Este trabalho teve como objetivo obter nanocompósitos de nanotubos de carbono de paredes múltiplas (NTCPMs) com TiO2, e caracterizá-los quanto a sua estrutura, características ópticas e atividade fotocatalítica. Os nanocompósitos foram obtidos a partir de NTCPMs comerciais (Baytubes®), e dois diferentes TiO2: um comercial (P25) e um obtido na síntese de TiO2 tendo tetra propóxido de titânio (TTP) como precursor. Foram utilizados dois diferentes sistemas líquidos para a obtenção dos nanocompósitos NTCPM-TiO2: um, em pH ácido e outro, em pH alcalino. Os nanocompósitos obtidos a partir do TTP foram posteriormente tratados termicamente a 400 °C, 500 °C, 600 °C e 700 °C para formação de fases cristalinas de TiO2. Os nanocompósitos foram investigados quanto a sua atividade fotocatalítica, empregando-os como catalisadores na degradação do corante orgânico alaranjado de metila, em solução aquosa, sob radiação ultravioleta. Os resultados foram associados a características da estrutura dos nanocompósitos, utilizando técnicas como difração de raios X, microscopia eletrônica de varredura, microscopia eletrônica de transmissão, espectroscopia Raman e espectroscopia por infravermelho e área superficial específica. A caracterização óptica foi obtida por espectroscopia fotoluminescente e espectroscopia por refletância difusa. A análise térmica foi empregada para quantificar a presença de NTCPMs no nanocompósito empregado como catalisador. O desempenho fotocatalítico dos nanocompósitos foi correlacionado com o efeito do pH dos sistemas líquidos empregados na sua obtenção, natureza da interação (química e/ou física) entre nanotubo de carbono e TiO2, fases presentes no TiO2, energia do gap óptico e presença de defeitos estruturais no TiO2. A maior eficiência na fotocatálise foi observada nos nanocompósitos NTCPMs-TiO2 obtidos a partir do TiO2 comercial, e nos obtidos a partir do precursor TTP tratado termicamente a 500 °C, ambos em meio ácido. Estes resultados puderam ser associados às menores energias de transição e nível de defeitos no TiO2 nesses nanocompósitos, quando comparados aos demais. / This study aimed to obtain nanocomposites from multi-walled carbon nanotubes (MWCNTs) with TiO2, and characterize them according to their structure, optical properties and photocatalytic activity. The nanocomposites were obtained from commercial MWCNTs (Baytubes®) and two different types of TiO2: a commercial one (P25) and one obtained by synthesizing TiO2 with titanium tetra propoxide (TTP) as a precursor. Two different fluid systems were used for obtaining the MWCNT-TiO2 nanocomposites: one with acid pH and the other with alkaline pH. The nanocomposites obtained from TTP were subsequently heat treated at 400 °C, 500 °C, 600 °C and 700 °C to form crystalline phases of TiO2. The nanocomposites were investigated for their photocatalytic activity, employing them as catalysts in the degradation of organic methyl orange dye in an aqueous solution under ultraviolet radiation. The results were associated with the characteristics of the nanocomposites’ structure, using techniques such as X-ray diffraction, scanning electron microscopy, transmission electron microscopy, Raman spectroscopy, infrared spectroscopy, and specific surface area. Optical characterization was obtained by photoluminescence spectroscopy and diffuse reflectance spectroscopy. Thermal analysis was used to quantify the presence of MWCNTs in the nanocomposite employed as catalyst. The photocatalytic performance of the nanocomposites were correlated with the effect of the pH of the liquid systems employed for obtaining them, the nature of the interaction (chemical and/or physical) between the carbon nanotube and TiO2, the phases present in the TiO2, the optical energy gap and the presence of structural defects in TiO2. The highest photocatalytic efficiency was observed in the MWCNT-TiO2 nanocomposites obtained from commercial TiO2, and in those obtained from the TTP precursor heat treated at 500 °C, both in an acid medium. These results could be associated with the lower transition energy and level of defects in the TiO2 of these nanocomposites when compared to the other samples.

Este trabalho teve como objetivo obter nanocompósitos de nanotubos de carbono de paredes múltiplas (NTCPMs) com TiO2, e caracterizá-los quanto a sua estrutura, características ópticas e atividade fotocatalítica. Os nanocompósitos foram obtidos a partir de NTCPMs comerciais (Baytubes®), e dois diferentes TiO2: um comercial (P25) e um obtido na síntese de TiO2 tendo tetra propóxido de titânio (TTP) como precursor. Foram utilizados dois diferentes sistemas líquidos para a obtenção dos nanocompósitos NTCPM-TiO2: um, em pH ácido e outro, em pH alcalino. Os nanocompósitos obtidos a partir do TTP foram posteriormente tratados termicamente a 400 °C, 500 °C, 600 °C e 700 °C para formação de fases cristalinas de TiO2. Os nanocompósitos foram investigados quanto a sua atividade fotocatalítica, empregando-os como catalisadores na degradação do corante orgânico alaranjado de metila, em solução aquosa, sob radiação ultravioleta. Os resultados foram associados a características da estrutura dos nanocompósitos, utilizando técnicas como difração de raios X, microscopia eletrônica de varredura, microscopia eletrônica de transmissão, espectroscopia Raman e espectroscopia por infravermelho e área superficial específica. A caracterização óptica foi obtida por espectroscopia fotoluminescente e espectroscopia por refletância difusa. A análise térmica foi empregada para quantificar a presença de NTCPMs no nanocompósito empregado como catalisador. O desempenho fotocatalítico dos nanocompósitos foi correlacionado com o efeito do pH dos sistemas líquidos empregados na sua obtenção, natureza da interação (química e/ou física) entre nanotubo de carbono e TiO2, fases presentes no TiO2, energia do gap óptico e presença de defeitos estruturais no TiO2. A maior eficiência na fotocatálise foi observada nos nanocompósitos NTCPMs-TiO2 obtidos a partir do TiO2 comercial, e nos obtidos a partir do precursor TTP tratado termicamente a 500 °C, ambos em meio ácido. Estes resultados puderam ser associados às menores energias de transição e nível de defeitos no TiO2 nesses nanocompósitos, quando comparados aos demais. / This study aimed to obtain nanocomposites from multi-walled carbon nanotubes (MWCNTs) with TiO2, and characterize them according to their structure, optical properties and photocatalytic activity. The nanocomposites were obtained from commercial MWCNTs (Baytubes®) and two different types of TiO2: a commercial one (P25) and one obtained by synthesizing TiO2 with titanium tetra propoxide (TTP) as a precursor. Two different fluid systems were used for obtaining the MWCNT-TiO2 nanocomposites: one with acid pH and the other with alkaline pH. The nanocomposites obtained from TTP were subsequently heat treated at 400 °C, 500 °C, 600 °C and 700 °C to form crystalline phases of TiO2. The nanocomposites were investigated for their photocatalytic activity, employing them as catalysts in the degradation of organic methyl orange dye in an aqueous solution under ultraviolet radiation. The results were associated with the characteristics of the nanocomposites’ structure, using techniques such as X-ray diffraction, scanning electron microscopy, transmission electron microscopy, Raman spectroscopy, infrared spectroscopy, and specific surface area. Optical characterization was obtained by photoluminescence spectroscopy and diffuse reflectance spectroscopy. Thermal analysis was used to quantify the presence of MWCNTs in the nanocomposite employed as catalyst. The photocatalytic performance of the nanocomposites were correlated with the effect of the pH of the liquid systems employed for obtaining them, the nature of the interaction (chemical and/or physical) between the carbon nanotube and TiO2, the phases present in the TiO2, the optical energy gap and the presence of structural defects in TiO2. The highest photocatalytic efficiency was observed in the MWCNT-TiO2 nanocomposites obtained from commercial TiO2, and in those obtained from the TTP precursor heat treated at 500 °C, both in an acid medium. These results could be associated with the lower transition energy and level of defects in the TiO2 of these nanocomposites when compared to the other samples.

A aplicação de Ensaios Não Destrutivos (END) é considerada uma alternativa para monitorar o estado das estruturas de concreto armado, principalmente quando ocorrem falhas de projeto, erros de dosagem, problemas na mistura, transporte ou lançamento do concreto. A medida da Velocidade de Pulso Ultrassônico (VPU) é um END que vem sendo usado comumente em diversas áreas da engenharia civil, devido à facilidade da operação, ao seu baixo custo, à rapidez na execução dos testes e à ausência de danos ao material analisado. O objetivo deste trabalho consiste em apresentar o resultado de uma pesquisa experimental, que visou estudar a influência que determinadas variáveis operacionais exercem sobre os resultados obtidos nos ensaios de VPU e qual a real capacidade de penetração e detecção de defeitos de diversas magnitudes. As medições foram realizadas utilizando o ultrassom pelo modo de transmissão indireta, pois esta é a alternativa que mais se adapta às condições verificadas no exame de estruturas reais de concreto armado. Primeiramente, foram moldados em laboratório dois blocos de concreto, com dimensões próximas às encontradas em estruturas reais de grande porte: um com armadura nas duas faces e outro sem armadura. Em ambos os elementos foram introduzidos objetos para simular possíveis falhas de concretagem. Em uma segunda etapa, foram realizadas perfurações visando identificar a capacidade de detecção de vazios da técnica de VPU para diferentes profundidades. Os resultados são apresentados na forma de imagens geradas através de software e por meio de analises estatísticas. Este estudo verificou que a escolha dos parâmetros de ensaio é um fator determinante para permitir a interpretação correta dos resultados, salientando-se que a experiência do operador e as dimensões de grid são os parâmetros que mais interferem na interpretação dos dados. / The use of Non-destructive Testing (NDT) methods is considered an alternative to monitor the behaviour of reinforced concrete structures, especially when there are design problems or errors related to concrete mixing, transport, placement and casting errors. The Ultrasonic Pulse Velocity (UPV) method is a Non-Destructive Test that is gaining acceptance. The test has been commonly used in various civil engineering areas due to the fact that it is easy to operate and rapid to perform, and to the low cost and lack of damage to the tested material. The objective of this work is to present the results of an experimental research that aimed to study the influence of certain technological variables on the measurements obtained from UPV tests. The work also aims to gather data to assess the penetration capacity of the US waves and their ability to detect defects of various dimensions. The measurements were performed using the indirect transmission mode, to simulate real work conditions on large structures. Firstly, two massive concrete blocks were cast in the laboratory, with dimensions close to real life structures: one with steel reinforcement on both sides and the other without reinforcement. Objects were introduced in both elements to simulate possible defects due to casting. In a second stage, perforations were conducted in the concrete elements to identify the void detection capability of the UPV technique for different depths. The results are presented by images generated by specific software and by using statistical analysis. This study found that the test parameters influence significantly the correct interpretation of the results. The operator experience and the grid dimensions are the parameters that most influence UPV data analysis.

A aplicação de Ensaios Não Destrutivos (END) é considerada uma alternativa para monitorar o estado das estruturas de concreto armado, principalmente quando ocorrem falhas de projeto, erros de dosagem, problemas na mistura, transporte ou lançamento do concreto. A medida da Velocidade de Pulso Ultrassônico (VPU) é um END que vem sendo usado comumente em diversas áreas da engenharia civil, devido à facilidade da operação, ao seu baixo custo, à rapidez na execução dos testes e à ausência de danos ao material analisado. O objetivo deste trabalho consiste em apresentar o resultado de uma pesquisa experimental, que visou estudar a influência que determinadas variáveis operacionais exercem sobre os resultados obtidos nos ensaios de VPU e qual a real capacidade de penetração e detecção de defeitos de diversas magnitudes. As medições foram realizadas utilizando o ultrassom pelo modo de transmissão indireta, pois esta é a alternativa que mais se adapta às condições verificadas no exame de estruturas reais de concreto armado. Primeiramente, foram moldados em laboratório dois blocos de concreto, com dimensões próximas às encontradas em estruturas reais de grande porte: um com armadura nas duas faces e outro sem armadura. Em ambos os elementos foram introduzidos objetos para simular possíveis falhas de concretagem. Em uma segunda etapa, foram realizadas perfurações visando identificar a capacidade de detecção de vazios da técnica de VPU para diferentes profundidades. Os resultados são apresentados na forma de imagens geradas através de software e por meio de analises estatísticas. Este estudo verificou que a escolha dos parâmetros de ensaio é um fator determinante para permitir a interpretação correta dos resultados, salientando-se que a experiência do operador e as dimensões de grid são os parâmetros que mais interferem na interpretação dos dados. / The use of Non-destructive Testing (NDT) methods is considered an alternative to monitor the behaviour of reinforced concrete structures, especially when there are design problems or errors related to concrete mixing, transport, placement and casting errors. The Ultrasonic Pulse Velocity (UPV) method is a Non-Destructive Test that is gaining acceptance. The test has been commonly used in various civil engineering areas due to the fact that it is easy to operate and rapid to perform, and to the low cost and lack of damage to the tested material. The objective of this work is to present the results of an experimental research that aimed to study the influence of certain technological variables on the measurements obtained from UPV tests. The work also aims to gather data to assess the penetration capacity of the US waves and their ability to detect defects of various dimensions. The measurements were performed using the indirect transmission mode, to simulate real work conditions on large structures. Firstly, two massive concrete blocks were cast in the laboratory, with dimensions close to real life structures: one with steel reinforcement on both sides and the other without reinforcement. Objects were introduced in both elements to simulate possible defects due to casting. In a second stage, perforations were conducted in the concrete elements to identify the void detection capability of the UPV technique for different depths. The results are presented by images generated by specific software and by using statistical analysis. This study found that the test parameters influence significantly the correct interpretation of the results. The operator experience and the grid dimensions are the parameters that most influence UPV data analysis.

Este trabalho teve como objetivo obter nanocompósitos de nanotubos de carbono de paredes múltiplas (NTCPMs) com TiO2, e caracterizá-los quanto a sua estrutura, características ópticas e atividade fotocatalítica. Os nanocompósitos foram obtidos a partir de NTCPMs comerciais (Baytubes®), e dois diferentes TiO2: um comercial (P25) e um obtido na síntese de TiO2 tendo tetra propóxido de titânio (TTP) como precursor. Foram utilizados dois diferentes sistemas líquidos para a obtenção dos nanocompósitos NTCPM-TiO2: um, em pH ácido e outro, em pH alcalino. Os nanocompósitos obtidos a partir do TTP foram posteriormente tratados termicamente a 400 °C, 500 °C, 600 °C e 700 °C para formação de fases cristalinas de TiO2. Os nanocompósitos foram investigados quanto a sua atividade fotocatalítica, empregando-os como catalisadores na degradação do corante orgânico alaranjado de metila, em solução aquosa, sob radiação ultravioleta. Os resultados foram associados a características da estrutura dos nanocompósitos, utilizando técnicas como difração de raios X, microscopia eletrônica de varredura, microscopia eletrônica de transmissão, espectroscopia Raman e espectroscopia por infravermelho e área superficial específica. A caracterização óptica foi obtida por espectroscopia fotoluminescente e espectroscopia por refletância difusa. A análise térmica foi empregada para quantificar a presença de NTCPMs no nanocompósito empregado como catalisador. O desempenho fotocatalítico dos nanocompósitos foi correlacionado com o efeito do pH dos sistemas líquidos empregados na sua obtenção, natureza da interação (química e/ou física) entre nanotubo de carbono e TiO2, fases presentes no TiO2, energia do gap óptico e presença de defeitos estruturais no TiO2. A maior eficiência na fotocatálise foi observada nos nanocompósitos NTCPMs-TiO2 obtidos a partir do TiO2 comercial, e nos obtidos a partir do precursor TTP tratado termicamente a 500 °C, ambos em meio ácido. Estes resultados puderam ser associados às menores energias de transição e nível de defeitos no TiO2 nesses nanocompósitos, quando comparados aos demais. / This study aimed to obtain nanocomposites from multi-walled carbon nanotubes (MWCNTs) with TiO2, and characterize them according to their structure, optical properties and photocatalytic activity. The nanocomposites were obtained from commercial MWCNTs (Baytubes®) and two different types of TiO2: a commercial one (P25) and one obtained by synthesizing TiO2 with titanium tetra propoxide (TTP) as a precursor. Two different fluid systems were used for obtaining the MWCNT-TiO2 nanocomposites: one with acid pH and the other with alkaline pH. The nanocomposites obtained from TTP were subsequently heat treated at 400 °C, 500 °C, 600 °C and 700 °C to form crystalline phases of TiO2. The nanocomposites were investigated for their photocatalytic activity, employing them as catalysts in the degradation of organic methyl orange dye in an aqueous solution under ultraviolet radiation. The results were associated with the characteristics of the nanocomposites’ structure, using techniques such as X-ray diffraction, scanning electron microscopy, transmission electron microscopy, Raman spectroscopy, infrared spectroscopy, and specific surface area. Optical characterization was obtained by photoluminescence spectroscopy and diffuse reflectance spectroscopy. Thermal analysis was used to quantify the presence of MWCNTs in the nanocomposite employed as catalyst. The photocatalytic performance of the nanocomposites were correlated with the effect of the pH of the liquid systems employed for obtaining them, the nature of the interaction (chemical and/or physical) between the carbon nanotube and TiO2, the phases present in the TiO2, the optical energy gap and the presence of structural defects in TiO2. The highest photocatalytic efficiency was observed in the MWCNT-TiO2 nanocomposites obtained from commercial TiO2, and in those obtained from the TTP precursor heat treated at 500 °C, both in an acid medium. These results could be associated with the lower transition energy and level of defects in the TiO2 of these nanocomposites when compared to the other samples.

Structural Health Monitoring (SHM) techniques are widely used in a number of Non – destructive Evaluation (NDE) applications. There is a need to develop effective techniques for SHM, so that the safety and integrity of the structures can be improved. Two most widely used SHM methods for plates and rods use either the spectrum of the impedances or monitor the propagation of lamb waves. Piezoelectric wafer – active sensors (PWAS) were used for excitation and sensing. In this study, surface response to excitation (SuRE) and Lamb wave propagation was monitored to estimate the integrity of the round objects including the pipes, tubes and cutting tools. SuRE obtained the frequency response by applying sweep sine wave to surface. The envelope of the received signal was used to detect the arrival of lamb waves to the sensor. Both approaches detect the structural defects of the pipes and tubes and the wear of the cutting tool.

Malgré une vaste littérature concernant les propriétés structurelles, électroniques et ther- modynamiques du silicium amorphe (a-Si), la structure microscopique de ce semi-cond- ucteur covalent échappe jusqu’à ce jour à une description exacte. Plusieurs questions demeurent en suspens, concernant par exemple la façon dont le désordre est distribué à travers la matrice amorphe : uniformément ou au sein de petites régions hautement déformées ? D’autre part, comment ce matériau relaxe-t-il : par des changements homo- gènes augmentant l’ordre à moyenne portée, par l’annihilation de défauts ponctuels ou par une combinaison de ces phénomènes ? Le premier article présenté dans ce mémoire propose une caractérisation des défauts de coordination, en terme de leur arrangement spatial et de leurs énergies de formation. De plus, les corrélations spatiales entre les défauts structurels sont examinées en se ba- sant sur un paramètre qui quantifie la probabilité que deux sites défectueux partagent un lien. Les géométries typiques associées aux atomes sous et sur-coordonnés sont extraites du modèle et décrites en utilisant les distributions partielles d’angles tétraédriques. L’in- fluence de la relaxation induite par le recuit sur les défauts structurels est également analysée. Le second article porte un regard sur la relation entre l’ordre à moyenne portée et la relaxation thermique. De récentes mesures expérimentales montrent que le silicium amorphe préparé par bombardement ionique, lorsque soumis à un recuit, subit des chan- gements structuraux qui laissent une signature dans la fonction de distribution radiale, et cela jusqu’à des distances correspondant à la troisième couche de voisins.[1, 2] Il n’est pas clair si ces changements sont une répercussion d’une augmentation de l’ordre à courte portée, ou s’ils sont réellement la manifestation d’un ordonnement parmi les angles dièdres, et cette section s’appuie sur des simulations numériques d’implantation ionique et de recuit, afin de répondre à cette question. D’autre part, les corrélations entre les angles tétraédriques et dièdres sont analysées à partir du modèle de a-Si. / Based on a detailed study of the radial distribution function (RDF) of a model for amorphous sili- con (a-Si), we address the relation between short-range rearrangements and an increase in medium- range order induced by thermal relaxation. Recent experimental measurements have shown that a small peak appears in the RDF around 4.7 Å upon annealing, along with other subtle changes, and this is attributed to ordering among the dihedral angles. We show that, although this is a possible explanation, an increase in short-range order (up to second neighbors) is not only necessary for these changes to occur, but could also be their sole cause. To clarify the nature of disorder in the amorphous system, correlations among dihedral and tetrahedral angles are examined. The bivariate probability distribution of these two variables reveals small correlations between dihedral and te- trahedral angles, associated with the staggered and eclipsed conformations. In the first case, bond angles around 112.5◦ are favored vs. 120◦ in the second case. Bond angles between 95◦ and 100◦ are less probable in both conformations. On another issue, the nature of disorder in amorphous silicon (a-Si) is explored by investigating the spatial ar- rangement and energies of coordination defects in a numerical model. Spatial correlations between structural defects are examined on the basis of a parameter that quantifies the probability for two sites to share a bond. Pentacoordinated atoms are found to be the dominant coordination defects. They show a tendency to cluster, and about 17% of them are linked through three-membered rings. As for tricoordinated sites, they are less numerous, and tend to be distant by at least two bond lengths. Typical local geometries associated to under and overcoordinated atoms are extracted from the model and described using partial bond angle distributions. An estimate of the formation energies of structural defects is provided. Using molecular-dynamics calculations, we simulate the implantation of high-energy atoms in the initial structure in order to study the effect of relaxation on the coordination defects and their environments.

Les semi-conducteurs III-V présentent des propriétés remarquables de mobilité électronique et d’émission optique. La croissance de ces matériaux par MOCVD sur substrat silicium (001) en 300 mm offre l’opportunité de réaliser les composants bas coût, et d’apporter de nouvelles fonctionnalités à la microélectronique silicium par l’intégration de composants électroniques et optiques. De ce fait, la croissance de semi-conducteurs III-V sur silicium a connu un engouement important au cours de ces dernières années. De telles intégrations monolithiques impliquent de développer des couches III-V de très bonne qualité épitaxiées sur silicium. Actuellement, le principal enjeu réside dans la réduction des densités de défauts cristallins tels que les fautes d’empilement, parois d’antiphase et dislocations. L’impact de ces défauts structuraux sur ces propriétés reste encore aujourd’hui principalement étudié à l’échelle micrométrique (effet Hall, photoluminescence,…) ne présentant pas de résolution spatiale suffisante permettant pas de dissocier les différents paramètres physico-chimiques. Ces développements nécessitent en conséquence des moyens adaptés de caractérisation pour contrôler et connaitre les propriétés physico-chimique de ces couches III-V épitaxiée sur Si.Le travail de cette thèse porte sur le développement d’une méthode destinée à corréler spatialement à l’échelle nanométrique les propriétés optiques et les caractéristiques morphologiques de couches III-As crues par MOCVD sur Si (001) en 300mm. Elle a permis d’étudier des puits quantiques d’InGaAs épitaxiés sur buffer GaAs et sur substrat Si avec pour méthode de réduction des défauts émergeants l’utilisation de couches tampons à base de GaAs et la croissance localisée entre murs d’oxydes (aspect ratio trapping (ART)). L’étude s’est appuyée sur deux techniques de caractérisation: La cathodoluminescence (CL) permettant d’observer spatialement sur l’échantillon bulk l’énergie d’émission et l’intensité correspondantes d’un puits quantique d’InGaAs, et la microscopie électronique en transmission (STEM/TEM) en lame mince donnant des informations quantitatives sur la morphologie de la structure (épaisseurs des couches, position des défauts, stoechiométrie…). Nous avons développé une méthode qui permet de corréler spatialement les résultats de ces deux techniques de caractérisation. La méthode consiste en plusieurs marquages spécifiques réalisés par faisceau d’électron pour repérer et extraire précisément les zones d’intérêt observés en cathodoluminescence. Des mesures de déformation (N-PED) sur lame mince ont également été réalisées.Cette méthode de caractérisation corrélée a permis de mettre en évidence des modifications des propriétés physico-chimique de puits quantiques d’InGaAs à l’échelle nanométrique directement liées aux conditions de croissance, et à la présence de défauts émergents. / III-V semiconductors have remarkable properties of electronic mobility and optical emission. Their growth by MOCVD on 300 mm (001) silicon substrate offers the opportunity of manufacturing at low cost and to get new functionalities in microelectronic devices for electronic components and optical emitters. Thereby III-V growth on silicon has been a huge success in recent years. Monolithic integrations imply to develop very good quality of III-V layers epitaxied on silicon. That is why currently, the major issue is to reduce crystalline defect density such as stacking faults, antiphase boundaries and dislocations. Structural defects which impact these properties are still mainly studied at micrometric scale (Hall effect, photoluminescence,…) but they not have sufficient spatial resolution to dissociate the different physicochemical parameters. Consequently, these developments require adapted characterizations to control and to obtain physicochemical properties of III-V epitaxied layers on Si.The aim of this thesis is to develop a method to spatially correlate at nanoscale optical properties and morphological characteristics of III-As layers grown by MOCVD on 300 mm (001) Si. In the frame of this thesis, this method allowed to study InGaAs quantum wells (QW) epitaxied on GaAs buffer and on Si substrate with defect density reduction method the use of GaAs buffer layers and patterned growth (aspect ratio trapping (ART)). The study is based on two characterization techniques : cathodoluminescence (CL) which allows to spatially observe on bulk sample InGaAs QW energy emission and intensity, and transmission electronic microscopy (STEM/TEM) of thin lamella which give quantitative information on the morphology (layer thicknesses, defect positions, stoichiometry,…). The method developed allows to spatially correlate results of these two techniques. It is to realize several specific marks made by electron beam to localize and precisely extract interesting areas observed by CL. Strain measurements (N-PED) on TEM lamella is realized too.This correlated characterization method has highlighted physicochemical property modifications of InGaAs QW at nanoscale directly related to growth conditions and to threading dislocations presence.

La compréhension et le contrôle de la conductivité thermique des couches minces polycristallines est fondamentale pour améliorer la performance et la fiabilité des dispositifs micro- et optoélectroniques. Toutefois, une description et un contrôle précis de la performance thermique de ces matériaux bidimensionnels restent une tâche difficile en raison de leur anisotropie et structure hétérogène. En effet, les couches minces obtenues par diverses techniques et avec une large gamme de paramètres de dépôt, sont composées de petites cristallites à l'interface avec le substrat, qui coalescent et évoluent vers une structure colonnaire à proximité de la surface extérieure du film. Ces grains,ainsi que d'autres défauts cristallographiques, tels que les impuretés d'oxygène,augmentent les processus de dispersion diffuse des porteurs d'énergie dans les matériaux, ce qui en conséquence, réduit considérablement leur conductivité thermique. La caractérisation thermique expérimentale, la description théorique et la modulation contrôlée des propriétés thermiques de ces matériauxs ont, par conséquent, indispensables.Cette thèse est consacrée à l'étude de la conductivité thermique des couches polycristallines présentant une non-homogénéité structurelle et elle a pour but d'explorer la possibilité de moduler le transfert de chaleur à travers ces structures bidimensionnelles. Le nitrure d'aluminium a été sélectionné pour cette étude du fait de ses propriétés thermiques et piézoélectriques, particulièrement intéressantes pour des nouvelles applications technologiques. Réalisées par pulvérisation cathodique magnétron, des monocouches et multicouches d'AlN hautement texturées sur des substrats de silicium monocristallin ont été obtenues.Leur microstructure et distribution d'orientations cristallographiques le long de la normale à la surface, ont été caractérisées expérimentalement pour déterminer,avec précision, l'évolution de la structure et de la taille des grains.L'impact de l'oxydation locale et l'évolution de la morphologie de grains sur la conductivité thermique transversale a été étudiée par la méthode 3W différentielle.La dispersion diffuse des phonons due aux défauts liés à la présence d'atomes d'oxygène, localisés à l'interface entre deux couches d'AlN, a été étudiée par des mesures thermiques sur la configuration multicouche.Les caractéristiques structurelles des couches polycristallines ont été corrélées avec les propriétés thermiques à partir d'un modèle théorique, qui tient compte de la répartition et de la géométrie des grains, et considère les films comme un ensemble en série de trois zones, composées de grains parallélépipédiques. Les résultats de conductivité thermique obtenus par la mesure des monocouches et multicouches polycristallines d'AlN sont bien prédits par le modèle développé,avec une différence inférieure à 10%. Une description physique détaillée des phénomènes de dispersion diffuse à l'interface avec le substrat, aux joints de grains, et aux défauts liés à l'oxygène, en fonction de l'hétérogénéité structurelle caractéristique, a été réalisée en comparant les résultats expérimentaux aux prédictions théoriques. Enfin, pour explorer la modulation dynamique du transfert de chaleur, l'influence de la déformation du réseau cristallin, causée par des contraintes mécaniques, sur la conductivité thermique des monocouches et multicouches d'AlN, a été étudiée en utilisant une nouvelle approche expérimentale qui couple un système de flexion 4-points avec la méthode 3W. / The understanding and control of the thermal conductivity of nano and microscale polycrystalline thin films is of fundamental importance for enhancing the performance and reliability of micro- and optoelectronic devices. However, the accurate description and control of the thermal performance of these bidimensional materials remain a difficult task due to their anisotropic and heterogeneous structure. Indeed, thin films obtained with a large number of deposition techniques and parameters, are composed of small crystallites at the interface with the substrate, which coalesce and evolve towards a columnar structure near the outer surface. These grains along with various crystallographic defects, such as oxygen impurities, increase the scattering processes of the energy carriers inside the materials, which in turn, reduce significantly their thermal conductivity. Experimental thermal characterization, accurate theoretical description and controlled modulation of the thermal properties of these materials are therefore desirable.This work is devoted to the investigation of the thermal conductivity of nanoscale polycrystalline films and explores the possibility to modulate heat transfer across these low dimensional structures. Because of its great interest in new technological applications, and its outstanding thermal and piezoelectric properties,aluminum nitride (AlN) served as a test material in this study. Highlytextured AlN mono- and multilayers were obtained by reactive radio-frequency magnetron sputtering on single-crystal silicon substrates. The microstructure and distribution of crystallographic orientations along the cross plane were characterized by transmission electron microscopy to accurately determine the grain structure and size evolution. The impact of local oxidation and structural inhomogeneity along the cross plane on the thermal conductivity was investigatedby thickness-dependent measurements performed by the differential 3Wtechnique. The diffusive scattering caused by oxygen-related defects, localized at the interface between two AlN layers, was studied by thermal measurements on the multilayered configuration. Structural features of the polycrystalline films were correlated with their thermal properties using a theoretical model,which takes into account the distribution of the grain geometry and considers the films as a serial assembly of three layers, composed of parallele piped grains.The experimental values of the thermal conductivity of the mono- and multilayerAlN polycrystalline films are well predicted by the developed model, witha deviation of less than 10%. Physical description of scattering phenomena at the interface, grain boundaries, and oxygen related defects, as a function of the characteristic structural heterogeneity, was achieved by comparing the experimental results to the theoretical predictions. It was found that grain mean sizes that evolve along the cross-plane direction, and structural features at the interface and transition domains, are key elements to understand and tailor thermal properties of nanocrystalline films with inhomogeneous structures. The results demonstrate that the structural inhomogeneity and oxygen-related defects in polycrystalline AlN films can be efficiently used to statically tune their cross-plane thermal conductivity. Finally, dynamic modulation of heat transfer bymeans of externally induced elastic strain on mono- and multilayer AlN films was investigated using a novel experimental approach consisting of a 4-pointsbending system coupled to the 3W method.

Cette thèse a été initialement motivée par la compréhension de l’activité électrocatalytique particulière de particules PtNi/C creuses pour l’electroréduction du dioxygène (ORR). L’étude des mécanismes de formation et croissance de ces particules creuses, grâce à des techniques operando basées sur l’interaction rayons X- et/ou électrons-matière, a permis de montrer que les défauts cristallins améliorent les propriétés électrocatalytiques de nanomatériaux bimétalliques pour l’ORR. Le concept de « catalyseur défectueux » a pu être progressivement étendu à d’autres nanostructures PtNi, notamment des nanocatalyseurs à forme contrôlée, grâce à de fructueuses collaborations avec d’autres laboratoires européens. Ce travail montre, qu’en raison de la distorsion de leur surface, les nanomatériaux défectueux présentent des propriétés de chimisorption uniques, définissant ainsi une nouvelle classe de catalyseurs prometteurs et stables. / This PhD thesis was initially motivated by the understanding of the peculiar electrocatalytic activity of hollow PtNi/C nanoparticles for the oxygen reduction reaction (ORR). Investigations on the formation and growth mechanism of this novel class of nanocatalysts using operando X-ray and electron-based techniques revealed that, beyond alloying effects, structural disorder is a lever to boost the ORR kinetics on bimetallic nanomaterials. The ‘defects do catalysis’ concept was progressively extended to various PtNi catalyst nanostructures, namely to advanced shape-controlled nanocatalysts from the ORR electrocatalysis landscape thanks to fruitful collaborations with European laboratories. This work shows that, through their distorted surface, microstrained nanomaterials feature unprecedented adsorption chemisorption properties and represent a viable approach to sustainably enhance the ORR activity.