Orientadores: Abner de Siervo e Flavio Garcia / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin. / Made available in DSpace on 2018-08-12T11:00:44Z (GMT). No. of bitstreams: 1 Figueiredo_JoseJadsomSampaiode_M.pdf: 4194889 bytes, checksum: 489e23ac30c4fa90808ef2de00ae46d4 (MD5) Previous issue date: 2008 / Resumo: Nanopartículas (NPs) metálicas, filmes ultrafinos e sistemas nano-estruturados têm motivado vários estudos teóricos e experimentais devido a novas propriedades eletrônicas, óticas e magnéticas destes materiais. Nas NPs, tais propriedades estão diretamente relacionadas aos efeitos de tamanho (size effect) e superfície, os quais afetam suas estruturas eletrônica e cristalográfica, concentrando um maior número de átomos na superfície em relação ao seu volume. Estudos recentes têm sugerido que alguns metais nobres sem nenhum tipo de ordenamento ferromagnético espontâneo podem apresentá-lo, mesmo a temperatura ambiente, quando segmentados em forma de nanopartículas. Exemplos desses metais são Au e Pd. Apesar dos esforços para provar a existência de comportamento ferromagnético nestes materiais, não existem medidas que demonstrem de forma não ambígua a polarização ferromagnética nos átomos de Pd. Na maioria dos trabalhos mostrados na literatura, o ferromagnetismo é quantificado por magnetometria SQUID. Faz-se necessário, portanto, verificar a existência de ferromagnetismo nestes materiais utilizando-se de técnicas elemento específicas, tais como XMCD (dicroísmo circular magnético de raios X). Esta técnica é muito importante para determinação quantitativa e seletiva dos momentos magnéticos de spin e orbital para cada elemento separadamente. O nosso trabalho se diferencia em relação a estes, pois ao invés de estudarmos cada elemento individualmente trabalhamos com nanopartículas do composto bimetálico Aux Pd1-x (x =0,00, 0,25, 0,50, 0,75) passivadas por tiol. Medidas de SQUID indicaram claramente um sinal magnético nestas nanopartículas, inclusive à temperatura ambiente. Neste trabalho apresentaremos os resultados preliminares da caracterização destas nanopartículas por medidas de SQUID, XANES, XPS, bem como XMCD nas bordas L2,3 do Pd e L3 do Au . Ressaltamos também, o desenvolvimento de uma instrumentação para medidas de XMCD em ambientes de UHV onde é possível submeter às amostras a altos campos magnéticos e condições de baixas temperaturas. Com base nos resultados obtidos discutiremos a possibilidade de ferromagnetismo nestas nanopartículas e sua origem. / Abstract: Metallic Nanoparticles (NP's), ultra-thin films and nano-structured systems have motivated several theoretical and experimental works because of their new electronic optical and magnetic properties of these materials. In NP's, such properties are directly related to the size and surface effects, which affect its electronic and crystallographic structures, by concentrating a larger number of atoms on the surface in relation to their volume. Recent studies have suggested that some noble metals without any kind of spontaneous ferromagnetic order can exhibit it, even at room temperature, when segmented as nanoparticles. Examples of these metals are Au and Pd. Despite the efforts to prove the existence of a ferromagnetic behavior in such materials, there are no measurements that demonstrate unambiguously the ferromagnetic polarization in Pd atoms, for example. Most of the works shown in the literature have used SQUID measurements to characterize the ferromagnetism in these materials. It is therefore necessary to verify the existence of ferromagnetism in these materials by using element specific techniques, such as XMCD (X-ray magnetic circular dichroism). XMCD is a very important technique, which is element selective and performs quantitative determination of the spin and orbital magnetic contributions for each element separately. Our work has a particular difference when compared to others in literature because instead of studying each element individually we have worked with NP's of the bimetallic compound Aux Pd1-x (x = 0.00, 0.25, 0.50, 0.75) capped by thiol. SQUID measurements clearly showed ferromagnetic behavior in these nanoparticles, even at room temperature. In this thesis we will present results of a systematic characterization of these nanoparticles by a multi-technique approach which involves measurements with SQUID, XANES, XPS and XMCD in the L2,3 edges of the Pd and Au L3 edge. We also emphasize the development of a new instrumentation for XMCD measurements at UHV where samples can be submitted to high magnetic fields and conditions of low temperatures. Based on these findings we discuss the possibility of ferromagnetism in these nanoparticles and its origin. / Mestrado / Física da Matéria Condensada / Mestre em Física

L'importance des effets relativistes dans la chimie a été reconnu depuis les années 1980. Par exemple, sans la relativité (a) l'or aurait la même couleur que l'argent (b) le mercure ne serait pas liquide à la température ambiante et (c) nos voitures ne démarrent pas avec une batterie de plomb. Pour une description théorique de la structure et la réactivité des éléments lourds, la relativité est un ingrédient essentiel. Le hamiltonien pour les calculs moléculaires relativistes à 4 composantes est construit en remplaçant la partie mono-électronique de l'hamiltonien électronique non-relativiste par le hamiltonien de Dirac. La partie bi-électronique est approchée par le terme de r Coulomb comme dans le cas non relativiste, ce qui donnel'hamiltonien de Dirac-Coulomb (DC). Pour réduire le coût de calcul, on peut utiliser des hamiltoniens relativistes à 2 composantes. Parmi eux, l'hamiltonien exact à 2 composantes (X2C) est le plus précise. La corrélation électronique est, cependant, une contribution très importante pour obtenir une description théorique à la fois qualitative et quantitative des spectroscopies moléculaires, réactions, etc. Dans cette thèse, nous avons étudié l'interaction entre la relativité et de la corrélation. à la fois par des développements méthodologiques et par des applications moléculaires. Dans la première partie de la thèse, nous avons calculé les constantes spectroscopiques dimères des gaz rares lourds. La liaison faible de ces dimères ne peut être décrit que par l'inclusion de la corrélation électronique. Les dimères des gaz rares les plus lourds, le radon et l'eka-radon, nécessite de plus un traitement adéquat de la relativité. Nos calculs sont basés sur l'hamiltonien X2Cmmf, à la fois avec des méthodes de corrélation basés sur une fonction d'onde et séparation de porte (srDFT). La deuxième partie de cette thèse concerne la simulation de la spectroscopie des rayons X, où l'on sonde la région du cœur d'une molécule, ou la relativité joue un rôle très important. Nous avons étudié la spectroscopie L-edge de la série isoélectronique: UO22 +, UNO+, et UN2, où le couplage spin-orbite joue un rôle majeur. Au niveau des méthodes, nous avons considéré MP2 à couches ouvertes et la théorie de la fonctionnelle de la densité dépendante de temps (TDDFT). Dans un autre étude, nous avons simulé la spectroscopie K-edge de la série H2X (X = O, S, Se, Te) et XH3 (X = N, P, As) ainsi que les molécules N2 et N2O2. Pour ces systèmes, l'interaction spin-orbite est moins important. Par conséquent, nous avons utilisé un hamiltonien DC sans spin (SF). Certains des systèmes pris en compte dans ce travail sont de caractère multi-référentielles ; nous avons utilisé la methode Coupled Cluster Multi-référentielle de type State Universal et adapté au groupe unitaire (UGA-SUMRCC) comme une méthode de corrélation. Dans la troisième et partie principale de la thèse, l'attention est de nouveau sur la relativité et de la corrélation, mais pour le calcul des propriétés électriques et magnétiques moléculaires. Nous avons développé et mis en œuvre un module pour le calcul des valeurs moyennes au niveau relativiste à 4-composantes coupled cluster monoréferentiel. Les propriétés qui sondent la densité électronique près de noyaux (lourds), telles que la résonance paramagnétique électronique (RPE), les paramètres des gradients de champ électrique et la non-conservation de la parité (NCP) des molécules chirales ,sont parfaitement adaptés pour l'application de cette méthode. Pour l'instant, nous avons étudié que la NCP. Ce module dans le logiciel DIRAC pour les calculs moléculaires relativistes fournit un cadre propice pour la mise en œuvre de méthodes de CC relativistes employant la symétrie de groupes doubles et de permutation de manière très efficace. En perspective, nous ciblons la mise en œuvre de la réponse linéaire CC pour le calcul des énergies d'excitation et propriétés moléculaires de second ordre tels que les paramètres de RMN. / The importance of relativistic effects in chemistry has been recognized since the 1980s. Without relativity (a) gold would have the same colour as silver (b) mercury would not be liquid at room temperature (c) our cars would not start (lead-battery). For a theoretical description of the structure and reactivity of heavy-elements, relativity is considered as an essential ingredient. The Hamiltonian for the 4-component relativistic molecular calculations is constructed by replacing the one-electronic part of the non-relativistic molecular Hamiltonian by the Dirac Hamiltonian. The two-electronic part of the Hamiltonian is approximated by the Coulombic repulsion term as in the non-relativistic case. The resulting Hamiltonian is called the Dirac-Coulomb (DC) Hamiltonian. For chemical applications there exist a class of relativistic Hamiltonians, where one-electronic part of the DC Hamiltonian is transformed to a 2-component one. Among them the eXcact 2-component (X2C) Hamiltonian is the most accurate one. Electron correlation, however, is a very important contribution to achieve a both qualitative and quantitative correct description of molecular spectroscopies, reactions etc. It is, therefore, essential to study the interplay between relativity and correlation. In this thesis, we have studied this interplay both in terms methodological developments and molecular applications. In the first part of the thesis we have studied the spectroscopic constants of the heavy rare gas dimers. The weak bonding of those dimers can only be described by the inclusion of electron correlation. The heavier analogues in the rare gas series i.e, Radon and eka-Radon, in addition require adequate treatment of relativity. Our calculations are based on the eXact 2-Component molecular-mean field (X2Cmmf) Hamiltonian both with wave function methods and range-separated DFT methods. The second part of this thesis simulates X-ray spectroscopy, where one probes the core region of a molecule. In the core region relativity plays a very significant role. Removal and excitation of electrons from that region involve various processes, which are beyond a mean-field description. We have studied L-edge spectroscopy of the isoelectronic series: UO22+, UNO+, and UN2, where spin-orbit coupling plays a major role. For the theory we have considered single reference open-shell MP2 and Time Dependent Density functional Theory (TDDFT). In another work, we have studied K-edge spectroscopy of the H2X (X= O, S, Se, Te) and XH3 (X= N, P, As) series as well as N2, N2O2 molecules. For this study spin-orbit coupling is less important, therefore, we have treated them with the Spin-Free (SF) DC Hamiltonian. Some of the systems considered in this work are Multi-Reference in nature; we have used Unitary Group Adapted (UGA) State Universal Multi-reference Coupled Cluster (UGA-SUMRCC) theory as a correlation method. In the third and major part of the thesis, the thrust is again on relativity and correlation, but for the calculation of molecular electric and magnetic properties. We have developed and implemented a module for the calculation of expectation values at the 4-component Relativistic Single Reference Coupled Cluster level. Properties that probe the electron density near (heavy) nuclei, such as Electron Paramagnetic Resonance (EPR) parameters, electric field gradients and parity non-conservation (PNC) in chiral molecules are ideally suited for the application of this method. However, we have only studied PNC so far. This module in the DIRAC software for relativstic molecular calculations provides a convenient framework for the implementation of relativistic CC methods employing double group and permutation symmetry very efficiently. In the near future we therefore target the implementation of Linear Response CC for the calculation of excitation energies and second-order molecular properties such as NMR parameters.

Concrete substructures are often subjected to environmental deterioration, such as sulfate and acid attack, which leads to severe damage and causes structure degradation or even failure. In order to improve the durability of concrete, the High Performance Concrete (HPC) has become widely used by partially replacing cement with pozzolanic materials. However, HPC degradation mechanisms in sulfate and acidic environments are not completely understood. It is therefore important to evaluate the performance of the HPC in such conditions and predict concrete service life by establishing degradation models. This study began with a review of available environmental data in the State of Florida. A total of seven bridges have been inspected. Concrete cores were taken from these bridge piles and were subjected for microstructural analysis using Scanning Electron Microscope (SEM). Ettringite is found to be the products of sulfate attack in sulfate and acidic condition. In order to quantitatively analyze concrete deterioration level, an image processing program is designed using Matlab to obtain quantitative data. Crack percentage (Acrack/Asurface) is used to evaluate concrete deterioration. Thereafter, correlation analysis was performed to find the correlation between five related variables and concrete deterioration. Environmental sulfate concentration and bridge age were found to be positively correlated, while environmental pH level was found to be negatively correlated. Besides environmental conditions, concrete property factor was also included in the equation. It was derived from laboratory testing data. Experimental tests were carried out implementing accelerated expansion test under controlled environment. Specimens of eight different mix designs were prepared. The effect of pozzolanic replacement rate was taken into consideration in the empirical equation. And the empirical equation was validated with existing bridges. Results show that the proposed equations compared well with field test results with a maximum deviation of ± 20%. Two examples showing how to use the proposed equations are provided to guide the practical implementation. In conclusion, the proposed approach of relating microcracks to deterioration is a better method than existing diffusion and sorption models since sulfate attack cause cracking in concrete. Imaging technique provided in this study can also be used to quantitatively analyze concrete samples.

Un-doped, Tb3+ as well as Yb3+ doped ZnO nanocrystals with different concentrations of RE3+ (Tb3+, Yb3+) ions were successfully synthesized via sol-gel method to produce rare earth activated zinc oxide nanophosphors. The phosphor powders were produced by drying the precursor gels at 200˚C in ambient air. Based on the X-ray diffraction results, it was found that the pure and RE3+ doped ZnO nanophosphors were highly polycrystalline in nature regardless of the incorporation of Tb3+ or Yb3+ ions. Moreover, the diffraction patterns were all indexed to the ZnO Hexagonal wurtzite structure and belong to P63mc symmetry group. The Raman spectroscopy confirmed the wurtzitic structure of the prepared samples. Elemental mapping conducted on the as prepared samples using Scanning electron microscope (SEM) equipped with energy dispersive X-ray spectrometer (EDX) revealed homogeneous distribution of Zn, O, and RE3+ ions. The high resolution transmission electron microscope (HR-TEM) analyses indicated that the un-doped and RE3+ doped samples were composed of hexagonal homogeneously dispersed particles of high crystallinity with an average size ranging from 4 to 7 nm in diameter, which was in agreement with X-ray diffraction (XRD) analyses. ZnO:Tb3+ PL study showed that among different Tb3+ concentrations, 0.5 mol% Tb3+ doped ZnO nanoparticles showed clear emission from the dopant originating from the 4f-4f intra-ionic transitions of Tb3+ while the broad defects emission was dominating in the 0.15 and 1 mol% Tb3+doped ZnO. Optical band-gap was extrapolated from the Ultraviolet Visible spectroscopy (UV-Vis) absorption spectra using TAUC‟s method and the widening of the optical band-gap for the doped samples as compared to the un-doped sample was observed. The PL study of ZnO:Yb3+ samples was studied using a 325 nm He-Cd laser line. It was observed that the ZnO exciton peak was enhanced as Yb3+ions were incorporated in ZnO matrix. Furthermore, UV-VIS absorption spectroscopic study revealed the widening of the band-gap in Tb3+ doped ZnO and a narrowing in the case of Yb3+ doped ZnO system. X-ray photoelectron spectroscopy demonstrated that the dopant was present in the doped samples and the result was found to be consistent with PL data from which an energy transfer was evidenced. Energy transfer mechanism was evidenced between RE3+ and ZnO nanocrystals and was discussed in detail. / Physics / M.Sc. (Physics)

With the development of new generation synchrotron facilities, the performances of various X-ray spectroscopies have become more advanced. In order to interpret the X-ray spectrum experiments of various novel materials related to transition metal and rare earth elements, new advanced theoretical methods are required. The present thesis incorporates four modus operandi based on the classic multiplet theory to study the core level X-ray spectroscopy of transition metal and rare earth element. The four approaches consist of new methods developed from classic multiplet approach to high level first-principles method assisted multiplet calculation. Some methods are selected from previous researches and some are invented by original researches. These methods are integrated together to form a complete set of multiplet computational methods. This set of multiplet computational methods can perform calculations on various X-ray spectroscopies such as XAS, XPS, XES and RIXS related to the core-level electron. These wide range of spectroscopic methods coupled to different multiplet theory approaches serve as efficient tools to understand the electronic structure of metal sites and their unique contribution to the physical/chemical properties of the materials. The thesis creatively improves the classic multiplet theory on several aspects: (1) the relation between crystal field parameters and local structure factors; (2) the difficulty of processing point group symmetry branching chain in low symmetric structure; (3) the first-principles calculation of semi-empirical parameters. Four modus operandi are presented in this thesis: the first is the classic multiplet theory consisting of the multiplet effect, crystal field effect and charge transfer effect via several semi-empirical parameters as description for these effects. The second level multiplet theoretical approach analyze the crystal field potential matrix in various symmetries according to the point group symmetry branching rules. Then the crystal field effect parameters used in classic multiplet theory are linked analytically to the specific structural factors such as bond length and angles. This approach is a good tool to study the structural distortion from higher to lower order symmetry with analysis of X-ray spectral feature changes in experiment. The third modus operandi adopts large cluster model consisting of point charges at equivalent atoms position to simulate the crystal field effect on the center metal site. This approach handles low order symmetric crystal field with long range effect in multiplet calculation in an easier way than the classic multiplet theory. The fourth modus operandi initially studies the system of interest in first-principles calculation for the electronic wavefunctions. Then the electronic wavefunctions are used to derive the maximally localized Wannier functions at metal/ligand sites. The analysis of these Wannier functions provide a lot of semi-empirical parameters required in the classic multiplet calculation approach in a first-principles way. This modus operandi has substantially resolved the problem of finding the best set of semi-empirical parameters to fit the calculated X-ray spectrum with experimental data. In order to study the core electrons of the light elements (such as C/N/O) around center metal ions, a theoretical calculation method used to study the core electrons' vibrationally-resolved X-ray spectroscopy is also introduced as a complementary research and applied to C1s core ionized XPS calculation as an example. / Med utvecklingen av nya generationens synkrotronanläggningar har prestandan för olika röntgenspektroskopier blivit mer avancerad. För att tolka röntgenspektrumexperiment av olika nya material relaterade till övergångsmetaller och sällsynta jordartsmetaller krävs nya avancerade teoretiska metoder. Denna avhandling inkorporerar fyra metoder baserade på den klassiska multiplettläran för att studera röntgenspektroskopi på kärnenerginivån för övergångsmetaller och sällsynta jordartsmetaller. De fyra metoderna består av nya metoder som utvecklats från den klassiska multiplettläran till multiplettberäkning med hjälp av högnivå-först-principer-metoder. Vissa metoder har valts från tidigare forskning och några har uppfunnits genom originalforskning. Dessa metoder integreras för att bilda en komplett uppsättning av multiplettberäkningsmetoder. Denna uppsättning av multiplettberäkningsmetoder kan utföra beräkningar av olika röntgenspektroskopier som XAS, XPS, XES och RIXS relaterade till elektroner på kärnenerginivån. Denna breda spektrum av spektroskopiska metoder som kopplas till olika multipletteteoretiska tillvägagångssätt fungerar som effektiva verktyg för att förstå den elektroniska strukturen hos metalliska platser och deras unika bidrag till materialets fysikaliska/kemiska egenskaper. Avhandlingen förbättrar på ett kreativt sätt den klassiska multiplettläran på flera områden: (1) förhållandet mellan kristallfältparametrar och lokala strukturfaktorer, (2) svårigheten med att hantera förgreningskedjor i punktgruppsymmetri för strukturer med låg symmetri, (3) först-principer-beräkning av semi-empiriska parametrar. Fyra metoder presenteras i denna avhandling: den första är den klassiska multiplettläran som omfattar multiplett-effekten, kristallfältseffekten och laddningsöverföringseffekten genom flera semi-empiriska parametrar som beskriver dessa effekter. Den andra multiplett-teoretiska tillvägagångssättet analyserar kristallfältspotensialmatrisen i olika symmetrier enligt reglerna för punktgruppsymmetri. Därefter kopplas de kristallfältseffektparametrar som används i den klassiska multiplettläran analytiskt till specifika strukturella faktorer som bindningslängd och vinklar. Detta tillvägagångssätt är ett bra verktyg för att studera strukturell distorsion från högre till lägre ordningssymmetri med analys av förändringar i röntgenspektrala funktionsändringar i experiment. Det tredje tillvägagångssättet använder en stor klustermetod som består av punktladdningar på ekvivalenta atompositioner för att simulera kristallfältsinverkan på metallcentralen. Detta tillvägagångssätt hanterar kristallfält med låg symmetri och lång räckvidd i multiplettberäkningar på ett enklare sätt än den klassiska multiplettläran. Det fjärde tillvägagångssättet studerar initialt det intressanta systemet med först-principer-beräkningar av elektronvågfunktioner. Därefter används dessa elektronvågfunktioner för att härleda maximalt lokaliserade Wannier-funktioner vid metall-/ligandpositioner. Analysen av dessa Wannier-funktioner ger många av de semi-empiriska parametrar som krävs i den klassiska multiplettberäkningsmetoden på ett först-principer-sätt. Detta tillvägagångssätt har i stor utsträckning löst problemet med att hitta den bästa uppsättningen semi-empiriska parametrar för att anpassa de beräknade röntgenspektrumen med experimentella data. För att studera de lättas elektroner (som C/N/O) runt centrala metalljoner introduceras även en teoretisk beräkningsmetod för att studera röntgenspektrum med vibrationell upplösning för kärnelektroner som ett kompletterande forskningsområde och tillämpas på beräkningar av C1s-kärnenerginivåns XPS som ett exempel. / <p>QC 2023-05-23</p>

No description available.

In strongly correlated electron systems the intricate interplay between electronic correlation effects and the tendency to form bands leads to a wealth of fascinating physical phenomena. The theoretical description of such systems is extremely complex and cannot be tackled exactly, so that ingenious modelling based on approximations must be utilized. Experiments are of utmost importance in this regard, since they provide a way to test and verify models, or to help devise better ones. This Dissertation deals with strongly correlated uranium intermetallic systems. The interesting phases they can adopt include heavy fermion behaviour, unconventional superconductivity, hidden and multipolar order, and exotic induced magnetism. Here the hybridization between the 5f states and the conduction electrons drives the physics. The description of the 5f states is therefore of utmost importance. However, since there is no clear hierarchy of interactions like Coulomb repulsion, spin-orbit coupling, hopping and crystal-field, the modelling is difficult. This is in strong contrast to the more spatially localized 4f states of, e.g., cerium compounds. It is far from clear how to quantitatively describe the electronic structure of uranium intermetallics and whether, for example, an itinerant band approach or an impurity-type model, taking local degrees of freedom explicitly into account, would be a better starting point. In intermetallics, the situation is aggravated by the fact that the modelling lacks important pieces of information. This is not least due to the fact that understanding the formal valence, the filling of the 5f shell, and the relevant symmetries of the $5f$ electrons are experimentally demanding tasks. This Dissertation, therefore, aims at developing new methods and Ans\'atze in this direction. We use x-ray spectroscopy to investigate the electronic structure, and in particular element-specific Inelastic X-ray Scattering (IXS); resonant (RIXS) at the U M(5) edge and non-resonant (NIXS) at the U O(4,5) edge. Both methods are innovative. For the first time, valence band RIXS measurements with sufficient resolution (150 meV) can be carried out at the U M(5) edge to measure ff excitations in intermetallic uranium compounds. Their existence, if present, provides information about the formal valence or main atomic configuration that determines the symmetry. The orientation dependence of the mutipolar excitations in NIXS (with restrictions also in RIXS), in turn, provides information about the orbital occupation. Atomic full-multiplet calculations are indispensable here. In addition, photoelectron spectroscopy (PES) is applied, both in the soft as well as in the hard x-ray regime (HAXPES), to investigate the hybridization and localization of the 5f electrons. The energy dependence of the cross-sections allows to determine the orbital contributions in the valence band, so that parameters like, e.g., the double-counting correction for the LDA+DMFT calculations, performed by Prof. A. Hariki from the Osaka Metropolitan University, can be determined from tuning the calculations to the experimental data. This combination of PES and DFT+DMFT provides a reliable new quantitative insight into the number of electrons in the 5f shell and their degree of delocalization. We consider UGa(2) and UB(2), respectively, as benchmark localized and itinerant systems and investigate them with IXS and PES. UGa(2) is a high-moment ferromagnet, with U-U distances above the Hill limit, while UB(2) is paramagnetic and clearly below the Hill limit. We observe sharp multiplet excitations of the 5f2 configuration in the IXS spectra of UGa(2), but none in the spectra of UB(2). The comparison of the spectra with full-multiplet calculations shows that in UGa(2) the U 5f2 configuration dominates and, from the orientation dependence (RIXS and NIXS), the crystal-field ground state can be determined. We show that the magnetism of this compound is of the induced type. The cross-section based analysis of the valence band PES data with the LDA+DMFT approach shows that the filling of the 5f shell is similar in both compounds, but that the distribution among different configurations is considerably wider in UB(2). Also the time-dependent charge correlation functions of UGa(2) and UB(2) show a larger itinerancy in the latter compound. The peculiarity and novelty about this combined study is that a reliable quantitative description of the electronic structure is achieved. This allows, for the first time, an accurate estimation of the 5f occupation and a quantitative description of the U\ 4f core-level PES spectra. This study paves the way to a systematic classification of uranium intermetallics. We further investigate the substitution series URu(2-x)Fe(x)Si(2) with PES. The systematic study of isostructural and/or isoelectronic series of compounds is crucial in unveiling the origin of their physical properties. URu(2)Si(2) exhibits hidden-order as well as superconductivity, and becomes antiferromagnetic upon Fe doping. Fe substitution seemingly involves the application of chemical pressure to the system. We measure the U 4f PES core-level of the URu(2-x)Fe(x)Si(2) substitution series and observe a non-monotonic shift of spectral weight. We argue that, besides chemical pressure, the Fe density of states at the Fermi level also plays a central role and we propose and extended Doniach diagram where the two effects compete. We also measure NIXS, confirming that the ground state symmetry is a singlet or quasi-doublet of the 5f2 configuration. The magnetic properties must then be understood, as in UGa(2), in terms of induced magnetism. We extend our study of the UT(2)Si(2) compounds to the case where T = Os, Ir, Pt and Au, i.e. 5d transition metals. Although the T= 3d and 4d transition metal systems have been extensively investigated, the T = 5d compounds lack systematic studies. The comparison of the NIXS spectra with multiplet calculations shows that also here the 5f2 dominates, only for T= Au it is not so clear. The absence of a strong directional dependence impairs the identification of the ground state symmetry. Valence band hard x-ray PES allows to probe the transition metal 5d states directly. U 4f core-level hard x-ray PES gives a qualitative indication of the filling of the 5f shell across the series. We then focus on hexagonal UNi(2)Al(3) and apply NIXS. Like isoelectronic and isostructural UPd(2)Al(3), it is a prototypical U heavy-fermion compound, showing antiferromagnetic order and unconventional superconductivity. We observe a strong directional dependence of the NIXS spectra at low temperatures. The possible ground-state symmetries of the 5f2 configuration that fit the NIXS data are in contradiction to previous proposals from fits of the static magnetic susceptibility. We put forward a new crystal-field model that describes the high temperature magnetic susceptibility and the NIXS data at low temperatures, and that explains the magnetism.

Nanotechnology offers a great potential do design fertilizers with unique properties capable to boost the plant productivity. However, the nanoparticles environmental fate and their toxic responses still need to be deeply investigated to their safe use. This study aims to investigate the effect of copper oxide (nCuO) and magnetite nanoparticles (nFe3O4) on the germination and seedling development of Phaseolus vulgaris L. Seeds were treated in nanoparticles dispersions in a wide range of concentrations (1, 10, 100 and 1 000 mg L-1) and incubated in a germination chamber during 5 days. Different sized nCuO (25, 40 and <80 nm) and polyethylene glycol (PEG) coated nFe3O4 were evaluated. Although both nCuO and nFe3O4 treatments did not affected the germination rate, seedling weight gain was promoted by 40 nm CuO at 100 mg Cu L-1 and inhibited by 1 000 mg Cu L-1 of 25 nm CuO and positive control (CuSO4). Among the tested nCuO, the higher chemical reactivity was found for the 25 nm CuO, and this may partially explain the observed deleterious effects. Seeds treated in nFe3O4-PEG at 1 000 mg Fe L-1 increased radicle elongation compared to the negative control (water), while Fe2+/Fe3+ (aq) (positive control) and bare nFe3O4 at 1 000 mg Fe L-1 treatments reduced the radicle of the seedlings. The growth promoted by the PEG-coated nanoparticles can be justified by the higher water uptake induced by the PEG, and also by its lower chemical reactivity compared to the bare nanoparticles. This was reinforced by enzymatic assays since nFe3O4-PEG treatment was also the least harmful to the alpha-amylase activity. X-ray fluorescence spectroscopy (XRF) showed that most of the Cu and Fe incorporated by the seeds remained in the seed coat, specially in the hilum region, and X-ray tomography indicated that Fe3O4-PEG penetrated in this structure. X-ray absorption spectroscopy (XAS) unraveled that the Cu and Fe chemical environment of the nCuO and nFe3O4-PEG treated seeds persisted mostly in its primitive form. These results contribute to the understanding of how nCuO, nFe3O4 and nFe3O4-PEG interact with common bean seeds and seedlings and highlights its potential use in seed priming / A nanotecnologia oferece um grande potencial para o desenvolvimento de fertilizantes com propriedades únicas, capazes de impulsionar a produtividade das plantas. Contudo, o destino ambiental e os efeitos tóxicos das nanopartículas ainda necessitam ser profundamente investigados para o seu uso seguro. Este estudo visa investigar o efeito das nanopartículas de óxido de cobre (nCuO) e magnetita (nFe3O4) na germinação e desenvolvimento das plântulas de Phaseolus vulgaris L. As sementes foram tratadas em dispersões de nanopartículas em diversas concentrações (1, 10, 100 and 1 000 mg L-1) e incubadas em uma câmara de germinação durante 5 dias. Diferentes tamanhos de nCuO (25, 40 e <80 nm) e nFe3O4 recoberta com polietileno glicol (PEG) e foram avaliados. Embora ambos tratamentos de nCuO e nFe3O4 não afetaram a taxa de germinação, o ganho de massa das plântulas foi promovido pela nCuO de 40 nm à 100 mg Cu L-1 e inibido pelos tratamentos de nCuO de 25 nm e controle positivo (CuSO4) à 1 000 mg Cu L-1. Dentre as nCuO testadas, a maior reatividade química foi encontrada para a nCuO de 25 nm, e isso pode explicar parcialmente os efeitos deletérios desta nanopartícula. Sementes tratadas com nFe3O4-PEG à 1 000 mg Fe L-1 aumentaram o alongamento das radículas em comparação ao controle negativo (água), enquanto que os tratamentos Fe2+/Fe3+ (aq) (controle positivo) e nFe3O4 sem recobrimento à 1 000 mg Fe L-1 reduziram as radículas das plântulas. O crescimento promovido pelas nanopartículas recobertas com PEG pode ser justificado pela maior absorção de água induzido pelo PEG, e também pela sua baixa reatividade química comparada às nanopartículas sem recobrimento. Isso foi reforçado por ensaios enzimáticos uma vez que o tratamento de nFe3O4-PEG foi também o menos prejudicial à atividade da alfa-amilase. A espectroscopia de fluorescência de raios-X (XRF) mostrou que a maior parte do Cu e do Fe incorporados pelas sementes permaneceu no tegumento, especialmente na região do hilo, e a tomografia de raios-X indicou que nFe3O4-PEG penetrou nesta estrutura. A espectroscopia de absorção de raios-X (XAS) revelou que o ambiente químico do Cu e do Fe das sementes tratadas com nCuO e nFe3O4-PEG persistiram majoritariamente em sua forma primitiva. Estes resultados contribuem para o entendimento de como nCuO, nFe3O4 e nFe3O4-PEG interagem com sementes de feijão e destaca seu potencial uso no tratamento de sementes

Diese Dissertation widmet sich der spektroskopischen Untersuchung verschiedener Aspekte der Strahlungsemis\-sion hochgeladener Krypton-Ionen mit Relevanz für die Fusionsforschung. Die Experimente hierzu erfolgten an der Berliner Elektronenstrahl-Ionenfalle (EBIT). Der erste Teil der Arbeit hat die Messung kanalspezifischer Wirkungsquerschnitte für die dielektronische Rekombination (DR) der KL$n$-Resonanzserie ($n$=2, \ldots, 5) von Helium- bis Kohlenstoff-ähnlichen Kr-Ionen ($\mbox{Kr}^{(34\, \ldots\,30)+}$) zum Inhalt, die relativ zum Wirkungsquerschnitt der nichtresonanten strahlenden Rekombination (RR) bestimmt wurden. Die Anpassung der Anregungskurven durch eine Modellfunktion aus berechneten Resonanzst ärken ermöglichte den Vergleich mit theoretischen DR-Wirkungsquerschnitten. Es zeigt sich, dass Vorhersagen des HULLAC-Atomstrukturcodes für die Resonanz\-st"ar\-ken der Kr-Ionen durch das Experiment innerhalb der Me"sunsicherheiten best"a\-tigt werden. Darüber hinaus wurde auch die Relaxation der einfach angeregten Ionen nach erfolgtem DR-Stabilisierungsübergang analysiert. Die zur Auswertung der DR-Anre\-gungs\-kurven angewandte Technik eröffnet gleichzeitig eine spektroskopische Methode für die Bestimmung der relativen Konzentration hochgeladener Ionen in EBIT. Die Messung der Strahlungskühlungsrate von Krypton, die den zweiten inhaltlichen Schwerpunkt der Dissertation darstellt, wäre ohne diese in situ Diagnostik der Ladungbilanz nicht möglich gewesen. Hier wurde die Ionenfalle so eingestellt, dass sich eine Ladungsverteilung herausbildet, die dem Ionisationsgleichgewicht eines Plasmas bei einer Temperatur von etwa $5\;\mbox{keV}$ entspricht. Die Bestimmung der Strahlungsk"uhlungsrate profitierte von dem Potential einer EBIT, die gefangenen Ionen mit Elektronenenergien aus einem weiten Bereich abzutasten und einzelne Strahlunsprozesse selektiv anzuregen. Die Röntgenemission verschiedener Strahlungskanäle, wie Bremsstrahlung, strahlende Rekombination, dielektronische Rekombination und Linienstrahlung nach direkter Anregung wurde separat erfaßt. Hieraus konnten erstmals kanalspezifische Strahlungskühlungsraten bestimmt werden. Es stellte sich heraus, dass der dominante Beitrag zur Strahlungskühlungsrate durch die direkt angeregte Linienstr ahlung des L-Schalen-Spektrums zustande kommt, die etwa 75\% der gesamten Verlustleistung ausmacht. Beim Vergleich der totalen Strahlungsverlustleistung mit Vorhersagen der Theorie sind Abweichungen festzustellen. Die berechneten Werte sind je nach Modell um einen Faktor 1.5 - 2.0 kleiner als das Ergebnis der Messung. Dieser Unterschied liegt außerhalb der experimentellen Unsicherheit von maximal 30\%. / This thesis deals with the spectroscopic investigation of various aspects of the x-ray emission of highly charged krypton ions with relevance for fusion research. The experiments have been performed at the Berlin electron beam ion trap (EBIT). One part of the work was devoted to the measurement of channel-specific cross sections for dielectronic recombination (DR) via the KL$n$ ($n$=2, \ldots, 5) resonance series of He- to C-like krypton ions ($\mbox{Kr}^{(34\, \ldots\,30)+}$). The DR cross sections were determined relative to the cross section for non-resonant radiative recombination (RR). A fit procedure was used to compare the measured data with theoretical calculations. Predictions of the HULLAC atomic structure code are confirmed within the experimental uncertainties. Additionally, the radiative relaxation mechanism following the stabilizing transition in the DR process was analyzed. The approach used to obtain the DR excitation function opens up a spectroscopic method to determine the relative abundance of the highly charged ions in the trap. This in situ diagnostic of the charge state balance allowed for the measurement of the radiative cooling rates of krypton being the second focus of the thesis. For this purpose EBIT was tuned to a charge state distribution approaching the ionization balance of a plasma at a temperature of about $5\;\mbox{keV}$. EBIT's capability to sample a wide range of electron-beam energies and distinguish between different radiation channels was utilized to determine the cooling rate. The x-ray emission from the various plasma radiation channels, like bremsstrahlung, radiative recombination, dielectronic recombination, and line radiation following electron-impact excitation was analyzed. For the first time, channel-specific cooling rates could be obtained from these data. It was found, that the dominant contribution to the cooling rate is made up by the directly excited x-rays of the L-shell spectra of krypton, producing more than 75\% of the total radiation loss. A difference with theoretical calculations is noted for the total cooling rate. The predicted values are lower by a factor of 1.5 - 2.0, depending on the theoretical model. This discrepancy is clearly beyond the experimental uncertainty of 30\% at maximum.

Les analogues du bleu de Prusse sont intéressants pour le stockage de l’information car ils présentent des propriétés de photo-commutation. Afin d’intégrer et d’exploiter ces propriétés photomagnétiques dans d’éventuelles applications, il est nécessaire de faire une étape de mise en forme. Cette dernière consiste à contrôler la taille, la forme ainsi que l’arrangement tridimensionnel des nanoparticules. Une stratégie consiste à élaborer des oxydes de silice nanostructurée par le procédé sol-gel. Une méthode a consisté faire précipiter les analogues du bleu de Prusse photomagnétiques dans la nanoporosité. Ainsi, il a fallu obtenir un monolithe ayant une unique organisation et non un mélange de phases. Il a été possible d’optimiser la méthode pour obtenir un monolithe hexagonal ayant des ions cobalt. Cette méthode a pu être étendue à différentes organisations : lamellaire, cubique, hexagonale, vermiculaire. Un traitement thermique permettant d’éliminer le copolymère afin de former le nanoréacteur. Une étude a permis de montrer que l’organisation etait conservée après un traitement thermique. Les ions cobalt subissaient une thermo hydrolyse durant le traitement thermique.Il a été possible de mettre en place une méthode d’imprégnation permettant d’obtenir différents nanocomposites CoFe et de contrôler la stœchiométrie des particules d’ABP confinées.Cette méthode a permis d’élaborer des nanocomposites CoFe sans cations alcalin et des nanocomposites CoFe contenant deux cations rubidium par maille. Les propriétés magnétiques ont permis de montrer l’effet de la mise en forme. De plus cette étude a permis de montrer que les propriétés de commuation sont concervées mêmes sur des nanoparticules de 5 nm. / The Prussian blue analogues (CoFe PBA) have attracted growing interest owing to the tunability of their magnetic properties by external stimuli that make them good candidates for future optical memories or switching devices.This study need a processing step to elaborate a nanocomposite material with full control on the stoichiometry, the size, the shape and the organisation of PBA particles.Our strategie consit to precipitate photomagnetic Prussian blue analogue in the nanoporosite. Our main idea is to use mesoporous silica monoliths as template for the precipitation of the PBA in order to control the size and the shape of the nanoparticles. A silica monolith contaigning cobalt ions in the hexogonal structuraction have been obtained. The method have been extended to another structuration ( wormlike, cubic and lamallar). A termal traitement at 500°C in air have been optimised to remove the copolymer and to obtain the nanoreacteur. During the termal traitement the cobalt ion change the symmetry. A thermo-hydrolysis of the cobalt drives to the formation of monomer tetrahedron of cobalt.The precipitation of Prussian Blue Analogues is realized by impregnation with a solution of potassium hexacyanoferrate (III). The parameters of the impregnation have to be finely controlled in order to obtain nanocrystals of Co-Fe wihout alkalin cations and nanocrystrals of CoFe with two cation rubidium. A photomagnetic nanocrystrals have been obtained.