Fases topológicas da matéria são caracterizadas por terem uma degenerescên- cia do estado fundamental que depende da topologia da variedade em que o sistema físico é definido, além disso apresentam estados excitados no interior do sistema que são interpretados como sendo quase-partículas com estatística de tipo anyonica. Estes sistemas apresentam também excitações sem gap de energia em sua borda. Fases topologicamente ordenadas distintas não podem ser distinguidas pelo esquema usual de quebra de simetria de Ginzburg-Landau. Nesta dissertação apresentamos como exemplo o modelo mais simples de um sistema com Ordem Topológica, a saber, o Toric Code (TC), introduzido originalmente por A. Kitaev em [1]. O estado fundamental deste modelo ap- resenta degenerescência igual a 4 quando incorporado à superfície de um toro. As excitações elementares são interpretadas como sendo quase-partículas com estatística do tipo anyonica. O TC é um caso especial de uma classe mais geral de models chamados de Quantum Double Models (QDMs), estes modelos podem ser entendidos como sendo uma implementação de Teorias de gauge na rede em (2 + 1) dimensões na formulação Hamiltoniana, em que os graus de liberdade vivem nas arestas da rede e são elementos do grupo de gauge G. Nós generalizamos estes modelos com a inclusão de campos de matéria nos vértices da rede. Também apresentamos uma construção detalhada de tais modelos e mostramos que eles são exatamente solúveis. Em particular, exploramos o modelo que corresponde à escolher o grupo de gauge como sendo o grupo cíclico Z2 e os graus de liberdade de matéria como sendo elementos de um espaço vetorial bidimensional V2. Além disso, mostramos que a degenerescência do estado fundamental não depende da topologia da variedade e obtemos os estados excitados mais elementares deste modelo. / Topological phases of matter are characterized for having a topologically dependent ground state degeneracy, anyonic quasi-particle bulk excitations and gapless edge excitations. Different topologically ordered phases of matter can not be distinguished by te usual Ginzburg-Landau scheme of symmetry breaking. Therefore, a new mathematical framework for the study of such phases is needed. In this dissertation we present the simplest example of a topologically ordered system, namely, the \\Toric Code (TC) introduced by A. Kitaev in [1]. Its ground state is 4-fold degenerate when embedded on the surface of a torus and its elementary excited states are interpreted as quasi-particle anyons. The TC is a particular case of a more general class of lattice models known as Quantum Double Models (QDMs) which can be interpreted as an implementation of (2+1) Lattice Gauge Theories in the Hamiltonian formulation with discrete gauge group G. We generalize these models by the inclusion of matter fields at the vertices of the lattice. We give a detailed construction of such models, we show they are exactly solvable and explore the case when the gauge group is set to be the abelian Z_2 cyclic group and the matter degrees of freedom to be elements of a 2-dimensional vector space V_2. Furthermore, we show that the ground state degeneracy is not topologically dependent and obtain the most elementary excited states.

Simulationen von Gittereichtheorien spielen eine grundlegende Rolle bei First-Principles-Rechnungen im Kontext der Hochenergiephysik. Diese Arbeit zielt darauf ab, aktuelle Simulationsmethoden für First-Principle-Berechnungen zu verbessern und diese Methoden auf relevante physikalische Modelle anzuwenden. Wir gehen dieses Problem mit drei verschiedenen Ansätzen an: maschinelles Lernen, Quantencomputing und Tensornetzwerke. Im Rahmen des maschinellen Lernens haben wir eine Methode zur Schätzung thermodynamischer Observablen in Gitterfeldtheorien entwickelt. Genauer gesagt verwenden wir tiefe generative Modelle, um den absoluten Wert der freien Energie abzuschätzen. Wir haben die Anwendbarkeit unserer Methode durch die Untersuchung eines Spielzeugmodells demonstriert. Unser Ansatz erzeugt genauere Messungen im Vergleich mit dem Standard-Markov-Ketten-Monte-Carlo-Verfahren, wenn wir einen Phasenübergangspunkt überqueren. Im Kontext des Quantencomputings ist es unser Ziel, die aktuellen Algorithmen für Quantensimulationen zu verbessern. In dieser Arbeit haben wir uns mit zwei Themen moderner Quantencomputer befasst: der Quantenrauschunterdrückung und dem Design guter parametrischer Quantenschaltkreise. Wir haben eine Minderungsroutine zum Auslesen von Bit-Flip-Fehlern entwickelt, die Quantensimulationen drastisch verbessern kann. Wir haben auch eine dimensionale Aussagekraftanalyse entwickelt, die überflüssige Parameter in parametrischen Quantenschaltkreisen identifizieren kann. Darüber hinaus zeigen wir, wie man Expressivitätsanalysen mit Quantenhardware effizient umsetzen kann. Im Kontext des Tensornetzwerks haben wir ein Quantenbindungsmodell U(1) und 2+1-Dimensionen in einer Leitergeometrie mit DMRG untersucht. Unser Ziel ist es, die Eigenschaften des Grundzustands des Modells in einem endlichen chemischen Potential zu analysieren. Wir haben unterschiedliche Windungszahlsektoren beobachtet, als wir chemisches Potential in das System eingebracht haben. / Simulations of lattice gauge theories play a fundamental role in first principles calculations in the context of high energy physics. This thesis aims to improve state-of-the-art simulation methods for first-principle calculations and apply those methods to relevant physical models. We address this problem using three different approaches: machine learning, quantum computing, and tensor networks. In the context of machine learning, we have developed a method to estimate thermodynamic observables in lattice field theories. More precisely, we use deep generative models to estimate the absolute value of the free energy. We have demonstrated the applicability of our method by studying a toy model. Our approach produces more precise measurements in comparison with the standard Markov chain Monte Carlo method when we cross a phase transition point. In the context of quantum computing, our goal is to improve the current algorithms for quantum simulations. In this thesis, we have addressed two issues on modern quantum computers: the quantum noise mitigation and the design of good parametric quantum circuits. We have developed a mitigation routine ffor read-out bit-flip errors that can drastically improve quantum simulations. We have also developed a dimensional expressiveness analysis that can identify superfluous parameters in parametric quantum circuits. In addition, we show how to implement expressivity analysis using quantum hardware efficiently. In the context of the tensor network, we have studied a quantum bond model U(1) and 2+1 dimensions in a ladder geometry with DMRG. Our goal is to analyze the properties of the ground state of the model in a finite chemical potential. We have observed different winding number sectors when we have introduced chemical potential in the system.

Nesta dissertação estudamos as teorias de gauge acoplada com campos de matéria em variedades bidimensionais. Para isso, descrevemos primeiro um formalismo em duas e três dimensões o qual é baseado na ideia de Kuperberg de definir um invariante topológico em três dimensões usando álgebras de Hopf e diagramas de Heegaard. O uso do formalismo é útil para este trabalho pois é fácil a identificação de limites topológicos sem resolver o modelo. Também escrevemos o modelo de gauge com campos de matéria usando uma fixação de gauge chamada de gauge unitário. Trabalhamos com o grupo abeliano $\\mathbb_$ e explicamos com detalhe o caso $\\mathbb_$. Calculamos as funções de partição e loops de Wilson para este grupo nos diferentes limites topológicos. Mostramos que existem casos nos quais os resultados dependem da triangulação mas de maneira trivial, estes casos foram chamados de quase-topológicos. / In this thesis we study gauge theories coupled with matter fields in two-dimensional manifolds. In order to proceed we first describe a formalism in two and three dimensions which is based on the idea of Kuperberg of defining a topological invariant in three dimensions using Hopf algebras and Heegaard diagrams. The use of this formalism is useful here because it is easy to identify topological limits without solving the model. Furthermore, we write the gauge model with matter fields choosing the unitary gauge. We work with abelians groups Z(n) and explain the Z(2) case in detail. We calculate partition functions and Wilson loops for this group in the different topological limits. We show that, there were cases in which the results depended on the triangulation but in a trivial way, these cases are called quasi-topological.

Diese Arbeit befasst sich mit Quantenfeldtheorien auf nicht-kommutativen Räumen. Solche Modelle treten im Zusammenhang mit der Stringtheorie und mit der Quantengravitation auf. Ihre nicht-störungstheoretische Behandlung ist üblicherweise schwierig. Hier untersuchen wir jedoch drei nicht-kommutative Quantenfeldtheorien nicht-perturbativ, indem wir die Wirkungsfunktionale in eine äquivalente Matrixformulierung übersetzen. In der Matrixdarstellung kann die jeweilige Theorie dann numerisch behandelt werden. Als erstes betrachten wir ein regularisiertes skalares Modell auf der nicht-kommutativen Ebene und untersuchen den Kontinuumslimes bei festgehaltener Nicht-Kommutativität. Dies wird auch als Doppelskalierungslimes bezeichnet. Insbesondere untersuchen wir das Verhalten der gestreiften Phase. Wir finden keinerlei Hinweise auf die Existenz dieser Phase im Doppelskalierungslimes. Im Anschluss daran betrachten wir eine vier-dimensionale U(1) Eichtheorie. Hierbei sind zwei der räumlichen Richtungen nicht-kommutativ. Wir untersuchen sowohl die Phasenstruktur als auch den Doppelskalierungslimes. Es stellt sich heraus, dass neben den Phasen starker und schwacher Kopplung eine weitere Phase existiert, die gebrochene Phase. Dann bestätigen wir die Existenz eines endlichen Doppelskalierungslimes, und damit die Renormierbarkeit der Theorie. Weiterhin untersuchen wir die Dispersionsrelation des Photons. In der Phase mit schwacher Kopplung stimmen unsere Ergebnisse mit störungstheoretischen Berechnungen überein, die eine Infrarot-Instabilität vorhersagen. Andererseits finden wir in der gebrochenen Phase die Dispersionsrelation, die einem masselosen Teilchen entspricht. Als dritte Theorie betrachten wir ein einfaches, in seiner Kontinuumsform supersymmetrisches Modell, welches auf der "Fuzzy Sphere" formuliert wird. Hier wechselwirken neutrale skalare Bosonen mit Majorana-Fermionen. Wir untersuchen die Phasenstruktur dieses Modells, wobei wir drei unterschiedliche Phasen finden. / This work deals with three quantum field theories on spaces with noncommuting position operators. Noncommutative models occur in the study of string theories and quantum gravity. They usually elude treatment beyond the perturbative level. Due to the technique of dimensional reduction, however, we are able to investigate these theories nonperturbatively. This entails translating the action functionals into a matrix language, which is suitable for numerical simulations. First we explore a scalar model on a noncommutative plane. We investigate the continuum limit at fixed noncommutativity, which is known as the double scaling limit. Here we focus especially on the fate of the striped phase, a phase peculiar to the noncommutative version of the regularized scalar model. We find no evidence for its existence in the double scaling limit. Next we examine the U(1) gauge theory on a four-dimensional spacetime, where two spatial directions are noncommutative. We examine the phase structure and find a new phase with a spontaneously broken translation symmetry. In addition we demonstrate the existence of a finite double scaling limit which confirms the renormalizability of the theory. Furthermore we investigate the dispersion relation of the photon. In the weak coupling phase our results are consistent with an infrared instability predicted by perturbation theory. If the translational symmetry is broken, however, we find a dispersion relation corresponding to a massless particle. Finally, we investigate a supersymmetric theory on the fuzzy sphere, which features scalar neutral bosons and Majorana fermions. The supersymmetry is exact in the limit of infinitely large matrices. We investigate the phase structure of the model and find three distinct phases. Summarizing, we study noncommutative field theories beyond perturbation theory. Moreover, we simulate a supersymmetric theory on the fuzzy sphere, which might provide an alternative to attempted lattice formulations.

In dieser Dissertation untersuchen und entwickeln wir neue Methoden, die dabei helfen sollen eine effektive semiklassische Beschreibung der kanonischen Loop-Quantengravitation und der Spinfoam-Gravitation zu bestimmen. Einer kurzen Einführung in die Loop-Quantengravitation folgen drei Forschungsartikel, die die Resultate der Doktorarbeit präsentieren. Im ersten Artikel behandeln wir das Problem der Zeit und einen neuen Vorschlag zur Implementierung von Eigenzeit durch Randbedingungen an Pfadintegrale: wir untersuchen eine konkrete Realisierung dieses Formalismus für die freie Skalarfeldtheorie. Im zweiten Artikel übersetzen wir semiklassische Zustände der linearisierten Gravitation in Zustände der Loop-Quantengravitation. Deren Eigenschaften deuten an, wie sich Semiklassizität im Loop-Formalismus manifestiert, and wie man dies benützen könnte, um semiklassische Entwicklungen herzuleiten. Im dritten Teil schlagen wir eine neue Formulierung von Spinfoam-Modellen vor, die vollständig Triangulierungs- und Hintergrund-unabhängig ist: mit Hilfe einer Symmetrie-Bedingung identifizieren wir Spinfoam-Modelle, deren Triangulierungs-Abhängigkeit auf natürliche Weise entfernt werden kann. / In this Ph.D. thesis, we explore and develop new methods that should help in determining an effective semiclassical description of canonical loop quantum gravity and spin foam gravity. A brief introduction to loop quantum gravity is followed by three research papers that present the results of the Ph.D. project. In the first article, we deal with the problem of time and a new proposal for implementing proper time as boundary conditions in a sum over histories: we investigate a concrete realization of this formalism for free scalar field theory. In the second article, we translate semiclassical states of linearized gravity into states of loop quantum gravity. The properties of the latter indicate how semiclassicality manifests itself in the loop framework, and how this may be exploited for doing semiclassical expansions. In the third part, we propose a new formulation of spin foam models that is fully triangulation- and background-independent: by means of a symmetry condition, we identify spin foam models whose triangulation-dependence can be naturally removed.

Das Potential zwischen einem statischen Quark und Antiquark in der reinen SU(3) Yang-Mills Theorie wird auf dem Gitter in der Region von kurzen bis mittleren Abstaenden (0.05 fm < r < 0.8fm) nichtperturbativ ausgewertet. Renormalisierte dimensionslose Observablen werden zum Kontinuumslimes extrapoliert und bestaetigen damit die theoretische Erwartung, dass die fuehrenden Gitterartifakte quadratisch im Gitterabstand sind. Bei hohen Energien werden die Resultate mit der parameterfreien Vorhersage der Stoerungtheorie verglichen; diese wird erreicht, indem man die Renormierungsgruppengleichung in zwei- und drei-Loop-Ordnung loest. Die Wahl des Renormierungschemas fuer die Definition der laufenden Kopplung ist wichtig fuer die Genauigkeit der perturbativen Vorhersage. Wenn man die laufende Kopplung durch die Kraft definiert, ist Stoerungstheorie bis zu alpha ~ 0.3 anwendbar, waehrend mit dem statischen Potential nur bis zu alpha ~ 0.15. In der Region, in der Stoerungstheorie zuverlaessig sein sollte, wird kein grosser unerwarteter nichtperturbativer Term beobachtet: im Gegenteil, man findet eine gute uebereinstimmung zwischen Stoerungtheorie und unseren nicht-perturbativen Daten. Fuer grosse Quark-Antiquark Abstaende werden unsere Ergebnisse mit den Vorhersagen einer effektiven bosonischen Stringtheorie verglichen, und man findet bereits eine ueberraschend gute Uebereinstimmung fuer Abstaende > 0.5 fm. Im zweiten Teil dieser Arbeit sind Universalitaet und Skalierungsverhalten von unterschiedlichen Formulierungen der Yang-Mills Theorie auf dem Gitter diskutiert. Insbesondere werden Iwasaki- und DBW2- Wirkungen untersucht, die durch Renormierungsgruppe (RG) Argumente formuliert wurden. Die Laengenskala r_0 ~ 0.5 fm wird bei einigen Gitterabstaenden ausgewertet und die Skalierung der kritischen Deconfinement Temperatur T_c * r_0 wird mit den Resultaten analysiert und konfrontiert, die mit der ueblichen Wilson Plaquette Wirkung erreicht werden. Da sie im Kontinuumslimes uebereinstimmen, wird die Universalitaet bestaetigt. Die Groesse die man benutzt, um die Skala einzustellen, muss mit Vorsicht gewaehlt werden, um grosse systematische Ungenauigkeiten zu vermeiden. Fuer diesen Zweck zeigt sich r_0 als angebracht. Fuer die kritische Temperatur zeigen die Daten, die mit RG Wirkungen erhalten werden, verringerte Gitterartifakte, vor allem mit der Iwasaki Wirkung. Schliesslich wird die Masse der 0^{++}- und 2^{++}-Glueballs ausgewertet, indem man die Observablen m_0^{++} *r_0 und m_2^{++}*r_0 betrachtet. Jedoch kann keine genaue Schlussfolgerung ueber das Scalingverhalten fuer diese Observablen gezogen werden. Eine besondere Aufmerksamkeit ist der Verletzung der physikalischen Positivitaet, die in diesen Wirkungen auftritt und den Konsequenzen in der Extraktion der physikalischen Groessen aus euklidischen Korrelationsfunktionen gewidmet. / The potential between a static quark and antiquark in pure SU(3) Yang-Mills theory is evaluated non-perturbatively through computations on the lattice in the region from short to intermediate distances (0.05 fm < r 0.5 fm. In the second part of this work, universality and scaling behavior of different formulations of Yang-Mills theory on the lattice are discussed. In particular, the Iwasaki and DBW2 action are investigated, which were obtained by following renormalization group (RG) arguments. The length scale r_0 ~ 0.5 fm is evaluated at several lattice spacings and the scaling of the critical deconfinement temperature T_c*r_0 is analyzed and confronted with the results obtained with the usual Wilson plaquette action. Since they agree in the continuum limit, the universality is confirmed. We remark that the quantity to use to set the scale has to be chosen with care in order to avoid large systematic uncertainties and $\rnod$ turns out to be appropriate. For the critical temperature the data obtained with RG actions show reduced lattice artifacts, above all with the Iwasaki action. Finally the mass of the glueballs 0^{++} and 2^{++} is evaluated by considering the quantities m_0^{++}*r_0 and m_2^{++}*r_0; however for those observables no clear conclusion about the scaling behavior can be drawn. Particular attention is dedicated to the violation of physical positivity which occur in these actions and the consequences in the extraction of physical quantities from Euclidean correlation functions.

Nesta dissertação estudamos as teorias de gauge acoplada com campos de matéria em variedades bidimensionais. Para isso, descrevemos primeiro um formalismo em duas e três dimensões o qual é baseado na ideia de Kuperberg de definir um invariante topológico em três dimensões usando álgebras de Hopf e diagramas de Heegaard. O uso do formalismo é útil para este trabalho pois é fácil a identificação de limites topológicos sem resolver o modelo. Também escrevemos o modelo de gauge com campos de matéria usando uma fixação de gauge chamada de gauge unitário. Trabalhamos com o grupo abeliano $\\mathbb_$ e explicamos com detalhe o caso $\\mathbb_$. Calculamos as funções de partição e loops de Wilson para este grupo nos diferentes limites topológicos. Mostramos que existem casos nos quais os resultados dependem da triangulação mas de maneira trivial, estes casos foram chamados de quase-topológicos. / In this thesis we study gauge theories coupled with matter fields in two-dimensional manifolds. In order to proceed we first describe a formalism in two and three dimensions which is based on the idea of Kuperberg of defining a topological invariant in three dimensions using Hopf algebras and Heegaard diagrams. The use of this formalism is useful here because it is easy to identify topological limits without solving the model. Furthermore, we write the gauge model with matter fields choosing the unitary gauge. We work with abelians groups Z(n) and explain the Z(2) case in detail. We calculate partition functions and Wilson loops for this group in the different topological limits. We show that, there were cases in which the results depended on the triangulation but in a trivial way, these cases are called quasi-topological.

Neste trabalho estudamos as teorias de gauge puras (sem campo de matéria) na rede em três dimensões. Em especial, estudamos a subclasse das teorias topológicas. A maneira como denimos e tratamos as teorias de gauge e diferente, mas equivalente, à forma usual apresentada em [2, 3]. Definimos estas teorias via o formalismo de Kuperberg, que é um formalismo puramente matemático de um invariante topológico de variedades tridimensionais. Este formalismo, embora bastante abstrato, pode ser adaptado para descrever as classes de modelos das teorias de gauge na rede, e traz várias vantagens, pois possibilita que tratemos de teorias topológicas e não topológicas, além da fácil identicação dos limites topológicos da função de partição. Estudamos também a classe das teorias chamadas quase topológicas, que podem ser pensadas como deformações de teorias topológicas. Em particular, consideramos teorias de gauge com grupo de gauge Z2, que é o grupo de gauge mais simples possível com dinâmica não trivial. Dentro das teorias de gauge, identicamos as classes de modelos que são quase topológicos, além de outras classes nas quais a função de partição pode ser trivialmente calculada. A função de partição foi calculada explicitamente no caso quase topológico em duas situações: sobre a esfera tridimensional S3 e sobre o toroS1x S1x S1x, que representa uma rede com condições periódicas de contorno. Dois modelos físicos de teorias de gauge, ainda com grupo de gauge Z2, foram estudados: o modelo com ação de Wilson SW = Pfaces [Tr(g) - 1] e o modelo com ação spin-gauge SSG = Pfaces Tr(g). No limite de baixa temperatura ambos os modelos mostram-se ser topológicos, enquanto que no limite de alta temperatura mostraram-se ser trivialmente calculáveis. / In this work we studied the class of models of pure lattice gauge theories (without matter elds) in three dimensions. Especially, we studied the subclass of topological theories. Lattice gauge theories were dened in an unusual way, unlike the description shown in [2, 3]. We dened lattice gauge theories via the Kuperberg\'s formalism [4], which is a mathematical model for a topological invariant of 3-manifolds. Such formalism, although completely abstract, can describe the class of models of lattice gauge theories because it can describe both topological and non topological theories, besides it provides an easy identication of the partition function topological limits. We also studied the class of theories called quasi topological, which can be thought as deformations of topological theories. As an example, we consider Z2 as gauge group, because it is the simplest group that does not imply trivial dynamics. Inside this class of models we identify the subclasses of quasi topological theories and also other classes in which the partition function can be trivially computed. The partition function was explicitly computed in two situations: on the 3-sphere S3 and on the 3-manifold S1 x S1 x S1 that represents periodic boundary conditions. Two physical models were studied: the model with Wilson\'s action SW(conf)1 and the model with spin-gauge action SSG(conf)2. In the low temperature limit both models shown to be topological and in the high temperature limit they could be trivially computed.

O comportamento infra-vermelho dos propagadores de glúons e de ghosts é de fundamental importância para o entendimento do limite de baixas energias da cromodinâmica quântica (QCD), especialmente no que diz respeito ao problema do confinamento de quarks e de glúons. O objetivo desta tese é implementar um novo método para o estudo do propagador de glúons no calibre covariante linear para a QCD na rede. Em particular, analisamos em detalhe a nova implementação proposta e estudamos os algoritmos para fixação numérica deste calibre. Note que a fixação numérica da condição de calibre de Feynman apresenta vários problemas não encontrados nos casos de Landau e de Coulomb, o que impossibilitou por longo tempo o seu estudo adequado. De fato, a definição considerada inicialmente, por Giusti et. al., é de difícil implementação numérica e introduz condições espúrias na fixação de calibre. Como consequência, os únicos estudos efetuados anteriormente referem-se aos propagadores de glúons e de quarks em redes relativamente pequenas, não permitindo uma análise cuidadosa do limite infra-vemelho da QCD neste calibre. A obtenção de novas soluções para a implementação do calibre de Feynman na rede é portanto de grande importância para viabilizar estudos numéricos mais sistemáticos dos propagadores e dos vértices neste calibre e, em geral, no calibre covariante linear. / The infrared behavior of gluon and ghost propagators is of fundamental importance for the understanding of the low-energy limit of quantum chromodynamics (QCD), especially with respect to the problem of the confinement of quarks and gluons. The goal of this thesis is to implement a new method to study the gluon propagator in the linear covariant gauge in lattice QCD. In particular, we analyze in detail the newly proposed implementation and study the algorithms for numerically fixing this gauge. Note that the numerical fixing of the Feynman gauge condition poses several problems that are not present in the Landau and Coulomb cases, which prevented it from being properly studied for a long time. In fact, the definition considered initially, by Giusti et. al., is of difficult numerical implementation and introduces spurious conditions into the gauge fixing. As a consequence, the only studies carried out previously involved gluon and quark propagators on relatively small lattices, hindering a careful analysis of the infrared limit of QCD in this gauge. Obtaining new solutions for the implementation of the Feynman gauge on the lattice is therefore of great importance to enable more systematic numerical studies of propagators and vertices in this gauge and, in general, in the linear covariant gauge.

Neste trabalho estudamos as teorias de gauge puras (sem campo de matéria) na rede em três dimensões. Em especial, estudamos a subclasse das teorias topológicas. A maneira como denimos e tratamos as teorias de gauge e diferente, mas equivalente, à forma usual apresentada em [2, 3]. Definimos estas teorias via o formalismo de Kuperberg, que é um formalismo puramente matemático de um invariante topológico de variedades tridimensionais. Este formalismo, embora bastante abstrato, pode ser adaptado para descrever as classes de modelos das teorias de gauge na rede, e traz várias vantagens, pois possibilita que tratemos de teorias topológicas e não topológicas, além da fácil identicação dos limites topológicos da função de partição. Estudamos também a classe das teorias chamadas quase topológicas, que podem ser pensadas como deformações de teorias topológicas. Em particular, consideramos teorias de gauge com grupo de gauge Z2, que é o grupo de gauge mais simples possível com dinâmica não trivial. Dentro das teorias de gauge, identicamos as classes de modelos que são quase topológicos, além de outras classes nas quais a função de partição pode ser trivialmente calculada. A função de partição foi calculada explicitamente no caso quase topológico em duas situações: sobre a esfera tridimensional S3 e sobre o toroS1x S1x S1x, que representa uma rede com condições periódicas de contorno. Dois modelos físicos de teorias de gauge, ainda com grupo de gauge Z2, foram estudados: o modelo com ação de Wilson SW = Pfaces [Tr(g) - 1] e o modelo com ação spin-gauge SSG = Pfaces Tr(g). No limite de baixa temperatura ambos os modelos mostram-se ser topológicos, enquanto que no limite de alta temperatura mostraram-se ser trivialmente calculáveis. / In this work we studied the class of models of pure lattice gauge theories (without matter elds) in three dimensions. Especially, we studied the subclass of topological theories. Lattice gauge theories were dened in an unusual way, unlike the description shown in [2, 3]. We dened lattice gauge theories via the Kuperberg\'s formalism [4], which is a mathematical model for a topological invariant of 3-manifolds. Such formalism, although completely abstract, can describe the class of models of lattice gauge theories because it can describe both topological and non topological theories, besides it provides an easy identication of the partition function topological limits. We also studied the class of theories called quasi topological, which can be thought as deformations of topological theories. As an example, we consider Z2 as gauge group, because it is the simplest group that does not imply trivial dynamics. Inside this class of models we identify the subclasses of quasi topological theories and also other classes in which the partition function can be trivially computed. The partition function was explicitly computed in two situations: on the 3-sphere S3 and on the 3-manifold S1 x S1 x S1 that represents periodic boundary conditions. Two physical models were studied: the model with Wilson\'s action SW(conf)1 and the model with spin-gauge action SSG(conf)2. In the low temperature limit both models shown to be topological and in the high temperature limit they could be trivially computed.