Sejam G um grupo topológico e X um espaço topológico. Existe uma questão natural associada ao par (G; X) sobre a existência de ações livres e contínuas de G em X. Se tal ação existe, outra questão natural é o estudo de propriedades do espaço de órbitas X / G e, nesse contexto, temos o problema usualmente difícil de se calcular o anel de cohomologia de X / G. Este trabalho é dedicado a essas questões quando X são variedades de Dold P(m;n) especiais e G = Z2. A variedade fechada e suave P(m;n), de dimensão m+2n, é o espaço de órbitas da involução livre T : Sm × CPn → Sm × CPn (x; [z]) → (-x; [ z̄ ]) e foi introduzida por Albrecht Dold em 1956, sendo bastante estudada na literatura e desempenhando papel fundamental na teoria de cobordismo. A principal ferramenta utilizada nesse estudo foi a sequência espectral de Leray-Serre associada à fibração de Borel X → XG → BG; onde XG = (X × EG) / G é a construção de Borel associada ao G-fibrado universal EG → BG. / Let G be a topological group and X be a topological space. There is a natural question associated with the pair (G; X) about the existence of a continuous free action of G on X. If such an action exists, other natural question is the study of properties of the orbit space X / G and, in this setting, the study of the cohomology ring of X / G. This thesis is devoted to these questions when X are special Dold manifolds P(m;n) and G = Z2. The closed smooth (m+2n)-dimensional manifold, P(m;n), is the orbit space of the free involution T : Sm × CPn → Sm × CPn (x; [z]) → (-x; [ z̄ ]) and was introduced by Albrecht Dold in 1956, being well studied in literature and playing a fundamental role in cobordism theory. The main tool used in this study was the Leray-Serre spectral sequence associated with the Borel fibration X → XG → BG; where XG = (X × EG) / G is the Borel construction associated with the universal G-bundle EG → BG.

Para oferecer mais informações aos cirurgiões no planejamento de intervenções cirúrgicas maxilofaciais, o presente estudo teve como objetivo avaliar a presença, localização e tamanho dos canais alveolares superiores (CAS), anterior (CASA) e posterior (CASP), canal infra-orbital (CI) e forame infra-orbital (FI) em imagens de tomografia computadorizada de feixe cônico (TCFC) de 100 pacientes adultos (&#x2265; 20 anos), obtidas pelo aparelho i-CAT Classic®. Um examinador calibrado observou a presença dos CAS, CI e FI bilateralmente. Quando presentes, eram medidos comprimento (parte descendente, parte do assoalho da órbita e total), largura e altura do CI; altura, largura e distância ao rebordo inferior da órbita (RIO) do FI; distância da borda inferior dos CAS a crista do rebordo alveolar em 5 regiões: canino (CAN), primeiro pré-molar (1PM), segundo pré-molar (2PM), primeiro molar (1M) e segundo molar (2M); e os diâmetros dos CAS. Foram também observados a localização dos CAS em relação ao seio maxilar e a presença de reparos anatômicos como canal duplo/múltiplo, anastomose intra-óssea e extensão do CASA para a abertura piriforme. Associações entre homens e mulheres, lados direito e esquerdo e regiões foram investigadas para todas as medidas e visibilidade, utilizando separadamente os métodos estatísticos. A presença do CI e FI foi de 100%. A visibilidade do CASA/CASP foi de: 99%/0% (CAN), 99%/21% (1PM), 89%/41% (2PM), 46%/44% (1M) e 7%/61% (2M). As médias dos comprimentos da parte descendente, da parte do assoalho da órbita e total do CI foram de 8,82mm, 19,44mm e 28,35mm, respectivamente. Altura e largura médias do CI foram, respectivamente, 2,08mm e 3,91mm. As médias da altura, largura e distância ao RIO do FI foram: 4,43mm, 5,18mm e 7,52mm, respectivamente. As distâncias médias do CASA ao rebordo alveolar foram: 18,54mm (CAN), 25,47mm (1PM), 28,43mm (2PM), 30,78mm (1M) e 33,21mm (2M); e do CASP: 22,3mm (1PM), 17,65mm (2PM), 15,34mm (1M) e 16,87mm (2M). As médias dos diâmetros do CASA e do CASP foram 0,90mm e 0,83mm, respectivamente, sendo que 77,5% dos CASA e 82% dos CASP eram &#x2264; 1mm e 22,5% dos CASA e 18% dos CASP eram entre 1-2 mm. Em relação ao seio maxilar, 53% e 44% dos CASA foram encontrados no terço superior e médio, respectivamente, e 64% e 36% dos CASP foram encontrados no terço inferior e médio, respectivamente. Houve diferença estatisticamente significativa (p<0,05) entre homens e mulheres para o comprimento, altura e largura do CI; para a altura, largura e distância ao RIO do FI; e para as distâncias dos CAS ao rebordo alveolar. CASA duplo, anastomose intra-óssea e extensão do CASA para a abertura piriforme foram detectados em 24,5%, 38,5% e 84% dos casos, respectivamente. A localização da anastomose foi teve maior ocorrência entre o CAN e 1PM (43%). Os resultados sugerem que a TCFC seja uma ferramenta adequada para examinar os CAS, CI e FI no planejamento de procedimentos cirúrgicos do terço médio da face, a fim de evitar danos a estruturas neurovasculares importantes. / To provide more information to clinicians in planning maxillofacial surgical interventions, the present study evaluated the presence, location and size of the superior alveolar canals (SAC), anterior (ASAC) and posterior (PSAC), infraorbital canal (IC) and infra-orbital foramen (IF) in cone beam computed tomography (CBCT) of 100 adult patients (&#x2265; 20 years old), obtained by i-CAT Classic®. One calibrated examiner observed SAC, IC and IF presence, bilaterally. When present, were measured: length (descending part, orbital floor part and total), width and height of the IC, height, width and distance from the IF to the inferior orbital rim (IOR), distance from the lower border of SAC to the alveolar crest in 5 regions: canine (CAN), first premolar (1PM), second premolar (2PM), first molar (1M) and second molar (2M); and the diameters of SAC. It was also observed the location of SAC in relation to the maxillary sinus floor and the presence of anatomical landmarks such as doble/multiple canal, intraosseous anastomosis and ASAC extension to the piriform aperture. Associations between men and women, right and left sides and regions were investigated for all measurements and presence using statistical methods separately. The presence of the IC and IF was 100%. The presence of ASAC/PSAC was: 99%/0% (CAN), 99%/21% (1PM), 89%/41% (2PM), 46%/44% (1M) e 7%/61% (2M). The mean lengths of the IC descending part, IC orbital floor part and total lenght of the IC were 8.82mm, 19.44mm and 28.35mm, respectively. The mean height and width of the IC were, respectively, 2.08mm and 3.91mm. The mean height, width and distance to the IOR of IF were: 4.43mm, 5.18mm and 7.52mm, respectively. The mean distances of ASAC to the alveolar crest were: 18.54mm (CAN), 25.47mm (1PM), 28.43mm (2PM), 30.78mm (1M) and 33.21mm (2M); and of PSAC: 22.3mm (1PM), 17.65mm (2PM) 15.34mm (1M) and 16.87mm (2M). The mean diameters of ASAC and PSAC were 0.90mm and 0.83mm, respectively, and that, 77.5% of ASAC and 82% of PSAC were &#x2264; 1mm and 22.5% of ASAC and 18% of PSAC were between 1-2 mm. In relation to the maxillary sinus, 53% e 44% of the ASAC were found in the upper and middle third of maxillary sinuses, respectively, and 64% and 36% of the PSAC were found in lower and middle third, respectively. There was a statistically significant difference (p<0.05) between men and women for the length, height and width of the IC, for the height, width and distance to the IOR of IF, and for distances of SAC to the alveolar crest. Double ASAC, intraosseous anastomosis and ASAC extension to the piriform aperture were present in: 24.5%, 38.5% and 84% of the cases, respectively. The location of the anastomosis had more incidence between the CAN and 1PM (43%). The results suggest that CBCT is a suitable tool to evaluate the SAC, IC and IF, assisting surgeons in planning surgical procedures of the midface, in order to avoid damage to important neurovascular structures.

[pt] Este trabalho está relacionado ao problema da utilização eficiente da órbita de satélite geoestacionários. Neste problema, os satélites devem ser alocados adequadamente na órbita, obedecendo a limites de interferência máxima, de modo a satisfazer os requisitos de desempenho dos sistemas. Algumas técnicas para a determinação das posições orbitais de satélites existentes e planejados são mencionadas. Um novo modelo matemático para a otimização das posições orbitais dos satélites, baseado na minimização de uma função objetivo associada à interferência total que afeta cada um dos sistemas, é proposto. Finalmente, alguns resultados obtidos através da aplicação deste modelo são analisados e comparados a outros resultados disponíveis. / [en] This work is related to the problem of efficiently using the geoestationary satellite orbit in an interference limited environment. Some techniques to determine the orbital positions of existing and planned satellites are mentioned. A new mathematical model for optimizing satellite orbital positions based on the minimization of an objective function associated to the total interference affecting each system is proposed. Finally, results obtained through the application of this model are analysed and compared to other available results.

Cette thèse est principalement consacrée à des problèmes de dynamique linéaire dans les espaces de Banach. Répondant à une question récente de Hajek et Smith, on construit notamment, dans tout espace de Banach séparable, un opérateur borné tel que ses orbites tendent vers l'infini sur une partie ni vide, ni dense. On relie également, à l'aide d'un autre résultat, le module de lissité asymptotique au comportement des opérateurs bornés. / This work is mainly devoted to some problems of linear dynamics in Banach spaces. In particular, we answer a recent question of Hajek and Smith by constructing, in any separable Banach space, a bounded operator such that its orbits tending to infinity form a set which is neither empty, nor dense. We also connect the behaviour of bounded operators with the asymptotic modulus of smoothness.

Cette thèse est consacrée à l'étude des propriétés de transport statique et dynamique du gaz d'électrons bidimensionnel supraconducteur à l'interface LaAlO3/SrTiO3. Dans un premier temps, nous étudions l'effet du désordre microscopique induit par le dopage en Chrome, sur la supraconductivité et le couplage spin-orbite en fonction de la densité de porteur modulée par effet de champ. Dans une géométrie de grille locale au-dessus du gaz, nous montrons le contrôle électrostatique de la transition supraconducteur-isolant. De même, nous analysons l'ajustement du couplage spin-orbite contrôlé par effet de champ. A l'aide de méthodes de nanofabrication par lithographie électronique, nous démontrons la première réalisation d'un point critique quantique dans LaAlO3/SrTiO3. En changeant le confinement latéral et le niveau de Fermi par effet de champ, nous sommes capables de régler le nombre de canaux conducteurs dans l'état normal et de mesurer la quantification de la conductance. Enfin, nous présentons des mesures radio-fréquence qui donnent accès aux propriétés dynamiques du gaz supraconducteur. L'évolution de la conductivité en fonction de la densité de porteurs et de la température est comparée avec la théorie standard BCS/Mattis-Bardeen d'une part, et avec la théorie BKT d'autre part. / This thesis is devoted to the study of static and dynamical transport properties of the superconducting two-dimensional electron gas at the LaAlO3/SrTiO3 interface. Under strong 2D confinement, the degeneracy of the t$_{2g}$ bands of SrTiO$_3$ is lifted at the interface, generating a rich and complex band structure. Starting from a free electron model, we derive numerically a self-consistent calculation of the potential well and the band structure (chapter 1). These simulations highlight the presence of two types of bands d$_{xy}$ and d$_{xz/yz}$ with very different transport properties. We investigate first the effect of microscopic disorder introduced by Cr doping, on superconductivity and spin-orbit coupling over a wide range of back-gate doping (chapter 3). We also describe the first implementation of a field-effect device where the superconductor-insulator transition could be continuously tuned with a top-gate. The presence of a strong spin-orbit coupling that could be controlled with the top-gate voltage is also demonstrated by analyzing the magneto-transport measurements. The gate dependence of the spin-splitting energy, of the order of a few meV, is found to be consistent with Rashba spin-orbit coupling. Going one step further in nanofabrication, we report on the first realization of a quantum point contact in LaAlO$_3$/SrTiO$_3$ using split gates (chapter 6). To go further in the understanding of the LaAlO$_3$/SrTiO$_3$ interface, we present high frequency measurements of the conductivity $\sigma$ (chapter 5). This measurement gives us access to the superfluid stiffness and to the gap energy via the BCS theory. We show that the competition between these two energy scales controls the superconducting Tc in the phase diagram.

Les éphémérides planétaires jouent un rôle important pour la navigation des missions spatiales actuelles et la mise en place des missions futures ainsi que la réduction et l'analyse des observations astronomiques les plus précises. Les éphémérides planétaires actuelles (DE, INPOP, EPM)L'objectif de la thèse est d'utiliser des archives de données de missions spatiales passées et présentes et de fournir des outils d'analyse pour l'amélioration de l'éphéméride de données pour l'amélioration de l’éphéméride planétaire planétaire INPOP, ainsi que pour une meilleure utilisation des éphémérides pour effectuer des teste de la physique tels que la relativité générale, les études de la couronne solaire [...] / The planetary ephemerides play a crucial role for spacecraft navigation, mission planning, reductionand analysis of the most precise astronomical observations. The construction of suchephemerides is highly constrained by the tracking observations, in particular range, of the spaceprobes collected by the tracking stations on the Earth. The present planetary ephemerides (DE,INPOP, EPM) are mainly based on such observations. However, the data used by the planetaryephemerides are not the direct raw tracking data, but measurements deduced after the analysisof raw data made by the space agencies and the access to such processed measurements remainsdifficult in terms of availability.The goal of the thesis is to use archives of past and present space missions data independentlyfrom the space agencies, and to provide data analysis tools for the improvement of theplanetary ephemerides INPOP, as well as to use improved ephemerides to perform tests ofphysics such as general relativity, solar corona studies, etc.The first part of the study deals with the analysis of the Mars Global Surveyor (MGS)tracking data as an academic case for understanding. The CNES orbit determination softwareGINS was used for such analysis. The tracking observations containing one-, two-, and threewayDoppler and two-way range are then used to reconstruct MGS orbit precisely and obtainedresults are consistent with those published in the literature. As a supplementary exploitationof MGS, we derived the solar corona model and estimated the average electron density alongthe line of sight separately for slow and fast wind regions. Estimated electron densities arecomparable with the one found in the literature. Fitting the planetary ephemerides, includingadditional data which were corrected for the solar corona perturbations, noticeably improves theextrapolation capability of the planetary ephemerides and the estimation of the asteroid masses(Verma et al., 2013a).The second part of the thesis deals with the complete analysis of the MESSENGER trackingdata. This analysis improved the Mercury ephemeris up to two order of magnitude comparedto any latest ephemerides. Such high precision ephemeris, INPOP13a, is then used to performgeneral relativity tests of PPN-formalism.[...]

An important part of the fusion research program is to understand and control the large number of plasma instabilities that a fusion plasma can exhibit. One such instability is known as the “sawtooth” instability, which is a perturbation in the plasma electric and magnetic fields that manifests itself as periodic relaxations of the temperature and density in the plasma center. The aim of this project was to investigate how the fuel ions in a fusion plasma react to the sawtooth instability. We were able to implement a model of the plasma electromagnetic field during a sawtooth relaxation into an existing code that computes the orbits of the fuel ions in the tokamak magnetic field. To this end, it was necessary to modify the orbit code to allow for non-zero electric fields, and for time-varying fields. In order to validate the new additions to the code, we compared simulated results to analytical ones. The model of the sawtooth electromagnetic fields required for our simulations was set up within a different student project. However, due to unforeseen complications, only the magnetic (not the electric) field contribution was available to us during our project, but once the electric field is available it is straightforward to include in our code. Our simulations did not exhibit any noticeable perturbation to the particle orbit during a sawtooth crash. However, before the electric field contribution is included it is not possible to draw any physics conclusions from these results. Our code could also be used as a foundation for future projects since it is possible (with further implementations to the existing code) to simulate how the spatial profile of the neutron emission is expected to vary during the sawtooth. These simulations can be compared against experimental measurements of the neutron emission profile in order to investigate the accuracy of the sawtooth model under consideration.

L’objectif de ce travail de thèse est d’étudier les possibilités d’excitation, de propagation et de détection d’ondes de spin dans des couches ultraminces de Y₃Fe₅O₁₂(YIG). Pour ce faire, nous proposons d’accéder à la dynamique d’aimantation du YIG à travers l’exploitation de phénomènes de transport électronique polarisé en spin liés aux effets Hall de spin (SHE) existant au sein de métaux normaux (MN) présentant un fort couplage spin-orbite.Nous avons pour cela fait croître, par ablation laser pulsée, des films de YIG épitaxiés d’épaisseur nanométrique. En effet, le fait de pouvoir disposer de films magnétiques ultraminces est nécessaire pour ouvrir le champ d’étude du YIG à la lithographie et aux techniques de micro-fabrications de la micro-électronique en vue notamment d’applications magnoniques.L’étude de la dynamique d’aimantation de ces couches magnétiques par résonance ferromagnétique (FMR) nous a permis de quantifier la constante d’amortissement magnétique de Gilbert pour ces films. Pour les meilleurs films ayant une épaisseur de 20 nm celle-ci est de de 2*10^-4, soit une valeur comparable avec celle des films de YIG d’épaisseur micrométrique obtenus par la technique standard d’épitaxie en phase liquide et seulement un ordre de grandeur supérieure à l’amortissement du YIG massif (a=3*10^-5). Nous avons pu montrer par des mesures inductives à l’analyseur de réseau que les ondes de spin pouvaient se propager dans de tels films sur plusieurs centaines de microns, ce résultat valide la pertinence de l’emploi de couches de YIG de 20 nm d’épaisseur pour des applications magnoniques.Afin de procéder à des mesures de détection/excitation d’ondes de spin par exploitation du SHE, nous avons déposé différents métaux (Pt, Pd, CuBi) sur nos films de YIG, formant ainsi des bicouches YIG|MN. Nous avons mis en évidence la détection non locale d’ondes de spin due aux phénomènes d’effet Hall de spin inverse dans le MN et de pompage de spin dont nous avons caractérisé le paramètre de mixing conductance g à partir d’analyse de résonance ferromagnétique.Afin d’apporter un élément de réponse sur l’origine d’effets de la magnétorésistance observés dans des bicouches YIG|MN, origine intrinsèque ou extrinsèque (induite par une polarisation magnétique du MN), nous avons effectué des mesures de dichroïsme magnétique circulaire de rayons X au niveau du seuil K du Pd dans des bicouches YIG|Pd. Ces mesures indiquent l’absence de moments magnétiques induits par le YIG dans le Pd avec un seuil de détection estimé à 10^-5µB/atome. Ce résultat, associé à des mesures de magnétotransport réalisées sur les mêmes échantillons, nous a permis de montrer que le pompage de spin entre un isolant ferromagnétique et un métal non magnétique, permettait seul (sans effets de proximité magnétique) d’expliquer les effets de magnétorésistance d’effet Hall de spin observés dans ces bicouches.Nous avons étudié la dynamique d’aimantation de nanodisques de YIG et YIG|MN par microscopie à force de résonance magnétique. Cette étude a permis de montrer que la nanostructuration de ces films ne dégrade pas les paramètres a et g et permet même, grâce au confinement géométrique des modes propres de résonance, d’annuler la contribution inhomogène (DH0) à largeur de raie.Finalement, à partir de l’injection dans le MN d’un courant électrique continu, nous avons mis en évidence la possibilité de modifier (augmenter/diminuer) la largeur de raie du mode FMR d’un microdisque de YIG|MN, exploitant ainsi le couple de transfert de spin d’un courant de spin créé dans le MN par effet de Hall de spin direct. Passé une valeur critique de courant, nous avons pu activer (en l’absence d’excitation RF) un régime d’auto oscillations de l’aimantation du YIG. Ce dernier résultat qui confirme la possibilité d’exciter la dynamique d’aimantation dans le YIG par un couple de transfert de spin est très prometteur pour l’intégration des dispositifs magnoniques au sein de circuits électroniques. / The aim of this thesis is to study the excitation, the propagation and the detection of spin waves into ultrathin Y₃Fe₅O₁₂ (YIG) films. Our approach consists in interacting with the YIG magnetization dynamics thanks to spin polarized electronic transport phenomena related to spin Hall effect (SHE) that occurs within normal metals (NM) which have a large spin orbit coupling.To do so, we first grew epitaxially, by pulsed laser deposition, nanometer thick YIG films. Having ultrathin magnetic films is indeed necessary to be able to use lithography and microfabrication technics in order to develop magnonic devices.Studying these films by ferromagnetic resonance (FMR) allowed us to quantify the Gilbert magnetic damping constant, which is down to 2∙〖10〗^(-4) for the best 20 nm thick films. This value is comparable to which is reported for micrometer thick YIG films grown by liquid phase epitaxy and is only one order of magnitude higher than the bulk YIG damping (α = 3∙〖10〗^(-5)). Inductive measurements performed with a vectorial network analyzer show that spin waves can propagate over hundreds microns through such thin films. The suitability of 20 nm thick YIG films for magnonic applications is therefore confirmed by this result.In order to detect/excite spinwave by exploiting SHE, we have deposited different metals (Pt, Pd, CuBi) on our YIG films, thus obtaining YIG|NM bilayers. We have performed a nonlocal detection of spin waves based on inverse SHE and spin pumping phenomena. From the FMR analysis, we also measured the spin mixing conductance g_(↑↓) that quantify the efficiency of spin pumping.To answer the question of the origin of magnetoresistance observed into YIG|NM bilayers, intrinsic or extrinsic origin (ie. induced by the magnetic polarization of the NM), we have measured the X-ray magnetic circular dichroism of YIG|Pd bilayers at the K-edge of Pd. These measurements do not show any induced magnetic moment for YIG to Pt for a resolution of 〖10〗^(-5) μ_B/atom. Combining this result with magnetotransport measurements performed on the same samples, allowed us to show that the spin pumping between a ferromagnetic insulator and a nonmagnetic metal could explain (without any magnetic proximity effects) spin Hall magnetoresistance effects observed into these bilayers.We studied the magnetization dynamics of YIG and YIG|NM nanodisks by magnetic resonance force microscopy. This study has shown that the nanostructuration performed on these films doesn’t affect such parameters as α and g_(↑↓), the geometrical confinement even leads to suppress the linewidth inhomogeneous broadening (ΔH_0).Then, we injected a continuous electrical current through the NM. Using the spin transfer torque of the spin current thus created into the NM by direct spin Hall effect, it allowed us to modified (increased/decreased) the FMR mode’s linewidth of a YIG|NM microdisk. After reaching a critical current, we managed to enable (without any RF excitation) a regime of magnetization auto-oscillations into the YIG. This last result, which confirms the ability of exciting the YIG magnetization dynamics by spin transfer torque, is very promising for the integration of magnonic devices into electronic circuits.

L’interaction spin-orbite a permis de reproduire dans les modèles nucléaires théoriques, les nombres magiques N=28 et 50 observés dans les noyaux atomiques. Ces dernières décennies, l’étude expérimentale de noyaux exotiques a mis en évidence une évolution des nombres magiques loin de la vallée de stabilité. On peut alors se poser la question de l’évolution des potentiels d’interaction eux-mêmes, et en particulier de l’interaction spin-orbite. Si cette interaction a été historiquement incluse « à la main » dans les modèles de champ moyen « classiques », elle émerge cependant naturellement dans les modèles relativistes. La description de l’interaction spin-orbite est très similaire dans ces deux types de modèles, mais il subsiste a priori un désaccord du point de vue de sa dépendance en isospin : les modèles non-relativistes de type Hartree-Fock présentent en effet un potentiel spin-orbite dépendant fortement de l’isospin, contrairement aux modèles de type Relativistic Mean Field.En 2009, des calculs mettant en œuvre différents modèles théoriques ont prédit l’existence d’une « bulle », caractérisée par une déplétion en densité protonique centrale, dans le ³⁴Si. Ce dernier aurait une densité protonique très exotique, et bien différente de sa densité neutronique. Le ³⁴Si constituerait alors une sonde idéale de l’évolution du potentiel spin-orbite dans les systèmes présentant une forte asymétrie protons-neutrons. L’émergence d’un tel effet trouverait son origine dans la déplétion de l’orbitale protonique2s½, les orbitales s étant les seules à contribuer à la densité nucléaire centrale.Une expérience réalisée en Septembre 2012 à NSCL (MSU, Etats-Unis), a permis de mettre en évidence pour la première fois un effet de bulle nucléaire dans le ³⁴Si. L’étude des facteurs spectroscopiques des états peuplés lors des réactions d’arrachage de proton ou de neutron ³⁴Si(-1p) ³³Al et ³⁴Si(-1n) ³³Si indique que sa structure neutronique est très proche d’un système sans corrélations au-delà du champ moyen, tandis que son orbitale protonique est très faiblement occupée : n(2s½) = 0,16(4).Les réactions ³⁶S(-1p) ³⁵P et ³⁶S(-1n) ³⁵S ont été étudiées dans les mêmes conditions expérimentales. L’évolution de l’occupation n(2s½) mesurée entre le ³⁶S et le ³⁴Si, ainsi que la variation de l’écart en énergie des partenaires spin-orbite neutroniques 2p½-2p^3/2, mesurée entre ces deux noyaux dans une expérience antérieure, sont en faveur des modèles de champ moyen non-relativistes. La partie théorique de cette thèse a cependant montré que la différence de comportement de l’interaction spin-orbite entre modèles relativistes et non-relativistes est en fait un artefact causé par l’omission du terme d’échange dans les calculs de type Relativistic Mean Field. En effet, l’inclusion du terme de Fock dans les modèles relativistes permet de rétablir la dépendance en isospin du potentiel spin-orbite observée dans le cas non-relativiste. / The spin-orbit interaction is essential for the reproduction of magic numbers N=28 and 50 in theoretical nuclear models. Over the past few decades, the experimental study of exotic nuclei has highlighted an evolution of magic numbers far from stability. One can then wonder about the evolution of nuclear potentials themselves, and in particular the one of spin-orbit interaction. Historically, this interaction was included « by hand » in mean field models, whereas it naturally arises in relativistic mean field models. The description of the spin-orbit interaction happens to be very similar in those two kinds of models, but there remains a disagreement regarding its isospin dependance. Indeed, Hartree-Fock models exhibit a spin-orbit potential which strongly depends on isospin, contrary to relativistic mean field models.In 2009, a proton bubble was predicted in ³⁴Si by means of several different nuclear models. This effect consists in a central proton central density depletion. ³⁴Si would exhibit a quite exotic proton density, and very different from its neutron density. This nucleus would then constitute an ideal probe to test the behaviour of the spin-orbit potential in systems with strong proton-neutron asymmetry. The appearance of such an effect would originate from the depletion of proton 2s½ orbitals, as s orbitals are the only ones contributing to the central density.An experiment which was performed in September 2012 at NSCL (MSU, United States) highlighted for the first time a proton bubble in ³⁴Si. The spectroscopic strengths of states populated in the knockout reactions ³⁴Si(-1p)³³Al and ³⁴Si(-1n)³³Si reveal that the neutron structure of ³⁴Si is close to the one of a system without beyond-mean-field correlations, whereas its proton orbital is only weakly occupied : n(2s½) = 0,16(4).The reactions ³⁶S(-1p)³⁵P and ³⁶S(-1n)³⁵S were studied in similar experimental conditions. The change in occupancy n(2s½) measured between ³⁶S and ³⁴Si, as well as the variation in the neutron spin-orbit splitting 2p½-2p^3/2 measured in an earlier experiment, suggest that non-relativistic models exhibit the right isospin dependance. The theoretical part of this thesis showed however that the difference in behaviour of the spin-orbit interaction between relativistic and non-relativistic model is actually an artefact caused by the omission of the exchange term in relativistic mean field calculations. Indeed, including the Fock term in relativistic models enables to restore the isospin dependance observed in the non-relativistic case.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第22582号 / 情博第719号 / 新制||情||123(附属図書館) / 京都大学大学院情報学研究科数理工学専攻 / (主査)教授 矢ヶ崎 一幸, 教授 中村 佳正, 教授 梅野 健 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM