Ce travail de thèse présente une étude théorique, numérique et expérimentale de l’intégration sur un guide d’onde diélectrique de chaînes de nanoparticules d’or supportant des résonances « plasmon de surface localisé ». Les guides d’onde à plasmon de surface localisé procurent un confinement sub-longueur d’onde de la lumière, ce qui permet d’envisager la réalisation de composants optiques ultra-compacts. Cependant, leurs pertes optiques élevées restreignent leur application à de courtes distances de propagation, contrairement aux guides d’onde diélectriques. Une combinaison judicieuse des deux types de guide doit donc permettre de bénéficier de leurs avantages respectifs. Dans un premier temps, nous avons étudié théoriquement les propriétés des chaînes des nanoparticules grâce à un modèle analytique basé sur l’approximation de dipôles ponctuels couplés, que nous avons développé. Cette étude a permis de déterminer la forme et les dimensions des nanoparticules qui ont ensuite été introduites dans un logiciel de FDTD pour simuler le couplage entre la chaîne de nanoparticules et le guide diélectrique (SOI ou en Si3N4). De cette étude numérique, nous avons déduit les géométries des structures à fabriquer. Les structures réalisées ont été caractérisées à l’aide d’un banc de transmission résolue spectralement, mis en place pendant cette thèse, et d’un système de mesures en champ proche optique en collaboration avec le LNIO (Troyes). Pour la première fois, nous avons montré expérimentalement les propriétés d’une chaîne courte de nanoparticules intégrée sur un guide SOI, ainsi que le phénomène de guides couplés entre une chaîne longue de nanoparticules et un guide SOI. Une valeur record de la constante de couplage a été obtenue, et ce, aux longueurs d’onde des télécoms (proche infrarouge). L’énergie lumineuse transportée par le mode TE du guide SOI peut ainsi être entièrement transférée au guide plasmonique en 4 ou 5 nanoparticules, soit une distance de propagation de moins de 600 nm. Nous avons également étudié les propriétés de réseaux de Bragg à base de plasmon de surface localisé en confrontant les résultats de mesures de transmission résolue spectralement aux résultats théoriques d’un modèle analytique basé à la fois sur l’approximation de dipôle ponctuel en régime quasi-statique et la théorie des modes couplés. Ces travaux ouvrent la voie à des applications de pinces optiques, de capteurs ou de spaser, qui bénéficieront de l’intégration de nanoparticules métalliques dans les circuits photoniques. / This PhD work presents a theoretical, numerical and experimental study of the integration of a gold nanoparticle chain supporting "localized surface plasmon resonances" on a dielectric waveguide. The localized surface plasmon allows a sub-wavelength confinement of light which could lead to the achievement of ultra-compact optical components. However, the high level of optical losses restricts their application to short propagating distances unlike dielectric waveguides. A judicious combination of both types of guides should therefore allow taking profit of their respective advantages. Firstly, we have theoretically studied the properties of nanoparticles chains using an analytical model that we have developed following the coupled dipoles approximation. This has helped us to determine the shape and size of nanoparticles, which have been further used in a FDTD software, to simulate the coupling between the chain and the dielectric waveguide (SOI or Si3N4). Using this numerical study, we have deduced the geometries of structures to be fabricated. The realized structures have been characterized using a spectrally resolved transmission set-up, built during this thesis, and an optical near field measurement set-up (collaboration LNIO Troyes). For the first time, we have experimentally shown the properties of short nanoparticle chains integrated on a SOI waveguide as well as the existence of a coupled waveguide phenomenon between long nanoparticle chains and SOI waveguides. A record value has been obtained for the coupling constant at telecom wavelengths (near infrared). The light energy carried by the TE mode of the SOI waveguide can be completely transferred into the plasmonic waveguide via the first 4 or 5 nanoparticles of the chain, which means a distance of less than 600 nm. We have also studied the properties of Bragg gratings based on localized surface plasmon. Experimental results from spectrally resolved transmission measurements have been compared to theoretical results obtained from an analytical model based on the point dipole approximation in quasi-static regime, on one hand, and using the coupled mode theory, on the other hand. This work opens the way for applications to optical tweezers, sensors or spasers, which will benefit from the integration of metal nanoparticles in photonic circuits.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / CerÃmica Aurivillius sÃo conhecidas por sua excelente resistÃncia à fadiga e a sua alta temperatura de Curie, a qual apresenta cÃlula unitÃria ortorrÃmbica. A cerÃmica a ser estudada, SrBi2Nb2O9 (SBN), à matÃria-prima potencial na fabricaÃÃo de memÃrias ferroelÃtricas, com polarizaÃÃo espontÃnea e bons coeficientes piezelÃtricos. Dentro deste grupo cerÃmico, o SrBi2Nb2O9 (SBN) à o material com menor estrutura octaÃdrica distorcida. Apesar disto, o SBN sofre de perda dielÃtrica elevada, devido à evaporaÃÃo de Ãxido de bismuto durante a preparaÃÃo, vindo a limitar sua utilizaÃÃo. A ativaÃÃo mecÃnica usada foi a de moinho planetÃrio, que nos permite calcinar o composto por temperaturas mais baixas que as geralmente apresentadas na literatura. ApÃs a calcinaÃÃo, moemos por mais duas horas, e obtemos um pà de boa densidade e reduzida temperatura de sinterizaÃÃo. Este fato previne a perda de elementos volÃteis (Pb ou Bi) e controla o crescimento do grÃo que à produzido quando a alta temperatura à necessÃria para a obtenÃÃo dessa cerÃmica. No presente trabalho, esta cerÃmica Aurivillius serà estudada em suas propriedades estruturais e dielÃtricas com dopagem de Ãxido de lantÃnio, de bismuto e de chumbo, em diversas porcentagens. Adicionalmente, houve a necessidade de um estudo comparativo com respeito a diferentes tipos de ligantes, e os testes obedeceram Ãs dopagens mencionadas acima. Os resultados foram obtidos pelo mÃtodo Hakki & Coleman, pela Espectroscopia de ImpedÃncia e medidas de RadiofreqÃÃncia, para a caracterizaÃÃo elÃtrica. DifraÃÃo de Raios-X, juntamente com o Refinamento Rietveld, Espectroscopia Raman e de Infravermelho, para anÃlise estrutural, dentre outras medidas e mÃtodos. / Aurivillius ceramic are know by is excellent resistence to high use and itÂs high Curie temperature, in wich it presents unicell ortorrÃmbic. The ceramic to be study SrBi2Nb2O9 (SBN), is a potential prime-mater in the production of ironelectric memories with sponteneous polarization and good piezelectric coeficient. Within this ceramic group, SrBi2Nb2O9 (SBN) is the material with less twisted octaÃdric structure. Even so, SBN, have high loss of dielectric, due to the evaporation of bismut oxide during preparation, inposing a limit to is use. The mechanical activation used was of a planetarium will, that allows us to calcinate the compost by very low temperatures that are presented in literature. After the calcination, we will it for two more hours and have dust that have good density and low sinterization temperature. That fact privine us the lost of volatile elements ( Pb or Bi ) and controls the grow of the seed wich is produced when high Temperature is needed to obtain this ceramic . In the present work, this Aurivillius ceramic wil be studyed in is dielectric and structure properties with dopage of lantÃnium oxide, lead oxide and bismut oxide, on different percentages. In addiction there was necessety of a comparative study, regarding the different types of leagues and tests obay the dopage already menttionated. The results were obtained by the Hakki & Coleman method by the Impendence Espectroscopic and radiofrequense mesures, to the electric caracterization . X-rays Defraction with Rictveld refiament , Raman and infrared espectroscopic, to structural analises , within other mesures and methods .

Submitted by Rosivalda Pereira (mrs.pereira@ufma.br) on 2017-05-16T17:11:07Z No. of bitstreams: 1 ChristianCarvalho.pdf: 1874919 bytes, checksum: 3c4ecd041d9c80194ae2c47a229bbd6a (MD5) / Made available in DSpace on 2017-05-16T17:11:07Z (GMT). No. of bitstreams: 1 ChristianCarvalho.pdf: 1874919 bytes, checksum: 3c4ecd041d9c80194ae2c47a229bbd6a (MD5) Previous issue date: 2016-03-31 / Fundação de Amparo à Pesquisa e ao Desenvolvimento Científico e Tecnológico do Maranhão (FAPEMA) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPQ) / The use of biodiesel/diesel mixtures out of the quality standards specified by the National Agency of Petroleum and Natural Gas and Biofuels - ANP can be harmful to the environment and automotive engines. Therefore, it is important to develop methodologies capable of certifying the commercial quality of such products. In such context, this project aims to design a low-cost sensor to the quantification of the biodiesel content in biodiesel / diesel pseudo-binary blends based on the dielectric constant as a response. Thus, pseudo-binary samples were prepared using soybean or babassu methyl biodiesel with commercial diesel. The dielectric measurements were performed using a LCR meter operating in parallel circuit. In this way, a voltage of 1.0 V, frequency ranging from 100 Hz to 2000 Hz and a pitch of 100 Hz were employed. The dielectric profile of diesel and biodiesel showed remarkable variations in the frequency range of 100-300 Hz and less intense changes in under 400 and 2000 Hz, which confirms greater system stability in this measurement range. The results obtained by ANOVA of pure soybean biodiesel, pure babassu biodiesel and diesel oil showed that there was no significant difference between data treatment at the 95% significance level, once the Fcalculate was lower than the Fcritical. The frequency of 1000 Hz was adopted in biodiesel analysis of content in diesel, as it was verified that the system had a higher stability at this frequency. The correspondent values of dielectric constant demonstrate regularity and stability as well as a precision between the values found. Observing the values of the dielectric constants as a function of the percentage of biodiesel in the diesel, it was possible to plot the correlation curve for a linear model. Finally, according to the Kruskal-Wallis test reinforces the thesis that measuring dielectric constants (K) can be used to determine the concentration of biodiesel in mineral diesel. / A utilização de misturas biodiesel/diesel fora dos padrões especificados pela Agência Nacional de Petróleo e Gás Natural e Biocombustíveis – ANP podem ser prejudiciais ao meio-ambiente e causar prejuízos aos donos de veículos automotivos. Neste sentido, torna-se necessário o desenvolvimento de novas metodologias capazes de certificar a qualidade destes produtos no Brasil. O presente trabalho teve por objetivo desenvolver um sensor de baixo custo para determinar o teor de biodiesel em misturas pseudo-binárias biodiesel/diesel, utilizando constante dielétrica como resposta. Para tanto, foram preparadas misturas utilizando biodieseis metílicos de soja e/ou babaçu e óleo diesel A comercial. As medidas dielétricas foram realizadas em um medidor LCR configurado para funcionar com o circuito em paralelo. Desta forma, foram empregadas tensão de 1,0 V, frequência de 100 Hz - 2000 Hz e um passo de 100 Hz. O perfil dielétrico do óleo diesel e dos biodieseis demonstrou variações acentuadas no intervalo de frequência de 100 - 300 Hz e alterações de menor intensidade entre as frequências de 400 a 2000 Hz, confirmando maior estabilidade do sistema nessa faixa. Os resultados obtidos por ANOVA do biodiesel de soja puro, biodiesel de babaçu puro e o óleo diesel, demostraram que não há diferença significativa entre os tratamentos ao nível de significância de 95%, visto que o Fcalculado foi menor que o Fcrítico. A frequência de 1000 Hz foi adotada nas análises do teor de biodiesel no diesel, pois verificou-se maior estabilidade do sistema nessa frequência. Os valores de constante dielétrica em questão demonstram regularidade e estabilidade, além de uma precisão entre os valores encontrados. Observando os valores das constantes dielétricas em função da porcentagem de biodiesel no diesel, foi possível traçar a curva de correlação para um modelo linear. Por fim, o teste de Kruskal-Wallis demonstrou que a tese de que as medidas constantes dielétricas (𝐾) podem vir a ser utilizadas para determinar a concentração de biodiesel no diesel mineral.

Ce travail vise à étudier l'intérêt duchauffage micro-ondes pour l'inactivation desenzymes dans les jus de fruit, à travers desapproches numériques et aussi expérimentales.En premier lieu, une étude sur les propriétésdiélectriques des jus de fruits modèles est menée,démontrant leur forte sensibilité à la température,à la fréquence et à la composition du produit. Dansune seconde partie, l'inactivation de la peroxydaseest étudiée par chauffage conventionnel et lesdonnées sont ajustées par un modèle cinétique dupremier ordre. Dans la troisième et principalepartie de ce travail, un modèle tridimensionnel,résolu par éléments finis, est proposé pour simulerle chauffage par micro-ondes du jus, en couplantélectromagnétisme, transfert de chaleur etécoulement, avec la cinétique d'inactivation de laperoxydase précédemment déterminée.Cette simulation permet de prédire la distributionspatiale de la température, le profil d’écoulementet l'inactivation de la peroxydase. L’accord entre lemodèle et les expériences est très satisfaisant, cequi confirme la pertinence de l’approche. Dans ladernière partie, les réactivations de la peroxydaseaprès chauffage conventionnel et micro-ondessont évaluées et comparées. Enfin, l’éventuelleexistence d’effets non thermiques des microondesest discutée via des expériencesadditionnelles. En conclusion, ces travauxmontrent tout l’intérêt de la simulation numériquecomme outil de compréhension du processusmultiphysique du chauffage par micro-ondes pourl’inactivation des enzymes, ce qui peut êtreparticulièrement intéressant pour la conception etl’optimisation de traitements micro-ondes. / This work aims at studying the use ofmicrowave heating for enzyme inactivation in fruitjuices by means of numerical and experimentalapproaches. In the first part, a study on thedielectric properties of model fruit juices isconducted, evidencing their high dependence onthe temperature, frequency and composition of theproduct. Then in the second part, the inactivation ofperoxidase is studied using conventional heatingand the data are fitted by a first order kinetic model.In the third and main part of this work, a threedimensionalfinite element model is developed tosimulate the microwave heating of juices, couplingelectromagnetics, heat transfer and fluid flow aswell as the peroxidase inactivation kineticspreviously determined.As a result, spatial temperature distribution, flowpattern and peroxidase inactivation are obtained.The model is experimentally validated and goodagreement is observed, confirming the relevanceof the approach. Finally, in the last part, thepotential peroxidase reactivations afterconventional and microwave heating areassessed and compared. Also, the possibleexistence of non-thermal effects of microwaves isdiscussed thanks to additional experimentations.In conclusion, this work shows the large interest ofcomputer simulation as a tool for understandingthe multiphysics process of microwave heating forenzyme inactivation, which can be particularlyinteresting for further design of optimizedmicrowave processing.

The field of ambient desorption/ionization mass spectrometry (ADI-MS) has existed for over a decade. ADI-MS is a technique that offers benefits including fast analysis time, simple ionization sources that are easily constructed, and little to no required sample preparation. The research presented here describes efforts to better understand plasma-based ADI-MS sources and to explore the use of hydrogen-doped plasma gases with these sources. The use of hydrogen-doped argon (H2/Ar) and hydrogen-doped helium (H2/He) as plasma gases for a dielectric barrier discharge (DBD) and an AC glow discharge is presented first. When using the DBD, the intensity of the signal obtained when analyzing organic molecules in positive ion mode was increased by factors up to 37 times. In negative ion mode, only H2/Ar was shown to enhance the signal of an organic analyte. The limits of detection for caffeine when using hydrogen-doped plasma gases were found to decrease by factors of 78 and 1.9 for H2/Ar and H2/He respectively. The same phenomenon was observed when using H2/Ar with the AC discharge, but no signal enhancement was observed when using hydrogen-doped helium with the AC discharge. Similarly, if the DBD was allowed to ground through a wire rather than through the air, no signal enhancement was observed for H2/He. Using H2/Ar with metal samples is presented second. By using the metal sample as the grounded electrode for the AC glow discharge, many different metals could be detected directly with a time-of-flight mass spectrometer (TOF-MS) in the form of atomic ions both on their own and in combination with water and ammonia from the discharge. Any refractory metals tested did not yield signal. In addition to direct analysis with a TOF-MS, the AC discharge was used as a sampling method for an inductively coupled plasma mass spectrometer (ICP-MS). When coupled with an ICP-MS, the AC glow discharge was found capable of sampling even refractory elements, though the power of the ICP was required for ionization and detection. Scanning electron microscope (SEM) images of a copper surface exposed to the plasma discharge showed signs of melting when using the H2/Ar. Finally, a computer simulation of the chemistry and flow dynamics of a DC glow discharge generated in helium is presented. The simulation explores many of the fundamental processes at work and how they depend on the composition of the plasma gas. The generation of important species in the plasma was found to depend more on the amount of N2 and H2O impurities in the plasma gas rather than on the humidity or air pressure.

The propagation of light beams in inhomogeneous dielectric media is considered. The derivation begins with first principles and remains general enough to include off-axis asymmetric multimode input beams in tapered lenslike media with spatial variations of gain or loss. The tapering of lenslike media leads to a number of important applications. A parabolic taper is proposed as a model for a heated axially stretched fiber taper, and beams in such media are fully characterized. Other models are proposed by the concatenation of a parabola with other taper functions.

La quantité et la distribution de l'eau interstitielle dans l'espace poral des milieux cimentaires sont des marqueurs fondamentaux de la durabilité des structures de Génie Civil en béton. La connaissance de ces grandeurs est également importante pour l'interprétation de certains essais non destructifs mis en œuvre pour évaluer les performances mécaniques des ouvrages ou détecter certains défauts. L'évaluation de la teneur en eau par méthode non-destructive requiert l'utilisation d'une grandeur intermédiaire telle que la permittivité diélectrique. La relation entre cette propriété électromagnétique et la teneur en eau dépend alors de la composition et donc de la formulation du béton. En électromagnétisme, les lois d'homogénéisation permettent de lier la permittivité effective d'un matériau hétérogène avec la permittivité intrinsèque et la fraction volumique de chaque hétérogénéité présente. Afin de pallier le temps important requis pour l'établissement d'une courbe de calibration expérimentale propre à chaque formulation, l'étude présentée propose la mise en place d'une démarche d'homogénéisation de la permittivité pour lier quantité d'eau présente dans un béton et permittivité macroscopique du matériau. Les travaux présentés rapportent la fabrication, la modélisation et l'utilisation de sondes coaxiales ouvertes pour la mesure de la permittivité complexe de matériaux solides et liquides. Le concept d'estimation de la teneur en eau par utilisation de lois d'homogénéisation est validé pour le cas d'un sable partiellement saturé en eau. Au vu des résultats prometteurs obtenus par modélisation analytique, des schémas d'homogénéisation sont combinés lors d'un processus de remontée d'échelle depuis celle des hydrates jusqu'à celle des granulats, en tenant compte de la morphologie de la microstructure. Les propriétés intrinsèques des principaux constituants d'un béton (granulats, hydrates, ciment anhydre) sont alors mesurées par sonde coaxiale et utilisées en données d'entrée du modèle construit. Une bonne cohérence est observée entre parties réelles de la permittivité simulées et mesurées, pour des échantillons de pâtes de ciment, mortiers et bétons. A la différence des lois expérimentales et empiriques, le modèle construit se caractérise par un temps de calcul quasi-instantané et peut être adapté d'une formulation de béton à une autre en fonction du type de ciment utilisé, de la nature et de la quantité de granulats ou encore de la porosité accessible à l'eau du matériau. / Prediction of delayed behavior in concrete can be significantly improved by monitoring the amount and spatial distribution of water within a concrete structure over time. Water content of cement-based materials can also be required to interpret non-destructive tests such as ultrasonic and radar measurements. Electromagnetic properties of heterogeneous and porous materials, such as dielectric permittivity, are closely related to water content. Measurement of these properties is thus a common non-destructive technique used to assess the moisture content, but a calibration curve is required to link the measured permittivity to the saturation degree. This curve can be determined experimentally, or from empirical models. However, the first approach is tedious and time consuming, while the second one is not adapted to concrete. Hence, this contribution proposes an alternative route, relying on electromagnetic homogenization schemes, to connect the macroscopic permittivity of cement-based materials with the water content of the structure. Therefore, different open-ended coaxial probes were designed, modelled and tested in order to perform complex permittivity measurements of both solids and liquids. The homogenization approach is first validated on unsaturated sand. Then, the permittivity of concrete components (aggregates, hydrates, interstitial liquid, anhydrous cement) was assessed by means of coaxial probe measurements. Finally, a specific combination of analytical homogenization laws taking into account the microstructure's morphology of the material is built. Results show that there is a good correlation between the model and measurements acquired on different cement pastes, mortars and concretes, at different saturation degrees. The model is characterized by a quasi-instantaneous calculation time and can be adapted to different concretes depending on cement type, nature and quantity aggregates or porosity.

<div>Cables play an important role in the design of a power system. DC cable design presents unique challenges due to the fact that space charge can accumulate within the dielectric over time. Space charge accumulation is a function of temperature, electric field, and dielectric properties. Of particular concern is that the space charge leads to electric fields that are sufficient to break down the cable, particularly during transient conditions such as voltage reversal.</div><div><br></div><div>In this research, a focus is on the development of a coupled thermal- and electricalequivalent-circuit model that is general and provides the ability to predict the electric fields and space charge accumulation within single and multi-conductor DC cables. In contrast to traditional analytical models, the approach is more general, allowing for exploration of a wide spectrum of geometries. In contrast to traditional numerical methods, including finite element or finite difference, apriori knowledge of the electric field behavior is used to discretize the dielectric into a small number of electric flux tubes. The electric field dynamics within each tube are then modeled using a first order nonlinear differential equation. The relatively coarse discretization enables the solution to be computed rapidly. This is useful in population-based design where a large number of candidate evaluations is necessary to explore a design space. The modeling approach has been validated using several examples presented in the literature. In addition, its usefulness has been highlighted in the optimization of a 20 kV cable wherein objectives include minimization of mass and loss. </div>

Des études dans le domaine plasma-biologie ont démontré l'effet biocide des jets plasma à pression atmosphérique (JPPAs), faisant de ces dispositifs une alternative aux traitements classiques de décontamination microbiologique. Le gaz (ou mélange de gaz) utilisé et la puissance injectée dans le plasma ont été identifiés comme des paramètres importants influençant l'efficacité biocide des jets plasmas. Récemment, il a été rapporté qu'un autre facteur pouvait influencer l'effet biocide des jets plasmas : la nature de la cible en interaction avec le jet. Cette thèse propose d'étudier l'influence de la nature d'une cible sur les propriétés physiques et biocides des jets plasmas. Dans un premier temps, le jet plasma a été caractérisé en l'absence de cible à l'aide d'outils de diagnostics électriques (sonde de tension, sonde de courant, oscilloscope) et optiques (caméra ICCD, spectromètre optique). Ainsi, la caméra ICCD a permis l'observation de la propagation d'une onde d'ionisation à des vitesses très élevées (de l'ordre du km.s-1). Dans un second temps, la caractérisation a porté sur l'interaction du jet plasma avec d'une part une cible diélectrique (10-10 S.m-1) et d'autre part une cible conductrice (6.0 107 S.m-1). Les résultats montrent que lorsque l'onde d'ionisation atteint la cible diélectrique, celle-ci se propage de manière radiale sur la surface de la cible formant une onde d'ionisation surfacique. Pour la cible conductrice, un canal s'établit entre la source plasma et la surface de la cible, suivi éventuellement par la propagation d'une onde d'ionisation secondaire de la surface de la cible en direction de la sortie de la source plasma. Enfin, afin d'évaluer l'influence de la nature de la cible sur l'efficacité biocide du jet plasma, des suspensions d'endospores de Bacillus atrophaeus et de bactéries Escherichia Coli ont été utilisées comme indicateur biologique. Les endospores et les bactéries ont été inoculées sur la surface des cibles diélectriques et conductrices, puis exposées au jet plasma. Les résultats montrent des modifications différentes sur le manteau des spores et une efficacité biocide plus élevée pour la cible conductrice comparée à la cible diélectrique. En résumé, cette thèse montre que les effets biocides des jets plasmas doivent être évalués dans les conditions expérimentales qui correspondent à l'application envisagée. / Studies in the plasma-biology field have demonstrated the biocidal effect of atmospheric pressure plasma jets (JPPAs), making these devices an alternative to conventional biological decontamination treatments. The gas (or gas mixture) used and the power injected into the plasma were identified as the parameters influencing the biocidal efficacy of plasma jets. Recently, it has been reported that another factor may influence the biocidal effect of plasma jets: the nature of the target in interaction with the jet. This thesis proposes to study the influence of the nature of a target on the biocidal properties of plasma jets. Initially, the plasma jet was characterized in the absence of a target using electrical (voltage probe, current probe, oscilloscope) and optical (ICCD camera, optical spectrometer) diagnostic tools. The ICCD camera allowed the observation of the propagation of an ionization wave at very high velocities (of the order of km.s-1). Secondly, the characterization focused on the interaction of the plasma jet with a dielectric target (10-10 S.m-1) and a conductive target (6.0 107 S.m-1). The results show that when the ionization wave reaches the dielectric target, it propagates radially on the surface of the target forming a surface ionization wave. For the conductive target, a channel is established between the plasma source and the target surface, possibly followed by propagation of a secondary ionization wave from the target surface towards the plasma source outlet. Finally, the influence of the target nature on the biocidal efficacy of the plasma jet was studied. Endospores suspensions of Bacillus atrophaeus and Escherichia coli bacteria were used as biological indicator. Endospores and bacteria were inoculated on the surface of the dielectric and conductive target and exposed to the plasma jet. The results show different changes on the spores coat's and higher biocidal efficiency for the conductive target compared to the dielectric target. In summary, this thesis shows that the biocide effects of plasma jets should be evaluated under the experimental conditions corresponding to the intended applications.

Le sujet de cette thèse porte sur la compréhension et la modélisation du comportement électrique de système d'isolation soumis à un fort champ électrique continu. Les propriétés électriques du polymère sont directement pilotées par ses hétérogénéités chimiques et physiques présentes à plusieurs échelles. Dans cette étude, un nouveau modèle est développé ayant pour but de simuler l'évolution de la microstructure de polymère avec la température, le champ électrique et le temps et de simuler l'impact de cette évolution sur les propriétés électriques du polymère. Dans ce modèle, des matrices sont utilisées pour décrire la distribution de chacune des hétérogénéités et propriété électriques d'un polymère semi-cristallin. L'évolution de ces matrices de microstructure suit des lois génétiques dont la définition a été obtenue à partir d'une caractérisation fine des propriétés physicochimiques et électriques de matériaux spécifiques en fonction de la température et du champ électrique. Ces lois implémentées sont basés sur des calculs simples permettant un temps de résolution plus rapide comparativement aux autres modèles préexistants. Basée sur ces lois d'évolutions génétiques, le comportement électrique sous champ électrique continue de polymère isolant peut être simulé uniquement à partir d'une caractérisation physique et chimique de ce polymère. Le modèle est ainsi capable de reproduire le comportement électrique de plusieurs polymères semi-cristallins et de suivre les données expérimentales mesurées par ailleurs. Le modèle intègre plusieurs physiques tels que la diffusion, le transport ionique et le transport électronique, lui permettant ainsi de prendre en compte l'influence d'un grand nombre d'hétérogénéités. / This thesis presents a research work on understanding and modeling the electrical behavior of insulation system in cables subjected to high DC constraints. Electrical properties of polymeric insulation are directly related to their chemical and physical heterogeneities present at several scales. In this work, a new model is developed to simulate the modification over time of the microstructure in insulation polymers under electric field and temperature as well as the subsequent impacts on electrical properties. In this model, matrices are used to describe the distribution of each heterogeneity and electrical property in semi-crystalline polymer. When stressed under electric field and at temperature, matrices of microstructure evolve from implemented genetic laws. This simulated microstructure evolution yields to the simulation of electrical property changes over time at transient and steady state. To define these genetic laws, a detailed characterization of the physical, chemical and electrical properties of specific materials as a function of temperature and electric field is experimentally performed. Genetic laws are notably implemented to take into account the impact of the semi-crystalline structure and the presence of chemical residues in polymer electrical properties. Based on these genetic evolution laws, this modeling approach allows simulating DC electrical behavior of polymers only from their physical and chemical characterizations and reproduce accurately experimental electrical behavior with a faster solving time compared to other simulation methods.