show how RIXS spectra of quasiparticle excitations in superconductors can measure the superconducting gap magnitude, and reveal the presence of nodal points and phase differences of the superconducting order parameter on the Fermi surface. This can reveal the properties of the underlying pairing mechanism in unconventional superconductors, in particular cuprates and iron pnictides, discriminating between different superconducting order parameter symmetries, such as s, d (singlet pairing) and p wave (triplet pairing).

This thesis covers different problems that arise due to crystal and pseudospin anisotropy present in 3d and 5d transition metal oxides. We demonstrate that the methods of computational quantum chemistry can be fruitfully used for quantitative studies of such problems. In Chapter 2, Chapter 3, and Chapter 7 we show that it is possible to reliably calculate local multiplet splittings fully ab initio, and therefore help to assign peaks in experimental spectra to corresponding electronic states. In a situation of large number of peaks due to low local symmetry such assignment using semi-empirical methods can be very tedious and non-unique. Moreover, in Chapter 4 we present a computational scheme for calculating intensities as observed in the resonant inelastic X-ray scattering and X-ray absorption experiments. In our scheme highly-excited core-hole states are calculated explicitly taking into account corresponding orbital relaxation and electron polarization. Computed Cu L-edge spectra for the Li2CuO2 compound reproduce all features present in experiment. Unbiased ab initio calculations allow us to unravel a delicate interplay between the distortion of the local ligand cage around the transition metal ions and the anisotropic electrostatic interactions due to second and farther coordination shells. As shown in Chapter 5 and Chapter 6 this interplay can lead to the counter intuitive multiplet structure, single-ion anisotropy, and magnetic g factors. The effect is quite general and may occur in compounds with large difference between charges of metal ions that form anisotropic environment around the transition metal, like Ir 4+ in plane versus Sr 2+ out of plane in the case of Sr2IrO4. An important aspect of the presented study is the mapping of the quantum chemistry results onto simpler physical models, namely extended Heisenberg model, providing an ab initio parametrization. In Chapter 5 we employ the effective Hamiltonian technique for extracting parameters of the anisotropic Heisenberg model with single-ion anisotropy in the case of quenched orbital moment and second-order spin-orbit coupling. Calculated strong easy-axis anisotropy of the same order of magnitude as the symmetric exchange is consistent with experimentally-observer all-in/all-out magnetic order. In Chapter 6 we introduce new flavour of the mapping procedure applicable to systems with first-order spin-orbit coupling, such as 5d 5 iridates based on analysis of the wavefunction and interaction with magnetic field. In Chapter 6 and Chapter 7 we use this new procedure to obtain parameters of the pseudospin anisotropic Heisenberg model. We find large antisymmetric exchange leading to the canted antiferromagnetic state in Sr2IrO4 and nearly ideal one-dimensional Heisenberg behaviour of the CaIrO3, both agree very well with experimental findings.

In the last decade, a large number of studies have been devoted to the peculiarities of correlated physics found in the quasi-two-dimensional square lattice iridium oxides. It was shown that this 5d family of transition metal oxides has strong structural and electronic similarities to the famous 3d family of copper oxides. Moreover, a delicate interplay of on-site spin-orbit coupling, Coulomb repulsion and crystalline electric field interactions is expected to drive various exotic quantum states. Many theoretical proposals were made in the last decade including the prediction of possible superconductivity in square-lattice iridates emerging as a sister system to high-Tc cuprates, which however met only limited experimental confirmation. One can, therefore, raise a general question: To what extent is the low-energy physics of the quasi-two-dimensional square-lattice iridium oxides different from other transition metal oxides including cuprates? In this thesis we investigate some of the effects which are usually neglected in studies on iridates, focusing on quasi-two-dimensional square-lattice iridates such as Sr2IrO4 or Ba2IrO4. In particular, we discuss the role of the electron-phonon coupling in the form of Jahn-Teller interaction, electron-hole asymmetry introduced by the strong correlations and some effects of coupling scheme chosen to calculate multiplet structure for materials with strong on-site spin-orbit coupling. Thus, firstly, we study the role of phonons, which is almost always neglected in Sr2IrO4, and discuss the manifestation of Jahn-Teller effect in the recent data obtained on Sr2IrO4 with the help of resonant inelastic x-ray scattering. When strong spin-orbit coupling removes orbital degeneracy, it would at the same time appear to render the Jahn-Teller mechanism ineffective. We show that, while the Jahn-Teller effect does indeed not affect the antiferromagnetically ordered ground state, it leads to distinctive signatures in the spin-orbit exciton. Second, we focus on charge excitations and determine the motion of a charge (hole or electron) added to the Mott insulating, antiferromagnetic ground-state of square-lattice iridates. We show that correlation effects, calculated within the self-consistent Born approximation, render the hole and electron case very different. An added electron forms a spin-polaron, which closely resembles the well-known cuprates, but the situation of a removed electron is far more complex. Many-body configurations form that can be either singlets and triplets, which strongly affects the hole motion. This not only has important ramifications for the interpretation of angle-resolved photoemission spectroscopy and inverse photoemission spectroscopy experiments of square lattice iridates, but also demonstrates that the correlation physics in electron- and hole-doped iridates is fundamentally different. We then discuss the application of this model to the calculation of scanning tunneling spectroscopy data. We show that using scanning tunneling spectroscopy one can directly probe the quasiparticle excitations in Sr2IrO4: ladder spectrum on the positive bias side and multiplet structure of the polaron on the negative bias side. We discuss in detail the ladder spectrum and show its relevance for Sr2IrO4 which is in general described by more complicated extended t-J -like model. Theoretical calculation reveals that on the negative bias side the internal degree of freedom of the charge excitation introduces strong dispersive hopping channels encaving ladder-like features. Finally, we discuss how the choice of the coupling scheme to calculate multiplet structure can affect the theoretical calculation of angle-resolved photoemission spectroscopy and scanning tunnelling spectroscopy spectral functions.

This thesis concerns studies of aqueous soft matter systems, especially surfactant micelle systems. The aim has been to study the molecular arrangement and electronic structure of the constituents of, as well as transport properties in such a system. The molecular arrangement and electronic structure has been studied by means of X-ray absorption spectroscopy (XAS) and resonant inelastic X-ray spectroscopy (RIXS). The transport properties have been investigated by low-frequency dielectric spectroscopy (LFDS) and small angle X-ray scattering (SAXS) as well as a theoretical modelling. The latter was based on Fick’s laws of the release from binary surfactant system and was validated by experiments. The RIXS and XAS measurements show the electronic structure in bulk water and the influence of the chemical surrounding of the water molecule in bulk water and of the water molecules confined in a micelle lattice. The spectra are highly dependent on the molecular arrangement in such systems. For glycine and sodium polyacrylate RIXS and XAS spectra show features which are unique for carboxyl and carboxylate groups and such measurements can thus be used for fingerprinting. The LFDS and SAXS measurements show a strong correlation between structure in a surfactant/poly-ion system and apparent mobility of surfactants. This conclusion is in line with earlier observations. By the theoretical modelling a predictive model for the surfactant release from a binary surfactant micelle system has been obtained and the importance of different factors for surfactant release has been further clarified.

Based on a previously completed design for a soft X-ray (50-1100 eV) emission spectrometer, an endstation was constructed for the Resonant Elastic and Inelastic X-ray Scattering (REIXS) beamline at the CLS. The optical design employed techniques and software tools developed in-house using ray-tracing and diffraction grating efficiency calculations to analyze and compare existing designs and to propose a new design with superior performance. This design employs Rowland circle geometry to achieve a theoretical resolving power in excess of 2,500 in our range of interest. In addition, a novel optical design for a larger extremely high resolution spectrometer has been completed to provide theoretical resolving powers exceeding 10,000 throughout the higher end of the spectrum. To accommodate this optical design a completely new mechanical design was required, involving significant mechanical, electrical, vacuum and software engineering. Countless custom fabricated parts were required along with numerous o -the-shelf secondary instruments and systems. All told, this entirely student-managed project has cost over $1.5M and taken over 5 years. Construction is finally complete and the endstation is currently being commissioned. Necessary design changes made during the mechanical design process resulting in the selection of a more suitable, but lower resolution, detector. This reduced the theoretical maximum resolving power to 1,800 for the first order gratings and roughly 5,000 for the third order gratings. Commissioning is still underway, but first order resolutions in the range of 1,000 - 2,000 have been recorded as have third order resolutions exceeding 4,000. Publication quality data has been collecting by members of this research group and invited external users have successfully grown and measured samples here. Two of the optical elements required rework and upon their delivery the system commissioning will be completed and peer-reviewed access will begin.