The process of laser-induced brazing constitutes a potential option for connecting several ceramic components (n- and p-type ceramic bars and ceramic substrate) of a thermoelectric generator (TEG) unit. For the construction of the TEGs, TiOₓ and BₓC were used as thermoelectric bars and AlN was used as substrate material. The required process time for joining is well below that of conventional furnace brazing processes and, furthermore, establishes the possibility of using a uniform filler system for all contacting points within the thermoelectric unit. In the work reported here, the application-specific optimization of the laser-joining process is presented as well as the adapted design of the thermoelectric modules. The properties of the produced bonding were characterized by using fatigue strength and microstructural investigations. Furthermore, the operational reliability of the modules was verified.

L'électrolyse de la vapeur d'eau à haute température et sous pression avec des électrolytes à conduction protonique est un moyen envisagé pour produire massivement de l'hydrogène à un coût compétitif. Pour être fonctionnelle, une cellule d'électrolyse doit être hermétique vis-à-vis de l'extérieur et entre le compartiment anodique (O2 et H2O) et le compartiment cathodique (H2). Ainsi, la réalisation de liaisons céramique-métal est nécessaire, soit pour l'intégration de l'électrolyte (liaison perovskite-métal), soit pour l'intégration d'alimentations électriques (liaison alumine-métal). Compte-tenu des conditions de fonctionnement de la cellule (600°C, 50 bars, vapeur d'eau), seul un procédé de brasage est envisageable pour réaliser des jonctions résistantes mécaniquement et hermétiques. L'expertise des processus de corrosion sous vapeur d'eau a permis de définir la partie métallique et la brasure pour cette application et a montré que des alliages spécifiques étaient requis. L'étude de métallisation des céramiques par PVD pour permettre le mouillage de la brasure au moment du brasage a montré que la présence d'une barrière de diffusion était nécessaire, pour éviter la dissolution du dépôt lors du passage à l'état liquide de la brasure. Après une étude de mouillabilité à chaud de la brasure sélectionnée sur les céramiques métallisées et les substrats métalliques, les liaisons ont été optimisées en appliquant la méthodologie des plans d'expériences sur des éprouvettes normalisées. L'ensemble des résultats a ainsi conduit au développement de liaisons céramique-métal possédant de bonnes propriétés mécaniques (Rm = 60 MPa à 20°C) et hermétiques. / Steam electrolysis at high temperature and under pressure with protonic conduction electrolyte is a way to produce hydrogen massively at low cost. To operate, an electrolysis cell must be hermetic especially between the anodic compartment (O2 and H2O) and the cathodic one (H2). Thus, ceramic to metal assemblies are needed, either to insert the electrolyte (perovskite-metal junction) or to realize power supply (alumina-metal junction).Considering the operating conditions (873°K, 725 PSI, steam), only a brazing process is possible to realize strong and hermetic junctions.The evaluation of corrosion process under steam pressure has enabled to select a metallic component and a brazing alloy for this application, and has shown that specific alloys are needed.The studies of ceramics metallization by PVD to assure the brazing alloy wetting throughout the brazing process have shown that a barrier diffusion was necessary to avoid the dissolution of coatings when the brazing alloy become liquid.After wetting experimentations of the chosen brazing alloy over the metalized ceramics and the metallic substrate, ceramic to metal assemblies have been optimized in applying the experimental design methodology on standard samples (ASTM).The whole results have enabled to develop hermetic ceramic to metal assemblies having good mechanical properties (Rm = 8700 PSI at 300°K).

Alumina-to-alumina brazed joints were formed using 96.0 and 99.7 wt.% Al2O3 ceramics in as-ground and in ground and heat treated conditions using TICUSIL® (68.8Ag-26.7Cu-4.7Ti wt.%) braze preforms of thicknesses ranging from 50 to 250 μm. Brazing was conducted in a vacuum of 1 x 10-5 mbar at 850 °C for 10 minutes. Joint strengths were evaluated using four-point bend testing and were compared to the flexural strengths of standard test bars according to ASTM C1161-13. Post-grinding heat treatment, performed at 1550 °C for 1 hour, did not affect the average surface roughness or grain size of either grade of alumina but affected their average flexural strengths, with a small increase for 96.0 wt.% Al2O3 and a small decrease for 99.7 wt.% Al2O3. Post-grinding heat treatment led to secondary phase migration, creating a fissured 96.0 wt.% Al2O3 surface. This affected the reliability of 96.0 wt.% Al2O3 brazed joints, in which braze infiltration was observed. As the TICUSIL® braze preform thickness was increased from 50 to 150 μm, the average strengths of both 96.0 and 99.7 wt.% Al2O3 brazed joints improved. This occurred due to a microstructural evolution, in both sets of joints, which was studied using SEM, TEM and nanoindentation techniques. An increase in the TICUSIL® braze preform thickness increased the amount of Ti which was available to diffuse to the joint interfaces. This led to increases in both, reaction layer and braze interlayer thicknesses. Excess Ti in joints that were made using TICUSIL® braze preforms thicker than 50 μm, led to relatively hard Cu-Ti phases in an Ag-Cu braze interlayer. Cu-Ti phase formation, which may have reinforced joint strength whilst also reducing CTE mismatch at the joint interface, also led to Ag-rich braze outflow at the joint edges. Brazed joints made using as-ground 96.0 wt.% Al2O3 consistently outperformed brazed joints made using as-ground 99.7 wt.% Al2O3, due to the formation of Ti5Si3 phases at locations where the Ti-rich reaction layer intersected with the triple pocket grain boundary regions of the as-ground 96.0 wt.% Al2O3 surface.

Das Löten wird als ein Fügeverfahren charakterisiert, bei dem die thermische Beeinflussung der Fügepartner minimal ist. Gleichzeitig ist damit jedoch der prinzipielle Mangel einer Lötverbindung verbunden, denn durch den Verzicht auf das Aufschmelzen des Grundwerkstoffs erreichen die Festigkeiten in der Regel nicht die der Grundwerkstoffe. Diesem Mangel läßt sich einerseits durch Überlapp- oder Steckverbindungen begegnen andererseits bilden sich bei sehr kleiner Lötspaltbreiten ( < 0,02 mm) und in Verbindung mit einem geeigneten Lot-Grundwerkstoff-System sog. Stengelkristallite aus, die über die Lötnaht wachsen und die Festigkeit des Lötverbundes nachhaltig verbessern. Es wird ein Konzept vorgestellt, mit dem der Anwendungsbereich der Stengelkristallite auf industrieübliche Lötspaltbreiten von bis zu 0,2 mm ausgebreitet wird. Die theoretischen Grundlagen werden durch umfangreiche Versuchsplanung und -durchführung bestätigt. Die Bewertung dieses Verbindungstyps im Vergleich zu Verbindungen ohne Stengelkristallite wird in umfangreichen statischen und dynamischen Festigkeitsuntersuchungen dargelegt. Metallographische Untersuchungen bestätigen den Einfluss auf die Festigkeitserhöhung.

The thesis deals with study of high-temperature alloying process of power semiconductor devices. Mechanical durability and loss heat extraction is achieved via contacting silicon devices on molybdenum electrode in vacuum alloying furnace. Properties of alloy contact determine electrical, geometrical and mechanical parameters of device and are related with durability of device under heavy load during operation in various industrial machines or transport vehicles. Process parameters are evaluated in terms of input materials and their preparation along with temperature ramp profile and conditions in vacuum furnace. Test samples are analyzed to assess their electrical and geometrical parameters and selected samples underwent element analysis of alloy contact. Optimal process parameters are estimated according to analysis of experiments.

O objetivo deste trabalho foi comparar, por meio de teste de microdureza e técnicas de caracterização metalúrgica, que inclui a metalografia, microscopia óptica, microscopia eletrônica de varredura (MEV) e espectroscopia por dispersão de energia (energy dispersive spectroscopy EDS), propriedades mecânicas e metalúrgicas de juntas soldadas pelos métodos TIG e brasagem em uma liga odontológica à base de Ni-Cr (FIT-CAST SB). As regiões estudadas em cada amostra foram o metal de base (MB), a zona termicamente afetada (ZTA) e a zona de fusão (ZF). Foram confeccionados 40 corpos-de-prova cilíndricos, em forma de bastão, com 2,5mm de diâmetro e 18,0mm de comprimento, obtidos apartir de fundição odontológica por centrifugação (técnica da cera perdida). Foram separados vinte cilindros para cada técnica, que após a soldagem formaram grupos de dez amostras para cada procedimento. A microestrutura do MB apresentou-se como uma matriz de uma solução sólida à base de Ni-Cr contendo um típico arranjo dendrítico de estrutura eutética. A liga apresentou precipitados metálicos na região interdendrítica da estrutura eutética. A análise química semi-quantitativa obtida através da técnica de EDS apresentou resultados compatíveis com a composição química quantitativa fornecida pelo fabricante da liga. Na brasagem, o MB e a ZF apresentaram microestruturas distintas. A ZF apresentou uma morfologia dendrítica grosseira com a presença de precipitados e porosidades. A análise química semiquantitativa por EDS da ZF mostrou-se bastante semelhante aos resultados obtidos para o MB. Nos corpos de prova unidos pelo processo de brasagem, a média de microdureza Vickers para o MB, a ZTA e a ZF foram respectivamente 83,501, 87,782 e 127,520. A análise estatística t-Student revelou diferenças significativas entre as regiões MB e ZF. No processo TIG, a microestrutura da ZF apresentou aspecto semelhante a do MB, porém, com o arranjo dendrítico mais refinado. A análise química semiquantitativa por EDS da ZF apresentou resultados semelhantes aos obtidos para o MB. Os corpos de prova apresentaram média de microdureza Vickers para o MB, a ZTA e a ZF respectivamente de: 74,607, 71,649 e 75,409. A análise estatística t-Student não revelou diferenças significativas entre as regiões analisadas. De acordo com a metodologia aplicada neste estudo e apartir da análise estatística dos dados obtidos, pode-se concluir que: o processo de união por brasagem gerou diferenças microestruturais e de microdureza Vickers entre a ZF e o MB; nas juntas soldadas pelo método TIG, os ensaios de microdureza Vickers revelaram uniformidade no comportamento mecânico entre a ZF, a ZTA e o MB; o processo de soldagem TIG gerou uma microestrutura bem semelhante entre a ZF e o MB e acarretou naquela apenas um arranjo dendrítico de estrutura eutética mais refinado; a análise química por EDS revelou que não houve variação de composição química significativa entre o MB e a ZF das juntas soldadas por ambos os métodos. / The purpose of this study was to compare, by means of microhardness testing and metallurgical characterization techniques as metallography, optical microscopy, scanning electron microscopy (SEM) and energy dispersion spectroscopy (EDS), mechanical and metallurgical properties of welded joints by TIG and brazing methods in a dental alloy based on Ni-Cr (FIT-CAST SB). The studied regions in each sample were the base metal (MB), the heat affected zone (ZTA) and the fusion zone (ZF). Forty cylindrical bodies-of-proof, stick-shaped, with 2.5mm diameter and 18.0mm in length were made from dental casting by centrifugation (lost wax technique). Twenty cylinders were used for each technique, which formed, after the welding, groups of ten samples for each procedure. The microstructure of the MB is presented as a matrix of a solid solution based on Ni-Cr with a typical dendritic arrangement of eutectic structure. The alloy presented metallic precipitates in the interdendritic region of the eutectic structure. The semiquantitative chemical analysis obtained by EDS technique showed results consistent with the quantitative chemical composition provided by the alloy manufacturer. In the brazing method, the MB and ZF showed different microstructures. The ZF showed a rough dendritic morphology with the presence of precipitates and porosities. The semiquantitative chemical analysis by EDS of the ZF had very similar results to those obtained for the MB. In the bodies-of-proof joined by the brazing process, the average Vickers microhardness for the MB, ZTA and ZF were respectively 83.501, 87.782 and 127.520. The t-Student statistical analysis revealed significant differences between the MB and ZF regions. In the TIG process, the microstructure of ZF presented similar aspect to the MB, but with a more refined dendritic arrangement. The semiquantitative chemical analysis by EDS of ZF presented similar results to those obtained for the MB. The bodies-of-proof had an average Vickers microhardness of the MB, ZTA and ZF respectively: 74.607, 71.649 and 75.409. The t-Student statistical analysis revealed no significant differences between the studied regions. According to the methodology applied in this study and from the statistical analysis of the obtained data, we can conclude that the process of union by brazing created microstructural and Vickers microhardness differences between the ZF and MB; in welded joints by the TIG method, the Vickers microhardness tests showed uniformity in the mechanical behavior among the ZF, ZTA and MB; the TIG welding process generated a very similar microstructure between the ZF and MB and made on that just a dendritic arrangement of more refined eutectic structure; the EDS chemical analysis revealed that there was no significant change in the chemical composition between the MB and ZF of welded joints by both methods.

O objetivo deste trabalho foi comparar, por meio de teste de microdureza e técnicas de caracterização metalúrgica, que inclui a metalografia, microscopia óptica, microscopia eletrônica de varredura (MEV) e espectroscopia por dispersão de energia (energy dispersive spectroscopy EDS), propriedades mecânicas e metalúrgicas de juntas soldadas pelos métodos TIG e brasagem em uma liga odontológica à base de Ni-Cr (FIT-CAST SB). As regiões estudadas em cada amostra foram o metal de base (MB), a zona termicamente afetada (ZTA) e a zona de fusão (ZF). Foram confeccionados 40 corpos-de-prova cilíndricos, em forma de bastão, com 2,5mm de diâmetro e 18,0mm de comprimento, obtidos apartir de fundição odontológica por centrifugação (técnica da cera perdida). Foram separados vinte cilindros para cada técnica, que após a soldagem formaram grupos de dez amostras para cada procedimento. A microestrutura do MB apresentou-se como uma matriz de uma solução sólida à base de Ni-Cr contendo um típico arranjo dendrítico de estrutura eutética. A liga apresentou precipitados metálicos na região interdendrítica da estrutura eutética. A análise química semi-quantitativa obtida através da técnica de EDS apresentou resultados compatíveis com a composição química quantitativa fornecida pelo fabricante da liga. Na brasagem, o MB e a ZF apresentaram microestruturas distintas. A ZF apresentou uma morfologia dendrítica grosseira com a presença de precipitados e porosidades. A análise química semiquantitativa por EDS da ZF mostrou-se bastante semelhante aos resultados obtidos para o MB. Nos corpos de prova unidos pelo processo de brasagem, a média de microdureza Vickers para o MB, a ZTA e a ZF foram respectivamente 83,501, 87,782 e 127,520. A análise estatística t-Student revelou diferenças significativas entre as regiões MB e ZF. No processo TIG, a microestrutura da ZF apresentou aspecto semelhante a do MB, porém, com o arranjo dendrítico mais refinado. A análise química semiquantitativa por EDS da ZF apresentou resultados semelhantes aos obtidos para o MB. Os corpos de prova apresentaram média de microdureza Vickers para o MB, a ZTA e a ZF respectivamente de: 74,607, 71,649 e 75,409. A análise estatística t-Student não revelou diferenças significativas entre as regiões analisadas. De acordo com a metodologia aplicada neste estudo e apartir da análise estatística dos dados obtidos, pode-se concluir que: o processo de união por brasagem gerou diferenças microestruturais e de microdureza Vickers entre a ZF e o MB; nas juntas soldadas pelo método TIG, os ensaios de microdureza Vickers revelaram uniformidade no comportamento mecânico entre a ZF, a ZTA e o MB; o processo de soldagem TIG gerou uma microestrutura bem semelhante entre a ZF e o MB e acarretou naquela apenas um arranjo dendrítico de estrutura eutética mais refinado; a análise química por EDS revelou que não houve variação de composição química significativa entre o MB e a ZF das juntas soldadas por ambos os métodos. / The purpose of this study was to compare, by means of microhardness testing and metallurgical characterization techniques as metallography, optical microscopy, scanning electron microscopy (SEM) and energy dispersion spectroscopy (EDS), mechanical and metallurgical properties of welded joints by TIG and brazing methods in a dental alloy based on Ni-Cr (FIT-CAST SB). The studied regions in each sample were the base metal (MB), the heat affected zone (ZTA) and the fusion zone (ZF). Forty cylindrical bodies-of-proof, stick-shaped, with 2.5mm diameter and 18.0mm in length were made from dental casting by centrifugation (lost wax technique). Twenty cylinders were used for each technique, which formed, after the welding, groups of ten samples for each procedure. The microstructure of the MB is presented as a matrix of a solid solution based on Ni-Cr with a typical dendritic arrangement of eutectic structure. The alloy presented metallic precipitates in the interdendritic region of the eutectic structure. The semiquantitative chemical analysis obtained by EDS technique showed results consistent with the quantitative chemical composition provided by the alloy manufacturer. In the brazing method, the MB and ZF showed different microstructures. The ZF showed a rough dendritic morphology with the presence of precipitates and porosities. The semiquantitative chemical analysis by EDS of the ZF had very similar results to those obtained for the MB. In the bodies-of-proof joined by the brazing process, the average Vickers microhardness for the MB, ZTA and ZF were respectively 83.501, 87.782 and 127.520. The t-Student statistical analysis revealed significant differences between the MB and ZF regions. In the TIG process, the microstructure of ZF presented similar aspect to the MB, but with a more refined dendritic arrangement. The semiquantitative chemical analysis by EDS of ZF presented similar results to those obtained for the MB. The bodies-of-proof had an average Vickers microhardness of the MB, ZTA and ZF respectively: 74.607, 71.649 and 75.409. The t-Student statistical analysis revealed no significant differences between the studied regions. According to the methodology applied in this study and from the statistical analysis of the obtained data, we can conclude that the process of union by brazing created microstructural and Vickers microhardness differences between the ZF and MB; in welded joints by the TIG method, the Vickers microhardness tests showed uniformity in the mechanical behavior among the ZF, ZTA and MB; the TIG welding process generated a very similar microstructure between the ZF and MB and made on that just a dendritic arrangement of more refined eutectic structure; the EDS chemical analysis revealed that there was no significant change in the chemical composition between the MB and ZF of welded joints by both methods.

Gegenstand der Arbeit ist die Suche nach einer Technologie zum stoffschlüssigen Fügen von Leichtbaustrukturen, die aus Aluminiumblechen bestehen. Hohe Anforderungen an die Oberflächenqualität der Fügestelle können dabei nur durch den Einsatz eines Lötprozesses erfüllt werden. Es wird theoretisch untersucht, welche werkstoffspezifischen und technischen Schwierigkeiten das Hartlöten von Aluminium im allgemeinen hervorruft. Unter deren Berücksichtigung werden Lösungsansätze getroffen und bewertet. Im Ergebnis dieser Bewertung wird ein Hartlötprozeß modelliert, bei dem unter Verzicht auf Flußmittel ein Festkörperlaser als Wärmequelle dient. Zur Abschätzung der Vorgänge im Fügespalt, in der Oxidhaut an der Aluminumoberfläche sowie der Temperaturverläufe in den Fügeteilen erfolgen Simulationsrechnungen. Einer ersten experimentellen Überprüfung folgen Modifikationen des Prozesses und deren erneute Simulation. Technologische Versuche testen den Einsatz besonderer Lotwerkstoffe sowie eines zusätzlichen Aktivierungslasers. Metallographische Untersuchungen gelöteter Probebleche geben Aufschluß über den Stoffschluß, der trotz sehr kurzer, lötuntypischer Prozessdauer und ohne Anschmelzung des Grundwerkstoffs eintritt. Zum Abschluß wird in einigen grundlegenden Versuchen überprüft, inwiefern anhand der gewonnenen Erkenntnisse auch ohne den hohen technischen Aufwand des Hochleistungslasers ein flußmittelfreier Hartlötprozeß realisierbar ist.

The diploma work is focused to evaluation of contemporary state of sealing system solution of high speed turbine engines in the world and, newly suggesting nontraditional design of labyrinth sealing lining for turbine side of small turbojet engine, including the production technology, technical-economic classification and experimental verification.

HABÁSKO Jakub: Production process optimization. The Master´s thesis of Master degree, second grade, school year 2008/2009, FME Brno University of Technology, Institute of Metrology and Quality Assurance Testing, October 2009, pages 68, pictures 40, tables 2, supplements 4. The project elaborated in frame of the Master degree, study branch Metrology and Quality Assurance Testing. This Master´s thesis deals with optimization of process plan. In virtue of submission was created an analysis present condition using of statistical methods and it was created suitable statistical methods applicable in production organization. Then is here described the problem of brazing bells, because in this place is originating the most of uptightness. Then the thesis brings recommendation and findings, which can help with optimization this process.