Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-05-04T18:03:22Z No. of bitstreams: 1 juliatristaodocarmorocha.pdf: 2765026 bytes, checksum: 7a9e6fcc24e70e565d8b382aeff776df (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-05-17T13:33:05Z (GMT) No. of bitstreams: 1 juliatristaodocarmorocha.pdf: 2765026 bytes, checksum: 7a9e6fcc24e70e565d8b382aeff776df (MD5) / Made available in DSpace on 2017-05-17T13:33:05Z (GMT). No. of bitstreams: 1 juliatristaodocarmorocha.pdf: 2765026 bytes, checksum: 7a9e6fcc24e70e565d8b382aeff776df (MD5) Previous issue date: 2016-05-25 / Os produtos petrolíferos em geral são altamente complexos e é exigido um esforço considerável para a caracterização de suas propriedades químicas e físicas. Às vezes tem-se urgência no resultado de determinadas análises e isto fica prejudicado pela forma como as análises são feitas. Assim, a quimiometria, associada à espectroscopia molecular (NIR e MIR em particular) vem gerando métodos alternativos para a caracterização e avaliação de propriedades físicas e químicas de petróleos e seus derivados com elevada exatidão, confiabilidade e rapidez. Para melhorar o desempenho previsor têm sido utilizados procedimentos apropriados para a seleção das regiões espectrais associadas com a propriedade de interesse. Desta forma, face às suas aplicabilidades, foi proposto neste trabalho a utilização das ferramentas quimiométricas com seleção de variáveis (método dos mínimos quadrados parciais por intervalos, iPLS, e por sinergismo de intervalos, siPLS; método de eliminação de variáveis não informativas por mínimos quadrados parciais, UVE; e algoritmo genético, GA), associada ao MIR e ao NIR, para a determinação das seguintes propriedades em frações de petróleo: Grau API, Índice de cetano, Índice de refração (a 20°C), Teor de Enxofre (%m/m), Ponto de fuligem (mm), Ponto de anilina (°C), Ponto de congelamento (°C), Ponto de entupimento (°C), Ponto de névoa (°C) e Ponto de fluidez (°C), avaliando, assim, a performance dos modelos obtidos, bem como as técnicas utilizadas na seleção de variáveis. Essa avaliação se deu pela determinação e análise do coeficiente de determinação (R2), de diversos erros calculados para os conjuntos de calibração e previsão. Os modelos foram, ainda, submetidos a testes estatísticos (α=0,05), e tiveram suas figuras de mérito calculadas. Os melhores modelos para a previsão do Grau API e do ponto de névoa foram criados aplicando-se iPLS a dados de MIR, enquanto que para a previsão do teor de enxofre e pontos de refração, de fuligem e de anilina foram criados aplicando-se siPLS também ao MIR. Já para a previsão do índice de cetano e do teor de enxofre e do ponto de entupimento, os melhores modelos foram criados aplicando-se iPLS a dados de NIR. Nesse contexto, o melhor modelo para a predição do ponto de fluidez foi o GA. Finalmente, para a previsão do ponto de congelamento, nenhum método de seleção de variáveis melhorou a capacidade preditiva, quando comparados ao modelo criado aplicando-se PLS a dados de MIR. Dessa forma, conclui-se que houve um melhor desempenho dos modelos criados a partir de dados de MIR. Quanto aos métodos de seleção de variáveis, iPLS e siPLS obtiveram o melhor desempenho. / Petroleum products are, in general, highly complex and a considerable effort is needed to characterize their chemical and physical properties, though sometimes the results of several analyses are urgent and this is compromised by the way the analyses are carried out. Thus, chemometrics associated with molecular spectroscopy (particularly NIR and MIR) has good potential as a tool in analytical chemistry, creating alternative methods to characterize and evaluate physical and chemical properties of petroleum and its derivates with high precision, reliability and rapidity. To improve the predictor performance, appropriate procedures are being used to select spectral regions associated with the property of interest. In face of their applicabilities, this work proposes the use of chemometric tools, with variable selection (Interval Partial Least Square, iPLS and Sinergism Interval Partial Least Square, siPLS; Elimination of Uninformative Variables, UVE and Genetic Algorithm, GA), associated with mid infrared (MIR) and near infrared (NIR) spectroscopies to determine the following properties in petroleum fractions: API gravity, Cetane index, Refractive index (at 20°C), Sulfur content (%m/m), Smoke point (mm), Aniline point (°C), Freezing point (°C), Plugging point (°C), Cloud point (°C) and Pour point (°C), enabling the evaluation of performance of the obtained models, as well as the techniques used in variable selection. This evaluation was performed by determination and analyses of the following requirements: coefficient of determination (R2), several calculated errors for the calibration and prediction set. The models were also subjected to statistical tests (α=0,05), and the figures of merit were calculated. The best models to predict API gravity and cloud point were created by applying iPLS to the MIR data, whereas for prediction of sulfur content, refractive index, and smoke and aniline points the models were created by applying iPLS to NIR data. In this context, the best model to predict the pour point was the GA. Finally, to predict freezing point, none of the variable selection methods improved the predictive capability when comparing to the model created using only PLS in MIR data. Thus, the conclusion is that a better performance was obtained for the models created from MIR data. Regarding efficiency of variable selection methods, the iPLS and siPLS methods resulted in a best performance.

Orientador: Ronei Jesus Poppi / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Química / Made available in DSpace on 2018-08-24T05:27:34Z (GMT). No. of bitstreams: 1 Breitkreitz_MarciaCristina_D.pdf: 4580305 bytes, checksum: 77a15e51b9d1be694cc7c5db18fa63bb (MD5) Previous issue date: 2013 / Resumo: Este trabalho teve como objetivo demonstrar como a espectroscopia NIR e Raman de imagem, juntamente com diversos métodos quimiométricos podem auxiliar a tomada de decisões durante o desenvolvimento farmacêutico de formulações de fármacos pouco solúveis em água, tendo como modelo a atorvastatina cálcica. Foram avaliados dois excipientes inovadores, o Gelucire® 44/14 e o Soluplus® que serviram de base para o desenvolvimento das formulações. O Gelucire® 44/14 foi utilizado para o preparo de dispersões sólidas pelos métodos da fusão e da evaporação do solvente e formulações do tipo SEDDS (Self-Emulsifying drug Delivery Sytems), enquanto o Soluplus® foi utilizado para o preparo de dispersões sólidas pelo método da evaporação do solvente. Nas dispersões preparadas pelo método da fusão foram observados aglomerados do fármaco, enquanto o método de evaporação do solvente apresentou melhor homogeneidade na distribuição dos componentes, porém a amostra se apresentou enrijecida após a secagem. Com as formulacoes SEDDS foi possível contornar estes problemas. Foram obtidas dispersões sólidas homogêneas de atorvastatina em Soluplus® utilizando etanol como solvente e lactose como diluente. As imagens foram geradas por calibração univariada, análise de componentes principais (PCA), regressão em mínimos quadrados clássicos (CLS) e regressão em mínimos quadrados parciais (PLS) e os resultados foram comparados. O método CLS foi estudado com mais detalhes devido às suas vantagens para utilização na pesquisa farmacêutica. Os principais fatores que levaram à problemas de exatidão com este método foram identificados e procedimentos para contorná-los foram apresentados e discutidos. / Abstract: The aim of this work was to demonstrate how NIR and Raman image spectroscopy associated with different chemometric methods can support decision making during pharmaceutical development of formulations of low water soluble drugs, using atorvastatin calcium as a model drug. Two innovative excipientes were used, Gelucire® 44/14 and Soluplus®. The former was used to prepare solid dispersions by both hot melt and solvent evaporation methods and to formulate Self Emulsifying Drug Delivery Systems (SEDDS). The latter was used to prepare solid dispersions by the solvent evaporation method. In solid dispersions prepared by the hot melt method it was observed lumps of the drug, whereas the solvent evaporation method presented more homogeneous distribution of the components, even though the sample became stiffened after drying. By preparing SEDDS formulations, it was possible to overcome these problems. A homogeneous solid dispersion of atorvastatin in Soluplus® was achieved by using ethanol as solvent and lactose as diluent. Chemical images were generated by univariate calibration, principal component analysis (PCA), classical least squares (CLS) and partial least squares (PLS). The capability of these methods to generate chemical images were compared. CLS method was studied more carefully due to its advantages for pharmaceutical research use. The main features that lead to accuracy issues were identified both in Raman and in NIR spectroscopy and procedures to overcome them were presented and discussed. / Doutorado / Quimica Analitica / Doutora em Ciências

Orientador: Adriana Vitorino Rossi / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Química / Made available in DSpace on 2018-08-20T02:46:28Z (GMT). No. of bitstreams: 1 Favaro_MarthaMariaAndreotti_D.pdf: 3734314 bytes, checksum: 08002efe51b2f18e9a942c3b818270b7 (MD5) Previous issue date: 2012 / Resumo: Antocianinas (ACYS) são corantes naturais responsáveis pela coloração de frutas, hortaliças, flores e grãos. Novas perspectivas de usos de antocianinas em diversos segmentos industriais estimulam estudos analíticos para sistematizar a identificação e a classificação de fontes e extratos desses corantes. Neste trabalho foram utilizadas fontes de ACYS como frutas típicas brasileiras: AMORA (Morus nigra), amora preta (Rubus sp.), jabuticaba (Myrciaria cauliflora), jambolão (Syzygium cumini), jussara (Euterpe edulis Mart.), morango (Fragaria x ananassa Duch) e uva (Vitis vinífera e Vitis vinífera L. Brasil); hortaliças: alface roxa (Lactuca sativa), berinjela (Solanum melongena), cebola roxa (Allium cepa), rabanete (Raphanus sativus), repolho roxo (Brassica oleraceae) e flores: beijo-turco (Impatiens walleriana), gerânio (Pelargonium hortorum e Pelargonium peltatum L.), hibisco (Hibiscus sinensis e Hibiscus syriacus) e hortênsia (Hydrangea macrophylla). A literatura descreve diversas técnicas para análise de ACYS em vegetais e seus extratos, com destaque para cromatografia líquida de alta eficiência (HPLC), espectrometria de massas (MS) e espectrofotometria (UV-Vis), sendo que todas elas foram aplicadas neste trabalho, incluindo-se espectrofotometria de reflectância e a técnica de eletromigração em capilares cromatografia eletrocinética micelar (MEKC). As ferramentas quimiométricas utilizadas no tratamento dos dados foram análise de componentes principais (PCA) e método do vizinho mais próximo (KNN). Os modelos quimiométricos de classificação obtidos apresentaram-se robustos com erros de previsão de menos de 30 % sendo possível identificar as fontes de ACYS, o solvente extrator, a idade dos extratos e dados sobre sua estabilidade e condições de armazenamento. Os resultados apontaram que dados obtidos de técnicas analíticas simples como espectrofotometria de absorção e sem necessidade de preparo de amostra como reflectância difusa na região do visível são comparáveis a resultados de técnicas mais sofisticadas e caras como HPLC e MEKC e até superam o potencial de algumas informações obtidas por MS / Abstract: Anthocyanins (ACYS) are natural dyes responsible for color in fruits, vegetables, flowers and grains. New perspectives for use of anthocyanins in various industries stimulate analytical studies to systematize the identification and classification of sources and extracts of these dyes. In this work, typical Brazilian fruits: mulberry (Morus nigra), blackberry (Rubus sp), jaboticaba (Myrciaria cauliflora), jambolan (Syzygium cumini), jussara fruit (Euterpe edulis Mart.), strawberry (Fragaria x ananassa Duch) and grapes (Vitis vinifera and Vitis vinifera L. 'Brazil'); vegetables: red lettuce (Lactuca sativa), eggplant (Solanum melongena), purple onion (Allium cepa), radish (Raphanus sativus), red cabbage (Brassica oleracea) and flowers, Buzy Lizzie (Impatiens walleriana), geranium (Pelargonium hortorum and Pelargonium peltatum L.), hibiscus (Hibiscus sinensis and Hibiscus syriacus) and hydrangea (Hydrangea macrophylla) were used as sources of ACYS. The literature describes several techniques for analyzing ACYS in vegetables and their extracts, with emphasis on high performance liquid chromatography (HPLC), mass spectrometry (MS) and spectrophotometry (UV-VIS). All of these techniques were applied in this work, including reflectance spectrophotometry and micellar electrokinetic chromatography (MEKC) which is one of the capillary electromigration techniques. The chemometric tools used in data handling were the principal component analysis (PCA) and the K-nearest neighbor method (KNN). The chemometric classification models obtained are robust with predict errors of less than 30 %. It is possible to identify the sources of ACYS, the extractor solvent, the age of the extracts, their stability and storage conditions. The results show that data obtained from simple analytical techniques such as absorption spectroscopy and diffuse reflectance in the visible region (sample preparation is not needed) are comparable to results of those obtained from sophisticated and expensive techniques such as HPLC and MEKC. These techniques also surpass the information obtained by MS / Doutorado / Quimica Analitica / Doutor em Ciências

Les systèmes plastiques à usage unique sont destinés aux industries biopharmaceutiques et biotechnologiques. Ils sont composés de polymères et sont appréciés pour leurs propriétés barrières au gaz et à la vapeur d’eau assurant une préservation des solutions contenues dans ces systèmes. Ces dispositifs présentent de nombreux avantages pour les industriels : ils permettent d’éviter un nettoyage des cuves en inox et ainsi de limiter des risques de contamination croisée. En vue de leur utilisation, ces dispositifs sont stérilisés par irradiation γ. Le but de cette étude est de mettre en évidence l'impact de la stérilisation γ sur certains polymères et sur certains antioxydants utilisés dans la fabrication des polymères. Plusieurs échantillons sont analysés avant et après la stérilisation en utilisant différentes doses de rayonnement γ et au cours du temps (vieillissements naturels). Les modifications de structure des polymères γ-irradiés (EVA, PE et EVOH) ont été caractérisées par diverses méthodes analytiques : IRTF, spectroscopie Raman, RPE, chromatographie… Les résultats sont traités en utilisant des méthodes chimiométriques, comme l’ACP (Analyse en Composantes Principales), l’AComDim (ANOVA Common Dimensions) et SIMPLISMA (Simple-to-use interactive self-modeling mixture analysis). L’ACP a permis de mettre en évidence l’impact de la dose d’irradiation sur le polymère. L’AComDim a conduit à identifier les facteurs responsables de l’évolution des structures (la dose γ, les temps et l’interaction dose γ × temps), alors que les différents lots n’induisent pas de déviation. Enfin, la méthode de résolution de courbe SIMPLISMA a été appliquée pour l’étude du cœur du matériau. / The single use plastic systems are designed for biopharmaceutical and biotechnology industries. They are principally made of polymers and they are very appreciated for their barrier properties to gas and water vapor ensuring preservation of the solutions contained in these single-use systems. These devices have many advantages for industrial: they avoid cleaning stainless steel tanks and thus limit the risk of cross-contamination. In view of their future use, these devices are sterilized by γ-irradiation. The purpose of this study is to highlight the impact of γ sterilization on some polymers and on some antioxidants used in the manufacture of polymers. The samples are analyzed before and after sterilization using different γ-doses and over time (natural ageing). The structural modifications of γ-irradiated polymers (EVA, EVOH and PE) were characterized by various analytical methods: FTIR, Raman spectroscopy, ESR, chromatography… Some results are analyzed using chemometric methods such as PCA (Principal Component Analysis), AComDim (ANOVA Common Dimensions), and SIMPLISMA (Simple-to-use interactive self-modeling mixture analysis). The PCA allowed to highlight the impact of the irradiation dose on the polymer. The AComDim method led to the identification of the factors contributing to the evolution of structures (γ-dose, ageing and γ × ageing interaction), whereas the several lots do not involve variation. Finally, the curve resolution method, SIMPLISMA, was applied to the study of the core of the material.

Orientadores: Rubens Maciel Filho, Maria Izabel Maretti Silveira Bueno / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Química / Made available in DSpace on 2018-08-18T14:39:31Z (GMT). No. of bitstreams: 1 Henriques_ClaudeteBernardo_D.pdf: 9506603 bytes, checksum: 875e508c591361025ae37d3b9ac0f432 (MD5) Previous issue date: 2011 / Resumo: Esta tese tem como objetivo apresentar um estudo detalhado para a caracterização de petróleos, tanto no âmbito do desenvolvimento de modelos capazes de prever propriedades físico-químicas clássicas, como também avaliar os fenômenos associados a instabilidades decorrentes de características químicas. Para alcançar este objetivo, duas ações foram executadas: (1) Obter a curva de destilação do ponto de ebulição verdadeiro (PEV) através de simulação, por espectrometria de espalhamento de raios-X (EERX) aliada ao algoritmo denominado máquina vetor de suporte (SVM), com o propósito de prever o rendimento do processo de refino, e (2) Caracterizar o petróleo em relação ao fenômeno de agregação dos asfaltenos, para assim possibilitar a previsão da qualidade do processo de refino. Foram avaliadas amostras de petróleos com propriedades distintas, com o intuito de fornecer amplitude e robustez ao estudo. A primeira abordagem foi à análise classificatória dos petróleos em relação às suas propriedades intrínsecas (teor de enxofre, Grau API e ponto de ebulição verdadeiro), com aplicação de métodos multivariados de classificação. Com estas classificações em mãos, foram desenvolvidos modelos para previsão da curva PEV e do Grau API, utilizando o algoritmo SVM (segunda abordagem). Os dados tratados por meio deste algoritmo apresentaram um perfil de não linearidade. Os resultados obtidos para a caracterização das cargas a partir dos modelos não foram significativamente diferentes dos valores experimentais, permitindo o desenvolvimento de estudos computacionais usando o software comercial Petro-SIM®, para caracterização, aquisição das curvas das misturas de petróleos e, finalmente, previsão dos rendimentos dos cortes na simulação de uma torre de destilação atmosférica. A partir destes resultados, é muito importante ressaltar que, ao contrário da maioria das modelagens quimiométricas, um pequeno grupo de amostras foi necessário para o desenvolvimento do modelo (13 amostras). Vale também serem mencionadas a alta capacidade de previsão do modelo, a rápida e eficiente previsão de curvas PEV de amostras de petróleos puros, processados nas refinarias e, finalmente, a aplicação com sucesso destes modelos em simuladores de processo. A terceira abordagem do trabalho foi à avaliação dos aspectos relacionados à existência de formas coloidais dos asfaltenos, peptizados por resinas e dos efeitos relacionados à presença de outros solventes nos petróleos, que podem causar mudanças na cinética de floculação dos asfaltenos. A composição dos solventes (petróleos diluentes) e dos petróleos pesados foi correlacionada com os resultados do estudo cinético, obtido por meio da avaliação da transmitância por varredura óptica da camada de precipitação dos particulados e da floculação dos asfaltenos, como também por modelos dinâmicos. Para isto, foram elaboradas misturas de petróleos e observou-se que os efeitos associados ao fenômeno de agregação estão diretamente correlacionados com a composição dos petróleos, assim como com a presença de particulados nas misturas. A caracterização adequada dos constituintes de petróleo é uma informação indispensável para garantir o processamento adequado nas refinarias. A reprodução do cenário real do processo, por meio da simulação, foram promissores, mostrando que a caracterização da carga é possível ao se trabalhar com os modelos desenvolvidos, reforçando a relevância das técnicas aplicadas para descrição dos fenômenos químicos associados ao cenário de produção nacional / Abstract: This work aims to present a detailed study for petroleum characterization, through model developments able to preview physical chemical properties and also to evaluate the associate phenomena due to instability in current chemical characteristics. For this purpose two actions were executed: (1) To obtain the curve true boiling point (TPB) through the simulation by spectrometry X-ray scattering (EERX) combined with the algorithm called support vector machine (SVM), in order to predict the yield of the refining process, and (2) to characterize the petroleum in respect to asphaltene aggregation phenomena, making possible the prediction of the quality of the refining process. For these purposes, the petroleum samples were evaluated with different properties, providing amplitute and robustness to the study. The first approach was the classificatory analisis of petroleum in relation to intrinsic properties (sulfur contents, API degrees and true boiling point curves) with application of classification multivariate methods. With these classifications in hands, models were developed to predict the TBP curve and API gravity, using the algorithm SVM (second approach). The data were processed by this algorithm, showing nonlinear profiles. The results for the blend characterization using the models were not significantly different from the experimental values, allowing the development of computational studies using the commercial software Petro-SIM® for characterization, acquisition of mixtures curves and finally, prediction of the cuts yield in the simulation of an atmospheric distillation tower. From these results, it is important to emphasize that a small group of samples was necessary for the model development (13 samples). It is also worth to mention the high predictive ability of the model, the rapid and efficient prediction of TBP curves of crude oil samples, processed in the refineries and finally, the successful application of these models in process simulators. The third approach of this work was to evaluate the aspects related to the existence of colloidal forms of asphaltenes, peptized by resins and of effects related to the presence of other solvents in oils, which can cause changes in the kinetics of flocculation of asphaltenes. The composition of solvents (thinner oil) and of heavy crude oils were correlated with the results of kinetic study, obtained by evaluating the transmittance for optical scanning of the layer of precipitation of particles and the flocculation of asphaltenes, but also by dynamic models. For this, petroleum mixtures were prepared and it was observed that the effects associated to aggregation phenomena are correlated with the petroleum composition as well as the particulate presence in mixtures. The adequate characterization of petroleum constituents is essential information to ensure proper processing in refineries. The reproduction of real scenario of the process, through the simulation, were promising, showing that the blend characterization is possible when working with these developed models, reinforcing the relevance of the techniques applied for describing the chemical phenomena associated with the setting of national production / Doutorado / Desenvolvimento de Processos Químicos / Doutor em Engenharia Química

This disertation deals with geographical authentification of different types of spices and spice preparations. Investigated spice species were garlic (Allium sativum), ground pepper (Capsicum anuum), and dried carrot (Daucus carota). Theoretical part of the disertation describes main qualitative parameters of the examined species and production technology, means of food adulteration and statistic methods used in chemometrics. The aim of this dissertation was to verify the hypothesis whether it is possible to use targeted analytical techniques commonly used in quality control of spices and condiments, and statistical processing of measured data to distinguish samples of spices and condiments of different geographical origin. The use of non-targeted analysis was also investigated. Samples of garlic and ground pepper were used for targeted analysis. The examined parameters were dry matter and moisture of the sample, ash content, total phenolic content according to Folin-Ciocaulteu, carbohydrate content, alliin concentration, total nitrogen content, total color pigment (ASTA), pH of aqueous extract, total fat content, and concentration of selected elements (Ca, Cu, Fe, K, Mn, Mg, Na, P, Zn). The instrumental techniques used were molecular absorption spectrometry, inductively coupled plasma optical emission spectrometry and high performance liquid chromatography. The obtained data were statistically processed by analysis of variance (ANOVA) and principal component analysis (PCA). Using statistical analysis significant differences between samples that came from more distant areas were found. However, samples from closer areas could not be distinguished. The researched hypothesis could not be unequivocally confirmed or refuted. Metabolic fingerprint of carrot samples was determined using non-targeted analysis. Metabolic analysis was performed using the tandem LC-TOF-MS technique. The data were processed by recursive peak extraction (BRE) and subsequently uvaluated with PCA. The samples were divided into clusters according to their origin. Targeted and non-targeted techniques have great potential in verifying the geographical authenticity of different types of spices. However, the main condition is consistent and sufficient sampling, guaranteed information on the origin of the sample and obtaining a sufficient amount of input data for statistical analysis.

Les cellules photovoltaïques (PV) des modules ou panneaux solaires, sont protégées des agressions extérieures par des résines polymères encapsulantes qui, pour la plupart, sont des matériaux élastomériques réticulés. L’optimisation et le contrôle de l’étape d’encapsulation peut permettre un gain de productivité et augmenter la durée de vie des modules, ce qui réduit les coûts de l’électricité générée. Deux voies ont ainsi été explorées dans ce travail : 1) La première concerne l’étude de l’état de réticulation de l’encapsulant majoritairement utilisé actuellement, un copolymère d’éthylène et d’acétate de vinyle connu sous le nom d’EVA ; celui-ci est mis en œuvre sous forme de feuilles ou films. Une faible réticulation de l’encapsulant peut mener, entre autre, à son fluage lors de son utilisation, impactant directement la durée de vie du module. Il est donc important de suivre le niveau de réticulation de l’encapsulant lors des opérations de contrôle. La comparaison de différentes méthodes d’évaluation du degré de réticulation ont donc été menées ; 2) La deuxième voie concerne des études autour d’un nouveau procédé d’encapsulation. En effet, le procédé industriel actuel, inclut une étape dite de « lamination », pendant laquelle l’encapsulant est fondu et réticulé autour des cellules. Cette étape relativement longue crée des contraintes thermomécaniques pouvant limiter la durée de vie des modules PV. Le développement d’un nouveau procédé d’encapsulation où l’encapsulant en film est remplacé par un encapsulant liquide photopolymérisable permettrait de diminuer les coûts de production tout en augmentant potentiellement la durée de vie des modules. Les propriétés rhéologiques et la cinétique de polymérisation de ce nouvel encapsulant sont ainsi étudiées. Enfin les deux voies d’encapsulation sont comparées. Il a été montré que d’un point de vue des performances électriques le nouveau procédé présente un avantage potentiel et que d’un point de vue de la tenue au vieillissement il est équivalent au procédé industriel actuel / Photovoltaic (PV) cells, for solar modules or panels, are protected from environmental stresses by polymeric encapsulants, which are mostly crosslinked elastomers. The optimization and the control of the encapsulation step have a twofold interest by increasing PV module lifetime and productivity, thus leading to a decrease of the cost of generated electricity. Two main directions have been investigated in this work: 1) The first one is related to the study of the crosslinking degree of the main industrial PV polyolefin encapsulant, EVA, which is a copolymer composed of ethylene and vinyl acetate, used currently in film form. Indeed, poor crosslinking level can lead to its creep, impacting directly the module lifetime. To overcome this problem, the quality control needs to be improved, by the evaluation of the crosslinking degree obtained while using the conventional encapsulation process (through lamination of encapsulant foils). Thus, the comparison of several methods to evaluate this degree are led ;2) The second direction concerns the study of a new encapsulation process. Indeed, the conventional lamination process potentially creates mechanical stresses in the PV cells, which as a consequence may limit the PV module lifetime. Moreover, lamination requires a relatively long processing time. To overcome this problem, the development of a new encapsulation process using a photopolymerizable encapsulant, initially liquid, decreases the production costs of PV modules and potentially increases their lifetime. The rheology properties and the polymerization kinetics of the new encapsulant are studied. At the end of the present work, both encapsulation processes are compared. Electrical performances of PV cells are measured before and after encapsulation as well as before and after ageing cycles. It has been revealed that the new encapsulation process presents at least as good, if not better performances than the standard process, thus highlighting its big potential for the manufacturing of PV modules

L’identification des auteurs d’incendies criminels où un accélérant a été utilisé demeure à ce jour un domaine de recherche en développement. Les traces biologiques reines pour l’identification de personnes comme l’ADN et les traces papillaires sont généralement détruites, donc rarement recherchées. Pourtant, lier l’auteur des faits au lieu de l’incendie est un réel besoin. Ce lien pourrait être établi par l’inférence de source des traces d’accélérant détectées sur les lieux avec une source potentielle souvent amenée par l’enquête comme des objets saisis en possession du suspect (vêtements), un jerrican ou encore des prélèvements effectués sur ses mains. Dès lors, la question qui se pose consiste à déterminer si les traces d’accélérant détectées sur les lieux et les traces détectées sur l’élément de comparaison partagent une source commune. Ainsi, l’inférence de source de l’accélérant constitue une alternative au manque de traces matérielles de sorte à fournir des éléments de preuve à la suite d’un incendie volontaire. En ce sens, cette recherche propose une approche chimiométrique non ciblée pour l’inférence de source de liquides inflammables en science forensique. Cette approche a été appliquée à un échantillonnage conséquent d’essences non altérées et à un échantillonnage réduit d’essences altérées de 0 à 99% par évaporation et par combustion. L’évaluation des résultats a validé l’hypothèse selon laquelle il est possible de lier des échantillons d’essences altérés ou non, par évaporation ou combustion, indépendamment du mode et du degré d’altération. / The identification of arsonists when an accelerant was used is still a challenging and ongoing research area. Golden standards in forensic human identification such as DNA and fingermarks are usually destroyed during the fire, hence not often looked for. It is yet obvious that the need to link the perpetrator to the arson site exists. This link could be made through a source inference process of the traces of an accelerant detected on site. These traces could be compared with a potential source often brought by the police investigation such as seized items in possession of a suspect (clothes), a jerrican or even hand sampling. Thenceforward, the question arising would be to determine if the traces of an accelerant from unknown source share a common source with the seized item. Thus, the source inference of accelerants constitutes an alternative to the lack of material traces in order to provide evidence in arson cases. To tackle this question, the present research proposes an untargeted chemometric approach for the source inference of ignitable liquids in forensic science. This approach was applied to a large dataset of unaltered gasoline samples and to a reduced one of altered samples by evaporation and combustion between 0 and 99%. The evaluation of results shows that it is possible to link gasoline samples altered or not by evaporation and combustion independently of the alteration mode and degree..

Indiana University-Purdue University Indianapolis (IUPUI) / Solid phase micro-extraction is a powerful and versatile technique, well-suited to the analysis of numerous samples of forensic interest. The exceptional sensitivity of the SPME platform, combined with its adaptability to traditional GC-MS systems and its ability to extract samples with minimal work-up, make it appropriate to applications in forensic laboratories. In a series of research projects, solid phase micro-extraction was employed for the analysis of explosives, commercial tire treatments, and entomological specimens. In the first project, the volatile organic compounds emanating from two brands of pseudo-explosive training aids for use in detector dog imprinting were determined by SPME-GC-MS, and the efficacy of these training materials was tested in live canine trials. In the second project, the headspace above various plasticizers was analyzed comparative to that of Composition C-4 in order to draw conclusions about the odor compound, 2- ethyl-1-hexnaol, with an eye toward the design of future training aids. In the third, automobile tires which had participated in professional race events were analyzed for the presence of illicit tire treatments, and in the fourth, a novel SPME-GC-MS method was developed for the analysis of blowfly (Diptera) liquid extracts. In the fifth and final project, the new method was put to the task of performing a chemotaxonomic analysis on pupa specimens, seeking to chemically characterize them according to their age, generation, and species.

No description available.