No description available.

No description available.

AN ABSTRACT OF THE DISSERTATION OFDinush Lanka Panditharathna Jayasooriya Arachchilage, forthe Doctor of Philosophy degree in MATHEMATICS, presented on June 22, 2020 at SouthernIllinois University Carbondale.TITLE: SPIN NETWORK EVALUATION AND THE ASYMPTOTIC BEHAVIORMAJOR PROFESSOR: Dr. Jerzy KocikGraphically, a spin network is a trivalent graph with weights on each edge. At anyof the vertices, the sum of all three weights is even and the sum of any two weights isgreater than or equal to the remaining weight. If the spin network has no free ends, thenwe can evaluate the spin network. Here, we propose a method to evaluate some basic spinnetworks using the idea of Stirling triangle.Tangent circles with integer curvatures are a natural source to make a spin network.In particular, there are spin networks corresponding the Apollonian circle packing and theFord circle packing. We obtain the recurrence relations using the Descartes circle theoremand we evaluate the Apollonian spin network and the Ford circle spin network. We alsodiscuss the asymptotic behavior of the Ford circle spin network.

Models of geometry that are intrinsically quantum-mechanical in nature arise from the recoupling theory of space-time symmetry groups. Roger Penrose constructed such a model from SU(2) recoupling in his theory of spin networks; he showed that spin measurements in a classical limit are necessarily consistent with a three-dimensional Euclidian vector space. T. Regge and G. Ponzano expressed the semi-classical limit of this spin model in a form resembling a path integral of the Einstein-Hilbert action in three Euclidian dimensions. This thesis gives new proofs of the Penrose spin geometry theorem and of the Regge-Ponzano decomposition theorem. We then consider how to generalize these two approaches to other groups that give rise to new models of quantum geometries. In particular, we show how to construct quantum models of four-dimensional relativistic space-time from the re-coupling theory of the Poincare group.

La gravité quantique à boucles a fourni un cadre d’étude particulièrement bien adapté aux théories de jauge définies sans métrique fixe et invariante sous difféomorphismes. Les excitations fondamentales de cette quantification sont appelées réseaux de spins, et dans le contexte de la relativité générale donnent un sens à la géométrie quantique au niveau canonique. Les mousses de spins constituent une sorte d’intégrale de chemins adaptée aux réseaux de spins, et donc destinée à permettre le calcul des amplitudes de transition entre ces états. Cette quantification est particulièrement efficace pour les théories des champs topologiques, comme Yang-Mills 2d, la gravité 3d ou les théories BF, et des modèles ont aussi été proposés pour la gravité quantique en dimension 4.Nous discutons dans cette thèse différentes méthodes pour l’étude des modèles de mousses de spins.Nous présentons en particulier des relations de récurrence sur les amplitudes de mousses de spins. De manière générique, elles codent des symétries classiques au niveau quantique, et sont susceptible de permettre de faire le lien avec les contraintes hamiltoniennes. De telles relations s’interprètent naturellement en termes de déformations élémentaires sur des structures géométriques discrètes, telles que simplicielles. Une autre méthode intéressante consiste à explorer la façon dont on peut réécrire les modèles de mousses de spins comme des intégrales de chemins pour des systèmes de géométries sur réseau, en s’inspirant à la fois des modèles topologiques et du calcul de Regge. Cela aboutit à une vision très géométrique des modèles, et fournit des actions classiques sur réseau dont on étudie les points stationnaires. / Loop quantum gravity has provided us with a canonical framework especially devised for back-ground independent and diffeomorphism invariant gauge field theories. In this quantization the funda-mental excitations are called spin network states, and in the context of general relativity, they give ameaning to quantum geometry. Spin foams are a sort of path integral for spin network states, supposed to enable the computations of transition amplitudes between these states. The spin foam quantization has proved very efficient for topological field theories, like 2d Yang-Mills, 3d gravity or BF theories. Different models have also been proposed for 4-dimensional quantum gravity.In this PhD manuscript, I discuss several methods to study spin foam models. In particular, I present some recurrence relations on spin foam amplitudes, which generically encode classical symme-tries at the quantum level, and are likely to help fill the gap with the Hamiltonian constraints. These relations can be naturally interpreted in terms of elementary deformations of discrete geometric struc-tures, like simplicial geometries. Another interesting method consists in exploring the way spin foam models can be written as path integrals for systems of geometries on a lattice, taking inspiration from topological models and Regge calculus. This leads to a very geometric view on spin foams, and gives classical action principles which are studied in details.