Vingt années se sont écoulées depuis la découverte de l'expansion accélérée de l'Univers, ravivant l'intérêt pour les théories alternatives de la gravité. Ajouter un champ scalaire à la métrique habituelle de la relativité générale est l'une des manières les plus simples de modifier notre théorie de la gravité. En parallèle, nos connaissances sur les trous noirs et les étoiles à neutrons sont en plein essor, grâce notamment au développement de l'astronomie par ondes gravitationnelles. Cette thèse se situe au carrefour entre les deux domaines : elle étudie les propriétés des objets compacts dans les théories tenseur-scalaire généralisées. Je commence par rappeler les théorèmes d'unicité essentiels établis depuis les années soixante-dix. Après avoir présenté le théorème d'unicité pour les trous noirs en théorie de Horndeski, je l'étends aux étoiles. La deuxième partie de cette thèse détaille les différentes manières de contourner ce théorème. Parmi elles, je présente des solutions où la dépendance temporelle du champ scalaire permet de le raccorder à une solution cosmologique, mais aussi des trous noirs statiques et asymptotiquement plats. Dans la troisième partie, j'établis un critère important pour la stabilité de ces solutions, qui s'appuie sur leur structure causale. C'est aussi l'occasion d'étudier la propagation des ondes gravitationnelles au voisinage de trous noirs, et de sélectionner les théories dans lesquelles les ondes gravitationnelles se propagent à la même vitesse que la lumière. / Twenty years have passed since the discovery of the accelerated expansion of the Universe, reviving the interest for alternative theories of gravity. Adding a scalar degree of freedom to the usual metric of general relativity is one of the simplest ways to modify our gravitational theory. In parallel, our knowledge about black holes and neutron stars is booming, notably thanks to the advent of gravitational wave astronomy. This thesis is at the crossroads between the two fields, investigating the properties of compact objects in extended scalar-tensor theories. I start by reviewing essential no-hair results established since the seventies. After discussing the no-hair theorem proposed for black holes in Horndeski theory, I present its extension to stars. The second part of the thesis investigates in detail the various ways to circumvent this theorem. These notably include solutions with a time-dependent scalar field in order to match cosmological evolution, but also static and asymptotically flat configurations. In a third part, I establish an important stability criterion for these solutions, based on their causal structure. It is also the occasion to study the propagation of gravitational waves in black hole environments, and to select the theories where gravitational waves travel at the same speed as light.

Les trous noirs sont produits par effondrement gravitationnel d'étoiles supermassives et contiennent en leur centre une singularité de l'espace-temps habillée d'un horizon auquel rien ne peut s'échapper. Ils se situent à la frontière théorique commune entre la Relativité Générale et la Mécanique Quantique, ce qui en fait le principal laboratoire théorique et expérimental pour tester les théories quantiques de la gravité comme la Théorie des Cordes. L'entropie d'un trou noir est énorme, de l'ordre de sa masse au carré. Comme tout objet entropique, une description microscopique en termes de dégénérescence d'états devrait exister. De plus, le trou noir s'évapore par rayonnement d'Hawking et l'information à l'intérieur semble perdue, ce qui compromet la principe d'unitarité, pierre angulaire de la Mécanique Quantique. Par conséquent, la Théorie des Cordes doit fournir les degrés de liberté nécessaires pour décrire la nature de micro-état de trous noirs, elle doit également trouver un mécanisme résolvant la singularité et le paradoxe de la perte d'information. Cette thèse porte sur la physique des trous noirs à travers le "fuzzball proposal" et le "microstate geometry program". La majeure partie de la discussion se déroulera dans la limite de basse énergie de la Théorie des Cordes, c'est-à-dire en Supergravité. Le ``proposal" stipule qu'il existe "eS" solutions non singulières sans horizon qui ressemblent à un trou noir à large distance mais qui diffèrent à proximité de l'horizon. Sur la base de cette affirmation, la solution de trou noir classique correspond à la description statistique d'un système de solutions qui ont la même géométrie que le trou noir à l'extérieur de l'horizon, mais qui se terminent par des géométries régulières, dites "fuzzy". La proposition soulève plusieurs questions : Comment la singularité est-elle résolue ? De telles géométries peuvent-elles être construites en Supergravité ? Comment l'information s'échappe-t-elle de l'ensemble des micro-états ? La thèse est décomposée en trois parties. La première partie présente les bases et donne un aperçu du "microstate geometry program". La deuxième partie regroupe cinq travaux qui se consacrent à construire de larges familles de micro-états de trous noirs supersymétriques ou non supersymétriques. La dernière partie passe en revue deux travaux. L'un d'eux étudie le processus de diffusion dans les micro-états. Cela permet d'élucider comment le principe d'unicité est restaurée et comment l'information s'échappe des micro-états. La seconde traite du rôle des micro-états dans le contexte de la correspondance AdS2/CFT1 et donne l'ébauche d'une preuve pour le "fuzzball proposal". / Black holes are produced by gravitational collapse of supermassive stars and consist of a spacetime singularity dressed by a horizon from which nothing can escape. They lie at the common theoretical border between General Relativity and Quantum Mechanics, making them the main theoretical and experimental laboratory for testing quantum theories of gravity as String theory. The entropy of a black hole is huge, of the order of its mass squared. As any entropic object, a microscopic description in terms of large degeneracy of states should exist. Moreover, black hole evaporates through thermal Hawking's radiation and the information in the interior seems lost, that compromises the unitary principle, a cornerstone of Quantum Mechanics. Therefore, String Theory must provide the degrees of freedom necessary to describe the microstate nature of black holes, it must also find a mechanism resolving the singularity and the information loss paradox. This thesis addresses black-hole physics through the lens of the fuzzball proposal and the microstate geometry program. The major part of the discussion will be conducted in the low-energy limit of String Theory, that is in Supergravity. The proposal states that there exist "eS" horizonless non-singular solutions that resemble a black hole at large distance but differ in the vicinity of the horizon. Based on this statement, the classical black-hole solution corresponds to the average description of a system of solutions which match the black-hole geometry outside the horizon but cap off as ``fuzzy" smooth geometries in the infrared. The proposal leads to several questions: How is the singularity resolved? Can "eS" such geometries be built in Supergravity? How does the information escape from the ensemble of microstates?The thesis is decomposed in three parts. The first part introduces the basic materials and gives a review of the microstate geometry program. The second part gathers five works that all consist in constructing large classes of smooth horizonless microstate geometries of supersymmetric or non-supersymmetric black holes. The last part review two works. One is investigating the scattering process in microstate geometries. This helps to elucidate how unitarity is restored and how information escapes from black-hole backgrounds. The second one addresses the role of microstate geometries in the context of the AdS2/CFT1 correspondence and gives a beginning of proof for the fuzzball proposal.

As a non-linear theory of space-time, general relativity deals with interesting dynamical systems which can be expected more prone to chaos than their Newtonian counter-parts. In this thesis, we study the dynamics of time- like geodesics in the static and axisymmetric field of a Schwarzschild black hole surrounded, in a concentric way, by a massive thin disc or ring. We reveal the rise (and/or decline) of geodesic chaos in dependence on parameters of the sys- tem (the disc/ring mass and position and the test-particle energy and angular momentum), (i) on Poincaré sections, (ii) on time series of position and their power spectra, (iii) by applying two simple yet powerful recurrence methods, and (iv) by computing Lyapunov exponents and two other related quantifiers of or- bital divergence. We mainly focus on "sticky" orbits whose different parts show different degrees of chaoticity and which offer the best possibility to test and compare different methods. We also add a treatment of classical but dissipative system, namely the evolution of a class of mechanical oscillators described by non-standard constitutive relations.

Stars can be stretched and ripped apart by the super-massive black hole at the core of a galaxy. The remnant gaseous trail gradually circularizes in a ring of mass that spreads by the viscous forces into an accretion disc. In this thesis we have studied the spectral line profle time evolution of radiation refected by the accretion disc located around a super-massive black hole. We assume the central body to be a slowly rotating or non-rotating super-massive black hole with no charge, in the frst approximation represented by the Schwarzschild solution. In a sense of Shakura-Sunyaev standard accretion disc model with the kinematic viscosity parameter α ≈ 1 we allow the accretion disc evolution to be guided by the angular momentum transfer equation with the initial mass ring located at the tidal radius being the product of tidal disruption of a star passing by a super-massive black hole. During the simulations we keep varying the mass of the central body while we keep the mass and the radius of the star constant (M = 1M⊙ and R = 1R⊙), i.e. taking into account the solar-type stars only. We defer the prospects of the full analysis involving spin (and charge) of the central body for the future study as it will be necessary to use the equations for the redshift factor and the accretion disc evolution...

This thesis is devoted to various aspects of electric-magnetic duality and its gravitational generalization, with an emphasis on the case of maximal supergravity. It is divided into three parts.In the first part, we review the symmetries of maximal supergravity in various dimensions, with a particular focus on the exceptional ``hidden" symmetries that appear upon toroidal dimensional reduction of eleven-dimensional supergravity. Two new results are obtained. First, we prove in detail that these hidden symmetries appear if and only if the Chern-Simons coupling of the eleven-dimensional theory takes the value predicted by supersymmetry. Second, we obtain a manifestly E7(7)-invariant formula for the entropy of non-extremal black holes in four-dimensional N = 8 supergravity.The second part of the thesis concerns the gaugings of extended supergravities in four dimensions. We first show that the embedding tensor formalism does not allow for deformations that cannot be reached by working with the usual Lagrangian in the duality frame picked by the embedding tensor. We then examine through BRST methods the deformations of a large class of non-minimally coupled scalars and abelian vector fields in four dimensions, of which ungauged supergravities offer a prime example. We prove that all local deformations of these models which modify the gauge transformations are of the usual Yang-Mills type, i.e. correspond to the gauging of some rigid symmetries of the undeformed theory. Combined with the first result, this shows that the embedding tensor formalism correctly captures the most general local deformations of these theories.In the third part, we construct self-contained action principles for several types of free fields in six dimensions, whose field strengths satisfy a self-duality condition. These fields are motivated by two considerations. First, their existence allows for a remarkable geometric interpretation of the electric-magnetic duality symmetries of vector fields and linearized gravity in four dimensions. Second, they appear in the spectrum of the chiral N = (4,0) and N = (3,1) ``exotic supergravities" in place of the usual metric. The free action and supersymmetry transformations for those theories are explicitly constructed. We also check that they reduce to linearized maximal supergravity in five dimensions, thus completing the picture of higher-dimensional parents of N = 8 supergravity, at least at the free level. Along the way, we also generalize previous works on linearized supergravity by other authors, in which the graviton and its dual appear on the same footing at the level of the action.We conclude with some open questions, perspectives for future work, and a series of technical appendices. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

The cores of galaxy clusters contain the most massive galaxies in the Universe, the brightest cluster galaxies. These galaxies are unique compared to their counterpart galaxies outside of clusters as they have much brighter cores, and vast spatially- extended stellar envelopes. The theoretical picture of how they reached their huge masses relied on the idea of gradual stellar mass growth during the second half of the history of the Universe. However, recent observational evidence of highly-overdense protoclusters, the progenitors of these galaxies, demonstrates that some brightest cluster galaxies may have assembled within the first few billion years after the Big Bang – seemingly contradicting our theoretical predictions. I include my theoretical work that shows the short timescales over which these observed protoclusters trans- form into the brightest cluster galaxies and discuss the likelihood of finding these rare protoclusters in the early Universe. To push our understanding of the rapid evolution of these galaxies even further for- ward demands the use of numerical simulations due to the highly coupled, non-linear astrophysical processes that occur during the process. In this dissertation, I include improvements to our numerical models of hydrodynamical turbulence and supermas- sive black holes that I incorporated into a state-of-the-art hydrodynamical+gravity simulation code, in effort to provide the groundwork to improving our understanding of the build-up of the brightest cluster galaxies in the early Universe, and galaxy evolution in general. / Graduate

Dans l’Univers, on observe des galaxies lointaines ne formant plus d’étoiles, mais les astrophysiciens n’ont pas encore identifié avec certitude les phénomènes physiques à l’origine de leur “mort”. Pour apporter des éléments de réponse, je me suis penchée sur l’étude de phénomènes qui pourraient y jouer un rôle : les processus de rétroaction des étoiles et des trous noirs supermassifs actifs, la formation stellaire, et les vents galactiques. Le Chapitre 1 présente toutes les notions nécessaires à la compréhension du problème : les caractéristiques des galaxies typiques de l’Univers proche et lointain ; les vents galactiques ; la mort des galaxies; les trous noirs supermassifs actifs (noyaux actifs de galaxies, AGN) et les étoiles ; et leur rétroaction. Dans le Chapitre 2, je présente les techniques numériques utilisées : le code de simulations astrophysiques RAMSES et le code de transfert radiatif Cloudy, que j’ai utilisé pour développer une méthode de calcul de l’état d’ionisation d’une galaxie, détaillée au Chapitre 3. Le Chapitre 4 étudie le couplage entre les trous noirs actifs et les étoiles, avec le projet POGO, Origines Physiques des Vents Galactiques. Durant cette thèse, j’ai montré que les trous noirs actifs n’étaient pas en mesure de tuer subitement leur hôte, même en prenant en compte la rétroaction des étoiles, et que leur couplage peut réduire ou renforcer les vents dans les galaxies en fonction de leur masse. Le Chapitre 5 fait un état de l’art du domaine avant et pendant mon doctorat, reprend les conclusions de cette thèse et donne quelques perspectives, notamment en ce qui concerne le rôle additionnel des rayons cosmiques dans la mort des galaxies / In the Universe, we observe galaxies forming no, or almost no, stars anymore, but astrophysicists do not know yet what physical mechanisms cause their “death”. To give clues to solve the problem, I studied feedback processes from stars and active supermassive black holes, star formation and galactic outflows. Chapter 1 presents all the notions to understand the problem: the characteristics of typical galaxies in the local and distant Universe, galactic outflows, galaxy death, active supermassive black holes, stars, and their feedback processes. In Chapter 2, I describe the numerical techniques I used: the simulation code RAMSES, and the radiative transfer code Cloudy, which I used to develop a computation method to get the ionization state of an entire galaxy. This method is presented in Chapter 3. Chapter 4 studies the coupling between the feedback processes of active supermassive black holes and stars, with the POGO project, Physical Origins of Galactic Outflows. During this thesis, I showed that typical active supermassive black hole cannot suddenly kill their host, even when stellar feedback processes are accounted for, and that their coupling either reduces or enhances the mass outflow rate depending on the mass of the host. In Chapter 5, I give a state-of-the-art about active supermassive black holes before and during my thesis, sum up the conclusions of the work, and give perspectives to enlarge the scope of the study, especially regarding the additional role of cosmic rays in the death of galaxies

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第18796号 / 理博第4054号 / 新制||理||1583(附属図書館) / 31747 / 京都大学大学院理学研究科物理学・宇宙物理学専攻 / (主査)准教授 上田 佳宏, 教授 太田 耕司, 教授 長田 哲也 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第20179号 / 理博第4264号 / 新制||理||1613(附属図書館) / 京都大学大学院理学研究科物理学・宇宙物理学専攻 / (主査)准教授 上田 佳宏, 教授 嶺重 慎, 教授 長田 哲也 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第20903号 / 理博第4355号 / 新制||理||1625(附属図書館) / 京都大学大学院理学研究科物理学・宇宙物理学専攻 / (主査)教授 佐々木 節, 教授 柴田 大, 教授 川合 光 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM