Single self-assembled semiconductor quantum dots (QDs) are able to emit single-photons and entangled-photons pairs. They are therefore considered as potential candidate building blocks for quantum information processing (QIP) and communication. To exploit them fully, the ability to precisely control their optical properties is needed due to several reasons. For example, the stochastic nature of their growth ends up with only little probability of finding any two or more QDs emitting indistinguishable photons. These are required for two-photon quantum interference (partial Bell-state measurement), which lies at the heart of linear optics QIP. Also, most of the as-grown QDs do not fulfil the symmetries required for generation of entangled-photon pairs. Additionally, tuning is required to establish completely new systems, for example, 87Rb atomic-vapors based hybrid semiconductoratomic (HSA) interface or QDs with significant heavy-hole (HH)-light-hole (LH) mixings. The former paves a way towards quantum memories and the latter makes the optical control of hole spins much easier required for spin- based QIP. This work focuses on the optical properties of a new type of QDs optimized for HSA experiments and their broadband tuning using strain. It was created by integrating the membranes, containing QDs, onto relaxor-ferroelectric actuators and was quantified with a spatial resolution of ~1 µm by combining measurements of the µ-photoluminescence of the regions surrounding the QDs and dedicated modeling. The emission of a neutral exciton confined in a QD usually consists of two fine-structure-split lines which are linearly polarized along orthogonal directions. In our QDs we tune the emission energies as large as ~23meV and the fine-structure-splitting by more than 90 µeV. For the first time, we demonstrate that strain is able to tune the angle between the polarization direction of these two lines up to 40° due to increased strain-induced HH-LH mixings up to ~55%. Compared to other quantum emitters, QDs can be easily integrated into optoelectronic devices, which enable, for example, the generation of non-classical light under electrical injection. A novel method to create sub-micrometer sized current-channels to efficiently feed charge carriers into single QDs is presented in this thesis. It is based on focused-laserbeam assisted thermal diffusion of manganese interstitial ions from the top GaMnAs layer into the underlying layer of resonant tunneling diode structures. The combination of the two methods investigated in this thesis may lead to new QDbased devices, where direct laser writing is employed to preselect QDs by creating localized current-channels and strain is used to fine tune their optical properties to match the demanding requirements imposed by QIP concepts.

Transition metal oxides have proven to be a fertile research area for condensed matter physicists due to the fascinating array of superconducting, magnetic and electronic properties they exhibit. A particular resurgence of intense activity in investigating the properties of these systems followed the discovery of high temperature superconductivity in the cuprates, colossal magnetoresistance in the manganites, ferroelectricity in the cobaltites and simultaneous ferroelectric and ferromagnetic ordering in the manganites. These diverse properties of transition metal compounds arise due to the presence of strong electron-electron interactions within the transition element 3d states. Indeed, it is the competition between the localizing effects of such interactions and the comparable hopping strengths driving the system towards delocalization, that is responsible for these wide spectrum of interesting properties. In terms of theoretical and fundamental issues, electronic structure of transition metal oxides play a most important role, providing a testing ground for new many-body theoretical approaches treating the correlation problem at various levels of approximations. In addition to this rich spectrum of properties, metal-insulator transitions often occur and can even be coincident with structural or magnetic changes due to the strong coupling between charge, magnetic and lattice degrees of freedom. However, in spite of the immense activities in this area, the underlying phenomena is not yet completely understood. A careful investigation of the electronic structure of these systems will help in the microscopic understanding of these and photoelectron spectroscopy has been established as the most powerful tool for investigating the electronic structures of these systems. In this thesis we investigate the electronic structures of some of these transition metal oxides and the metal-insulator transition as a function of electron correlation strength and doping of charge carriers by means of photoelectron spectroscopy; we analyze the experimental results using various theoretical approaches, in order to obtain detailed and quantitative understandings. This thesis is organized into seven chapters. Chapter 1 is a general introduction to the various concepts discussed in this thesis. Here we briefly describe the various mechanisms and theoretical formalisms used for understanding the metal-insulator transitions in strongly correlated systems and the evolution of the electronic structure across the transition. The experimental and the calculational techniques used in this thesis is described in Chapter 2. This includes different sample synthesis techniques and the characterization tools used in the present study. Photoelectron spectroscopic techniques used for probing the electronic structure of various systems are also discussed in this chapter. In Chapter 3, we discuss the coexistence of ferromagnetism and superconductivity in ruthenocuprates by looking at the electronic structures of RuSr2Eu1.5Ce0.5Cu2O10 which is a ferromagnetic superconductor having the ferromagnetic TC ~ 100 K and a superconducting transition of ~ 30 K compared with RuSr2EuCeCu2O10 which is a ferromagnetic (TC ~ 150 K) insulator in conjunction with two reference systems, RuSr2GdO6and Sr2RuO4. The coexistence of ferromagnetic order with superconductivity below the superconducting temperature is an interesting issue since the pair-breaking due to magnetic interactions is not significant in these cases. Extensive photoelectron spectroscopic measurements were performed on these systems and our results show that Eu and Ce in both the ruthenocuprates exists in 3+ and 4+ states, respectively. Also the analysis of the Ru 3d and 3p core levels suggests that Ru remains in the pentavalent state in both the cases. The constancy of Ru valency with doping of charge carriers that bring about an insulator to metal transition and the superconducting state suggests that the electronic structure and transport properties of these compounds are not governed by the Ru-O plane, but by the Cu-O plane, much as in the case of other high TC cuprates. Analysis of the Cu 2p core level spectra in terms of a cluster model, including configuration interaction and multiplet interactions between Cu 3d and 2p as well as that within the Cu 3d states, establish a close similarity of the basic electronic structure of these ruthenocuprates to those of other high TC cuprates. Here the charge transfer energy, Δ << Udd,Cu 3d multiplet-averaged Coulomb repulsion energy, establishing the compounds to be deep in the charge transfer regime. Continuing with the ruthenocuprate systems in Chapter 4, we look at the electronic structure of hole doped La2CuRuO6systems using various photoemission techniques. It was expected that since the substitution of La3+by Sr2+changes the d electron count, the system will undergo a metal to insulator transition, but the transport properties show that all of them remain semiconducting through out the lowest temperature of measurement. A careful analysis of the Ru 3d and 3p core level spectra shows that Ru exists in Ru 4+state in La2CuRuO6and goes towards Ru 5+state with hole doping. This suggests that the doped holes affects the electronic structure of the Ru levels in these systems. A spectral decomposition of the Ru 3d core level suggests the existence of a spin orbit split doublet having two peaks, a main core level peak and a satellite peak at the higher binding energy side of the main peak and the intensity ratio of the satellite peak to the main peak increases with the insulating nature of the compounds as reported for other Ru 4d strongly correlated systems. This observation is also consistent with the transport properties. Cu 2p core level spectra also shows variations in the satellite-to-main peak Cu 2p intensities suggesting that the electronic structure of the Cu levels are also getting affected with Sr doping. Valence band spectral features near the Fermi level shows that the spectral weight is highest for La2CuRuO6and depletes slowly with Sr doping consistent with the expected d electron count as suggested by the Ru valencies. In Chapter 5 and Chapter 6 we discuss the electronic structure investigations of two early transition metal oxide series, namely Ca1−xSrxVO3and Ce1−xSrxTiO3. Surface sensitivity dependence of photoemission experiments has been explored to show that the surface and the bulk electronic structures of Ca1−xSrxVO3system is different. Photoemission spectra of this system using synchrotron radiation reveal a hither to unnoticed polarization dependence of the photoemission matrix elements for the surface component leading to substantial underestimation. Extracted bulk spectra from experimentally determined electron escape depth and underestimation of surface contributions resolve the puzzling issues that arose due to the recent diverse interpretations of the electronic structure in Ca1−xSrxVO3. Keeping in mind the above-mentioned caveat, the present results still clearly establish that the linear polarization of synchrotron radiation plays a key role in determining the spectral lineshape in these systems. The experimentally-determined bulk spectra provide an understanding of the electronic structure in Ca1−xSrxVO3, consistent with experimental γ values, calculated change in the d-bandwidth and the geometrical/structural trends across the series, thereby resolving the puzzle concerning the structure-property relationship in this interesting class of compounds. In Chapter 6 we discuss the issues of metal-insulator transition close to the d0limit as well as the evolution of the electronic structure of a strongly correlated system as a function of electron occupancy, by investigating the family of Ce1−xSrxTiO3compounds by recording core level as well as valence band photoemission spectra using lab source as well as synchrotron radiations. Core level Ce 3d spectra from Ce1−xSrxTiO3samples establish a trivalent state of Ce in these compounds for all values of x confirming that charge doping in the present system does not alter the electronic structure of Ce. Hence the change in valency due to Sr substitution and thus, the carrier number, takes place only in the Ti 3d-O 2p manifold. We also carried out extensive VUV photoemission experiments on these samples with the photon energy varying between 26-122 eV. From the difference spectrum obtained by subtracting the off-resonance spectrum from the on-resonance one, we obtain the Ce 4f spectral signature; thus obtained Ce 4f spectrum which has a peak at about 3 eV binding energy and shows no intensity at EF even for the metallic samples, consistent with a Ce3+state. In order to study the states near EF responsible for the metal-insulator transition in these compounds, we recorded the valence band spectra at the Ce 4f off-resonance condition so that the coherent and the incoherent spectral features arising from the Ti 3d states could be clearly resolved, allowing us to investigate the metal insulator transition in the Ce1−xSrxTiO3system as a function of Sr or hole doping. The experimental spectra of the metallic compounds exhibit an intensity of the incoherent feature considerably larger than that predicted by theory. This discrepancy is possibly due to a difference in the surface and the bulk electronic structures of these compounds. Chapter 7 is divided into two parts. In the first part we discuss the extended x-ray absorption fine structure (EXAFS) studies performed on two transition metal oxide series, La1−xSrxCoO3and La1−xSrxFeO3to look at the local structure distortions happening around the transition metal ions and its role in bringing out metal to insulator transitions in transition metal oxide systems. Here we chose to investigate these two systems since La1−xSrxCoO3undergoes an insulator to metal transition for x ∼ 0.15 and La1−xSrxFeO3remains insulating for the entire range of doping. The static mean square relative displacement, which we believe to be a representation of the disorder present in the system, extracted by fitting the experimental data by a correlated Einstein model, as a function of composition in La1−xSrxCoO3saturates beyond the critical composition where as the disorder parameter continues to increase through out the entire doping range in the case of La1−xSrxFeO3where metal-insulator transition is absent. In the second part of Chapter 7 we discuss the x-ray absorption near edge structure (XANES) studies performed on the above mentioned series of systems. Co K-edge XANES spectra of La1−xSrxCoO3show that there is a systematic shift of the main absorption peak with hole doping suggesting that the Co valency changes systematically with Sr doping. Also, the pre-edge feature of LaCoO3shows the transitions to t2g level clearly showing that Co3+in LaCoO3is not in a pure low spin (t6 2g) state. The Fe K-edge XANES spectra of La1−xSrxFeO3also exhibit a systematic shift to the higher energy side with increase in Sr content, indicating an increase in the Fe valence. Also from the La L3edge analysis, it can be concluded that the oxygen environment around La and the electronic configuration of La are systematically changing with Sr doping.

This thesis work reports in depth analyses of measured µ-XRD and µ-XAS data from standard UO2, chromia (Cr2O3) doped UO2 and MOX fuels, and interpretation of the results considering the role of chromium as a dopant as well as several fission product elements. The lattice parameters of UO2 in fresh and irradiated samples and elastic strain energy densities in the irradiated UO2 samples have been measured and quantified. The µ-XRD patterns have further allowed the evaluation of the crystalline domain size and sub-grain formation at different locations of the irradiated fuel pellets. Attempts have been made to determine lattice parameter and next neighbor atomic environment in chromia-precipitates found in fresh chromia-doped fuel pellets. The local structure around Cr in as-fabricated chromia-doped UO2 matrix and the influence of irradiation on the state of chromium in irradiated fuel matrix have been addressed. Finally, for a comparative understanding of fission gases behavior and irradiation induced re-solution phenomenon in standard and chromia-doped UO2, the last part of the present work tries to clarify the fission gas Kr atomic environment in these irradiated fuels. The work performed on Kr, by micro-beam XAS, comprises the determination of Kr next neighbor distances, an estimation of gas atom densities in the aggregates, and apparent internal pressures in the gas bubbles.

Polarization entangled photon pairs on demand are considered to be an important building block of quantum communication technology. It has been demonstrated that semiconductor quantum dots (QDs), which exhibit a certain spatial symmetry, can be used as a triggered, on-chip source of polarization entangled photon pairs. Due to limitations of the growth, the as-grown QDs usually do not exhibit the required symmetry, making the availability of post-growth tuning techniques essential. In this work first the QD-morphology of hundreds of QDs is correlated with the optical emission of neutral excitons confined in GaAs/AlGaAs QDs. It is presented how elastic anisotropic stress can be used to partially restore the symmetry of self-assembled GaAs/AlGaAs and InGaAs/GaAs QDs, making them as candidate sources of entangled photon pairs. As a consequence of the tuning of the QD-anisotropy we observe a rotation of the polarization of the emitted light. The joint modification of polarization orientation and QD anisotropy can be described by an anticrossing of the so-called bright excitonic states. Furthermore, it is demonstrated that anisotropic stress can be used to tune the purity of the hole states of the QDs by modifying the degree of heavy and light hole mixing. This ability might be interesting for applications using the hole spin as a so-called quantum bit.

The thesis entitled “Local structure-property relationship in some selected Solid State Materials” mainly focuses on two fundamental topics: (a) evaluation of some standard global structural concepts in terms of local structure to provide a unique description of the crystal structure, and (b) the role of the crystal structure at different length-scales in controlling the properties in some selected materials.

Neste trabalho abordamos os processos de perdas colisionais por mudança de estrutura fina e por escape radiativo. Estes processos são importantes por serem os principais fatores limitantes da densidade e da permanência de átomos confinados em armadilhas magneto-ópticas. A utilização do formalismo de Wigner nos possibilitou tratar a dinâmica dos graus de liberdade internos do sistema de forma puramente quântica, enquanto tratamos os graus de liberdade translacionais de maneira semiclássica. Com este formalismo deduzimos equações inéditas para uma colisão unidimensional e desenvolvemos um algoritmo, também inédito, para a resolução numérica de tais equações. / In this work we study the collision loss processes caused by fine-structure change and radiative escape. These processes are important because they limit the density and the confinement time of atoms in magneto-optical traps. The use of the Wigner-function formalism allows us to treat the internal degrees of freedom purely quantum mechanically, while treating the dynamics of the external degrees of freedom quasi-classically. This dissertation employs this formalism to derive for the first time, quasi-classical equations describing one-dimensional cold collisions, including dissipation due to spontaneous emission, and an algorithm, also for the first time, to solve numerically the mentioned set of coupled equations.

Orientador: Gilberto Medeiros Ribeiro / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin / Made available in DSpace on 2018-08-09T10:48:50Z (GMT). No. of bitstreams: 1 GonzalezHernandez_FelixGuillermo_D.pdf: 12837677 bytes, checksum: 70e82c96ea88ab1de4fa785d908c9af6 (MD5) Previous issue date: 2007 / Resumo: Medidas experimentais foram realizadas para determinar as escalas de tempo de relaxação e decoerência do spin eletrônico como bit quântico. A estrutura dos estados de exciton foi investigada com o objetivo de servir como estados intermediários na manipulação do spin. O sistema utilizado para o estudo de decoerência é um ensemble de pontos quânticos auto-formados semicondutores. Dois temas servem como eixos centrais dos três experimentos desenvolvidos nesta tese: a polarização de spin e o fator g de Landé. No primeiro experimento, ao incluir o efeito do reservatório térmico, foi obtido o grau de polarização do spin (populações dos níveis up e down) para as camadas s e p. O desdobramento dos níveis orbitais em subníveis de spin permitiu obter a magnitude do fator g para estes estados. Mudando a orientação do campo magnético, foram observadas as anisotropias do tensor g e a sua relação com os detalhes do potencial de confinamento. Estas características permitiram inferir o tempo de relaxação T1. A medida da polarização resolvida no tempo foi realizada através de es-pectroscopia óptica de bombeio-prova. Os pulsos de luz e o campo magnético transverso permitem que uma polarização líquida seja inicializada. A rotação de Kerr permitiu observar oscilações desta polarização em torno do campo magnético com freqüência determinada pelo fator g. A perda da coerência de fase do spin resulta no decaimento destas oscilações numa escala de tempo T2. Medidas realizadas num ensemble de spins implicam em que o tempo de decoerência encontra-se limitado pela escala de defasagem T¤2< T2. Uma técnica semelhante à refocalização por spin-eco em experimentos de ressonância magnética nuclear, foi aplicada utilizando pulsos de laser para reverter a defasagem do ensemble. Tanto a possibilidade de medir o sinal de eco como o tempo de decoerência foram medidos como função da temperatura. A estrutura de níveis de exciton e a sua distribuição no ensemble foi estudada também com espectroscopia de bombeio-prova. Foram observados batimentos quânticos entre os níveis de estrutura fina do exciton para sis-temas 0D e 2D limitados pelo tempo de recombinação / Abstract: Experimental measurements were carried out to determine the scales of the relaxation and decoherence time for the electronic spin as quantum bit. The structure of the exciton states was investigated with the objective to serve as intermediate states in the spin manipulation. The system studied for the implementation of the quantum computation is an ensemble of self-assembled semiconductor quantum dots. Two subjects serve as central axes of the three experiments developed in this thesis: the spin polarization and the Landé g-factor. In the first experiment, when including the effect of the thermal reservoir, the degree of spin polarization (populations for the up and down levels) was measured for layers s and p. The splitting of the orbital levels in spin sublevels allowed to get the magnitude of factor g for these states. Changing the orientation of the magnetic field, the g-tensor anisotropies and its relation with the details of the confinement potential had been observed. These characteristics had allowed to infer the relaxation time T1. The time resolved polarization measurement was carried out by optical pump-probe spectroscopy. The pulses of light and the transverse magnetic field allow the initialization of a net polarization. The Kerr rotation allowed to observe oscillations of this polarization around the magnetic field with frequency determined for factor g. The loss of the spin phase coherence results in the decay of these oscillations in a time scale T2. Measurements carried out in an ensemble of spins imply that the decoherence time is limited by the ensemble dephasing time T¤2 < T2. A technique similar to the spin-echo refocalization in nuclear magnetic resonance experiments using laser pulses was applied to reverse the ensemble dephasing. The possibility to measure the echo signal and the decoherence time was measured as a function of the temperature. The structure of exciton levels and its distribution in ensemble were also studied with pump-probe spectroscopy. Quantum beats were observed be-tween the fine structure exciton levels for 0D and 2D systems, yet limited by the recombination time / Doutorado / Física da Matéria Condensada / Doutor em Ciências

Cette thèse est consacrée à l’étude expérimentale des propriétés magnéto-optiques intrinsèques des nanotubes de carbone mono-paroi par spectroscopie de photoluminescence résolue en temps.Un dispositif de microscopie optique confocale de grande ouverture numérique (NA = 0.95),incluant un cryostat magnétique, permet l’étude de nanotubes suspendus à l’échelle individuelle,à température cryogénique (jusqu’à 2 Kelvin) et sous champ magnétique (jusqu’à 7 Tesla). L’évolution des spectres et des déclins de photoluminescence avec le champ magnétique montre l’influence de l’effet Aharonov-Bohm sur les deux excitons singulets de plus basse énergie, c’est à-dire l’exciton fondamental qui est optiquement inactif (exciton noir) et un exciton d’énergie supérieure séparé de quelques milliélectronvolts qui est optiquement actif (exciton brillant). L’interprétation de ces résultats à partir d’un modèle d’équations de taux qui intègre le couplage Aharonov-Bohm entre ces deux excitons permet de déterminer séparément les durées de vie excitoniques et de fournir des informations quantitatives sur la relaxation de l’énergie depuis les niveaux supérieurs photo-excités. La relaxation de l’énergie suite à la photo-excitation de la transition S22 conduit à une efficacité de peuplement de l’état brillant quatre fois plus faible que celle de l’état noir, mais qui augmente significativement lorsque la relaxation se produit depuis les niveaux excitoniques KK’. D’autre part, le bon rapport signal à bruit obtenu dans les spectres de photoluminescence permet de révéler l’existence d’un couplage intrinsèque en champ nul entre l’exciton noir et l’exciton brillant ainsi que le maintien de la mobilité excitonique dans les nanotubes suspendus à la température de l’hélium liquide. / This thesis is dedicated to the experimental study of the intrinsic magneto-optical properties of single-walled carbon nanotubes through time-resolved photoluminescence spectroscopy. Measurements are performed on suspended nanotubes samples at the single-object level using a home-built confocal optical microscope with a large numerical aperture (NA = 0.95) operating at cryogenic temperature (down to 2K) and high magnetic field (up to 7T). The evolution of the photoluminescence spectra and decay signals with increasing magnetic fields shows the influence of the Aharonov-Bohm effect on the two lowest-energy singlet excitons, namely the ground exciton which is optically inactive (dark exciton) and an exciton lying a few millielectron volts higher in energy which is optically active (bright exciton). A model of these results based on rate equations and including the Aharonov-Bohm coupling between these two excitons enables to determine separately the excitons lifetimes and to derive quantitative information on the energy relaxation from the photo-excited higher levels. The energy relaxation following the photo-excitation of the S22 transition leads to a bright state population efficiency four times lower than that of the dark state, but it significantly increases when energy relaxation occurs from the KK’ excitonic levels. Thanks to a good signal to noise ratio, the photoluminescence spectra also reveal the presence of an intrinsic zero-field coupling between the dark and the brightexcitons, as well as an excitonic mobility preserved at liquid helium temperature in suspended nanotubes.

Exploiting coordination bonding of aromatic carboxylic acids at metal surfaces, this thesis explores new directions in the design and application of self-assembled monolayers (SAMs). The SAMs are investigated using a multi-technique approach comprising of a complementary combination of scanning tunneling microscopy (STM), X-ray photoelectron spectroscopy (XPS) and near-edge X-ray absorption fine structure (NEXAFS) spectroscopy. In addition, the X-ray standing wave technique (XSW) was used to characterise the substrates. The process of layer formation and the final structures of the SAMs are found to be strikingly dependent on the combination of molecule and substrate, which is discussed in terms of the intermolecular and molecule-substrate interactions, bonding geometries and symmetry of the organic molecules. This is illustrated by the dramatic difference between molecular adsorption on Ag and Cu for molecules such as biphenyl-3,4',5-tricarboxylic acid and biphenyl-4-acetic acid. In the case of self-assembly on Cu, the molecule-substrate interactions play a decisive role in the resulting SAM structure, whereas on Ag, the intermolecular interactions dominate over the weaker molecule-substrate binding. This exploration of the balance of interactions that lead to the formation of these SAM structures lays the foundation for a systematic design of the structures and properties of aromatic carboxylic acid based monolayers. Finally, different applications and properties of some SAMs were investigated, namely coordination of a Pd(II) complex to a pyridine/pyrazole terminated molecule adsorbed on Ag. Evidence of coordination of Pd(II) to single molecules was provided by STM, XPS and NEXAFS spectroscopy. Additionally, controlled STM tip induced modification of local areas of a 1,3,5-tris(4-carboxyphenyl)benzene SAM on Ag was performed, opening an exciting prospect for nanoscale molecular manipulation.

Sementes de Sesbania virgata (Cav.) Pers. acumulam suas reservas de carbono no endosperma na forma de um polissacarídeo de parede celular, o galactomanano. Os galactomananos são polissacarídeos constituídos de uma cadeia principal de resíduos de D-manose ligadas &beta;-1,4, ramificada por resíduos de D-galactose &alpha;-1,6 ligados. A mobilização deste ocorre após a germinação e envolve três enzimas hidrolíticas (&alpha;-galactosidase, endo-&beta;-mananase e exo-&beta;-manosidase). A &alpha;-galactosidase é a primeira enzima atuar sobre o galactomanano hidrolisando as ligações &alpha;-1,6 das galactoses ramificadas a cadeia principal de manano (ligados &beta;-1,4), permitindo a ação da endo-&beta;-mananase, que hidrolisará o polissacarídeo a oligossacarídeos, onde a &beta;-manosidase atuará (ligações &beta;-1,4), transformando oligossacarídeos a monossacarídeos a serem utilizados no desenvolvimento do embrião. Buscando a compreensão das características da &alpha;-galactosidase e modo de ação sobre o galactomanano, procedeu-se com a purificação, em três etapas,e caracterização bioquímica (pH ótimo, temperatura ótima e aspectos cinéticos) da &alpha;-galactosidase de sementes de Sesbania virgata (Cav.) Pers. Além disso, visando evidenciar a modulação da enzima endo-&beta;-mananase pela distribuição de ramificações de galactose no galactomanano (estrutura fina do galactomanano), procedeu-se com hidrólises enzimáticas do galactomanano de Sesbania virgata (Cav.) Pers. utilizando a enzima endo-&beta;-mananase de Aspergillus niger (Megazyme&reg;) somente ou em conjunto com a &alpha;-galactosidase semipurificada de Sesbania virgata (Cav.) Pers. (Capítulo 1) ou com a &alpha;-galactosidase comercial de Cyamopsis tetragonoloba (Megazyme&reg;), seguido de análise dos oligossacarídeos por HPAEC-PAD (High Performance Anion Exchange Chromatography with Pulsed Amperometric Detection). Também procedeu-se com hidrólises enzimáticas de galactomananos de 6 espécies com razão manose:galactose variando de 1:1 a 150:1 com endo-&beta;-mananase de Aspergillus niger (Megazyme&reg;) e análise dos oligossacarídeos produzidos por HPAEC-PAD. A &alpha;-galactosidase semipurificada possui, aproximadamente, 42 kDa de peso molecular em condições desnaturantes e, aproximadamente 72 kDa de peso molecular na forma nativa, sugerindo que a enzima assuma estrutura quartenária. A temperatura ótima apresentada se encontra na faixa de 50°C a 55°C, pH ótimo na faixa de 4,4 a 5,4, Km= 1,8276 mM e a velocidade máxima de 0,5024 &mu;molGal.min-1.mgprot-1. A espectrometria de massas gerou os fragmentos: ALADYV-HSK-RMPGSLGHEE-QDAK-TT-GDIEDNWNSM-TSIADS NDKW-ASYAGPGGWN-DPDMLEVGNG-GMTTEEYR-AP-LLVGCDIR-VAVIL-WNR, estando a proteína referente a esta sequência relacionada à mobilização de reserva. Durante a purificação e sequenciamento interno da &alpha;-galactosidase e demais proteínas foram detectadas isoformas da &alpha;-galactosidase de pesos moleculares variados (42 kDa a 20 kDa). Sugere-se que estas isoformas encontradas inicialmente na purificação estejam relacionadas com outras funções da &alpha;-galactosidase, enquanto as isoformas encontradas após todas as etapas de purificação e identificação por espectrometria de massas estejam relacionadas com ativação e adaptação da &alpha;-galactosidase durante todo o processo de mobilização de reservas. Os dados gerados das comparações dos oligossacarídeos produzidos em cada hidrólise sugerem que as ramificações do galactomanano podem modular o reconhecimento de sítios de clivagem pela endo-&beta;-mananase: (1) existe a produção de oligossacarídeos limites de digestão F1, F2 e F3 após hidrólise do galactomanano com endo-&beta;-mananase, como demonstrado para xiloglucanos; (2) os oligossacarídeos F1 possuem proporções distintas quando da hidrólise do galactomanano com endo-&beta;-mananase em diferentes concentrações (ExP I e EXP IV), evidenciando preferência por sítios com menor grau de galactosilação; (3) a presença da &alpha;-galactosidase diminui a produção dos oligossacarídeos F2 e F3, mostrando que estes não possuem resistência intrínseca a hidrólise e que a reação atinge o equilíbrio mesmo quando ainda existem sítios de clivagem ainda disponíveis (EXP III); (4) polissacarídeos com estruturas diferentes, razão manose:galactose variando entre 150:1 a 1:1, são digeridos em diferentes taxas de hidrólise pela mesma enzima, evidenciando que a ramificação com galactose dificulta a ação da endo-&beta;-mananase. Dessa forma, sugere-se que a estrutura do polissacarídeo galactomanano também contenha, pelo menos, parte da informação requerida para seu próprio metabolismo, código para a sua degradação, estando esta informação contida na distribuição das ramificações com resíduos de D-galactose. Sendo assim, sugere-se que as diferentes isoformas da &alpha;-galactosidase relacionadas à degradação da reserva de galactomanano de sementes de Sesbania virgata (Cav.) Pers. seriam produto da ação proteolítica da própria enzima a fim de melhorar a afinidade da &alpha;-galactosidase ao substrato durante o processo de mobilização de reserva. O aumento da afinidade da &alpha;-galactosidase ao substrato durante todo o processo de mobilização garantiria a liberação das ramificações com galactose de forma contínua, permitindo e aumentando a eficiência da ação da enzima endo-&beta;-mananase aos sítios de clivagem, garantindo a degradação do polissacarídeo a oligossacarídeos de forma regulada, passível de bloqueio, pelo acúmulo de oligossacarídeos e galactose livre que inibem a ação das enzimas endo-&beta;-mananase e &alpha;-galactosidase, respectivamente, e dificultando a ação de microorganismos, propiciando ao embrião a maior quantidade de açúcares para o seu desenvolvimento, aumentando as chances de sucesso no estabelecimento da plântula / The seeds of Sesbania virgata (Cav.) Pers. have an endosperm which accumulates galactomannan as a storage polysaccharide in the cell walls. Galactomannans are composed of a linear backbone of &beta;-(1,4)-linked D-mannose residues with D-galactose &alpha;-(1,6)-linkages substitutions. The galactomannans are hydrolysed after protrusion of the radicle. This process is perfomed by three enzymes (&alpha;-galactosidase, endo-&beta;-mannanase and exo-&beta;-manosidase). The &alpha;-galactosidase is the first enzyme to cleave the polysaccharides, removing the D-galactose residues, allowing the performance of the endo-&beta;-mannanase, which hydrolyses the mannan backbone to mannan oligosaccharides. The last part of the process includes exo-&beta;-manoside, that cleaves the mannan oligosaccharides to mannose residues, which could be used by the embryo during growth. Aiming at understanding the function of ?-galactosidase in the process of galatomanannan degradation, we studied its mode of action on mannans and galactomannans. The &alpha;-galactosidase of Sesbania virgata (Cav.) Pers. was purified and characterized (pH and temperature optimum and the enzyme kinetics). We found that the semipurified &alpha;-galactosidase molecular weight was 42kDa at denaturating conditions, but in native conditions was 72kDa, suggesting that the enzyme has a quaternary structure. The enzyme optimum pH was between 4,4-5,4, optimum temperature between 50°C-55°C, Km= 1,8276 mM and Vmáx= 0,5024 &mu;molGal.min-1.mgprot-1. Mass spectrometry measures resulted the following fragments: ALADYV-HSK-RMPGSLGHEE-QDAK-TT-GDIEDNWNSMTSIADS-NDKW-ASYAGPGGWN-DPDMLEVGNG-GMTTEEYR-AP-LLVGCDIR-VAVIL-WNR, being the protein from this sequence related with storage mobilization. Possible &alpha;-galactosidase isoforms were detected during the purification, suggesting other functions for the enzyme. The &alpha;-galactosidase isoforms detected after all purification steps and with measured mass spectrometry (from 42kDa to 20kDa) should be related to the storage mobilization. We suggest that the &alpha;-galactosidase isoforms in Sesbania virgata (Cav.) Pers. seeds represents products of the enzyme self-digestion, this process being correlated with the enzyme/polysaccharide affinity and at last, correlated to the galactomannan mobilization. An extract semipurified from Sesbania virgata (Cav.) Pers. and enriched with &alpha;-galactosidase activity, was used along with endo-&beta;-mannanase from Aspergillus niger (Megazyme&reg;) or both endo-&beta;-mannanase and &alpha;-galactosidase (semipurified from Sesbania virgata seeds - Chapter 1- or commercial enzyme from Cyamopsis tetragonoloba - Megazyme&reg;) were used to study the fine structure of galactomannans. Hydrolysis of galactomannans from six species with different mannose:galactose (1:1 to 150:1) ratio were performed with endo-&beta;-mananase from Aspergillus niger. The oligosaccharides from all hydrolysis were analyzed by HPAEC-PAD (High Performance Anion Exchange Chromatography with Pulsed Amperometric Detection). The hydrolysis fragments data (HPAEC-PAD) suggest that the side-chains of the polysaccharides can modulate the hydrolytic sites recognition on the galactomannan by the endo-&beta;-mannanase. This conclusion is supported by: (1) the presence of limited digest oligosaccharides F1 and dimmers (F2) and trimers (F3) of the F1 oligosaccharides; (2) the presence of different F1 oligosaccharides proportions after hydrolysis with endo-&beta;-mannanase at different concentrations, showing preference on less-branched hydrolytic sites; (3) the &alpha;-galactosidase digestion avoided the accumulation of oligosaccharides F2 and F3, showing that these oligosaccharides do not present intrinsic resistance to hydrolysis and that the reaction reaches an equilibrium even when sites of hydrolysis are still available; (4) polymers with different fine structure (ratio mannose:galactose 1:1 to 150:1) were hydrolysed at different rates by the endo-&beta;-mannanase, showing that galactose branching interferes on the enzyme action. Considering that, the branching pattern of the polysaccharide seems to have direct influence on the interaction of the enzyme with substrate; we suggest that the structure of the galactomannan holds part of information required for its own degradation. The higher enzyme x substrate affinity, ensure the galactose branches digestion, improving the endo-&beta;-mannanase action, ensuring the degradation of the polysaccharides to oligosaccharides. This highly regulated degradation process prevents microorganisms predation and increases the plantlet establishement