Une transition entre deux niveaux voisins d'un atome de Rydberg et une cavité micro-onde de très haut facteur de qualité constituent un excellent outil pour la recherche sur les interactions matière- rayonnement au niveau le plus fondamental. La simplicité du système (deux niveaux atomiques couplés à un seul mode du champ) permet un traitement analytique complet de la plupart des phénomènes. Dans ce mémoire nous étudions les effets dispersifs de l'interaction non-résonnante entre atomes et cavité. Nous avons mesuré la variation linéaire des déplacements des niveaux d'énergie atomiques avec le nombre moyen de photons dans la cavité. Des déplacements dus à une intensité moyenne inférieure au photon unique ont été observés. En l'absence de champ injecté, il a été possible de mesurer le déplacement résiduel d'un des deux niveaux de la transition atomique: un déplacement de Lamb dû à un seul mode du champ. Ces déplacements d'énergie sont mesurés de façon sensible par une méthode interférométrique: la technique des champs oscillants séparés de Ramsey. Des expériences futures, dans une situation de très faible relaxation du champ, sont proposées. Le caractère quantique du champ sera alors dominant et il sera possible de réaliser une mesure nondestructive du nombre de photons: le caractère non-résonnant de l'interaction assure que les atomes ne peuvent ni absorber ni émettre des photons dans la cavité. Les expériences réalisées démontrent la sensibilité de l'appareil et ouvrent la voie à ces mesures non-destructives ainsi qu'à l'étude de systèmes "mésoscopiques" (états "chat de Schrödinger" du champ), à la "frontière" entre les mondes classique et quantique.

This thesis discusses three bronze oil lamps found in the ancient city of Meroë, in the collection of the Museum of Fine Arts, Boston. Scholars have considered the lamps to be imported from Hellenistic Egypt, but careful examination has revealed that the lamps were not imported. The lamps were locally made in Meroë; the materials and technology needed to create bronze lamps were available to Meroitic craftsmen.

Nowadays, lighting represents 20% of the global electricity consumption. Light can be produced using different technologies but more than 100 years after its invention, the incandescent bulb is still the most sold and one of the more used light sources. Of the total energy input in an incandescent bulb more than 90% is lost as heat while less than 10% is converted into visible light. However, there are alternative technologies which use up to 85% less energy for conventional lighting and there are even more efficient light sources for other purposes that if they replace completely all incandescent lamps over the world could reduce dramatically the global electricity consumption and greenhouse gases emissions. One may identify these alternative technologies mainly as LEDs and discharge lamps, but are they the unique alternatives? This thesis is focused on a new lighting technology whose name is LightLab and which is based on the field emission and cathodoluminescence concepts. This technology is under the research and development stage but prototypes have already achieved energy savings over 85% compared to incandescent lamps with a great color performance and with the advantage that it does not use mercury or other hazardous substances compared with discharge lamps. Thus, in the first part of the project all technologies and last improvements are studied while the second part analyses the market applications possibilities for the LightLab lamp considering the environmental perspective regulations and comparing the lamp with the other light sources. The result is that despite there are still some unknown parameters that need to be developed or improved, the lamp has a great potential for different applications fields.

Large Area Maskless Photopolymerization (LAMP) is a direct digital manufacturing technology being developed at Georgia Tech to produce ceramic molds for investment casting of turbine airfoils. In LAMP, UV light incident on a spatial light modulator is projected in the form of a structured black and white bitmap image onto a platform supporting slurry comprising a ceramic particle loaded photocurable resin. Curing of the resin is completed rapidly with exposures lasting 20~160ms. Three-dimensional parts are built layer-by-layer by sequentially applying and selectively curing resin layers of 25-100 micron thickness. In LAMP, diacrylate-based ceramic particle-loaded resins with photoinitiators sensitive in the range of spectral characteristics of the UV source form the basis for an ultra-fast photopolymerization reaction. At the start of the reaction, the monomer molecules are separated by van der Waals distance (~10⁴Å). As the reaction proceeds, these monomer molecules form a closely packed network thereby reducing their separation to covalent bond lengths (~ 1 Å). This results in bulk contraction in the cured resin, which accumulates as the part is fabricated layer-by-layer. The degree of shrinkage is a direct measure of the number of covalent bonds formed. Thus, shrinkage in LAMP is characterized by estimating the number of covalent bonds formed during the photopolymerization reaction. Polymerization shrinkage and accompanying stresses developed during photopolymerization of ceramic particle-loaded resins in LAMP can cause deviations from the desired geometry. The extent of deviations depends on the photoinitiator concentration, the filler loading, the degree of monomer conversion, and the operating parameters such as energy dose. An understanding of shrinkage and stresses built up in a part can assist in developing source geometry compensation algorithms and exposure strategies to alleviate these effects. In this thesis, an attempt has been made to understand the curing kinetics of the reaction and its relation to the polymerization shrinkage. Realtime Fourier Transform Infrared Spectroscopy (RTFTIR) is used to determine the conversion of monomers into polymer networks by analyzing the changes in the chemical bonds of the participating species of molecules. The conversion data can further be used to estimate the curing kinetics of the reaction and the relative volumetric shrinkage strain due to polymerization.

The pre-symptomatic stage of neurodegenerative diseases such as Alzheimer’s disease (AD) and Parkinson’s disease (PD) occurs several decades before the clinical onset. Changes in the lysosomal network, i.e. the autophagosomal, endosomal and lysosomal vesicular system, are among the first alterations observed. There are currently no treatments to slow or cure neurodegenerative diseases, and there is a great need for discovery of treatment targets in cellular pathways where pathology pre-dates the neuronal death. It is also crucial to be able to diagnose neurodegenerative diseases earlier, both to enable early intervention treatment and aid in selecting clinical trial populations before the patient has widespread pathology. This thesis aims at investigating the potential of lysosomal network proteins as biomarkers and therapeutic targets in neurodegenerative disease. A targeted search for lysosomal network proteins was performed in cerebrospinal fluid (CSF) from AD patients, and seven proteins: early endosomal antigen 1 (EEA1), lysosomal-associated membrane proteins 1 and 2 (LAMP-1, LAMP-2), lysozyme, microtubule-associated protein 1 light chain 3 (LC3), Rab3 and Rab7, were elevated. The levels of EEA1, LAMP-1, LAMP-2, LC3, lysozyme and Rab3 were also measured in CSF from parkinsonian syndrome patients: PD, clinically diagnosed 4-repeat tauopathy, pathologically confirmed corticobasal degeneration (CBD) and pathologically confirmed progressive supranuclear palsy (PSP) patients. LAMP-1 and LAMP-2 were decreased in PD. LC3 and lysozyme levels were increased in 4-repeat tauopathy patients. EEA1 was decreased and lysozyme increased in PSP, and LAMP-1, LAMP-2, LC3 and lysozyme were increased in CBD. The lysosomal network proteins had different CSF protein profiles in all the parkinsonian syndromes, as well as in AD. It should be emphasized that only a select few of the lysosomal network proteins were observed to be changed, rather than a general change in lysosomal network proteins, which implicates the involvement of these seven proteins in specific pathological processes. The most interesting candidates, LAMP-2 and lysozyme, were selected for further study for their involvement in the pathology of AD. Lysozyme was found to co-localise with Aβ plaques in AD patients and overexpression prolonged survival and improved the activity in a Drosophila model of AD. Lysozyme was found to alter the aggregation pathway of Aβ1-42, to counteract the formation of toxic Aβ species and to protect from Aβ1-42 induced cell toxicity. Aβ1-42 in turn was found to increase the expression of lysozyme in both neuronal and glial cells. These data suggest that lysozyme levels rise in AD as a compensatory response which is protective against Aβ associated toxicity. LAMP-2 mRNA and protein were found increased in brain areas relevant for AD pathology and various cellular models showed complex involvement of LAMP-2 in Aβ related pathology, with extensive crosstalk between LAMP-2 and Aβ. Exposure to oligomeric Aβ1-42 caused an upregulation of LAMP-2 and in turn, overexpression of LAMP-2 caused a reduction in secreted levels of Aβ1-42, as well as changing the generation pattern of Aβ and affecting clearance and secretion of Aβ1-42. These data indicate that the increased levels of LAMP-2 in AD could be an attempt to regulate Aβ generation and secretion. In summary, this thesis reports that utilising lysosomal network proteins as biomarkers and novel therapeutic targets for neurodegenerative diseases holds great promise.

Thema der vorliegenden Arbeit war die Entwicklung und Konstruktion einer Gasentladungslampe unter Verwendung von transparenten, lumineszenten Dünnfilmen. Zunächst wurden Leuchtstoffe in nanoskaligem Maßstab hergestellt, um daraus anschließend dünne Filme herzustellen, die für Licht im sichtbaren Wellenlängenbereich durchlässig sind. Diese Filme können nach Anregung mit UV-Strahlung sichtbares Licht emittieren und eignen sich daher für den Bau einer transparenten Gasentladungslampe. Als Leuchtstoffe kommen Verbindungen in Frage, die auch in herkömmlichen Leuchtstoffröhren verwendet werden. Dazu zählt unter anderem YVO4:Eu als viel genutzter rot emittierender Leuchtstoff und LaPO4:Ce,Tb mit charakteristischer grüner Emission. Diese beiden Verbindungen wurden zunächst als funktionalisierte Nanopartikel hergestellt. Unter Verwendung einer Mikrowelle wurden agglomeratfreie, wässrige YVO4:Eu Dispersionen synthetisiert, deren Partikel einen hydrodynamischen Durchmesser im Bereich von 45,6 nm bis 51,7 nm aufwiesen. Diese Partikel kristallisieren im tetragonalen Kristallsystem und besitzen eine Kristallitgröße zwischen 9 nm bis 10 nm. Die in wässriger Dispersion ausgelöschte Lumineszenz ist nachweisbar, nachdem die Partikel mit Aceton destablisiert wurden. Die getrockneten Pulverproben zeigen dann nach Anregung mit UV-Licht die typische rote Emission des Eu3+ mit Intensitätsmaximum bei 618 nm und besitzen eine Quantenausbeute von ca. 17,3 %. Durch eine Temperaturbehandlung bei 600 °C konnte die Kristallinität der Partikel erhöht werden. Dies führte zu einer Erhöhung der Quantenausbeute auf bis zu 37,2 %. Die Stabilisierung der LaPO4:Ce,Tb Partikel erfolgte im Gegensatz zu YVO4:Eu aus einem Gemisch von Zitronensäure und Phosphorhexansäure. Auf diese Weise konnten stabile wässrige Dispersionen mit 39 nm (hydrodynamischer Durchmesser) großen Partikeln hergestellt werden. Allerdings konnte nicht eindeutig belegt werden, dass es sich LaPO4 mit hexagonalem Kristallsystem handelt, da die Partikel nur sehr schlecht kristallisieren. Daher ist auch die Bestimmung der Kristallitgröße nicht möglich gewesen. Fällt man die Partikel aus wässriger Dispersion aus, so ist anschließend im Fluoreszenzspektrum die charakteristische Tb3+ Lumineszenz mit Intensitätsmaximum bei  = 544 nm nachweisbar. Die Kristallisierung dieser Partikel durch eine Temperaturbehandlung bei 600 °C führte zur Bildung von monoklinem LaPO4. Dabei kam es jedoch zu einem unerwünschten Nebeneffekt. Das im LaPO4:Ce,Tb enthaltene Cer wurde von Ce3+ zu Ce4+ oxidiert und damit die komplette Tb3+-Emission ausgelöscht. Dieser Vorgang ist reversibel, so dass Tempern unter reduktiven Bedingungen die Lumineszenz wieder aktivierte. Um die Partikel in organischen Medien dispergieren zu können, wurde die Oberfläche beider Partikelsysteme mit Octylamin funktionalisiert. Die agglomeratfreien Dispersionen in THF konnten anschließend für Schleuderprozesse zur Dünnfilmerzeugung genutzt werden. Dünne Schichten von YVO4:Eu bzw. LaPO4:Ce,Tb auf Glassubstraten besaßen auch nach Tempern bei 600 °C eine sehr hohe Transmission im sichtbaren Wellenlängenbereich und emittierten charakteristische Fluoreszenz nach Anregung mit UV-Licht. YVO4:Eu beschichtete Substrate wiesen trotz geringer Schichtdicken von ca. 350 nm hohe Quantenausbeuten in Höhe von 22,9 % auf und eignen sich hervorragend für die Verwendung in Gasentladungslampen. Dagegen sind LaPO4:Ce,Tb beschichtete Substrate problematisch, da die Lumineszenz erst durch Reduktion im Wasserstoffstrom aktiviert werden muss (Reduzierung von Ce4+ zu Ce3+). Ziel war es, mit den hergestellten lumineszenten Dünnfilmen eine transparente Gasentladungslampe herzustellen. Bei der Planung der Konstruktion mussten viele Punkte beachtet werden. Dazu gehörte unter anderem die Art des Füllgases, das Elektrodenmaterial, die Abdichtung der Gaskammer, Messung des Kammerdrucks, Art des Dielektrikums und vor allem Spannungsversorgung. Im Verlauf mehrerer Serien wurden die hergestellten Lampen bis zu einem Prototyp laufend in Hinsicht Gasabdichtung, Gaszuführung und Druckkontrolle weiterentwickelt. Letztlich konnte eine funktionierende Gasentladungslampe gebaut werden, die nach Anlegen einer Spannung von 4 kV bei 50 kHz Licht in der Farbe des Leuchtstoffes der Schicht (rot – YVO4:Eu) emittierte.

Thermische Ausheilprozesse werden bei der Transistorformation im Wesentlichen eingesetzt, um die durch die Ionenimplantation entstandenen Kristallschäden auszuheilen und die eingebrachten Dotanden zu aktivieren. Besonders kritisch sind dabei die finalen Aktivierungsprozesse, bei denen die Source/Drain-Gebiete der Transistoren gebildet werden. Im Zuge der kontinuierlichen Skalierung der CMOSFET-Technologie ist es außerdem erforderlich, möglichst flache, abrupte Dotierungsprofile mit maximaler elektrischer Aktivierung zu erhalten, um die bei diesen Bauelementeabmessungen immer stärker auftretenden Kurzkanaleffekte zu unterdrücken und gleichzeitig eine höhere Leistungsfähigkeit der Transistoren zu gewährleisten. Zur maximalen Aktivierung bei minimaler Diffusion der eingebrachten Dotanden müssen dazu während der finalen Ausheilung extrem kurze Ausheilzeiten bei sehr hohen Temperaturen bewerkstelligt werden. Mit dem derzeitig angewandten Ausheilverfahren, der schnellen thermischen Ausheilung (RTA), bei der die minimale Ausheilzeit im Bereich von 1 s liegt, sind diese Vorgaben nicht mehr realisierbar. Nur durch den Einsatz von neuartigen thermischen Ausheilprozessen mit Ausheilzeiten im Millisekundenbereich können diese Forderungen erreicht werden. Das Thema der vorliegenden Arbeit ist die wissenschaftliche Untersuchung der neuartigen Ausheilprozesse und die experimentelle Realisierung von Integrationsmöglichkeiten in die planare Hochleistungs-SOI-CMOSFET-Technologie. Dazu wird zunächst die Notwendigkeit der Einführung der neuartigen Ausheilprozesse erläutert. Anschließend wird basierend auf experimentellen Untersuchungen der Einfluss der Kurzzeitausheilung auf die Diffusion und Aktivierung der Dotierstoffe für eine p- und n-Dotierung analysiert. Des Weiteren werden zwei unterschiedliche Technologien der Kurzzeitausheilung, die Blitzlampen- und Laser-Ausheilung, und deren Einfluss auf das Transistorverhalten sowohl auf Wafer- als auch auf Mikroprozessorebene untersucht. Der Schwerpunkt der vorliegenden Arbeit liegt auf der experimentellen Untersuchung zur Integration der Kurzzeitausheilung in den Herstellungsprozess von Hochleistungs-SOI-CMOSFETs. Zwei verschiedene Ansätze werden dabei näher betrachtet. Zum Einen wird der Einfluss der Kurzzeitausheilung als zusätzlicher Ausheilschritt im Anschluss an die herkömmliche RTA und zum Anderen als alleiniger Ausheilschritt ohne RTA untersucht. Die Ergebnisse der durchgeführten Experimente zeigen, dass durch die zusätzliche Kurzzeitausheilung nach Ansatz 1 ohne eine Veränderung des Herstellungsprozesses ein verbessertes Transistorverhalten erreicht werden kann. Demgegenüber ist die Integration der Kurzzeitausheilung nach Ansatz 2 nur durch eine Anpassung der Transistorarchitektur und eine Optimierung der Implantationsparameter für die Halo-, Source/Drain-Erweiterungs- und Source/Drain-Gebiete möglich. Ein Hauptaugenmerk bei der Herstellung diffusionsarmer p-MOSFETs nach Ansatz 2 liegt in der Implementierung von Si1-xGex-Source/Drain-Gebieten, um die Erhöhung der Leistungsfähigkeit durch diese Verspannungsquelle auch bei diesen Transistortypen zu gewährleisten. Die dazu durchgeführten experimentellen Untersuchungen zeigen, dass bei diffusionsarmen p-MOSFETs mit Si1-xGex in den Source/Drain-Gebieten des Transistors, die Wahl der richtigen Implantationsspezies von entscheidender Bedeutung ist. Abschließend erfolgt eine Gegenüberstellung der Ergebnisse von optimierten, diffusionsarmen n- und p-MOSFETs mit Transistoren der 45 nm-Technologie. Letztere basieren auf einem Prozess mit einer kombinierten Ausheilung von RTA und Kurzzeitausheilung. Dabei wird gezeigt, dass im Gegensatz zur herkömmlichen RTA-Ausheilung eine weitere Miniaturisierung der planaren Transistorstruktur mit Hilfe der Kurzzeitausheilung möglich ist.

The presence of arsenic as a contaminant in water and biological fluids has been studied in recent years due to the high toxicity of this element. In this work a method for the selective determination of arsenic species was developed, which is based on the determination of As(III) directly in the samples and on the determination of total arsenic after a photochemical pretreatment step. For the determination of As(III), different voltammetric methods have been used: the Adsorptive Cathodic Stripping Voltammetry (AdCSV), the Cathodic Stripping Voltammetry (CSV) and Anodic Stripping Voltammetry (ASV). Amongst the studied methods, the ASV, using a gold electrode, showed to be highly sensitive and reproducible, with a detection limit of 0,2 μg L−1 and quantification limit of 0,5 μg L−1 calculated for arsenic. For the speciation between As(III) and As(V), a new procedure based on the photochemical pretreatment for the reduction of As(V) to As(III) in the presence of reducing sugars such as glucose was developed. Regarding the proposed mechanism for the photochemical pretreatment, it is suggested that reduction of As(V) to As(III) does not occur by direct action of glucose and even by the direct action of UV radiation. An indirect reduction process, which is induced by the glucose decomposition under UV irradiation, has been proposed. The experimental parameters irradiation time and glucose concentration were systematically investigated. The reduction of As(V) to As(III) was achieved with about 100% efficiency in the presence of glucose 60 mg L−1, which was added to the solution containing As(V) and then it was irradiated for 90 minutes at a temperature of 85 °C. For the solutions with a high content of organic matter, a previous step prior to the reduction step was necessary for the sample decomposition and the elimination of the matrix interference. Recovery experiments for As(III) and As(V) were carried out in spiked samples. The recovery results for total arsenic total were above 80%, confirming the applicability of the procedure involving the photochemical pretreatment for the arsenic speciation in samples of waters and biological fluids. / A presença de arsênio como contaminante em águas e fluidos biológicos tem sido estudada nos últimos anos devido à alta toxicidade deste elemento. Neste trabalho desenvolveu-se um método para a determinação seletiva das espécies de arsênio, baseado na determinação de As(III) direto nas amostras e de arsênio total após uma etapa de pré-tratamento fotoquímico. Para a determinação de As(III) na faixa de 1 a 100 μg L−1 utilizou-se três métodos voltamétricos diferentes, a voltametria adsortiva de redissolução catódica (AdCSV), a voltametria de redissolução catódica (CSV) e a voltametria de redissolução anódica (ASV), sendo que com a ASV, usando um eletrodo de ouro, observou-se uma maior reprodutibilidade e sensibilidade das medidas, com limite de detecção de 0,2 μg L−1 e limite de quantificação de 0,5 μg L−1. Para a especiação entre As(III) e As(V) utilizou-se um novo procedimento baseado no pré-tratamento fotoquímico para a redução de As(V) a As(III) empregando uma etapa de redução na presença de glicose como açúcar redutor. Com relação ao mecanismo proposto para o pré-tratamento fotoquímico pode-se afirmar que a redução As(V) a As(III) não ocorre por ação direta da glicose e nem mesmo pela ação direta da radiação UV, mas sim por processos indiretos induzidos pela decomposição da glicose sob ação da radiação UV. Os parâmetros experimentais como tempo de irradiação e concentração de glicose foram sistematicamente investigados, sendo que a redução do As(V) para As(III) ocorre com aproximadamente 100% de eficiência na presença de glicose 60 mg L−1 adicionada à solução contendo As(V) e irradiada por 90 minutos a uma temperatura de 85 °C. Para soluções com um alto teor de matéria orgânica é preciso uma etapa prévia à etapa de redução, para que haja a decomposição da amostra eliminando a interferência matricial presente. Ensaios de recuperação para As(III) e As(V) adicionados às amostras em diferentes concentrações foram realizados com recuperações para arsênio total acima de 80%, comprovando a aplicabilidade do procedimento envolvendo o pré-tratamento fotoquímico para especiação de arsênio em amostras de águas e fluidos biológicos.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / In this paper, we present numerical results obtained from the FORTRAN language for physical quantities of interest such as velocity profile, pro�le, heat ow rate, particle ow rate,heat ux and pressure tensor component. The gas ow occur in the direction parallel to thesurface the gas is confined because of a constant gradient of pressure and a constant gradient of temperature are represented by Poiseuille Problem and Problem Creep Thermal, respectively. It also considers the Couette Problem where the gas moves from the motion of the plates in opposite directions. In order to describe the gas-surface interaction we use the kernel of Cercignani-Lamp, which as opposed to core scattering Maxwell has two accommodation coeficients to represent the physical properties of gas, leaving this interaction closer to reality. From the simplification of the Boltzmann equation has the kinetic theory for rarefied gas dynamics, which is developed analytically in a uni�ed approach to the BGK Model, S Model, Gross-Jackson (GJ) Model and MRS Model. Thus, we seek to model that most closely approximates the veracity, comparing the numerical values generated by the models and the linearized Boltzmann equation through numerical analysis, graphics and mathematical statistics with the procedure of the variance of two factors made by Friedman. A version of the analytical method of discrete ordinates (ADO) is used to solve the problems of Poiseuille, Creep Thermal and Couette for two plates infinte paralalelas with different chemical constitutions Boundary Conditions for the Cercignani-Lampis. / Neste trabalho, são apresentados resultados numéricos obtidos a partir da linguagem FORTRAN para quantidades físicas de interesse como perfil de velocidade, perfil fluxo de calor, taxa fluxo de partícula, taxa fluxo de calor e componente tensor de pressão. O fluxo gasoso ocorre na direção paralela a superfície que o gás esta confinado devido a um gradiente constante de pressão e um gradiente constante de temperatura que são representados pelo Problema de Poiseuille e Problema Creep térmico respectivamente. Além disso, também considera-se o Problema de Couette onde o gás se move a partir do movimento das placas em sentidos opostos. A �fim de descrever a interação gás-superfície utiliza-se o núcleo de Cercignani-Lampis, que ao contrário do núcleo de espalhamento de Maxwell tem dois coeficientes de acomodação para representar as propriedades físicas do gás, deixando esta interação mais próxima da realidade. A partir da simplificação da Equação de Boltzmann tem-se a teoria cinética para a dinâmica de gases rarefeito, que é desenvolvida analiticamente em uma abordagem unificada para o Modelo BGK, Modelo S, Modelo Gross-Jackson (GJ) e Modelo MRS. Dessa maneira, busca-se o modelo que mais se aproxima da veracidade, comparando os valores numéricos gerados pelos modelos e a Equação Linearizada de Boltzmann através de análises numéricas, gráficas e de matemática estatística com o procedimento da variância de dois fatores de Friedman por postos. Uma versão analítica do método de ordenadas discretas (ADO) é usada para resolver os Problemas de Poiseuille, Creep Térmico e Couette para duas placas paralalelas infintas com constituições químicas diferentes para Condições de Contorno de Cercignani-Lampis.

Objetivo: Este estudo se propõe a verificar a influencia do tipo de luz fotopolimerizadora na força de retenção de pinos intra-radiculares, tanto na técnica direta como na direta indireta. Métodos: 40 raízes de dentes bovinos com comprimento de 12 mm foram tratadas endodonticamente e divididas aleatoriamente em 4 grupos de acordo com a técnica de obtenção dos pinos e luz fotopolimerizadora (n=10): grupo 1 (técnica direta associada à lâmpada halogena); grupo 2 (técnica direta ao LED); grupo 3 (técnica direta-indireta associada à lampada halógena); e grupo 4 (técnica direta-indireta associada ao LED). A força de retenção foi determinada através do teste de tração usando Universal Testing Machine (Instron). Todos os dados foram analisados usando one-way analysis of variance (ANOVA) com a significância de p<0.05 e complementadas com teste de Tukey. Após o teste, as interfaces adesivas onde ocorreram as falhas foram examinadas e classificadas. Resultado: Os grupos 3 (246,05N ± 29,51) e 4 (241,60N ± 28,95) não apresentaram força de retenção estatisticamente diferente mas foram maiores que os grupos 1 (142,30N ± 25,60) e 2 (178,56N ± 25,67). A maior parte das fraturas ocorreu na interface dentina/resina. Conclusão: Com base nos resultados obtidos, concluímos que o método direto-indireto proporcionou a melhor retenção dos pinos de fibra de vidro. / Objectives: The purpose of this study is to verify the influence of light curing type on retention force of direct technique and direct-indirect technique. Methods: Fourth bovine single root teeth with 12mm of length were used in this study. The roots were endodontically treated and randomly divided in four groups according to the light curing unit and technique used: group 1(direct technique, halogen lamp), group 2(direct technique, LED), group 3(direct-indirect technique, halogen lamp), group 4(direct-indirect technique, lamp). The retention force was determined sing a Universal Testing Machine (Instron). All data were analyzed using one-way analysis of variance (ANOVA) and Tukey test. After the test, the failure was examined and classified according to the fracture place: post/ resin; resin/dentin; mix. Results: Group 3 (246,05N ± 29,51) and 4(241,60N ± 28,95)4 (95,18N) did not show statistically difference but presented higher retention force than group 1 (142,30N ± 25,60) and 2 (178,56N ± 25,67). Most of fracture occurred in interface between dentin/resin. Conclusion: Based on the obtained results, it was concluded that adhesive cementation technique influenced in the retention of glass fiber post.