L’électronique de puissance est un domaine en mutation. Les environnements et les conditions de fonctionnement des modules de puissance sont de plus en plus sévères : hautes températures, tension et courant élevés. De plus, le frittage d’argent a été introduit dans les modules, en remplacement des joints brasés principalement composés de plomb. C’est la combinaison de ces évolutions qui ont motivés nos travaux. Dans l’objectif d’améliorer la conception des modules de puissance, ces travaux proposent d’augmenter la dissipation des modules grâce aux structures verticales-3D. Un onduleur triphasé vertical-3D a été conçut avec des puces Si reportées par frittage d’argent. Des caractérisations électriques et thermiques ont été réalisées et ont permis de montrer l’apport de cette technologie. Cette étude est couplée à des simulations numériques thermiques et électrostatiques permettant de mettre en lumière les enjeux de cette méthode d’assemblage. / Power electronics is a changing field. The environments and operating conditions of power modules are more severe: higher temperature, higher voltage and higher current. In addition, silver sintering was introduced in power modules to replace solders composed by lead. The combinations of these developments have motivated our work. In order to improve the design of the power modules our researches purpose to increase the dissipation of power modules with a3D-vertical structure. A three-phase inverter with3D-vertical structure has been designed with a Si dice sintered. Thermal and electrical characterizations were performed and allowed to show the contribution of this technology. This study is coupled to thermal and electrostatic Finite Element Method simulations to highlight and improve the possible issues of this assembly.

Nanomaterials have attracted considerable attention during the past decades due to their unique and fascinating properties. However, this class of materials is not an invention of modern age. People have been using nanomaterials for centuries, although unwittingly. Probably the most famous example for the usage of nanomaterials in ancient times is the Lycurgus Cup, a Roman glass cage cup created in the 4th century which changes the colour of its glass from green to ruby depending on the illumination conditions. The foundation for the development of the field of nanotechnology was laid by the speech of Feynman “There is plenty of room at the bottom” in 1959, in which he spoke about the principles of miniaturisation as low as to the atomic level. Today, modern nanotechnology made it its business to purposefully develop and synthesise nanomaterials as well as to face their applications in various fields, such as microelectronics, catalysis or biomedicine. However, the term “nanomaterials” does not solely involve the nanoparticulate units itself, but also their arrangement into two- or three-dimensional structures. Thereby, the maintenance of the nanoscale properties is one of the main challenges. This task was focussed by this work implied the preparation and macroscale arrangement of fluorescent QDs while preserving their optical properties. The main achievement of this work was the development of a novel aerogel material with non-quenching PL behaviour by using silica coated QDs as nanoparticulate building units. In comparison to other monolithic silica-QD structures or aerogels from pure QDs, a defined and controllable distance between the fluorescent QDs is provided in these structures by the silica shell. The spacing was shown to efficiently disable energy transfers so that no spectral shifts, lifetime shortening or PL QY losses are observed during the colloid to gel transition. The silica shell, established by a standard reverse microemulsion approach, was found to exhibit a certain porosity, which was proven by gas adsorption measurements. Existing cavities in the micro- and mesoporous range were found to allow small species such as metal ions to pass through the shell and interact with the QD core causing a detectable change of the PL intensity, which makes these materials suitable for future sensing applications. The gel preparation was based on a metal ion assisted complexation approach, which requires tetrazole functionalisation of the nanoparticulate building units. A major development in this work that permitted this gelation approach for silica-QDs was the development of a novel tetrazole-silane ligand. TMSPAMTz was specifically designed to bind to the silica surface of silica-QDs in aqueous solution and was prepared by a covalent coupling of an alkyl chained silane with a 5-subsituted tetrazole ring. Network formation is subsequently achieved by the interconnection of negatively charged tetrazole rings with metal ions, which allows for a broad spectrum of aerogel materials from different NP species as well as their mixtures as long as tetrazole capping is provided. Considering this diversity and the disabling of energy transfers, straightforward colour tuning was demonstrated herein by mixing differently emitting silica-QD species which gives great prospects for lighting applications. Furthermore, the possibility of plasmon enhanced emission was presented for mixed Au NP/silica-QD gels. With respect to future sensing applications, thin porous films from silica-QDs gels were prepared, which showed a promising concentration dependant PL quenching for the model analyst hydrogen peroxide. However, the film reproducibility of the applied drop-cast coating method was insufficient. As a suggestion to this, a LbL method was presented, wherein a gel is subsequently grown with the metal ion assisted complexation approach. In addition to the tetrazole ligands on the NP surface, tetrazole-silane ligands were used in this approach to functionalise the glass substrate surface. By this, homogeneous gel films of distinct thickness can be grown while the use of organic polymers can be completely avoided. Besides the preparation of NP assemblies, standard Cd-based QD materials as well as Au NPs of different sizes and shape, recent progresses in the synthesis of InP-based QDs were presented in this work. A thorough investigation and understanding of the growth influencing parameters allowed for the establishment of preparation routes for In(Zn)P/GaP/ZnS core/shell/shell QDs with emission wavelengths tuneable within a large range from 500 to 650 nm, narrow peak widths of 45 to 70 nm and PL QYs up to 60%. Successful incorporation of these QDs into salt matrices was further demonstrated. The resulting composite materials are very photostable and suitable as colour conversion materials for solid state lighting, as was clearly pointed out by a self-prepared WLED that met the standard commercial LEDs.

Nanomaterials have attracted considerable attention during the past decades due to their unique and fascinating properties. However, this class of materials is not an invention of modern age. People have been using nanomaterials for centuries, although unwittingly. Probably the most famous example for the usage of nanomaterials in ancient times is the Lycurgus Cup, a Roman glass cage cup created in the 4th century which changes the colour of its glass from green to ruby depending on the illumination conditions. The foundation for the development of the field of nanotechnology was laid by the speech of Feynman “There is plenty of room at the bottom” in 1959, in which he spoke about the principles of miniaturisation as low as to the atomic level. Today, modern nanotechnology made it its business to purposefully develop and synthesise nanomaterials as well as to face their applications in various fields, such as microelectronics, catalysis or biomedicine. However, the term “nanomaterials” does not solely involve the nanoparticulate units itself, but also their arrangement into two- or three-dimensional structures. Thereby, the maintenance of the nanoscale properties is one of the main challenges. This task was focussed by this work implied the preparation and macroscale arrangement of fluorescent QDs while preserving their optical properties. The main achievement of this work was the development of a novel aerogel material with non-quenching PL behaviour by using silica coated QDs as nanoparticulate building units. In comparison to other monolithic silica-QD structures or aerogels from pure QDs, a defined and controllable distance between the fluorescent QDs is provided in these structures by the silica shell. The spacing was shown to efficiently disable energy transfers so that no spectral shifts, lifetime shortening or PL QY losses are observed during the colloid to gel transition. The silica shell, established by a standard reverse microemulsion approach, was found to exhibit a certain porosity, which was proven by gas adsorption measurements. Existing cavities in the micro- and mesoporous range were found to allow small species such as metal ions to pass through the shell and interact with the QD core causing a detectable change of the PL intensity, which makes these materials suitable for future sensing applications. The gel preparation was based on a metal ion assisted complexation approach, which requires tetrazole functionalisation of the nanoparticulate building units. A major development in this work that permitted this gelation approach for silica-QDs was the development of a novel tetrazole-silane ligand. TMSPAMTz was specifically designed to bind to the silica surface of silica-QDs in aqueous solution and was prepared by a covalent coupling of an alkyl chained silane with a 5-subsituted tetrazole ring. Network formation is subsequently achieved by the interconnection of negatively charged tetrazole rings with metal ions, which allows for a broad spectrum of aerogel materials from different NP species as well as their mixtures as long as tetrazole capping is provided. Considering this diversity and the disabling of energy transfers, straightforward colour tuning was demonstrated herein by mixing differently emitting silica-QD species which gives great prospects for lighting applications. Furthermore, the possibility of plasmon enhanced emission was presented for mixed Au NP/silica-QD gels. With respect to future sensing applications, thin porous films from silica-QDs gels were prepared, which showed a promising concentration dependant PL quenching for the model analyst hydrogen peroxide. However, the film reproducibility of the applied drop-cast coating method was insufficient. As a suggestion to this, a LbL method was presented, wherein a gel is subsequently grown with the metal ion assisted complexation approach. In addition to the tetrazole ligands on the NP surface, tetrazole-silane ligands were used in this approach to functionalise the glass substrate surface. By this, homogeneous gel films of distinct thickness can be grown while the use of organic polymers can be completely avoided. Besides the preparation of NP assemblies, standard Cd-based QD materials as well as Au NPs of different sizes and shape, recent progresses in the synthesis of InP-based QDs were presented in this work. A thorough investigation and understanding of the growth influencing parameters allowed for the establishment of preparation routes for In(Zn)P/GaP/ZnS core/shell/shell QDs with emission wavelengths tuneable within a large range from 500 to 650 nm, narrow peak widths of 45 to 70 nm and PL QYs up to 60%. Successful incorporation of these QDs into salt matrices was further demonstrated. The resulting composite materials are very photostable and suitable as colour conversion materials for solid state lighting, as was clearly pointed out by a self-prepared WLED that met the standard commercial LEDs.

The work presented in the thesis is focused on the synthesis of diverse colloidal semiconductor NCs in organic media, their surface design with tiny inorganic and hybrid capping species in solution phase, and subsequent assembling of these NC building units into two-dimensional close-packed thin-films and three-dimensional non-ordered porous superstructures.

The work presented in the thesis is focused on the synthesis of diverse colloidal semiconductor NCs in organic media, their surface design with tiny inorganic and hybrid capping species in solution phase, and subsequent assembling of these NC building units into two-dimensional close-packed thin-films and three-dimensional non-ordered porous superstructures.