Additiv tillverkning (AM) är en tillverkningsmetod som skapar komponenter genom att addera material där det tidigare inte fanns, detta möjliggör tillverkning av geometrier som annars hade varit omöjliga eller mycket tidskrävande. Electron Beam Melting är en pulverbaserad AM metod där ett metallpulver smälts samman av en elektronstråle. De två största nackdelarna med pulverbaserad AM är en ojämn yta och inre porositet i tillverkade komponenter. Den grova ytan avhjälps i de fall det behövs genom att efterbehandla komponenter genom skärande bearbetning och porositeten åtgärdas idag med HIP, Hot Isostatic Pressing. Arbetet i denna rapport har som syfte att minimera porositeten in situ för att öka tillförlitligheten och repeterbarhet hos materialegenskaperna i EBM-tillverkade komponenter genom optimerandet av smältstrategin. Detta har skett genom ett experiment där fem smältstrategier har använts vid tillverkning av fyra olika utformade provstavar varvid porositeten har granskats i avseendena porandel av ytarea och porfördelning. De fem strategierna var S0, Standard; S1, Enkelriktad ifyllnad före kontur; S2.0, Endast kontur utifrån och in; S2.1, Endast kontur inifrån och ut; S2.2, Som S2.1 utan MultiBeam, och de fyra provstavsutformningarna var ett rätblock, en cylinder, ett rör med 3 mm tjocka väggar och ett timglas. Lägst porositet gavs av S2.1 med en genomsnittlig densitet på 99,993 % och högst gavs av S2.0 med en denistet på 98,63 % där S0 resulterade i en genomsnittlig densitet på 99,94%. / Additive manufacturing (AM) is a manufacturing method that creates components by adding material where there previously was none, this enables fabrication of geometries which otherwise had been impossible or very time consuming. Electron Beam Melting is a powder based AM-method where a metallic powder is melted by an electron beam. The two largest issues with powder based AM is its high surface roughness and internal porosity of manufactured components. The uneven surface is remedied where necessary by making the part larger than its final dimensions and machining it to size while the porosity today is rectified with HIP, Hot Isostatic Pressing. This works aims to minimize the porosity in situ in order to improve the reliability and repeatability of the material properties of EBM-manufactured parts through the optimization of the melting strategy. This has been done through an experiment in which five melting strategies have been applied to four different test rods after which the porosity was examined in terms of porosity and pore distribution. The five strategies were S0, Standard; S1, One-way hatch before contour; S2.0, Only contour outside and in; S2.1, Only contour inside and out; S2.2, Like S2.1 without MultiBeam, and the four test bar designs were a cuboid, a cylinder, a tube with 3 mm thick walls and an hourglass. The lowest porosity was given by S2.1 with a mean average density of 99 993% and highest was S2.0 with 98.63% density whereas S0 resulted in a mean average density of 99.94%. / <p>Betyg 2020-08-02</p>

O gênero Yersinia pertence à família Enterobacteriaceae e compreende 17 espécies. Y. pestis, Y. pseudotuberculosis e Y. enterocolitica são reconhecidamente patógenos de humanos e animais. Y. pestis cause a peste. Y. pseudotuberculosis e Y. enterocolitica são agentes causadores, sobretudo, de gastroenterites transmitidas por água e alimentos. As demais 14 espécies são, usualmente, consideradas não-patogênicas, com exceção de Y. ruckeri sorogrupo O:1 que causa infecções em peixes. Nas últimas décadas, a tipagem molecular tornou-se uma importante ferramenta nos estudos filogenéticos de numerosos micro-organismos e o desenvolvimento de sistemas de tipagem rápidos e baratos pode facilitar os estudos epidemiológicos de infecções bacterianas. No presente estudo objetivou-se desenvolver um método de genotipagem de Yersinia spp. baseado em high-resolution melting analysis (HRMA) para diferenciar os single-nucleotide polymorphisms (SNPs) presentes nas sequências dos genes 16S rRNA, glnA, gyrB, hsp60 e recA e aplicá-lo na tipagem de 40 linhagens de Y. pseudotuberculosis e 50 linhagens de Y. enterocolitica, bem como separar por HRMA as espécies Y. pseudotuberculosis e Y. enterocolitica. Os SNPs foram determinados nas sequências dos loci acima citados a partir de um conjunto de 119 linhagens de Yersinia spp. depositadas no GenBank/EMBL/DDBJ. Foram encontrados nas sequências dos genes analisados de Y. pseudotuberculosis, Y. enterocolitica, Y. bercovieri, Y. rohdei, Y. intermedia, Y. mollaretii e Y. ruckeri 10, 10, 9, 6, 4, 1 e 1 SNPs, respectivamente. Nenhum SNP foi encontrado nas sequências analisadas de Y. pestis e um grande número de SNPs foi encontrado nas sequências analisadas de Y. frederiksenii, Y. kristensenii e Y. massiliensis, o que impossibilitou a genotipagem dessas espécies por HRMA. As demais espécies não foram analisadas. Foram desenhados pares de primers para flanquear os SNPs encontrados em cada espécie de Yersinia testada. Usando um conjunto de primers espécie-específicos, a diversidade genética de cada espécie de Yersinia foi determinada por HRMA e a análise filogenética foi baseada na sequência concatenada composta pelos nucleotídeos identificados em cada fragmento analisado. O agrupamento foi realizado com o software BioNumerics usando o método UPGMA com 1.000 replicatas de bootstrap. A árvore filogenética ii construída para Y. pseudotuberculosis agrupou as linhagens em clusters bio-sorogrupo específicos. As linhagens do bio-sorogrupo 1/O:1 foram agrupadas em um cluster e as linhagens do bio-sorogrupo 2/O:3 em outro. A árvore filogenética construída para Y. enterocolitica agrupou as linhagens em três grupos. As linhagens altamente patogênicas, do biotipo 1B, foram agrupadas em um cluster, as linhagens de média patogenicidade, dos biotipos 2, 3, 4 e 5, foram agrupadas em um segundo cluster e as linhagens consideradas nãopatogênicas, do biotipo 1A, foram agrupadas em um terceiro cluster. O agrupamento encontrado em Y. pseudotuberculosis e Y. enterocolitica foi consistente com o perfil patogênico característico dessas duas espécies. Nenhuma correlação epidemiológica significativa foi encontrada no agrupamento de Y. bercovieri, Y. rohdei, Y. intermedia, Y. mollaretii e Y. ruckeri de acordo com os resultados de HRMA. Ademais, o método de HRMA aqui desenvolvido foi capaz de separar as espécies Y. pseudotuberculosis e Y. enterocolitica. O método de HRMA desenvolvido nesse estudo pode ser usado como uma alternativa para a genotipagem e para a diferenciação de Y. pseudotuberculosis de Y. enterocolitica. Esse método também pode complementar os métodos baseados em sequências e facilitar os estudos epidemiológicos dessas duas espécies de Yersinia. / The genus Yersinia belongs to the family Enterobacteriaceae and comprises 17 species. Y. pestis, Y. pseudotuberculosis and Y. enterocolitica are well recognized human and animal pathogens. Y. pestis causes plague. Y. pseudotuberculosis and Y. enterocolitica are, usually, causative agents of food-waterborne gastroenteritis. The other 14 Yersinia species are considered to be non-pathogenic, with the exception of Y. ruckeri serogroup O:1 which causes infections in fishes. In the last few decades, molecular typing has become an important tool in phylogenetic studies of several microorganisms and the development of fast and inexpensive typing systems can facilitate epidemiological studies of bacterial infections. The present study aimed to develop a method of Yersinia spp. genotyping based on high-resolution melting analysis (HRMA) in order to differentiate the single-nucleotide polymorphisms (SNPs) present in the 16S rRNA, glnA, gyrB, hsp60 and recA sequences and apply it in the typing of 40 Y. pseudotuberculosis strains and 50 Y. enterocolitica strains, as well as, to separate by HRMA the Y. pseudotuberculosis and Y. enterocolitica species. The SNPs were determined in the sequences of the aforementioned loci using a set of 119 Yersinia strains deposited in the GenBank/EMBL/DDBJ database. It were found in the gene sequences analyzed of Y. pseudotuberculosis, Y. enterocolitica, Y. bercovieri, Y. rohdei, Y. intermedia, Y. mollaretii and Y. ruckeri 10, 10, 9, 6, 4, 1 and 1 SNPs, respectively. No SNPs was found in the analyzed sequences of Y. pestis and a large number of SNPs were found in the analyzed sequences of Y. frederiksenii, Y. kristensenii and Y. massiliensis what prevented their genotyping by HRMA. The remaining Yersinia species were not analyzed. It was designed primer pairs to flank the SNPs found in each Yersinia species tested. Using a specie-specific set of primers, the genetic diversity of each Yersinia species used was determined by HRMA and the phylogenetic analysis was based on the concatenated sequence composed by the nucleotides identified in each fragment analyzed. Clustering was performed with the software package BioNumerics using UPGMA method and 1,000 bootstrap replicates. The phylogenetic tree constructed for Y. pseudotuberculosis grouped the strains into bio-serogroups specific clusters. The strains of 1/O:1 bio-serogroup were grouped into one cluster and the strains of 2/O:3 bio-serogroup into iv other cluster. The phylogenetic tree constructed for Y. enterocolitica grouped the strains in three clusters. The highly pathogenic strains, of biotype 1B, were grouped into one cluster, the moderate pathogenic strains, of biotypes 2, 3, 4 and 5, were grouped into a second cluster and, the non-pathogenic strains, of biotype 1A, were grouped into a third cluster. The clusterization of Y. pseudotuberculosis and Y. enterocolitica were consistent with the pathogenic profile characteristic of these two Yersinia species. No significant epidemiological correlation was found in the grouping of Y. bercovieri, Y. rohdei, Y. intermedia Y. mollaretii and Y. ruckeri according to HRMA results. Moreover, the HRMA-based method develop here was able to separate the Y. pseudotuberculosis and Y. enterocolitica species. The HRMA assay developed in this study can be used as an alternative for the genotyping and the differentiation of Y. pseudotuberculosis and Y. enterocolitica. This method can also complement sequence-based methods and facilitate epidemiological studies of these two Yersinia species.

We develop a two dimensional model that simulates the response of the crust to prolonged mantle-derived intrusions in arc environments. The domain includes the entire crustal section and upper mantle and focuses on the evolving thermal structure due to intrusions and external tectonic forcing. We monitor the thermal response, melt fraction and volume for different environments after a definite time by considering geologically relevant melt flux and extensional tectonic rates. The amount of crustal melt versus fractionated primary mantle melts present in the crustal column helps determine crustal structure and growth through time. We observe that with a geophysically estimated flux and tectonic rate, the mantle-derived magma bodies can melt the surrounding volume of crust. We express the amount of crustal melting in terms of an efficiency; therefore we define the melting efficiency as the ratio of the melted volume of crustal material to the volume of melt expected from a strict enthalpy balance as explained by Dufek and Bergantz (2005). Melting efficiencies are less than 1.0 in real systems because heat diffuses to sections of the crust that never melt. The maximum calculated efficiency is 0.05 in our model while most of our simulations show zero efficiency. Additionally, maximum total melt amount is observed in relatively greater extensional environments (0.02 m/yr) and high intrusion rates (10⁻² m³/m²/yr) and in long time periods (2 x 10⁶ years). However, maximum crustal melting in the same environment is reached in 1.2 x 10⁶ years. The relative amounts of mantle-derived and crustal melts in the total volume of magma suggest that the majority of magma composition in crustal column is derived from the mantle material.

Temperature and precipitation are the most important indicators of climate change. Especially for the basins fed by snow, the shifts of melting to earlier times, affects the streamflow. Increase in temperature causes to shifts of melting of snow to shift to earlier times so that hydrologic regime of the river system changes, and leads to changes in climatic conditions of the region. In this study the shifts of snow melting times are analyzed for the selected 15 streamflow stations located in Euphrates, Tigris, Aras, and &Ccedil / oruh basins in Eastern Anatolia of Turkey along with period from 1970 to 2010. The shifts in snowmelt runoff are determined by Center Time (CT) method. Meteorological stations representing the stream gauge stations regarding the basin characteristics are also selected to be used in the analyses. In order to relate CT shifts to temperature and precipitation changes, trend analysis are applied to temperature, precipitation and streamflow data. In addition to these, days with daily average temperature less than freezing and wet days below freezing until CT for each station pair between stream gauge and meteorological stations and each year are also analyzed. These days till CT within a year for each station pair can be indirectly linked to snowy days and accumulated snow amount. Complete analyses show significant warming at each station in the region and no important trends in annual precipitation. However at a few stations meaningful seasonal changes in precipitation are observed. Regional warming and associated changes in precipitation and snowmelt runoff cause significant shifts to earlier times of snowmelt runoff. In the region eight out of fifteen stream gauge stations in Euphrates, Tigris and Aras basins showed significant time shifts according to statistical trend tests.

No description available.

Additive manufacturing (AM), also called 3D-printing, are technologies where parts are formed from the bottom up by adding material layer-by-layer on top of each other. Electron Beam Melting (EBM) is an AM technique capable of manufacturing fully solid metallic parts, using a high-intensity electron beam to melt powder particles in layers to form finished components. Compared to conventional machining, EBM offers enhanced efficiency for production of customized and patient specific parts such as e.g. dental prosthetics. However, dental prosthetics are challenging to produce by EBM, as their small sizes mean that mechanical and surface properties may be altered as part sizes decreases. The aim of this thesis is to gain new insights that could lead to optimization for production of small sized components in the EBM. The work is focused to understand the process-property relationships for small size components production. To improve the surface resolution and part detailing, a smaller sized powder was used for production and compared to parts made with standard sized powder. The surface-, chemical and mechanical properties were evaluated for parts produced with both types of powders. The results indicate that the surface roughness may be influenced by powder and build layer thickness size, whereas the mechanical properties showed no influence of the layer-wise production. However, the mechanical properties are dependent on part size. The outermost surface of the parts consists of a surface oxide dominated by TiO2, formed as a result of reaction between the surface and residual gases in the EBM build chamber. The surface oxide thickness is comparable to that of a conventionally machined surface, but is dependent on build height. This work concludes that the surface resolution and component detailing can be improved by various measures. Provided that proper process themes are used, the EBM manufactured material is homogenous with properties comparable to conventional produced titanium. It has also been shown that the material properties will be altered for small components. The results point towards different ways of optimizing manufacturing of dental prosthetics by EBM, which will make dental prosthetics available for an increased number of patients.

Le récent développement de la Fabrication Additive de pièces métalliques permet d'élaborer directement des structures à partir de modèles 3D. En particulier, la technologie "Electron Beam Melting" (EBM) permet la fusion sélective, couche par couche, de poudres métalliques. Elle autorise la réalisation de géométries très complexes mais apporte de nouvelles contraintes de fabrication.Ce travail se concentre sur la caractérisation géométrique et mécanique de structures treillis produites par cette méthode. Les pièces fabriquées sont comparées au design initial à travers des caractérisations par tomographie aux rayons X. Les propriétés mécaniques sont testées en compression uni-axiale. Pour les poutres de faibles épaisseur, la différence entre la structure numérique et celle fabriquée devient significative. Les écarts au design initial se traduisent pour chaque poutre par un concept de matière mécaniquement efficace. D'un point de vue modélisation, ce concept est pris en compte en remplaçant la poutre fabriquée par un cylindre avec un diamètre mécaniquement équivalent. Ce diamètre équivalent est utilisé dans des simulations et optimisations "réalistes" intégrant ainsi les contraintes de fabrication de la technologie EBM.Différentes stratégies sont aussi proposées pour réduire la proportion de volume "inefficace" et améliorer le contrôle de la taille des poutres, soit en jouant sur les paramètres procédé et les stratégies de fusion, soit en effectuant des post-traitements. / The recent development of Additive Manufacturing for the fabrication of metallic parts allows structures to be directly manufactured from 3D models. In particular, the "Electron Beam Melting" (EBM) technology is a suitable process which selectively melts a powder bed layer by layer. It can build very complex geometries but brings new limitations that have to be quantified.This work focuses on the structural and mechanical characterization of lattice structures produced by such technology. The structural characterization mainly rely on X-ray tomography whereas mechanical properties are assessed by uni-axial compression. The geometry and related properties of the fabricated structures are compared with the designed ones. For small strut size, the difference between the designed structure and the produced one is large enough to impact the desired mechanical properties. The concept of mechanical efficient volume is introduced. For the purpose of simulation, this concept is taken into account by replacing the struts by a cylinder with a textit{mechanical equivalent diameter}. After validation, it has been used into "realistic" simulation and optimization procedures, thus taking into account the process constraints.Post-treatments (Chemical Etching and Electro-Chemical Polishing) were applied on lattice structures to get rid of the inefficient matter by decreasing the surface roughness. The control of the size of the fabricated struts was improved by tuning the process strategies and parameters.

En vue de leur refroidissement, les parois de chambres de combustion aéronautiques sont perforées de trous à travers lesquels de l’air plus froid est injecté. La paroi est ainsi refroidie par convection et un film isolant est créé en surface chaude (film cooling). Cette thèse a pour objectif d’utiliser les possibilités de la fabrication additive pour proposer de nouvelles solutions architecturées qui permettraient d’augmenter les échanges de chaleur internes et d’obtenir ainsi de meilleures efficacités de refroidissement.La première approche consiste à élaborer de nouveaux designs de plaques multiperforées par Electron Beam Melting (EBM) et Selective Laser Melting (SLM) aux limites de résolution des procédés. Les architectures sont caractérisées en microscopie, en tomographie X et en perméabilité. Des simulations aérothermiques permettent de mettre en évidence l’effet de ces nouveaux designs sur l’écoulement et les échanges de chaleur, et de proposer des voies d’amélioration de la géométrie.La deuxième approche consiste à élaborer de façon simultanée une pièce architecturée par EBM, avec des zones denses et poreuses. A partir d’analyse d’images associée à une cartographie EBSD grand champ, il est possible de remonter aux mécanismes de formation du matériau poreux et de relier la perméabilité et la porosité aux paramètres procédé. Afin de favoriser le film cooling, il pourrait être avantageux que les zones microporeuses soient orientées dans le sens de l’écoulement. Pour ce faire, un nouveau procédé dénommé Magnetic Freezing, où des poudres métalliques forment une structure orientée par un champ magnétique, est mis au point.Les diverses solutions développées durant cette thèse sont testées sur un banc aérothermique. Les essais montrent qu’elles offrent un refroidissement plus efficace et plus homogène que la référence industrielle. Enfin, de premiers tests en combustion sur l’une des structures retenues, plus légère et plus perméable que la référence, montrent qu’il s’agit d’une solution aussi efficace à un débit traversant donné, et donc a priori plus efficace à une surpression donnée. / Combustion chamber walls are perforated with holes so that a cooling air flow can be injected through them. The wall is cooled by convection and an insulating film is created on the hot surface (film cooling). This PhD thesis aims to use the possibilities of additive manufacturing to provide new architectured solutions that could enhance the internal heat exchanges, and lead to a higher cooling effectiveness.The first approach is to develop new designs of multiperforated walls by Electron Beam Melting (EBM) and Selective Laser Melting (SLM) used at the resolution limits of the processes. They are characterized by microscopy, X-ray tomography and permeability tests. Some aerothermal simulations help understanding the effects of these new designs on the flow and on heat exchanges. These results lead to a geometry adaptation.The second approach is to simultaneously manufacture an architectured part with dense and porous zones by EBM. Thanks to image analysis combined with large field EBSD, it is possible to investigate the mechanisms leading to the porous zones and to link them to permeability and porosity. The film cooling effect could be favoured by the orientation of pores towards the cooling flow. Therefore, a new powder-based manufacturing process named Magnetic Freezing, where metallic powders organize into an oriented structure thanks to a magnetic field, is developed.The various solutions studied during this thesis are tested on an aerothermal bench. They all show a more efficient and homogeneous cooling than the industrial reference. Some first tests on one of the selected solutions are performed on a combustion bench. This lighter and more permeable structure proves to be a solution as efficient as the industrial reference at a given flow rate. It should therefore be a more efficient solution for a given overpressure.

O gênero Yersinia pertence à família Enterobacteriaceae e compreende 17 espécies. Y. pestis, Y. pseudotuberculosis e Y. enterocolitica são reconhecidamente patógenos de humanos e animais. Y. pestis cause a peste. Y. pseudotuberculosis e Y. enterocolitica são agentes causadores, sobretudo, de gastroenterites transmitidas por água e alimentos. As demais 14 espécies são, usualmente, consideradas não-patogênicas, com exceção de Y. ruckeri sorogrupo O:1 que causa infecções em peixes. Nas últimas décadas, a tipagem molecular tornou-se uma importante ferramenta nos estudos filogenéticos de numerosos micro-organismos e o desenvolvimento de sistemas de tipagem rápidos e baratos pode facilitar os estudos epidemiológicos de infecções bacterianas. No presente estudo objetivou-se desenvolver um método de genotipagem de Yersinia spp. baseado em high-resolution melting analysis (HRMA) para diferenciar os single-nucleotide polymorphisms (SNPs) presentes nas sequências dos genes 16S rRNA, glnA, gyrB, hsp60 e recA e aplicá-lo na tipagem de 40 linhagens de Y. pseudotuberculosis e 50 linhagens de Y. enterocolitica, bem como separar por HRMA as espécies Y. pseudotuberculosis e Y. enterocolitica. Os SNPs foram determinados nas sequências dos loci acima citados a partir de um conjunto de 119 linhagens de Yersinia spp. depositadas no GenBank/EMBL/DDBJ. Foram encontrados nas sequências dos genes analisados de Y. pseudotuberculosis, Y. enterocolitica, Y. bercovieri, Y. rohdei, Y. intermedia, Y. mollaretii e Y. ruckeri 10, 10, 9, 6, 4, 1 e 1 SNPs, respectivamente. Nenhum SNP foi encontrado nas sequências analisadas de Y. pestis e um grande número de SNPs foi encontrado nas sequências analisadas de Y. frederiksenii, Y. kristensenii e Y. massiliensis, o que impossibilitou a genotipagem dessas espécies por HRMA. As demais espécies não foram analisadas. Foram desenhados pares de primers para flanquear os SNPs encontrados em cada espécie de Yersinia testada. Usando um conjunto de primers espécie-específicos, a diversidade genética de cada espécie de Yersinia foi determinada por HRMA e a análise filogenética foi baseada na sequência concatenada composta pelos nucleotídeos identificados em cada fragmento analisado. O agrupamento foi realizado com o software BioNumerics usando o método UPGMA com 1.000 replicatas de bootstrap. A árvore filogenética ii construída para Y. pseudotuberculosis agrupou as linhagens em clusters bio-sorogrupo específicos. As linhagens do bio-sorogrupo 1/O:1 foram agrupadas em um cluster e as linhagens do bio-sorogrupo 2/O:3 em outro. A árvore filogenética construída para Y. enterocolitica agrupou as linhagens em três grupos. As linhagens altamente patogênicas, do biotipo 1B, foram agrupadas em um cluster, as linhagens de média patogenicidade, dos biotipos 2, 3, 4 e 5, foram agrupadas em um segundo cluster e as linhagens consideradas nãopatogênicas, do biotipo 1A, foram agrupadas em um terceiro cluster. O agrupamento encontrado em Y. pseudotuberculosis e Y. enterocolitica foi consistente com o perfil patogênico característico dessas duas espécies. Nenhuma correlação epidemiológica significativa foi encontrada no agrupamento de Y. bercovieri, Y. rohdei, Y. intermedia, Y. mollaretii e Y. ruckeri de acordo com os resultados de HRMA. Ademais, o método de HRMA aqui desenvolvido foi capaz de separar as espécies Y. pseudotuberculosis e Y. enterocolitica. O método de HRMA desenvolvido nesse estudo pode ser usado como uma alternativa para a genotipagem e para a diferenciação de Y. pseudotuberculosis de Y. enterocolitica. Esse método também pode complementar os métodos baseados em sequências e facilitar os estudos epidemiológicos dessas duas espécies de Yersinia. / The genus Yersinia belongs to the family Enterobacteriaceae and comprises 17 species. Y. pestis, Y. pseudotuberculosis and Y. enterocolitica are well recognized human and animal pathogens. Y. pestis causes plague. Y. pseudotuberculosis and Y. enterocolitica are, usually, causative agents of food-waterborne gastroenteritis. The other 14 Yersinia species are considered to be non-pathogenic, with the exception of Y. ruckeri serogroup O:1 which causes infections in fishes. In the last few decades, molecular typing has become an important tool in phylogenetic studies of several microorganisms and the development of fast and inexpensive typing systems can facilitate epidemiological studies of bacterial infections. The present study aimed to develop a method of Yersinia spp. genotyping based on high-resolution melting analysis (HRMA) in order to differentiate the single-nucleotide polymorphisms (SNPs) present in the 16S rRNA, glnA, gyrB, hsp60 and recA sequences and apply it in the typing of 40 Y. pseudotuberculosis strains and 50 Y. enterocolitica strains, as well as, to separate by HRMA the Y. pseudotuberculosis and Y. enterocolitica species. The SNPs were determined in the sequences of the aforementioned loci using a set of 119 Yersinia strains deposited in the GenBank/EMBL/DDBJ database. It were found in the gene sequences analyzed of Y. pseudotuberculosis, Y. enterocolitica, Y. bercovieri, Y. rohdei, Y. intermedia, Y. mollaretii and Y. ruckeri 10, 10, 9, 6, 4, 1 and 1 SNPs, respectively. No SNPs was found in the analyzed sequences of Y. pestis and a large number of SNPs were found in the analyzed sequences of Y. frederiksenii, Y. kristensenii and Y. massiliensis what prevented their genotyping by HRMA. The remaining Yersinia species were not analyzed. It was designed primer pairs to flank the SNPs found in each Yersinia species tested. Using a specie-specific set of primers, the genetic diversity of each Yersinia species used was determined by HRMA and the phylogenetic analysis was based on the concatenated sequence composed by the nucleotides identified in each fragment analyzed. Clustering was performed with the software package BioNumerics using UPGMA method and 1,000 bootstrap replicates. The phylogenetic tree constructed for Y. pseudotuberculosis grouped the strains into bio-serogroups specific clusters. The strains of 1/O:1 bio-serogroup were grouped into one cluster and the strains of 2/O:3 bio-serogroup into iv other cluster. The phylogenetic tree constructed for Y. enterocolitica grouped the strains in three clusters. The highly pathogenic strains, of biotype 1B, were grouped into one cluster, the moderate pathogenic strains, of biotypes 2, 3, 4 and 5, were grouped into a second cluster and, the non-pathogenic strains, of biotype 1A, were grouped into a third cluster. The clusterization of Y. pseudotuberculosis and Y. enterocolitica were consistent with the pathogenic profile characteristic of these two Yersinia species. No significant epidemiological correlation was found in the grouping of Y. bercovieri, Y. rohdei, Y. intermedia Y. mollaretii and Y. ruckeri according to HRMA results. Moreover, the HRMA-based method develop here was able to separate the Y. pseudotuberculosis and Y. enterocolitica species. The HRMA assay developed in this study can be used as an alternative for the genotyping and the differentiation of Y. pseudotuberculosis and Y. enterocolitica. This method can also complement sequence-based methods and facilitate epidemiological studies of these two Yersinia species.

Les procédés additifs, auxquels appartient la fusion laser de poudres métalliques, ont la capacité de créer des structures à géométries complexes, avec la possibilité d'intégrer des formes creuses, par exemple des canaux de refroidissement assurant un contrôle thermique optimum. Ce procédé permet de fabriquer des pièces réelles à partir de poudres métalliques, par fusion du matériau, couche par couche, en accord avec le modèle CAO. Au cours du procédé, de nombreux cycles thermiques et d'importants gradients thermiques se produisent dans la pièce au cours de sa fabrication. Ces gradients de température induisent des déformations plastiques hétérogènes et de ce fait des contraintes résiduelles. Ces contraintes peuvent nuire à la qualité de la pièce obtenue, par exemple sa résistance mécanique. Ces travaux ont pour objectifs de proposer un modèle numérique, s’appuyant sur la méthode des éléments finis afin d'étudier l'apparition des contraintes résiduelles lors du procédé de fusion laser de poudres métalliques. Le logiciel multiphysique ABAQUS® a été utilisé pour effectuer les analyses thermiques et mécaniques. La technique « d'ajout et de suppression des éléments » a été utilisée afin de simuler la fusion et la solidification de la matière au cours du procédé. Les propriétés mécaniques dépendantes de la température de l'acier maraging, utilisé dans notre cas, ont été obtenues à l’aide d’essais expérimentaux de caractérisations et intégrées dans le modèle. Les calculs sont réalisés de manière découplée, dans un premier temps le calcul thermique est effectué, puis les résultats sont utilisés pour réaliser le calcul mécanique et finalement prédire les champs de contraintes. Dans le cadre de ce travail, une méthode originale s'appuyant sur la technique de mesure des contraintes résiduelles par enlèvement de couches successives a été mise au point pour mesurer ces contraintes en direct au cours du procédé. Les résultats renseignent sur le niveau et la distribution des contraintes dans la pièce créée et le support. Deux paramètres ont été testés afin d'étudier leur influence sur le niveau des contraintes résiduelles : le temps d’étalement de la poudre entre deux couches successives et la hauteur des couches. Le modèle numérique paramétrable permet d'analyser les effets de paramètres liés au procédé sur la répartition des contraintes résiduelles dans les pièces fabriquées. Les résultats montrent que la variation de l'épaisseur du support n'affecte pas la répartition des contraintes dans la pièce créée. Le préchauffage du support à une température de 800°C réduit les contraintes résiduelles L'étude de quelques trajectoires laser montre leurs influences sur la répartition des déformations plastiques cumulées ainsi que la hauteur des couches de poudres ou de la forme du support (embase, colonnes). / The Selective Laser Melting process, belonging to Additive processes , have the ability to create structures with complex geometries , with the possibility of including cavities, such as cooling channels providing optimum temperature control. This process enables the manufacture of three-dimensional parts from metal powders by melting the material , layer by layer, in agreement with the CAD model. In the process , high temperatures and thermal gradients cycles occur in the part during the process. These temperature gradients induce heterogeneous plastic strain and residual stresses. These residual stresses may affect the quality of the part obtained, for example the fatigue life. This work aims to propose a numerical model , based on the finite element method to study the appearance of residual stresses during laser melting process of metallic powders . The ABAQUS® Multiphysics software was used to perform the thermal and mechanical analyzes. The movement of the laser beam and the resolution of the thermal problem can predict the evolution of the temperature in the part and support. The "birth and death elements" technique was used to simulate the melting and solidification of the material during the process. Dependent mechanical properties of the temperature of the maraging steel used in this case were obtained using experimental testing and characterization and were established in the model. The calculations are decoupled : initially thermal calculation is performed and the results are used to perform mechanical calculations and finally predict the residual stress fields. In this work, a novel method based on the technique of measuring residual stresses by removing layers was developed to measure these stresses directly in the process. The results provide information on the level and distribution of stresses in the created part and support. Two parameters were tested to study their influence on the level of residual stress : time to spread the powder between two successive layers and layer height. The model is used to analyze the effects of process parameters related to the distribution of residual stresses in the manufactured parts. The results show that the variation of the thickness of the support does not affect the distribution of stresses in the part created. Preheating the substrate to a temperature of 800 °C reduces the residual stresses. The study of some laser strategies shows their influence on the distribution of plastic strain thus the height of the layers of powder or in the form of support (base, columns).