Esse trabalho aborda a questão da segregação residencial no município de São Paulo, sendo o objetivo principal a avaliação do impacto da segregação sobre o acesso das camadas mais pobres da população à política de infra-estrutura urbana. Além da discussão conceitual a respeito da segregação – termo muitas vezes confundido na literatura com os conceitos de pobreza, desigualdade, falta de acesso a serviços urbanos –, o trabalho pretende apresentar estratégias de identificação de áreas segregadas e um modelo de análise capaz de medir o impacto da residência nesses locais sobre as condições de acesso à política de infra-estrutura urbana, tradicionalmente associada com a segregação. Desse modo, além de um esforço teórico de deslocamento do foco em direção às conseqüências da segregação, e não as suas causas, esse trabalho apresenta um esforço empírico de elaboração de indicadores capazes de identificar áreas segregadas e medir o impacto da residência nesses locais sobre as condições de acesso a políticas públicas. / This dissertation deals with the issues of residential segregation in the city of São Paulo. It argues that segregation has a significant impact on the access to public policies among the poorest population. Besides the discussion of the urban literature on segregation – a concept generally mingled with the concepts of poverty and inequality as well as for the lack of access to basic public services –, this dissertation presents some strategies for identifying segregated areas and a multivariate model of access to urban infrastructure policies. Besides the theoretical effort of changing the stress on the causes to the stress on the consequences of segregation, this dissertation presents an empirical effort: the construction of different indicators able to measure the impact of segregation over the access to public policies.

Made available in DSpace on 2015-04-11T14:03:12Z (GMT). No. of bitstreams: 1 DISSERTACAO ANDERSON.pdf: 945892 bytes, checksum: 13a676735f1c713ef1251e829d5e41bb (MD5) Previous issue date: 2006-09-13 / Fundação de Amparo à Pesquisa do Estado do Amazonas / More and more the demanding for mobility has been wanted by the users who want to have everything no matter where they are. The applications have been modified in order to attend this needs demanded by the new environment without any predictable infrastructure or topology. The adhoc networks are very popular when concerning about non-infrastructures environment and many studies have been started looking for solutions and strategies to understand and adapt the applications originally created to work over client/server architecture. Voice over IP applications also have to be adapted to work properly over this environment and this study try to identify the factors which can impact on application performance. It is made looking at the look up algorithms used by them to find other nodes around the net. / A medida que a mobilidade se torna uma exigência cada vez maior dos usuários de redes de computadores, as aplicações precisam adaptarem-se ao novo ambiente descentralizado, que por vezes, é totalmente imprevisível e mutável. As redes adhoc estão muito difundidas em cenários totalmente descentralizados e desprovidos de infra-estrutura fixa e por isso diversos estudos estão sendo realizados visando a adaptação de aplicações originalmente construídas sobre a luz da arquitetura cliente/servidor. As aplicações de voz sobre IP (VoIP) também precisam se adaptar a nova realidade, e para isso foram adotadas diversas estratégias de descentralização, sendo uma delas a utilização de algoritmos de busca P2P para prover a independência de pontos centralizadores. Este trabalho realiza um estudo do comportamento desse tipo de algoritmo sobre redes adhoc visando identificar os fatores que podem influênciar direta ou indiretamente a performance de uma aplicação VoIP que os utilize como m´método de localização de seus nós.

O dano por umidade à um defeito frequente e um dos principais responsÃveis pelo desgaste prematuro de pavimentos. Esta pesquisa apresenta uma avaliaÃÃo do potencial do LCC como modificador de ligante asfÃltico com relaÃÃo à resistÃncia ao dano por umidadeem misturas asfÃlticas.Propriedades fÃsico-quÃmicas do LÃquido da Castanha de Caju (LCC) mostram-se potencialmente Ãteis para conferir ao Cimento AsfÃltico de PetrÃleo (CAP) maior adesÃo com os agregados.Utilizou-se um CAP convencional classificado por penetraÃÃo como 50/70 e um CAP modificado com 2% de LCC caracterizados quÃmica e reologicamente. Os agregados foram caracterizados atravÃs da fluorescÃncia de Raio-X e com relaÃÃo à forma. Foram dosadas, segundo o mÃtodo Superpave, diferentes misturas asfÃlticas contendo como ligante: CAP, CAP modificado com LCC e CAP e cal como fÃler. A caracterizaÃÃo mecÃnica das misturas foi realizada atravÃs dos ensaios Lottman modificado, MÃdulo de ResiliÃncia (MR) e desgaste CÃntabro. Utilizou-se o Processamento Digital de Imagens (PDI) para verificar o deslocamento da pelÃcula de CAP. Constatou-se que o LCC diminuiu a viscosidade e reduziu as Temperaturas de Usinagem e CompactaÃÃo (TUC) do CAP. Ambas as amostras, CAP e CAP modificado com 2% de LCC, foram classificadas como PG 70-28. Os agregados foram classificados como granÃticos, sub-arredondados, com baixa esfericidade e de textura lisa. Os resultados mostraram que a mistura com CAP modificado com LCC apresentou menor percentual de descolamento da pelÃcula de CAP e desempenho mecÃnico satisfatÃrio se comparada Ãs outras misturas analisadas. Espera-se que o LCC possa ser utilizado como aditivo gerando misturas asfÃlticas menos suscetÃveis à aÃÃo deletÃria da Ãgua. / Moisture damage is a frequent distress on pavements and is considered a major contributor to premature deterioration. Physicochemical properties of the Cashew Nut Shell Liquid (CNSL) showed to be potentially useful to improve the adhesion between asphalt binder and aggregates.The main objective of this study is to evaluate the potential of the CNSL modified asphalt binder to increase the hot mixture asphalt (HMA) resistance to moisture damage, and as a promoter of adhesiveness between binder and aggregates.Pure and modified asphalt binders with different CNSL contents were characterized chemically and rheologically. The aggregates were characterized by X-ray fluorescence and using Aggregate Imaging Measurement System (AIMS). HMAs with binders modified with CNSL were designed using the Superpave methodology. HMA mechanical characterization consisted of modified Lottman test, Resilient Modulus and Cantabro. Results were compared with those obtained using two different HMAs: (i) one with conventional asphalt binder and natural aggregates, and (ii) another one using 2% of lime as a filer. The stripping resistance of asphalt binderswere characterizedthrough digital image processing (DIP). The findings showed the CNSL decreased the asphalt binder viscosity. Both, the pure and the modified asphalt binder, were classified as PG 70-28. The granitic aggregates were classified as sub rounded, with low sphericity and smooth texture. The findings showed that the HMA contain binder modified with CNSL had better stripping resistance and behavior mechanical performance than other evaluated HMAs. Itâs expected that the effectiveness of the CNSL can be used as an additive to prevent striping in asphalt mixtures and to provide a greater resistance to breakdown caused by water.

O desenvolvimento de uma infra-estrutura de informação só é possível se ela for sustentada e legitimada por atores organizacionais e pelas pessoas que contribuem para o seu desenvolvimento ou dela se utilizam. O principal objetivo desta tese é contribuir para um maior entendimento do papel desempenhado pelo exercício do poder no processo de desenvolvimento de uma infra-estrutura de informação. Este estudo baseia-se em uma análise empírica de um sistema de informação, cujo desenvolvimento foi sustentado por atores poderosos e por agentes que se tornaram poderosos graças a uma ampliação no seu escopo e à sua utilização como fonte de poder ou vigilância, dando origem a uma infra-estrutura de informação. Uma infra-estrutura de informação difere dos sistemas de informação tradicionais por ser um recurso compartilhado por uma grande comunidade de usuários e organizações não estando sujeito ao controle de uma única unidade organizacional. O seu desenvolvimento é comandado por vários atores, com interesses diversos e sem uma coordenação central que exerça controle sobre todos os agentes, não podendo, portanto, ser projetada e gerenciada de acordo com os princípios utilizados para o projeto de sistemas de informação, ou seja, de forma isolada e estanque. Utiliza-se como fundamentação teórica para esta pesquisa, a teoria de Hanseth e Lyytinen (2005) que estuda a evolução e a expansão de infra-estruturas de informação, a qual não aborda o papel do poder no desenvolvimento de infra-estruturas, conforme reconhecem os próprios autores. Nesta tese usa-se o modelo dos circuitos de poder de Clegg (1989) para analisar os efeitos do exercício do poder para o desenvolvimento da infra-estrutura de informação. O sistema estudado é o SIAFEM ? Sistema Integrado de Administração Financeira para Estados e Municípios, que foi adotado para o controle orçamentário e financeiro do Governo do Estado de São Paulo, sendo posteriormente expandido e integrado a outros sistemas, transformando-se, desta forma, em uma infra-estrutura de informação. Para o desenvolvimento da infra-estrutura analisada foi preciso de-institucionalizar um sistema contábil utilizado anteriormente e institucionalizar um novo sistema. Este processo trouxe profundas mudanças organizacionais e culturais, cuja análise foi baseada na teoria do novo institucionalismo. / The development of an information infrastructure is possible only if it is sustained and legitimated by organizational actors and individuals that contribute to its development or use. The main objective of this thesis is to develop a better understanding of the role played by power as a design constraint in the process of development of an information infrastructure. This study is based on an empirical analysis of an information system that once implemented was enlarged due to new requirements and the role played by it as an empowering tool for some actors. Moreover, powerful actors that sustained its expansion make use of it as an apparatus of control and surveillance. An information infrastructure differs from traditional information systems by being a resource shared by a large community of users and organizations, and not being under the control of a single organizational unit. Several actors command its development, with diverse interests and without a central coordination over the agents. The building of these solutions is significantly different from those of the autonomous information systems built previously. Therefore, it is necessary to develop a new understanding for the building of these solutions, i.e., systems that can be enlarged and become an information infrastructure. Hanseth and Lyytinen´s theory (2005), that studies the evolution and the expansion of information infrastructures, is used as a theoretical foundation for this research, supplemented by Clegg´s Circuits of Power framework (1989) which is used to study power as a design constraint for the development of the information infrastructure. The studied system is SIAFEM - Integrated System for State Financial Administration, a large governmental system adopted by the State of Sao Paulo for budget and financial control. Later on, it was enlarged and integrated with new systems, changing to an information infrastructure. For the development of the analysed infrastructure it was necessary to de-institutionalise an accounting system used previously and to institutionalise a new system. This process brought deep organizational and cultural changes, whose analysis was based on the Theory of New Institutionalism.

L’objectif de cette thèse est de réaliser des revêtements de nitrure de bore sur différents types de substrats comme le graphite, le quartz, le pyrex, en allant jusqu’aux métaux et en particulier le titane. Le choix de la voie PDCs s’avère intéressant grâce à la maîtrise du précurseur de départ au niveau atomique d’une part et à la facilité du procédé de dépôt, d’autre part. Nos objectifs étaient d’étudier la faisabilité de réaliser des revêtements BN sur différents types de substrat en utilisant un traitement thermique résistif et de mettre en place un dispositif qui nous permette de pyrolyser les films polymériques sur métaux sans dommage pour le substrat, en vue de leur protection contre l’oxydation ou d’autres applications mécaniques. Dans ce sens, nous avons démontré la possibilité d’utiliser un traitement thermique alternatif par lampe halogène émettant dans l’infra-rouge pour densifier les revêtement BN déposer sur substrats métalliques. / The aim of this work was to prepare boron nitride coatings onto different substrates using the Polymers Derived Ceramics (PDCs) approach. In that way, BN coatings were obtained onto graphite, pure silica and metal especially titanium. The first part of this thesis was to study parameters (of the solution used and of the dip-coating process), to obtain the best coatings in terms of morphology, cristallinity and chemical composition. The second part was dedicated to BN coatings obtained onto metal substrates using an alternative thermal treatment allowing the polymer-to-ceramic conversion without any damage for the metal. Actually, annealing by infrared irradiation allows heating only the coating, energy being reflected by the metal.

abstract: The realization of Silicon based photonic devices will enable much faster data transmission than is possible today using the current electronics based devices. Group IV alloys germanium tin (GeSn) and silicon germanium tin (SiGeSn) have the potential to form an direct bandgap material and thus, they are promising candidates to develop a Si compatible light source and advance the field of silicon photonics. However, the growth of the alloys is challenging as it requires low temperature growth and proper strain management in the films during growth to prevent tin segregation. In order to satisfy these criteria, various research groups have developed novel chemical vapor deposition (CVD) reactors to deposit the films. While these reactors have been highly successful in depositing high crystal quality high Sn concentration films, they are generally expensive set-ups which utilize several turbomolecular/cryogenic pumps and/or load-lock systems. An more economical process than the state-of-the art to grow group IV materials will be highly valuable. Thus, the work presented in this dissertation was focused on deposition of group IV semiconductor thin films using simplified plasma enhanced CVD (PECVD) reactors. Two different in-house assembled PECVD reactor systems, namely Reactor No. 1 and 2, were utilized to deposit Ge, GeSn and SiGeSn thin films. PECVD technique was used as plasma assistance allows for potentially depositing the films at growth temperatures lower than those of conventional CVD. Germane (GeH4) and Digermane (Ge2H6) were used as the Ge precursor while Disilane (Si2H6) and tin chloride (SnCl4) were used as the precursors for Si and Sn respectively. The growth conditions such as growth temperature, precursor flow rates, precursor partial pressures, and chamber pressure were varied in a wide range to optimize the growth conditions for the films. Polycrystalline Ge films and SiGeSn films with an Sn content upto 8% were deposited using Reactor No. 1 and 2. Development of epitaxial Ge buffers and GeSn films was accomplished using a modified Reactor No. 2 at temperatures <400oC without the aid of ultra-high vacuum conditions or a high temperature substrate pre-deposition bake thereby leading to a low economic and thermal budget for the deposition process. / Dissertation/Thesis / Doctoral Dissertation Materials Science and Engineering 2020

This master’s thesis deals with parameters of the TV Remote Controller. Aim of this work is studying standards of remote controling of TV receiver by the way of infra-red light. For succesful solving of this project is important to learn about the theory of IrDA. In the second part ot this thesis is designed measurement unit for measuring the parameters of the remote control. The measurement unit includes an integrated IR receiver and photodiode for receive IR signals from remote control. The description of measurement unit blocks and principles of measuring parameters are also placed there. The final part is devoted to processing the measurement results of measuring levels and time waveforms of signal in the receiver module, directional characteristics of remote control and receiver. For measuremented unit is created laboratory task including the role specification and protocol.

Cette thèse s’intéresse à l’application des méthodes d’apprentissage statistique pour la prédiction de température d’un composant électronique présent dans un radar. On étudie un cas simplifié des systèmes réels, le système étudié se limitant à un seul composant monté sur un système de refroidissement réduit. Le premier chapitre est consacré à la modélisation thermique. Après avoir présenté les principaux modes de transmission de l’agitation thermique, les modèles analytiques et numériques qui en découlent sont étudiés. En utilisant cette connaissance,le deuxième chapitre propose de choisir dans les méthodes de mesures les plus adaptées aux spécifications et aux contraintes de l’application choisie. Une fois que les bases de données ont été établies, nous pouvons utiliser dans le troisième chapitre les techniques de l’apprentissage statistique pour construire un modèle dynamique. Après un bref rappel sur les tenants et les aboutissants de la modélisation statistique, quatre familles de méthodes seront présentées : les modèles linéaires, les réseaux de neurones, les réseaux bayésiens dynamiques et les machines à vecteur support (SVM). Enfin, le quatrième chapitre est l’occasion de présenter une méthode de modélisation originale.En effet, après avoir détaillé la mise en oeuvre des méthodes d’identification de représentation d’état, nous verrons comment prendre en compte des a priori théoriques au cours de l’apprentissage de ce type de modèle, à savoir une contrainte de stabilité. / This thesis is focused on the application of statistical learning methods for the temperature prediction of an electronic component embedded in a radar. We study a simplified case of real systems, the system under study is limited to a single component mounted on a reduced cooling system. The first chapter is devoted to heat transfer modelisation. After presenting the major mechanisms of thermal agitation transmission, analytical and numerical models are studied. Using this knowledge, the second chapter offers a survey on the methods of temperature measurement, choosing the fittest according to the specifications and the constraints of the chosen application.Once databases have been established, we can use in the third chapter statistical learning techniques to build a dynamic model. After a brief reminder about the ins and outs of statistical modeling, four families of methods willbe presented : linear models, neural networks, dynamic bayesian networks and support vector machines (SVM).The fourth chapter is an opportunity to present a novel method of modeling. Indeed, after a presentation of themethods for the identification of state representation, we see how to take into account theoretical apriorism during learning of this model type, ie a stability constraint.

Thesis (PhD)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: The reactive absorption rate of CO2 into non-aqueous solvents containing the primary amine, mono-ethanolamine (MEA) is recognised as a suitable method for measuring the effective interfacial mass transfer area of separation column internals such as random and structured packing. Currently, this method is used under conditions where the concentration of MEA in the liquid film is unaffected by the reaction and the liquid phase reaction is, therefore, assumed to obey pseudo first order kinetics with respect to CO2. Under pseudo first order conditions, the effect of surface depletion and renewal rates are not accounted for. Previous research indicated that the effective area available for mass transfer is also dependent upon the rate of surface renewal achieved within the liquid film. In order to study the effect of surface depletion and renewal rates on the effective area, a method utilising a fast reaction with appreciable depletion of the liquid phase reagent is required. The homogeneous liquid phase reaction kinetics of CO2 with MEA n-Propanol as alcoholic solvent was investigated in this study. A novel, in-situ Fourier Transform Infra-Red (FTIR) method of analysis was developed to collect real time concentration data from reaction initiation to equilibrium. The reaction was studied in a semi-batch reactor set-up at ambient conditions (T = 25°C, 30°C and 35°C, P = 1 atm (abs)). The concentration ranges investigated were [MEA]:[CO2] = 5:1 and 10:1. The concentration range investigated represents conditions of significant MEA conversion. The reaction kinetic study confirmed the findings of previous research that the reaction of CO2 with MEA is best described by the zwitterion reactive intermediate reaction mechanism. Power rate law and pseudo steady state hypothesis kinetic models (proposed in literature) were found to be insufficient at describing the reaction kinetics accurately. Two fundamentally derived rate expressions (based on the zwitterion reaction mechanism) provided a good quality model fit of the experimental data for the conditions investigated. The rate constants of the full fundamental model were independent of concentration and showed an Arrhenius temperature dependence. The shortened fundamental model rate constants showed a possible concentration dependence, which raises doubt about its applicability. The specific absorption rates (mol/m2.s) of CO2 into solutions of MEA/n-Propanol (0.2 M and 0.08 M, T = 25°C and 30°C, P = ±103 kPa) were investigated on a wetted wall experimental setup. The experimental conditions were designed for a fast reaction in the liquid film to occur with a degree of depletion of MEA in the liquid film. Both interfacial depletion and renewal of MEA may be considered to occur. The gas phase resistance to mass transfer was determined to be negligible. An increase in liquid turbulence caused an increase in the specific absorption rate of CO2 which indicated that an increase in liquid turbulence causes an increase in effective mass transfer area. Image analysis of the wetted wall gas-liquid interface confirmed the increase in wave motion on the surface with an increase in liquid turbulence. The increase in wave motion causes an increase in both interfacial and effective area. A numerical solution strategy based on a concentration diffusion equation incorporating the fundamentally derived rate expressions of this study is proposed for calculating the effective area under conditions where surface depletion and renewal rates are significant. It is recommended that the reaction kinetics of CO2 with MEA in solvents of varying liquid properties is determined and the numerical technique proposed in this study used to calculate effective area from absorption rates into these liquids. From the absorption data an effective area correlation as a function of liquid properties may be derived in future. / AFRIKAANSE OPSOMMING: Die reaktiewe absorpsie van CO2 in nie-waterige oplossings van die primêre amien, monoetanolamien (MEA) word erken as ‘n geskikte metode om die effektiewe massaoordragsarea van gepakte skeidingskolomme te bepaal. Tans word die metode gebruik onder vinnige pseudo eerste orde reaksietoestande met betrekking tot CO2. Die pseudo eersteorde aanname beteken dat die konsentrasie van MEA in die vloeistoffilm onbeduidend beïnvloed word deur die reaksie en effektief konstant bly. Onder pseudo eerste orde toestande word oppervlakverarming- en oppervlakvernuwingseffekte nie in ag geneem nie, juis as gevolg van die konstante konsentrasie van MEA in die vloeistoffilm. Daar is voorheen bevind dat oppervlakverarming en oppervlakvernuwing ‘n beduidende invloed het op die beskikbare effektiewe massaoordragsarea. Hierdie invloed kan slegs bestudeer word met ‘n vinnige reaksie in die vloeistoffilm wat gepaard gaan met beduidende oppervlakverarming van die vloeistoffase reagens. Die homogene vloeistoffase reaksiekinetika van CO2 met MEA in die alkohol oplosmiddel, n- Propanol, is in hierdie studie ondersoek. ‘n Nuwe, in-situ Fourier Transform Infra-Rooi (FTIR) metode van analiese is ontwikkel in hierdie ondersoek. Die reaksie is ondersoek in ‘n semienkelladings reaktor met MEA wat gevoer is tot die reaktor om met die opgeloste CO2 te reageer. Die FTIR metode meet spesiekonsentrasie as ‘n funksie van tyd sodat die konsentrasieprofiele van CO2, MEA en een van die soutprodukte van die reaksie gebruik kan word om verskillende reaksiesnelheidsvergelykings te modelleer. Die reaksie is ondersoek onder matige toestande (T = 25°C, 30°C and 35°C, P = 1 atm (abs)). Die konsentrasiebereik van die ondersoek was [MEA]:[CO2] = 5:1 en 10:1. Hierdie bereik is spesifiek gebruik sodat daar beduidende omsetting van MEA kon plaasvind. Die reaksiekinetieka studie het, ter ondersteuning van bestaande teorie, bevind dat die reaksie van CO2 met MEA in nie-waterige oplosmiddels soos alkohole, beskyf word deur ‘n zwitterioon reaksiemeganisme. Die bestaande reaksiesnelheids modelle (eksponensiële wet en pseudo gestadigde toestand hipotese) kon nie die eksperimentele data met genoegsame akuraatheid beskryf nie. Twee nuwe reaksiesnelheidsvergelykings, afgelei vanaf eerste beginsels en gebaseer op die zwitterioon meganisme, word voorgestel. Hierdie volle fundamentele model het goeie passings op die eksperimentele data getoon oor die volledige temperatuur en konsentrasiebereik van hierdie studie. Die reaksiekonstantes van die fundamentele model was onafhanklik van konsentrasie en tipe oplosmiddel en het ‘n Arrhenius temperatuurafhanklikheid. Die verkorte fundamentele model se reaksiekonstantes het ‘n moontlike konsentrasieafhanlikheid gewys. Dit plaas onsekerheid op die fundamentele basis van hierdie model en kan dus slegs as ‘n eerste benadering beskou word. Die spesifieke absorpsietempos (mol/m2.s) van CO2 in MEA/n-Propanol oplossings (0.2 M en 0.08 M MEA, T = 25°C and 30°C, P = ±103 kPa) is ondersoek met ‘n benatte wand (‘wetted wall’) eksperimentele opstelling. Die eksperimentele toestande is gekies sodat daar ‘n vinnige reaksie in die vloeistoffilm plaasgevind het, met beide beduidende en nie-beduidende MEA omsetting. Die doel met hierdie eksperimentele ontwerp was om die invloed van intervlakverarming en intervlakvernuwing op die spesifieke absorpsietempo te ondersoek. Gas fase weerstand was nie-beduidend onder die eksperimentele toestande nie. Beide intervlakverarming en intervlakvernuwing gebeur gelyktydig en is waargeneem vanuit die eksperimentele data. ‘n Beeldverwerkingstudie van die gas-vloeistof intervlak van die benatte wand het bevind dat daar ‘n toename in golfaksie op die vloeistof oppervlak is vir ‘n toename in vloeistof turbulensie. Hierdie golfaksie dra by tot oppervlakvernuwing en ‘n toename in effektiewe massaoordragsarea. ‘n Numeriese metode word voorgestel om die effektiewe area van beide die benatte wand en gepakte kolomme te bepaal vanaf reaktiewe absorpsietempos. Die metode gebruik die fundamentele reaksiesnelheidsvergelykings, bepaal in hierdie studie, in a konsentrasie diffusievergelyking sodat oppervlakverarming en vernuwing in ag geneem kan word. Daar word voorgestel dat die reaksiekinetika van CO2 met MEA in oplossings met verskillende fisiese eienskappe (digtheid, oppervlakspanning en viskositeit) bepaal word sodat die numeriese metode gebruik kan word om ‘n effektiewe area korrelasie as ‘n funksie van hierdie eienskappe te bepaal.

Ce mémoire de maîtrise en recherche-création cherche à explorer de deux façons singulières les possibilités littéraires de la construction (ou de la déconstruction) d’un objet saisi comme une totalité. La création s’écrit selon une contrainte stricte : laisser des matériaux produire du récit afin d’en découvrir méthodiquement le vocabulaire, la symbolique et les résonances historiques. Ce sont encore les matériaux qui dictent l’ordre des textes selon leur propre chronologie. L’essai quant à lui, procède d’une recherche semblable par l’analyse de textes choisis de trois auteurs modernes : Georges Perec, François Bon et Francis Ponge. Par la diversité des genres qu’ils expérimentent, par leurs méthodes d’écriture singulières et par la richesse de leurs corpus, ils constituent un terrain privilégié pour des études de cas variées ainsi que pour une synthèse pertinente. Ces deux parties se conjuguent étroitement, dans une expérience ouverte de la littérature. / The following master’s thesis, in both theory and creative writing, aims to explore by two different paths the literary possibilities of the construction (or deconstruction) of an object considered as a totality. The creative writing is bent by a strict rule: to allow chosen materials to awaken a narrative universe where their vocabularies, symbolisms and historical meanings will be revealed. Those same materials determine the order in which the texts are listed. The essay is aiming at similar goals, but proceeds by analysis of three modern authors: Georges Perec, François Bon and Francis Ponge. By the vast number of literary genres they experiment on, by their unique styles and by the richness of their corpuses, they prove to be perfect subjects for both isolated analysis and a common synthesis. The two parts work closely together to provide an open literary experiment.