This study focused on determining the groundwater flow paths in the crystalline subsurface rocks of the Bulawayo metropolitan, Zimbabwe, through analysing the discontinuities of the electrical properties of the ground, as well as in the magnetic field of the underlying rocks. Further, borehole water quality was analysed by measuring and mapping several chemical parameters, specifically TDS, salinity and the electrical conductivity. The electrical and magnetic anomalies were measured at two field sites within the Bulawayo province, the Harry Allen Golf Course and the Barbour Fields dumpsite, while 120 boreholes were sampled for water quality in a large part the province. Two magnetometers were used to measure the magnetic field and the time and location of the measurements, which resulted, after processing in SURFER, into the magnetic field map of these areas. This was used, in conjunction with the geologic map of the Bulawayo province, in order to determine useful locations for the electrical resistivity surveys. These included electrical resistivity tomography and vertical electrical sounding and were performed with an earth resistivity/induced polarization (IP) meter in order to measure the electrical resistivity of the ground. Inverse modelling was used in the RES2DINV software program to produce the topographic image of resistivity. The results for the borehole sampling showed that the maps for the three chemical parameters were very similar, with the western and northern parts of the mapped area displaying higher concentration values. The results of the electrical resistivity surveying showed probable areas of groundwater flow with its relationship to electric conductivity.

Doctor of Philosophy / Department of Mechanical and Nuclear Engineering / Steven J. Eckels / The research presents the investigation of personal cooling systems (PCS) and their effects on humans from a thermodynamic perspective. The original focus of this study was to determine the most appropriate PCS for dismounted U.S. Army soldiers in a desert environment. Soldiers were experiencing heat stress due to a combination of interrelated factors including: environmental variables, activity levels, and clothing/personal protective equipment (PPE), which contributed to the buildup of thermal energy in the body, resulting in heat stress. This is also a common problem in industry, recreation, and sports. A PCS can serve as a technological solution to mitigate the effects of heat stress when other solutions are not possible. Viable PCS were selected from the KSU PCS database, expanded to over 300 PCS in the course of this study. A cooling effectiveness score was developed incorporating the logistical burdens of a PCS. Fourteen different PCS configurations were tested according to ASTM F2370 on a sweating thermal manikin. Four top systems were chosen for ASTM F2300 human subject testing on 22 male and 2 female soldiers in simulated desert conditions: dry air temperature = 42.2 ºC, mean radiant temperature = 54.4 ºC, air velocity = 2.0 m/s, relative humidity = 20%. Subjects wore military body armor, helmets and battle dress uniforms walking on treadmills at a metabolic rate of approximately 375-400W. All the PCS conditions showed significant reductions in core temperature rise, heart rate, and total sweat produced compared to the baseline (p<0.05). The expected mean body temperature was higher in the human subjects than expected based on the cooling obtained from the sweating manikin test. Lowered sweat production was determined to be the likely cause, reducing the body’s natural heat dissipation. The ASHRAE two-node model and TAITherm commercial human thermal models were used to investigate this theory. A method to account for fabric saturation from dripping sweat was developed and is presented as part of a new model. This study highlights that the response of the human body is highly complex in high-activity, high-temperature environments. The modeling efforts show the PCS moved the body from uncompensable to compensable heat stress and the body also reduced sweating rates when the PCS was used. Most models assume constant sweating (or natural heat loss) thus the PCS sweat reduction is the likely cause of the higher than expected core temperatures, and is an important aspect when determining the purpose of a PCS.

Thesis (MSc)--University of Stellenbosch, 2011. / ENGLISH ABSTRACT: A magneto optical trap (MOT) is capable of trapping a vapor cloud consisting of atoms cooled down to the micro Kelvin range. Three orthogonal pairs of counter-propagating laser beams of the correct circular polarisation form an optical molasses which facilitates the cooling of neutral atoms. Additionally a spatially non-uniform magnetic field produced by two current carrying coils in a Maxwell gradient configuration is used to trap the cooled atoms. In this report the effects of the trap parameters, including the laser beam intensity and frequency detuning, beam diameter and magnetic field gradient, on the number of trapped atoms are discussed. Secondly the development of an experimental setup for laser cooling and trapping of 87Rb atoms in vacuum with the aid of a MOT is presented. All trap components were implemented and characterised. The vacuum system and trapping chamber in which the cooling takes place were designed and constructed. A rubidium getter to act as a source of atoms was integrated into the vacuum system. The two external cavity diode lasers used for trapping and optical re-pumping were characterised. The optical setup required for the optical molasses was designed, constructed and characterised. Saturated absorption spectroscopy was performed to investigate the hyperfine structure of 87Rb and to frequency lock the lasers. We report on the current status of the project with regards to progress, results and future work. / AFRIKAANSE OPSOMMING: 'n Magneto-optiese val (magneto optical trap, MOT) kan 'n dampwolk van atome vang en afkoel tot in die mikro Kelvin bereik. Drie ortogonale pare laserbundels, elke paar voortplantend in teenoorgestelde rigtings, met die korrekte sirkelvormige polarisasie vorm 'n sogenaamde optiese molasse wat die afkoeling van neutrale atome moontlik maak. Bykomend word 'n ruimtelik nie-uniforme magneetveld geproduseer deur twee stroomdraende spoele in 'n Maxwell gradient-opstelling gebruik om die afgekoelde atome te vang. In hierdie verslag word die invloed van die val parameters, insluitend die laserbundel intensiteit en frekwensie afstemming, die laserbundel deursnit en magneetveld gradiënt, op die aantal atome in die val bespreek. Tweedens word die ontwikkeling van 'n eksperimentele opstelling vir laser afkoeling en vang van 87Rb atome in vakuum met die hulp van 'n MOT voorgelê. Alle komponente van die val is geïmplementeer en gekarakteriseer. Die vakuumsisteem en val-kamer waarin die afkoeling plaasvind is ontwerp en gebou. 'n Rubidium gasbinder is in die vakuumsisteem ingebou om as 'n bron van atome te dien. Die twee eksterne resonator diodelasers wat gebruik is vir die val en die optiese terugpomp is gekarakteriseer. Die optiese opstelling wat nodig is vir die optiese molasse is ontwerp, gebou en gekarakteriseer. Versadigde absorpsiespektroskopie is uitgevoer om die hiperfynstruktuur van 87Rb te ondersoek en om die lasers se frekwensies te stabiliseer. Verslag word gedoen oor die huidige stand van die projek wat betref vordering, resultate en toekomstige werk.

Doctor of Philosophy / Department of Mechanical and Nuclear Engineering / Steven J. Eckels / The research presents the investigation of personal cooling systems (PCS) and their effects on humans from a thermodynamic perspective. The original focus of this study was to determine the most appropriate PCS for dismounted U.S. Army soldiers in a desert environment. Soldiers were experiencing heat stress due to a combination of interrelated factors including: environmental variables, activity levels, and clothing/personal protective equipment (PPE), which contributed to the buildup of thermal energy in the body, resulting in heat stress. This is also a common problem in industry, recreation, and sports. A PCS can serve as a technological solution to mitigate the effects of heat stress when other solutions are not possible. Viable PCS were selected from the KSU PCS database, expanded to over 300 PCS in the course of this study. A cooling effectiveness score was developed incorporating the logistical burdens of a PCS. Fourteen different PCS configurations were tested according to ASTM F2370 on a sweating thermal manikin. Four top systems were chosen for ASTM F2300 human subject testing on 22 male and 2 female soldiers in simulated desert conditions: dry air temperature = 42.2 ºC, mean radiant temperature = 54.4 ºC, air velocity = 2.0 m/s, relative humidity = 20%. Subjects wore military body armor, helmets and battle dress uniforms walking on treadmills at a metabolic rate of approximately 375-400W. All the PCS conditions showed significant reductions in core temperature rise, heart rate, and total sweat produced compared to the baseline (p<0.05). The expected mean body temperature was higher in the human subjects than expected based on the cooling obtained from the sweating manikin test. Lowered sweat production was determined to be the likely cause, reducing the body’s natural heat dissipation. The ASHRAE two-node model and TAITherm commercial human thermal models were used to investigate this theory. A method to account for fabric saturation from dripping sweat was developed and is presented as part of a new model. This study highlights that the response of the human body is highly complex in high-activity, high-temperature environments. The modeling efforts show the PCS moved the body from uncompensable to compensable heat stress and the body also reduced sweating rates when the PCS was used. Most models assume constant sweating (or natural heat loss) thus the PCS sweat reduction is the likely cause of the higher than expected core temperatures, and is an important aspect when determining the purpose of a PCS.

Nesta dissertação, propomos um método que combina a técnica de varredura-Z e a luz branca supercontínua gerada em uma fibra fotônica para caracterizar o espectro de não-linearidades de amostras. O espectro não-linear fornece informações cruciais a respeito da característica, da origem e do potencial da não-linearidade para aplicação em dispositivos. Neste sentido, a luz branca supercontínua aliada à técnica de varredura-Z vem permitir uma rápida determinação de espectros numa banda larga. Neste trabalho, a luz branca foi gerada numa fibra fotônica comercial bombeada por pulsos de femtossegundo de um oscilador laser de Ti:safira. Tanto os lasers de femtosegundo quanto as fibras fotônicas estão se tornando cada vez mais acessíveis. Ademais, a técnica de varredura-Z é um método bastante conhecido para a determinação de não-linearidades devido à sua simplicidade e precisão. Dessa forma, aliamos essas vantagens para fazer um aparato simples, compacto e de baixo custo. Em geral, a determinação do espectro de não-linearidades necessita de sistemas lasers complexos que possam sintonizar o comprimento de onda. O método de varredura-Z de luz branca desenvolvido aqui foi usado para medir o espectro não-linear do corante disperse red 13 (DR13) em metanol. Esse corante possui um efeito não-linear absortivo (absorção saturada) bem caracterizado, sendo um bom material não apenas para calibração, mas também para testar nosso sistema experimental. Os resultados experimentais comprovaram a viabilidade do sistema para caracterização de amostras com efeitos não-lineares absortivos intensos. / In this dissertation, we proposed a method that combines the Z-scan technique and the white-light continuum generated by a photonic crystal fiber to characterize the nonlinear spectrum of materials. The nonlinear spectrum provides crucial information on the characteristic, the origin and the potential of the nonlinearity for devices application. In this sense, the white-light continuum allied to the Z-scan technique allows fast spectra determination in a broad spectral range. In this work, the white light was generated in a commercial photonic crystal fiber pumped by femtosecond pulses from a Ti:sapphire laser oscillator. Both the femtosecond lasers and the photonic crystal fibers are becoming more accessible. In addition, the Z-scan technique is a well-known method for nonlinearities determination due to its simplicity and precision. In this way, we merged these advantages to make a simple, compact and low-cost apparatus for studying nonlinear samples. In general, the nonlinear spectrum determination requires complex laser systems which are able to tune the wavelength. The white-light Z-scan method we developed here was used to measure the nonlinear spectrum of disperse red 13 (DR13) dye in methanol. This dye presents a well-characterized absorcive nonlinear effect (saturable absorption), being a good material not only to calibrate, but also to test our experimental setup. The experimental results confirmed the viability of the setup for characterizing samples with strong nonlinear absorption effect.

La phénoménologie de la donation se présente sous la forme d´un projet d´ouverture radicale du champ de la phénoménalité. Après avoir montré que la modernité a enfermé les phénomènes dans des conditions de possibilité dictées par un Sujet, et que la phénoménologie allemande a poursuivi ce geste en indexant les phénomènes sur l´horizon de l´objectité ou de l´étantité, Marion propose de reconduire le geste de la réduction phénoménologique à la seule donation. Ce faisant, il réintroduit en phénoménologie des phénomènes paradoxaux – phénomènes saturés – qui font violence, par excès, aux capacités réceptrices du Sujet. Marion introduit donc en phénoménologie une nouvelle modalité de l´invisibilité (et donc de la visibilité) que nous appelons « inexistence » (en opposition tout autant avec le non-être qu’avec la non-existence), dont la structure est paradoxalement fondée sur celle de l´écrit, puisqu´il s´agit ni plus ni moins, pour elle, que de présenter de façon positive une absence selon la logique de la signification. Ce faisant, et de par cette structure discursive de la visibilité, toute la phénoménologie de Marion peut être lue comme une véritable apologie de l´inexistence. / The phenomenology of givenness is presented in the form of a project of radical opening of the field of phenomenality. After having shown that modernity has enclosed phenomena under conditions of possibility dictated by a Subject, and that German phenomenology has pursued this gesture by indexing phenomena in the horizon of objectivity or of beingness, Marion proposes to reestablish the gesture of phenomenological reduction to single donation. In so doing, he reintroduces into phenomenology some paradoxical phenomena —saturated phenomena— which violate, by excess, the receptive capacities of the Subject. Marion thus introduces into phenomenology a new modality of invisibility (and thus of visibility) which we call "inexistance" (equally opposed to non-being as to non-existence), whose structure is paradoxically founded on that of writting, since it is neither more nor less, for inexistence, than to present in a positive way an absence according to the logic of meaning. In so doing, and through this discursive structure of visibility, all Marion's phenomenology can be read as a true apology for existence.

A condutividade hidráulica do solo K é uma propriedade que expressa a facilidade com que a água se movimenta no solo. É de extrema importância ao manejo agrícola e consequentemente, à produção das culturas agrícolas e à preservação do solo e do ambiente. A determinação da condutividade hidráulica pode ser feita por métodos de laboratório e de campo. Um com maior controle das condições experimentais e o outro com menor grau de perturbação no solo. Dentre os métodos de campo o mais utilizado é o do perfil instantâneo que foi simplificado por Libardi et al (1980) que obteve teoricamente uma relação entre o conteúdo de água e o tempo de redistribuição t, supondo, gradiente de potencial total unitário. No intuito de uma análise mais profunda do método, pretendeu-se, nesse estudo verificar a possibilidade de obter uma equação do conteúdo de água no solo em função do tempo de redistribuição de modo similar ao modelo de Libardi et al (1980), mas sem a suposição de gradiente de potencial total unitário. O estudo foi desenvolvido com dados da literatura de quatro solos (17 profundidades) e a validação da equação foi realizada por comparação da função K( ) obtida a partir dela com a obtida tradicionalmente. Com base na análise dos resultados, pode-se concluir que a equação proposta entre e t para o método do perfil instantâneo mostrou-se válida e que por ela o gradiente de potencial total pode influir tanto o parâmetro como o parâmetro K0 da função K( ) dada pela equação K = K0 exp [ ( - 0)]. No caso específico dos solos avaliados neste trabalho, a influência do gradiente no parâmetro foi irrelevante comparativamente a tal influência no parâmetro K0, levando à possibilidade de utilização de um gradiente de potencial total médio para a obtenção da função K( ). Além disso, pode-se concluir também que o procedimento para o cálculo da densidade de fluxo a partir da curva da armazenagem em função do tempo de redistribuição da água é muito mais simples que o procedimento proposto por Hillel et al (1972) para esse fim, no método do perfil instantâneo. / The soil water hydraulic conductivity K is a property that expresses the ability of the soil in conducting water. It is of extreme importance to the agricultural management and, consequently, to the soil and environment preservation. The soil water hydraulic conductivity can be determined by means of laboratory and field methods, one better experimentally controlled and the other with less degree of soil disturbing. Among the field methods, the instantaneous profile one is the most used. It was simplified by Libardi et al (1980) that obtained theoretically a relation between the soil water content and the soil water redistribution time assuming unit soil water total potential gradient. In order to analyse in more detail the method, the objective of this work was to obtain an equation similar of the Libardi et al (1980) model but without assuming unit gradient. The study was developed with data of four soils (17 soil depths) from literature and the validation of the equation was carried out by comparing the K( ) function obtained from it and the K( ) function tradicionally obtained. From the results, it could be conclude that the as a function of t proposed equation showed to be valid and according to it the potential gradient may affect both and K0 parameters of the K = K0 exp [ ( - 0)] equation. However, for studied soils the influence of the gradient on the parameter was irrelevant in comparison with the influence of the gradient on the K0 parameter, which leads to the possibility of utilization of a mean gradient to obtain K( ) function. Besides, it could also be conclude that the procedure to calculate the flux density from the equation of soil water storage as a function of redistribution time is much more simple than the procedure proposed by Hillel et al (1972) for this purpose, in the instantaneous profile method.

A previsão de áreas suscetíveis a escorregamentos translacionais rasos tem como fundamental etapa a investigação dos fatores condicionantes e dos mecanismos de ruptura. Algumas destas propriedades, como os parâmetros físicos e hidrológicos dos solos, são determinantes para a instabilização de encostas que estão sujeitas a eventos pluviométricos intensos. Uma destas propriedades é a condutividade hidráulica saturada, cujo comportamento pode influenciar no desenvolvimento de poro-pressão positiva ou levar à perda de sucção dos solos, ocasionando rupturas. Desta forma, o objetivo principal desta pesquisa foi analisar a distribuição lateral e vertical da condutividade hidráulica saturada (Ksat) dos solos e sua influência nos escorregamentos translacionais rasos na Serra do Mar (SP). Para tanto, foi selecionada uma bacia representativa da Serra do Mar e nela foram escolhidas três cicatrizes acessíveis e com limites bem preservados para a realização dos ensaios in situ. As profundidades onde foram realizados os ensaios foram definidas a partir da caracterização morfológica dos mantos de alteração nas trincheiras abertas no topo, no centro e na lateral das cicatrizes (0.25, 0.50, 1.00, 1.50, 2.00 e 2.50 m). Finamente, foram realizados os ensaios in situ utilizando-se o Permeâmetro de Guelph, utilizando-se duas cargas hidráulicas H para a que fosse possível calcular a Ksat empregando-se as equações de Richards, Laplace e a análise de Elricket al. (1989), embora tenha sido utilizada esta última para a análise da variação dos valores. A partir dos valores de Ksat, das propriedades físicas dos materiais (distribuição granulométrica, micro e macroporosidade, porosidade total. Além disso, foi analisada a tendência de variação entre 10-4 e 10-7 m/s, predominando da Ksat com a profundidade. Foram obtidos 41 valores de Ksat, os quais variaram entre as ordens de grandeza 10-5 e 10-6 m/s (82,5% dos valores), que mostraram a pequena variabilidade da Ksat nos perfis de alteração investigados, se comparada às diferenças observadas nas propriedades dos solos e corroboram com os trabalhos desenvolvidos na Serra do Mar. Quando confrontado com as propriedades físicas dos materiais (por exemplo, granulometria e porosidade total), observou-se, principalmente, a correlação positiva com a fração areia. De forma geral, foi possível identificar uma tendência de aumento da Ksat com a profundidade, assim como a existência de algumas descontinuidades hidráulicas significativas, tanto de redução quanto de aumento brusco da Ksat em uma pequena profundidade. / The prediction of landslides-prone areas has as fundamental step researching controlling factors and failure mechanisms. Some of these properties, such as the physical and hydrological soil properties are crucial to the instability of slopes subject to intense rainfall events. One of these properties is the saturated hydraulic conductivity, whose behavior can influence and lead to the development of positive pore-pressure or the loss of soil suction, causing failures. The main objective of this research was to analyze the lateral and vertical hydraulic conductivity (Ksat) of soils and their influence on shallow landslides in the Serra do Mar (SP). Therefore, we selected a representative basin in the Serra do Mar and there three accessible scars, well preserved, were chosen for in situ measurements. Was also performed the topographical characterization of the scars (slope angle, curvature and orientation and contributing area). The depths where the measurements were performed were defined from the morphological characterization of soil profiles in opened trenches at the top, at the lateral and inside the scars (0.25, 0.50, 1.00, 1.50, 2.00 and 2.50 m). At last, were performed the in situ measurements using the Guelph Permeameter, using two heads H for calculating Ksat using Richards equations, Laplace equations and Elrick et al. (1989) analysis, although the latter has been used to analyze the range of values. From the Ksat values and physical soil properties (e. g. grain size distribution, micro and macroporosity, total porosity) we discussed the role played by them in Ksat values. Moreover was analyzed the trend of Ksat variation with depth. We obtained 41 Ksat values, which varied between 10-4 e 10-7 m/s, mostly among the orders of magnitude 10-5 e 10-6 m/s (82.5% values), that showed the small variability of Ksat in the soil profiles if compared to the observed differences in soil properties, and corroborate with studies developed in the Serra do Mar. When confronted with the physical properties of materials (eg, grain size and total porosity), we observed mainly the positive correlation with the sand content. In general, we observed a tendency of increase of Ksat with depth, as well as the existence of some significant hydraulic discontinuities, both the sudden decrease as the increase in Ksat in a small depth.

O presente trabalho teve por objetivo avaliar o potencial do ensaio de ascensão capilar e de uma técnica de otimização de parâmetros para estimar os parâmetros hidráulicos dos solos. O ensaio de ascensão capilar é de execução simples e rápida e reflete um fenômeno condicionado pelas propriedades hidráulicas do solo. Na representação matemática da ascensão capilar, utilizou-se o modelo de Terzaghi (1943) que considerou tratar-se de uma condição de fluxo governada pela condutividade hidráulica saturada (ks), e o modelo de Lu & Likos (2004b), que utilizaram a equação de Gardner (1958) para representar a função condutividade hidráulica, considerando que o fluxo se dá em meio não saturado. O processo de otimização utilizou a rotina SOLVER componente do programa EXCEL, e foi testado utilizando-se os dados experimentais de Lane & Washburn (1946) e de Zhang & Fredlund (2009). Em seguida, resultados de ascensão capilar de oito amostras de solos típicos do Estado do São Paulo, Brasil, foram modelados com os parâmetros derivados da técnica. Observou-se que tanto o modelo de Terzaghi (1943) quanto o de Lu & Likos (2004b) conseguiram reproduzir os resultados experimentais de ascensão capilar, com melhores resultados associados ao modelo de Lu & Likos (2004b) para a maior parte dos solos analisados. A previsão da condutividade hidráulica saturada, de forma geral, diferiu menos de uma ordem de grandeza dos valores medidos experimentalmente, embora maiores discrepâncias tenham sido notadas para determinados tipos de solos. O parâmetro de Gardner da função condutividade hidráulica resultou em valores da mesma ordem de grandeza dos valores reportados na literatura para solos similares. Os resultados sobre a aplicação do ensaio de ascensão capilar e da técnica de otimização mostraram-se promissores para a determinação de parâmetros hidráulicos dos solos analisados, com a vantagem de ter-se um procedimento simples e rápido para a finalidade descrita. / This study evaluates the potential of capillary rise test and a parameter optimization technique to estimate soil hydraulic parameters. The capillary rise is a simple and expedite test that is conditioned by soil hydraulic properties. In the mathematical representation of the capillary rise we used the model of Terzaghi (1943), who considered that saturated hydraulic conductivity commands the phenomeno and the model of Lu & Likos (2004b) who used the Gardner equation (1958) to represent the hydraulic conductivity function, considering that the flow takes place in a non-saturated condition. The optimization process used the SOLVER routine, component of the EXCEL program, that was firstly tested using experimental data of Lane & Washburn (1946) and Zhang & Fredlund (2009). Then, results of capillary rise of eight samples of typical soils of the State of São Paulo, Brazil, were modeled with the parameters derived from the technique. It was observed that both the Terzaghi (1943) and Lu & Likos (2004b) models were able to reproduce the experimental results of capillarity, although the best results were yielded by the model Lu & Likos (2004b) for most soils. The prediction of saturated hydraulic conductivity, in general, differed less than one order of magnitude of experimentally measured values, although larger discrepancies have been noted for certain types of soils. The resulting Gardner parameter of the hydraulic conductivity function were of the same order of magnitude of the values reported in the literature for similar soils. The results on the application of capillary rise test and optimization technique proved promising for the determination of soil hydraulic parameters analyzed, with the advantage of having a simple and rapid procedure for the purpose described.

Objetivo: Nesta dissertação de mestrado, investigamos a associação entre o consumo de gordura saturada (GS), seus alimentos-fontes e os ácidos graxos saturados (AGS) de diferentes tamanhos de cadeia com a incidência de diabetes mellitus tipo 2 (DM2). Método: O Estudo Longitudinal da Saúde do Adulto (ELSA-Brasil) é uma coorte multicêntrica que tem por objetivo investigar fatores de risco e de progressão de diabetes, doenças cardiovasculares e outras doenças crônicas relacionadas. Um total de 15.105 servidores ativos e aposentados de instituições públicas de educação superior e pesquisa, localizadas em seis capitais (Salvador, Belo Horizonte, Rio de Janeiro, São Paulo, Vitória e Porto Alegre), foram recrutados entre agosto de 2008 e dezembro de 2010 para realizarem exames laboratoriais e clínicos e responderem uma série de questionários e entrevistas. Anualmente, todos os participantes são contatados por telefone para entrevista de seguimento, sendo questionados sobre novos diagnósticos médicos, incluindo diabetes. Adicionalmente, entre os anos de 2011 e 2015, os participantes do estudo foram convidados a retornar aos centros de pesquisa para realização de novos exames clínicos e laboratoriais. Todos os participantes assinaram termo de consentimento livre e esclarecido, concordando em participar do estudo, e o protocolo de pesquisa foi aprovado pelo comitê de ética de todas as instituições envolvidas. Para o presente estudo, foram analisados dados de participantes entre 35 e 74 anos, que relataram não possuir diabetes e outras doenças crônicas que pudessem influenciar no consumo alimentar que responderam a um questionário de frequência alimentar (QFA) com 114 itens e os demais questionários com variáveis confundidoras na linha de base (2008-2010), e que não apresentaram ingestão do valor energético total diário pouco plausível (<600 ou >6.000 Kcal/dia) resultando, assim, em uma amostra final de 8.187 participantes. Para a análise do consumo alimentar alguns dos alimentos considerados fontes de GS foram agrupados: laticínios integrais, laticínios desnatados, laticínios fermentados, carnes processadas, carnes não processadas, carne vermelha e peixes. Os diferentes AGS foram organizados conforme o tamanho de sua cadeia: de cadeia curta e média (≤ 12 carbonos) e de cadeia longa (>13 carbonos). Para estimar a composição nutricional dos alimentos incluídos no QFA, foi utilizado o software Nutrition Data System for Research (NDSR). As medidas antropométricas (peso e altura) foram obtidas através de protocolos internacionalmente padronizados. A incidência de DM2 foi identificada de forma autorreferida em entrevistas anuais de seguimento e através do resultado de exames de sangue (glicemia de jejum, hemoglobina glicada e/ou glicemia 2h após teste de tolerância 75g de glicose) realizados em nova visita aos centros de pesquisa (2013-2015). Na análise descritiva, as características dos participantes de acordo com categorias de consumo de GS são comparadas por proporções através do teste quiquadrado, médias por ANOVA e medianas por teste de Wilcoxon. Modelos de Riscos Proporcionais de Cox foram usados para estimar razões de risco e IC95% na relação entre o consumo dos diferentes ácidos graxos saturados, gordura saturada total e seus alimentos-fontes com a incidência de diabetes. Para minimizar a influência de potenciais fatores de confusão, foram utilizados três modelos multivariaveis. As análises estatísticas foram conduzidas com o software SAS versão 9.4. Resultados: Foram identificados 300 novos casos de DM2. Após ajuste para confundidores sociodemográficos, de estilo de vida e dietéticos, encontrou-se efeito protetor da DM2 para o consumo de gordura saturada proveniente de manteiga de 0,1 a 0,8 g/dia (HR=0,69, IC95%: 0,49-0,96, vs. nenhum consumo), proveniente de laticínios integrais ≥10,5 g/dia (HR=0,63; IC95%: 0,53-0,92, vs. ≤2,0g/dia) e proveniente de peixes acima de 1,2 g/dia (HR= 0,63; IC95%: 0,43-0,92, vs. ≤0,2 g/dia) e, para ingestão de alimentos, de 0,1 a 1,6 g/dia de manteiga (HR=0,69; IC95%: 0,49-0,96, vs. nenhum consumo) e de peixes acima de 56,8 g/dia (HR=0,63; IC95%: 0,44-0,91, vs. ≤16,8 g/dia). O consumo 0,5 a 1,3 g/dia de gordura saturada de carnes processadas apresentoumaior risco (HR = 1,46, IC95%: 1,06-2,01, vs. ≤0,4 g/dia). A ingestão de ácidos graxos saturados de cadeia média e curta, presentes em maior quantidade em lácteos e peixes, acima de 4,1 g/dia, mostrou-se protetora (HR=0,70; IC95%: 0,45-0,96), quando ajustada para fatores sociodemográficos, de estilo de vida e de consumo de alguns alimentos. Gordura saturadas totais, ácidos graxos saturados de cadeia longa, demais alimentos e fontes de gordura saturada não apresentaram associação significativa com incidência de DM2. Conclusão: As associações entre gordura saturada e incidência de DM2 dependem das diferentes fontes e dos ácidos graxos saturados que as compõem. / Objective: In these masters dissertation, we investigated the association between the intake of Saturated Fat (SF), its food sources, saturated fatty acids (SFA) of different chain sizes and the incidence of diabetes mellitus type 2 (DM2). Method: The Longitudinal Study of Adult Health (ELSA-Brazil) is a multicenter cohort whose objective is to investigate risk factors and progression of diabetes, cardiovascular diseases and other related chronic diseases. A total of 15,105 active and retired employees from public institutions of higher education and research, located in six capitals (Salvador, Belo Horizonte, Rio de Janeiro, São Paulo, Vitória and Porto Alegre) were recruited between August 2008 and December 2010 to perform laboratory and clinical exams and to answer a series of questionnaires and interviews. Each year, all participants are contacted by phone for follow-up interview, being questioned about new medical diagnoses, including diabetes. In addition, between the years 2011 and 2015, the study participants were invited to return to the research centers for further clinical and laboratory tests. All participants signed a free and informed consent form, agreeing to participate in the study, and the research protocol was approved by the ethics committee of all the institutions involved. For the present study, data of participants between 35 and 74 years of age, who reported not having diabetes and other chronic diseases that could influence food consumption, who answered a food frequency questionnaire(FFQ) with 114 items and the other questionnaires with confounding variables in the study at baseline (2008-2010) and did not present daily intakes of total daily energy value (<600 or> 6,000 Kcal / day) were analized, resulting in a final sample of 8,187 participants. For the analysis of food consumption, some of the foods considered GS sources were grouped: whole dairy products, skimmed milk products, fermented dairy products, processed meats, unprocessed meats, red meat and fish. The different AGS were organized according to the size of their chain: short and medium chain (≤ 12 carbons) and long chain (> 13 carbons). The Nutrition Data System for Research (NDSR) software was used to estimate the nutritional composition of the foods included in the F. Anthropometric measures (weight and height) were obtained through internationally standardized protocols. The incidence of DM2 was self-reported in follow-up annual interviews and through the results of blood tests (fasting glycemia, glycated hemoglobin and / or glucose 2h after 75g glucose tolerance test) performed on a new visit to the research centers (2013-2015). In the descriptive analysis, the characteristics of the participants according to categories of GS consumption are compared by proportions through the chi-square test, means by ANOVA and medians by Wilcoxon test. Cox proportional hazards models were used to estimate risk ratios and 95% CI in the relationship between intake of different saturated fatty acids, total saturated fat and their source foods with the incidence of diabetes. To minimize the influence of potential confounders, three multivariate models were used. Statistical analyzes were conducted with SAS software version 9.4. Results: Were identified 300 new cases of T2DM. After adjustment for sociodemographic confounders, lifestyle and diet, It wasfound a protective effect ofT2DM for the intake of saturated fat from butter 0,1 to 0,8 g/day (HR = 0.69, 95% CI: 0.49 to 0.96, vs. No consumption), from dairy integrals ≥10,5 g/day (HR = 0.63; 95% CI: .53-.92, vs. ≤2,0 g/day) andfrom fish intake above 1.2 g/day (HR = 0.63, 95% CI: 0.43 to 0.92, vs. ≤0,2 g/day), and for food intake of 0.1 to 1.6 g/day butter (HR = 0.69, 95% CI: 0.49 to 0.96, vs. No consumption) andfor fish above 56.8 g/day (RH = 0.63; 95% CI: 0.44-0.91, vs. ≤16,8 g/day). Consumption of0.5 to 1.3 g/processed meats saturated fat showed increased risk (HR: 1.46, 95% CI: 1.06 to 2.01, vs. ≤0,4 g/day) . The intake of Medium and Short chain saturated fatty acids, presented in greater quantities in dairy and fish, above 4.1 g/day, seems to be protective (HR = 0.70; 95% CI: 0.45 to 0, 96) When adjusted for sociodemographic, lifestyle factors and consumption of some foods. Total saturated fat, long chain saturated fatty acids, other foods and saturated fat sources were not significantly associated with incidence of type 2 diabetes. Conclusion: Associations between saturated fat and DM2 incidence depend on its different sources and saturated fatty acids.