A afinidade eletrônica (AE) é uma importante propriedade de átomos e moléculas, sendo definida como a diferença de energia entre a espécie neutra e seu respectivo íon negativo. Uma vez que a AE é uma fração muito pequena da energia eletrônica total das espécies neutra e aniônica, é necessário que tais energias sejam determinadas com elevado grau de precisão. A receita utilizada para o cálculo teórico acurado da AE atômica e molecular baseia-se na escolha de um conjunto adequado de funções de base juntamente com o emprego de teorias com altos níveis de correlação eletrônica. Durante o cálculo, o mesmo conjunto de base é utilizado para descrever o elemento neutro e seu respectivo ânion. Geralmente, os conjuntos de base para descrever propriedades de ânions possuem seus expoentes otimizados em ambiente neutro, e sua difusibilidade é conferida pela adição de funções difusas para cada valor de momento angular, l. A ideia deste trabalho está no desenvolvimento de conjuntos de base otimizados exclusivamente em ambiente aniônico para cálculos precisos de afinidade eletrônica. Deste modo, foram escolhidos os átomos para serem estudados: B, C, O e F. Os conjuntos de base foram gerados pelo Método da Coordenada Geradora Hartree-Fock, empregando a técnica da Discretização Integral Polinomial para a solução das integrais do problema. Os conjuntos de base obtidos são compostos por (18s13p) primitivas que foram contraídos para [7s6p] via esquema de contração geral proposto por Raffenetti. Os conjuntos contraídos foram polarizados para 4d3f2g e 4d3f2g1h, sendo os expoentes otimizados em ambiente CISD através do método SIMPLEX. Avaliaram-se as funções de base no cálculo de afinidades eletrônicas, tendo seus resultados comparados aos obtidos utilizando as bases aug-cc-pVQZ e aug-cc-pV5Z. A análise dos resultados demonstrou que os conjuntos de base difusos, gerados neste trabalho, reproduzem de maneira satisfatória as afinidades eletrônicas em relação ao valor experimental. Os conjuntos difusos polarizados para 4d3f2g1h apresentaram eficiência superior aos conjuntos aug-cc-pVQZ e, em alguns casos, aos conjuntos aug-cc-pV5Z que são consideravelmente maiores. / The electron affinity (EA) is an important property of atoms and molecules defined as the energy difference between the neutral species and its negative ion. Since the EA is a very small fraction of the total electronic energy of anionic and neutral species, one must determine these energies with high accuracy. The recipe used to calculate accurate atomic and molecular EAs is based on the choice of an adequate basis set and the use of high level of electron correlation calculations. In the computation of EAs, the same basis set is used to describe both neutral and negatively charged species. In general, the basis sets designed to describe anionic properties have their exponents optimized in neutral environment, and its diffuseness is acquired through the addition of diffuse functions for each angular momentum. The main idea of this work is to develop basis sets optimized exclusively in anionic environment that would be applied in accurate calculations of electron affinity. Thus, here follows the chosen atoms to be studied: B, C, O and F. The basis sets were generated by the Generator Coordinate Hartree-Fock Method through the Polynomial Integral Discretization Method. Basis sets were obtained containing (18s13p) primitives that were contracted to [7s6p] via Raffenetti\'s general contraction scheme. The contracted basis sets were polarized to 4d3f2g and 4d3f2g1h, and the exponents of polarization were optimized in a CISD environment through the Simplex algorithm. The basis sets quality was evaluated through the calculation of the electron affinities. The results were compared to those obtained by using the aug-cc-pVQZ and aug-cc-pV5Z basis-sets. The calculation showed that our diffuse basis sets reproduce satisfactorily the electron affinities when compared to the experimental data. The diffuse basis sets polarized to 4d3f2g1h showed to be more efficient than the aug-cc-pVQZ basis sets and in some cases also better than the aug-cc-pV5Z basis sets that are considerably larger.

O conjunto de funções de base gaussianas, o p-GCHF, foi gerado para os átomos Na, Al, Si, P, S e Cl pelo Método da Coordenada Geradora (MCG) através da expansão integral polinomial para discretizar (DIP) as equações de Griffin-Hill-Wheeler-Hartree-Fock (GHW-HF). A base p-GCHF, de qualidade 7Z, foi contraída por meio do programa Contract que opera segundo os preceitos de contração de Davidson. O processo de contração resultou em 9 funções do tipos e 7 funções p para os átomos de Na e Mg e de 9 funções s e 8 funções p para os átomos de Al, Si, P, S e Cl. Expoentes de polarização foram gerados através do programa Polarization em nível CISD para os átomos de H, B, C, N, O, F, Na, Mg, Al, Si, P, S e Cl. Foram necessários não mais do que 2p1d expoentes de polarização para compor a base p-GCHF para o átomo de H, 2d1f para os átomos da primeira fila e 3d2f, para os átomos da segunda fila da tabela periódica. Cálculos moleculares revelaram que a base p-GCHF é competitiva em energia com as bases cc-pVQZ e cc-pV5Z, entretanto apresentando custos computacionais bem menores que as mencionadas bases de Dunning. Análises das frequências vibracionais e das geometrias de otimização dos pontos estacionários, tanto mínimos quanto de estado de transição, também apontam similaridades entre o conjunto de base p-GCHF e a base cc-pVQZ, porém com diferenças de tempos de CPU que apontam a base gerada pelo MCG como computacionalmente vantajosa. Um conjunto de base capaz de descrever um dado sistema de maneira equivalente à célebres conjuntos de base da literatura, mas trazendo consigo o benefício da economia de tempo computacional é absolutamente oportuno principalmente àqueles que têm como objeto de estudo moléculas com um número considerável de átomos. / The gaussian basis sets p-GCHF was generated for a set of atoms from Na to Cl through the Generator Coordinate Method (GCM) based on a polynomial integral expansion to discretize the Griffin-Wheeler-Hartree-Fock equations. The p-GCHF is a 7z type basis sets and was contracted through the Contract program which works based on the Davidson\'s contraction model. The contraction process provided a set of 9s7p functions for Na and Mg atoms and 9s8p functions for Al, Si, P, S e Cl atoms. Polarizations exponents were acquired through the Polarization program at CISD level of theory for H, B, C, N, O, F, Na, Mg, Al, Si, P, S e Cl atoms. No more than 2p1d polarization exponents were necessary to compose the p-GCHF basis set for hydrogen atom, 2d1f for the first row of the periodic table and 3d2f for the second row one. Molecular calculations show that p-GCHF works like cc-pVQZ and cc-pV5Z basis sets but with less computational cost than the Dunning\'s ones. Vibrational frequency analysis and optimization geometry to the stationary points minimum as well as transition state, revealed similarities between p-GCHF and cc-pVQZ basis sets, but again with the GCM basis sets being computationally advantageous. A basis sets capable to describe a system like the main gaussian basis sets known in the literature but demanding less computational effort is very helpful above all for those who work with massive molecular systems.

Esta tese apresenta uma nova aproximação variacional baseada no Método da Coordenada Geradora e na Teoria do Funcional da Densidade. Nesta nova aproximação, a função de onda de muitos corpos é representada como uma superposição de determinantes de Slater Kohn-Sham não-ortogonais calculados a partir de Hamiltonianos diferentes que carregam uma coordenada geradora atuando como parâmetro de deformação. A discretização integral sobre o conjunto de coordenadas geradoras fornece a energia total variacional do sistema e a contribuição de cada determinante na combinação da respectiva função de onda de muitos corpos. A flexibilidade desta nova metodologia permitiu aplicá-la no estudo das energias totais do estado fundamental e excitado dos átomos da série isoeletrônica do Hélio, utilizando diferentes conjuntos de coordenadas geradoras, diferentes aproximações para o potencial de troca e correlação e diferentes maneiras de implementar a coordenada geradora dentro do Hamiltoniano Kohn-Sham. Em seguida, as bases desta nova metodologia foram estendidas para o caso dependente do tempo, permitindo estudar, por exemplo, processos não-lineares como excitações duplas, conhecidas por sua forte dependência dos efeitos de memória. A nova metodologia foi aplicada no estudo das oscilações paramétricas de um sistema de dois elétron sob um potencial harmônico, o átomo de Hooke. Os resultados demonstram que a escolha adequada das coordenadas geradoras reproduz com precisão os efeitos lineares e não-lineares dos elétrons do sistema que não podem ser descritos pela Teoria do Funcional da Densidade Dependente do Tempo utilizando a aproximação adiabática. Assim, a nova metodologia mostra-se: flexível, pois permite calcular propriedades do estado fundamental e excitado, estáticas e dinâmicas dos sistemas eletrônicos fornecendo ainda uma aproximação variacional para as respectivas funções de onda de muitos corpos em todos os casos; e também viável, pois fornece resultados promissores no caso independente do tempo constituindo uma ferramenta simples e computacionalmente barata de incluir os efeitos de memória em qualquer aproximação adiabática no caso dependente do tempo. / A new variational approach based on the Generator Coordinate Method and Density- Functional Theory is presented. It represents the interacting many-body wave function as a superposition of non-orthogonal Kohn-Sham Slater determinants arising from different Hamiltonians featuring a generator coordinate acting as a deformation parameter. An integral discretization procedure over the set of generator coordinates provides the variational total energy of the system and the weight of each determinant in the approximation of the respective interacting many-body wave functions. The method was used to calculate the ground and excited state total energies of the Helium isoelectronic serie of atoms using different sets of generator coordinates, different approximations to the exchange-correlation potential and different implementations of the generator coordinate whithin the Kohn-Sham Hamiltonian. Next, the time dependent extension of the method is presented allowing its application, for example, on the study of nonlinear processess as double excitations which are known to be strongly dependent of the memory effects. As an illustration, the method is sucessfully applied to driven parametric oscillations of a two interacting electrons in a harmonic potential, the Hooke\'s atom. It is demonstrated that a proper choice of time-dependent generator coordinates in conjunction with the adiabatic local-density approximation reproduces the exact linear and nonlinear twoelectron dynamics quite accurately, including features associated with double excitations that cannot be captured by Time-Dependent Density-Functional Theory in the adiabatic approximation. Therefore, the method is considered, flexible since it allows to calculate ground and excited-states, static and dynamic properties of the electronic systems yeilding a variational approach to the interacting many-body wave functions for all cases, and feasible, since it improves the results for ground and excited-states total energies in the time-independente case, besides to be a conceptually and computationally simple tool to build memory effects into any existing adiabatic exchange-correlation potential in the time-dependent case.

A afinidade eletrônica (AE) é uma importante propriedade de átomos e moléculas, sendo definida como a diferença de energia entre a espécie neutra e seu respectivo íon negativo. Uma vez que a AE é uma fração muito pequena da energia eletrônica total das espécies neutra e aniônica, é necessário que tais energias sejam determinadas com elevado grau de precisão. A receita utilizada para o cálculo teórico acurado da AE atômica e molecular baseia-se na escolha de um conjunto adequado de funções de base juntamente com o emprego de teorias com altos níveis de correlação eletrônica. Durante o cálculo, o mesmo conjunto de base é utilizado para descrever o elemento neutro e seu respectivo ânion. Geralmente, os conjuntos de base para descrever propriedades de ânions possuem seus expoentes otimizados em ambiente neutro, e sua difusibilidade é conferida pela adição de funções difusas para cada valor de momento angular, l. A ideia deste trabalho está no desenvolvimento de conjuntos de base otimizados exclusivamente em ambiente aniônico para cálculos precisos de afinidade eletrônica. Deste modo, foram escolhidos os átomos para serem estudados: B, C, O e F. Os conjuntos de base foram gerados pelo Método da Coordenada Geradora Hartree-Fock, empregando a técnica da Discretização Integral Polinomial para a solução das integrais do problema. Os conjuntos de base obtidos são compostos por (18s13p) primitivas que foram contraídos para [7s6p] via esquema de contração geral proposto por Raffenetti. Os conjuntos contraídos foram polarizados para 4d3f2g e 4d3f2g1h, sendo os expoentes otimizados em ambiente CISD através do método SIMPLEX. Avaliaram-se as funções de base no cálculo de afinidades eletrônicas, tendo seus resultados comparados aos obtidos utilizando as bases aug-cc-pVQZ e aug-cc-pV5Z. A análise dos resultados demonstrou que os conjuntos de base difusos, gerados neste trabalho, reproduzem de maneira satisfatória as afinidades eletrônicas em relação ao valor experimental. Os conjuntos difusos polarizados para 4d3f2g1h apresentaram eficiência superior aos conjuntos aug-cc-pVQZ e, em alguns casos, aos conjuntos aug-cc-pV5Z que são consideravelmente maiores. / The electron affinity (EA) is an important property of atoms and molecules defined as the energy difference between the neutral species and its negative ion. Since the EA is a very small fraction of the total electronic energy of anionic and neutral species, one must determine these energies with high accuracy. The recipe used to calculate accurate atomic and molecular EAs is based on the choice of an adequate basis set and the use of high level of electron correlation calculations. In the computation of EAs, the same basis set is used to describe both neutral and negatively charged species. In general, the basis sets designed to describe anionic properties have their exponents optimized in neutral environment, and its diffuseness is acquired through the addition of diffuse functions for each angular momentum. The main idea of this work is to develop basis sets optimized exclusively in anionic environment that would be applied in accurate calculations of electron affinity. Thus, here follows the chosen atoms to be studied: B, C, O and F. The basis sets were generated by the Generator Coordinate Hartree-Fock Method through the Polynomial Integral Discretization Method. Basis sets were obtained containing (18s13p) primitives that were contracted to [7s6p] via Raffenetti\'s general contraction scheme. The contracted basis sets were polarized to 4d3f2g and 4d3f2g1h, and the exponents of polarization were optimized in a CISD environment through the Simplex algorithm. The basis sets quality was evaluated through the calculation of the electron affinities. The results were compared to those obtained by using the aug-cc-pVQZ and aug-cc-pV5Z basis-sets. The calculation showed that our diffuse basis sets reproduce satisfactorily the electron affinities when compared to the experimental data. The diffuse basis sets polarized to 4d3f2g1h showed to be more efficient than the aug-cc-pVQZ basis sets and in some cases also better than the aug-cc-pV5Z basis sets that are considerably larger.

Dans cette thèse, nous développons un modèle collectif de la structure du noyau préservant les symétries, basé sur la théorie des fonctionnelles de la densité relativistes. Les états de référence à déformation quadrupole/octupole et à symétrie axiale sont générés en résolvant les équations de Hartree-Bogoliubov relativistes. Nous employons la fonctionnelle avec couplage ponctuel covariant DD-PC1 dans le canal particule-trou de l'interaction effective, tandis que la force d'appariement non-relativiste séparable dans l'espace des impulsions est utilisée dans le canal particule-particule. Les corrélations collectives relatives à la restauration des symétries brisées sont prises en compte en projetant les états de référence à la fois sur les bonnes valeurs du moment angulaire, de la parité et du nombre de particules. L'étape suivante consiste à combiner les états à symétries restaurées à l'aide du formalisme de la méthode de la coordonnée génératrice. Ceci nous permet d'obtenir des prédictions spectroscopiques détaillées, incluant les énergies d'excitation, les moments multipolaires électromagnétiques et les taux de transition, ainsi que les facteurs de forme élastique et inélastique. La méthode décrite est globale et peut être employée pour l'étude de la structure de nucléides très divers. Comme première application de ce modèle, nous étudierons la formation de clusters dans les noyaux légers. Le clustering nucléaire peut être considéré comme étant un phénomène de transition entre les phases liquide quantique et solide des noyaux finis. En contraste avec l'image conventionnelle du liquide quantique homogène, la localisation spatiale des particules alpha donne une image du noyau atomique similaire à une molécule. Nous réalisons en particulier une analyse complète de la collectivité quadrupole-octupole et des structures de cluster dans les isotopes du néon. Une attention particulière est accordée au cas de l'isotope ²⁰Ne, dans lequel il semble que les structures de cluster apparaissent dès l'état fondamental. Nous étudions également la structure à basse énergie de l'isotope ¹²C. Nous concentrons notre analyse sur la structure en bandes construite à partir d'états 0⁺ qui manifestent une grande variété de formes, notamment les configurations triangulaires de la bande de Hoyle ainsi que des chaînes linéaires 3-alpha dans des états de plus haute énergie. / In this thesis we develop a symmetry-conserving collective model for nuclear structure studies based on the relativistic energy density functional framework. Axially-symmetric quadrupole- and octupole-deformed reference states are generated by solving the relativistic Hartree-Bogoliubov equations. In the particle-hole channel of the effective interaction we employ the covariant point-coupling DD-PC1 functional, while the non-relativistic pairing force separable in momentum space is used in the particle-particle channel. Collective correlations related to restoration of broken symmetries are accounted for by simultaneously projecting reference states on good values of angular momenta, parity, and particle numbers. In the next step, symmetry-restored states are mixed within the generator coordinate method formalism. This enables us to obtain detailed spectroscopic predictions, including excitation energies, electromagnetic multipole moments and transition rates, as well as both the elastic and inelastic form factors. The described framework is global and it can be employed in various nuclear structure studies across the entire nuclide chart. As a first application, we will study formation of clusters in light nuclei. Nuclear clustering is considered to be a transitional phenomenon between quantum-liquid and solid phases in nuclei. In contrast to the conventional homogeneous quantum-liquid picture, spatial localization of alpha-particles gives rise to a molecule-like picture of atomic nuclei. In particular, we carry out a comprehensive analysis of quadrupole-octupole collectivity and cluster structures in neon isotopes. A special attention is paid to the case of self-conjugate ²⁰Ne isotope, where cluster structures are thought to form already in the ground state. Finally, we study the low-lying structure of ¹²C isotope. We focus on the structure of bands built on 0⁺ states that are known to manifest a rich variety of shapes, including the triangular configurations of the Hoyle band and 3-alpha linear chains in higher states.

Dans ce manuscrit de thèse, nous nous intéressons à la description des noyaux atomiques composés d’un nombre impair de nucléons dans la méthode dite de la fonctionnelle énergiede la densité (EDF). Plus précisément, nous présentons et appliquons dans le cas de ces noyaux,des extensions à cette méthode : (i) la projection sur les bons nombres quantiques, (ii) le mélange de configurations à travers la méthode des coordonnées génératrices (GCM), qui permettent deprendre en compte dans nos calculs des corrélations de type « au-delà du champ moyen » entre les nucléons constituant le noyau. De telles extensions n’avaient jusqu’alors été employées, dansleurs versions les plus générales, qu’aux noyaux ayant à la fois un nombre pair de neutrons etde protons, et nous nous proposons donc de démontrer leurs applicabilités également dans le cas des noyaux impairs. Dans la première partie de ce travail, nous présentons le formalisme mathématique de la méthode EDF, en mettant tout particulièrement l’accent sur le traitement des symétries dans cette approche. Dans la seconde partie du manuscrit, nous appliquons notre modèle au cas du noyau de 25Mg et analysons les résultats sous différents angles (ex : précision numérique des résultats, convergence du mélange de configurations, comparaison avec les données expérimentales connues). Les premiers résultats obtenus dans ce travail de thèse sontencourageants et démontrent l’intérêt de notre approche pour les calculs théoriques de structure nucléaire. / In this work, we are interested in the treatment of odd-mass atomic nuclei in energydensity functional (EDF) models. More precisely, the goal of this thesis is to develop and to applyto odd-mass nuclei, the theoretical extensions of the EDF method that are: (i) the projectiontehnique, and (ii) the configuration mixing by the generator coordinate method (GCM). Thesetwo extensions are part of the so-called multi-reference energy density functional (MR-EDF)formalism and allow for taking into account, within an EDF context, the "beyond-mean-field"correlations between the nucleons forming the nucleus. Until now, the MR-EDF formalism hasbeen applied, in its full-fledged version, only to the calculation of even-even nuclei. In this thesis,we want to demonstrate the applicability of such a model also for the description of odd-massnuclei. In the first part of this thesis, we describe the theoretical formalism of the EDF models,giving particular attention to the treatment of symmetries within our approach. In the secondpart of the manuscript, we apply our model to the nucleus 25Mg and investigate different aspectsof the method (e.g. numerical accuracy, convergence of the configuration mixing, comparison toknown experimental data). The results obtained in this work are encouraging and demonstratethe potential of our approach for theoretical nuclear structure calculations.

La fission induite par neutron, découverte il y a plus de 70 ans, a de nombreuses applications, par exemple industrielles pour la production d'énergie, et intervient dans la nucléosynthèse. Cependant, sa description microscopique reste un problème ouvert. En effet, les degrés de liberté qui interviennent dans ce processus dynamique sont complexes. De plus, les noyaux fissiles ont un nombre élevé de nucléons en interaction (>200). Il s'agit donc d'un problème à N-corps quantique. Or, une résolution directe de ce dernier n'est pas possible à l'heure actuelle. Dans ce contexte, la description microscopique de la fission considérée ici est la suivante : la première étape consiste à déterminer un ensemble de configurations de champ moyen qui représentent différentes déformations du noyau, incluant ainsi explicitement les degrés de liberté collectifs qui leur sont associés. Dans la seconde étape, la dynamique est décrite dans cet espace de configurations en utilisant la méthode de la coordonnée génératrice dépendante du temps (TDGCM). L'approximation des recouvrements gaussiens (GOA) est alors utilisée. Cependant, elle introduit une erreur de modèle et limite les extensions comme par exemple la prise en compte explicite de degrés de liberté intrinsèques. Ce travail de thèse a pour objectif de décrire le processus de fission avec la TDGCM sans recourir à la GOA. Cela implique de résoudre l'équation de la dynamique en TDGCM appelée équation de Hill-Wheeler dépendante du temps (TD-HW). Les méthodes d'évaluations des matrices des recouvrements et du hamiltonien collectif sont présentées dans le cas d'une interaction de Gogny. La matrice des recouvrements représente la métrique de l'espace des configurations, et la matrice du hamiltonien collectif contient les couplages énergétiques entre les configurations. Les configurations sont exprimées dans des bases de particules deux à deux distinctes, introduisant des instabilités numériques dans les méthodes d'évaluation standard. Un formalisme adapté à ces bases est proposé permettant d'éliminer ces instabilités. Deux méthodes de résolution de TD-HW sont présentées. La première consiste à calculer l'opérateur d'évolution associé à l'équation de Hill-Wheeler dépendante du temps. Elle est adaptée à un faible nombre de configurations. La seconde utilise un schéma de discrétisation en temps permettant l'inclusion d'un plus grand nombre de configurations dans le modèle. Ce formalisme est ensuite appliqué à la description de la réaction de fission induite par neutron sur le plutonium 239, et une comparaison avec la TDGCM+GOA est effectuée. / Nuclear fission, where an atomic nucleus separates into two fragments while emitting a large amount of energy, is at the core of many applications in society (energy production) and national security (deterrence, non-proliferation). It is also a key ingredient of the mechanisms of formation of elements in the universe. Yet, nearly 80 years after its experimental discovery its theoretical description in terms of the basic constituents of the nucleus (protons and neutrons) and their interaction remains a challenge. In this thesis, we describe the fission process as follows. In a first step, we use large supercomputers to compute the deformation properties of the nucleus based on our knowledge of nuclear forces. In a second step, we simulate the time evolution of the system from its ground state up to the fragments separation with a fully quantum-mechanical approach called the time-dependent generator coordinate method (TDGCM). While results are in good qualitative agreement with experimental data, the implementation of the TDGCM so far had been greatly simplified using what is known as the Gaussian overlap approximation (GOA). We also developed the formalism and a numerical implementation of the exact TDGCM - without the GOA. This will allow the first systematic validation of that approximation and an assessment of the resulting theoretical uncertainties. The second chapter presents the description of the neutron induced fission process using the TDGCM+GOA. The third one introduces the developments carried out in this thesis allowing the description of the fission process with the TDGCM without the GOA. The last chapter shows the first results obtained with this approach.