>Magister Scientiae - MSc / In the current study, we give a report when oxalic acid was used as an electrolyte to synthesize an AAO template with hexagonal pore array. Optimum parameters were observed as 0.4 M of oxalic acid, anodizing voltage of 45 V, temperature of approximately 8 °C and the period of 120 minutes. Atomic force microscope (AFM) and High resolution scanning electron microscope (HRSEM) showed that template has an average pore diameter of 103 nm. Co and MnOx (x = 1,2) nanostructures were selectively deposited in the pores of the template using a novel atomic layer deposition (ALD) technique. The diameter sizes and the array of the nanostructures and the template were corresponding. Energy dispersive xrays (EDX) and X-ray photoelectron spectroscopy (XPS) confirmed the presence of Co and MnOx (x =1,2) on the samples while x-ray diffraction (XRD) provided an indication of their orientations. Magnetic force microscopy as main characterization tool showed the existence of multi-domains on both Co and MnOx (x =1,2) nanostructures. / South Africa

No description available.

Graphene is a single layer of carbon atoms arranged in honeycomb lattice. Over a long period of time it was treated as a hypothetical material to understand the properties of other allotropes of carbon, such as graphite, carbon nanotube etc. Half decade back, a single layer of graphene was finally isolated and since then the field has observed a flurry of activities. Low energy excitations in graphene are massless Dirac Fermions and quantum electrodynamic effects can be observed at room temperature in graphene, which makes it very popular among the condensed matter community. In addition graphene also shows many interesting mesoscopic effects, which is the focus of the present work. We study the electronic, magnetic and structural properties of the graphene nanostructures. The entire thesis based on the results and findings obtained from the present investigation is organized as follows. Chapter 1: provides a general introduction to the properties of graphene and graphene based nanostructures. Chapter2:describes the theoretical tools used in this thesis to investigate the properties of graphene nanoribbons. The first two chapters are meant to give the reader an overview about the field of graphene and a few of the computational techniques commonly used to investigate the properties of graphene. The following chapters are the new findings reported in this thesis. Chapter3:shows how zigzag graphene nanoribbons respond in a non-linear fashion when edges are subjected to some external potential such as magnetic field. Such response originates from the edge states present in zigzag ribbons and thus not observed in armchair nanoribbons. In the limit of ribbon width W→∞, an edge magnetic field produces a moment of ~ 1/3 per edge atom even for an infinitesimally small field, which is clearly a signature of non-linear response. Response of a finite width nanoribbon is size dependent and also depends on ln(V), the applied field. This is akin to Weber-Fechner law of audio visual perceptions. It is interesting to note that nature does provide a “quantum realization” of this in the form of biological sensing organs like the ear and eye. The magnetic response is found to scale inversely with the ribbon width. Chapter4:deals with the magnetic properties of the zigzag graphene nanoribbon. This is also a special property of the geometry of the zigzag edges and not observed in armchair nanoribbons. Our investigation reveals that the electron-electron repulsion (Hubbard U) energy creates a delta function like edge magnetic field in zigzag graphene nanoribbons. Starting from this, magnetic properties of zigzag graphene nanoribbons can be qualitatively and quantitatively explained from the non-linear response of zigzag nanoribbons. Zigzag graphene nanoribbons can exist in two possible ‘magnetic states’: antiferro (AF) where the two opposite edges have antiparallel magnetic moment and ferro (FM) where moment is parallel in the two opposite edges. First we describe the properties of undoped zigzag nanoribbons. They have AF ground state. Continuum theory can explain the size dependent bandgap and magnetic moment of the ground state. We present the first explicit derivation of the gap. Then we show that hole doping can change the ground state to FM, which is metallic. Thus the system has the property of magnetoresistance, which can be exploited by doping zigzag graphene nanoribbons externally with some gate voltage or internally by some electron acceptor element, such as boron. The critical doping for transition depends inversely with the ribbon width. We have found that the ‘phase transition’ on hole doping is a common phenomena for zigzag terminated nanostructures, such as hexagonal nanodots. Chapter5:discusses the effects of random edge shapes and random potential (Anderson disorder) on the magnetic properties of zigzag graphene nanostructures. Defects and disorders in the form of edge shape randomness and random potentials arising from substrate are very common in graphene. Our study reveals that edge state magnetism is very robust to shape randomness of the terminating edges of nanostructures; as long as there are three to four repeat units of a zigzag edge, the edge state magnetism is preserved. We also discover some “high energy” edges (ones where the edge atoms have only one nearest neighbor) can have very large moments compared to even the zigzag edges. Edge magnetism is also found to be robust to relatively small Anderson disorders, because a slowly varying small potential does not scatter the edge states. Chapter6:reveals how edge functionalization by O atom and OHgroup changes the properties of the zigzag graphene nanoribbons. Functionalization by various different molecules is a very popular method of tuning the properties of graphene. We have shown that it is possible to tune the properties of zigzag graphene nanoribbons by edge functionalization. Further, we have found that structures with clustered functionalization leads to “spatially” varying electronic structure, which can lead to interesting possibilities for electronic devices. Chapter7:describes structural stability, electronic and magnetic properties of graphene nanoribbons in presence of topological defects such as Stone-Wales defects. Our study reveals that the sign of stress induced by a SW defect in a graphene nanoribbon depends on the orientation of the SW defect with respect to the ribbon edge and the relaxation of the structure to relieve this stress determines its stability. Local warping or wrinkles arise in graphene nanoribbon when the stress is compressive, while the structure remains planar otherwise. The specific consequences to armchair and zigzag graphene nanoribbon can be understood from the anisotropy of the stress induced by a SW defect embedded in bulk graphene. We also have found localized electronic states near the SW defect sites in a nanoribbon. However, warping results in delocalization of electrons in the defect states. We have observed that, in zigzag graphene nanoribbons magnetic ordering weakens due to the presence of SW defects at the edges and the ground state is driven towards that of a nonmagnetic metal.

The hybrid nanostructures exhibit excellent performances in various fields such as catalysis, sensing, and energy conversion as compared to their individual ones. The thesis deals with the new methods for the synthesis of different type of hybrids with doped/pristine carbon nanostructures in the form of graphene, multiwall carbon nanotubes (MWCNTs) as one component and metals nanostructures (Ag, Pd, Pt and Au), carbide (Fe3C), metal chalcogenides (Ni3S2 and Co9S8) and oxide (CoO) as the other components. Various synthesis techniques such as modified galvanic replacement reaction at room temperature, hydrothermal, microwave and pyrolysis have been explored for the synthesis of different hybrid nanostructures. Furthermore, various hybrid nanostructures have been explored for various catalytic activities such as oxygen reduction reaction (ORR), oxygen evolution reaction (OER) and 4-nitrophenol (4-NP) reduction. It may be noted that the ORR and OER which are undoubtedly vital for their applications in fuel cells, metal-air batteries and water oxidation reaction. Interestingly, the catalytic activities of these hybrid nanostructures are comparable or better as compared to the commercial benchmark precious catalysts.

Les propriétés électroniques et thermoélectriques de matériaux basés sur Mg2Si ont été étudiées par calculs DFT et semi-classiques (théorie de Boltzmann). Les effets d’abaissement de dimensionalité et de contraintes ont été étudiés. Les calculs ont été effectués sur les films monocristallins orientés 001, 110 et 111 et sur les films polycristallins. Seul le film monocristallin orienté 110 a montré des propriétés thermoélectriques intéressantes. Trois types de contraintes ont été investiguées: uniaxiale, biaxiale et isotrope. L’augmentation de la contrainte sur Mg2Si produit un décalage du maximum du facteur de puissance (PF) vers les basses températures. Comparé à Mg2Si non contraint, le coefficient Seebeck (S) augmente uniquement sous contrainte isotrope. On montre l’équivalence des propriétés thermoélectriques entre Mg2Si contraint dans la direction [110] et celles du film orienté 110. Les contraintes de tension isotropes ont été modélisées en insérant des atomes Sb dans Mg2Si massif conduisant aux structures Mg2Si:Sb, Mg2Si:3Sb and Mg2Si:4Sb. Seul Mg2Si:4Sb produit une contrainte isotrope. Les effets de substitutions de Sn pour Si dans Mg2Si massif sont similaires à ceux observés pour Mg2Si sujet à des contraintes en tension uniaxiales et biaxiales. Pour les films Mg2Si1−xSnx orientés 110 le S du matériau dopé p est supérieur à celui des massifs Mg2Si et Mg2Si1−xSnx. Concernant les nanostructures, le super-réseau Mg2Si/Mg2Sn est le plus intéressant lorsque faiblement dopé p et à basse température. Les assemblages de fils sont les meilleurs en tant que matériaux faiblement dopés n et à basse température: le PF est quasiment doublé par rapport à celui de Mg2Si massif. / The electronic and thermoelectric properties of Mg2Si-based materials have been investigated by means of DFT calculations and semi-classical Boltzmann theory. The low-dimensional and strain effects on these properties have been studied. The properties have been investigated on 001-, 110- and 111-oriented Mg2Si monocrystalline films, and on polycrystalline Mg2Si film. Only the 110-oriented monocrystalline film has been found to have interesting thermoelectric properties. Three types of strains have been investigated: uniaxial, biaxial and isotropic. Increasing the intensity of the strain on Mg2Si induces a shift of the power factor (PF) maximum towards low temperature. Compared with unstrained Mg2Si, the Seebeck coefficient (S) increases only under isotropic strain. We evidence an equivalence in the thermoelectric properties between Mg2Si material constrained in the [110] direction and the 110-oriented Mg2Si film. Isotropic tensile strains have been modeled by inserting Sb atoms in bulk Mg2Si leading to the stuctures Mg2Si:Sb, Mg2Si:3Sb and Mg2Si:4Sb. Only Mg2Si:4Sb is found to induces such type of constraints. The effects of the Sn for Si substitutions in bulk Mg2Si are very similar to those observed for Mg2Si subjected to uniaxial and biaxial tensile strains. For (110)-oriented Mg2Si1−xSnx films S of the n−doped material outperforms that of the bulk Mg2Si and bulk Mg2Si1−xSnx. Regarding nanostructures, the Mg2Si/Mg2Sn superlattice is most interesting as a p-doped material at low carrier concentration/low temperature. The stick assemblage is best as a n-doping material at low carrier concentration/low temperature where its PF is almost twice as high as that of bulk Mg2Si.

Le besoin incessant de débits toujours plus élevés dans les systèmes de télécommunications a un impact sur tous les éléments composant la chaine de transmission. Ainsi, pour faire face à l’augmentation croissante du volume de données échangées à travers le monde, le développement de nouvelles sources optiques semi-conductrices est absolument nécessaire. La modulation directe de lasers nanostructurés constitue une alternative bas coût et à faible consommation énergétique qui permettra de remplacer progressivement les diodes lasers à puits quantiques actuelles. De nombreux efforts en recherche ont été consacrés au cours des dernières années en vue d’améliorer les performances dynamiques des lasers nanostructurés notamment en terme de bande passante, de facteur de couplage phase-amplitude (facteur α) et de dérive de fréquence (chirp). Pour les applications aux très grands réseaux et systèmes de communication, la croissance d’îlots ou de fils quantiques déposés sur substrat InP permet de réaliser des dispositifs nanostructurés émettant dans le proche infra-rouge autours de 1550 nm. Dans ce mémoire, la dynamique de modulation des lasers nanostructuré est étudiée en régime de modulation directe. Les caractéristiques analysées comprennent: la modulation en amplitude (AM) et en fréquence (FM), le chirp, et les réponses en régime grandsignal. Grâce à une approche semi-analytique, il est démontré que la bande passante et l’amortissement sont fortement limités par les phénomènes de capture et de relaxation des porteurs de charge dans les nanostructures. Afin d’étudier les propriétés du facteur α et du chirp, un nouveau modèle dynamique a été proposé, prenant en compte la contribution à l’indice optique des porteurs de charge dans des états hors résonance. Il est ainsi montré que, contrairement au cas des lasers à puits quantiques, le facteur α dépend fortement du courant de pompe et de la fréquence de modulation. Le facteur α reste constant à basses fréquences (<0,1 GHz) et supérieur aux valeurs obtenues à hautes fréquences (au-delà de quelques GHz) à partir de la technique FM/AM. Ces caractéristiques sont essentiellement attribuées aux contributions des porteurs dans les états hors résonance. Les simulations montrent que le facteur α peut être réduit en augmentant la séparation énergétique entre l’état fondamental résonant (GS) et les états hors résonance. En particulier, un effet laser sur 1’état excité des nanostructures (ES) constitue une solution prometteuse pour améliorer les performances dynamiques, en accroissant notamment la bande passante de modulation et en réduisant le facteur α d’environ 40%. Les techniques d’injection optique sont également intéressantes pour régénérer les performances dynamiques des lasers. Le couplage phase-amplitude et le gain optique y sont substantiellement modifiés via le contrôle de l’amplitude et du désaccord en fréquence du faisceau injecté. Dans ce cadre, ce travail propose une nouvelle technique dérivée de la méthode Hakki-Paoli, permettant de mesurer, sous injection optique, le facteur α à la fois en dessous et au-dessus du seuil. Les lasers à cascade quantique (QCL) sont basés sur des transitions électroniques inter-sous-bandes dans des hétérostructures à puits quantiques. Ces lasers présentent une bande passante (AM) relativement de quelques dizaines de GHz et sans résonance ce qui est prometteur pour les transmissions en espace libre. De manière surprenante, les calculs montrent que les QCL présentent une largeur de bande FM extrêmement large de l’ordre quelques dizaines de THz, environ trois ordres de grandeur supérieure à la largeur de bande AM. L’injection optique dans ces lasers présente les mêmes avantages que ceux procurés dans les lasers à transitions interbandes. Des désaccords positifs ou négatifs en fréquence augmentent notamment la largeur de la bande passante. / High performance semiconductor lasers are strongly demanded in the rapidly increasing optical communication networks. Low dimensional nanostructure lasers are expected to be substitutes of their quantum well (Qwell) counterparts in the next-generation of energy-saving and high-bandwidth telecommunication optical links. Many efforts have been devoted during the past years to achieve nanostructure lasers with broad modulation bandwidth, low frequency chirp, and reduced linewidth enhancement factor. Particularly, 1.55-μm InP-based quantum dash (Qdash)/dot (Qdot) lasers are preferable for long-haul transmissions in contrast to the 1.3-μm laser sources. In this dissertation, we investigate the dynamic characteristics of InPbased nanostructure semiconductor lasers operating under direct current modulation, including the amplitude (AM) and frequency (FM) modulation responses, the linewidth enhancement factor (also known as α-factor), as well as large-signal modulation responses. Using a semi-analytical analysis of the rate equation model, it is found that the modulation bandwidth of the quantum dot laser is strongly limited by the finite carrier capture and relaxation rates. In order to study the α- factor and chirp properties of the quantum dot laser, we develop an improved rate equation model, which takes into account the contribution of carrier populations in off-resonant states to the refractive index change. It is demonstrated that the α-factor of quantum dot lasers is strongly dependent on the pump current as well as the modulation frequency, in comparison to the case of Qwell lasers. The α-factor remains constant at low modulation frequencies (<0.1 GHz) and higher than the value derived at high modulation frequencies (beyond several GHz) from the FM/AM technique. These unique features are mostly attributed to the carrier populations in off-resonant states. Further simulations show that the α-factor can be reduced by enlarging the energy separation between the resonant ground state (GS) and off-resonant states. Lasing from the excited state (ES) can be a promising alternative to enhance the laser’s dynamic performance. The laser exhibits a broader modulation response and the α-factor can be reduced by as much as 40%. The optical injection technique is attractive to improve the laser’s dynamical performance, including bandwidth enhancement and chirp reduction. These are demonstrated both theoretically and experimentally. The phase-amplitude coupling property is altered as well in comparison with the free-running laser and the optical gain depends on the injection strength and the frequency detuning. This work proposes a new method derived from the Hakki-Paoli method, enabling to measure the α-factor of semiconductor lasers under optical injection both below and above threshold. In addition, it is demonstrated theoretically that the α-factor in nanostructure lasers exhibits a threshold discontinuity, which is mainly attributed to the unclamped carrier populations in the off-resonant states. It is a fundamental limitation, preventing the reduction of the α-factor towards zero. Quantum cascade (QC) lasers rely on intersubband electronic transitions in multi-quantum well heterostructures. QC lasers show flat broadband AM response (tens of GHz) without resonance, which constitutes promising features for free-space communications. Surprisingly, calculations show that the QC laser exhibits an ultrabroad FM bandwidth on the order of tens of THz, about three orders of magnitude larger than the AM bandwidth. Optically injection-locked QC lasers also exhibit specific characteristics by comparison to interband semiconductor lasers. Both positive and negative frequency detunings enhance the modulation bandwidth.

Du fait de leur facilité de fabrication et de leurs propriétés physiques uniques, les nanofils (NFs) de semi-conducteurs présentent des potentialités d’application importantes elles pouvaient être comme briques élémentaires de nombreux dispositifs nano- et opto-électroniques. Différents procédés de fabrication ont été développés pour fabriquer et organiser ces nanofils en épitaxie sur silicium. Cependant, un des principaux problèmes réside dans le manque de reproductibilité des NFs produits naturellement. Pour obtenir un meilleur contrôle de leur périodicité, localisation, forme et taille, différents types de gravure ont été mis au point. Aujourd’hui, des incertitudes importantes persistent quant à leurs propriétés fondamentales, en raison d’un manque de corrélation entre les propriétés électroniques et optiques et les détails microscopiques (composition, structure, chimie ...etc.). L’objectif de ce travail est de développer deux types de procédés de fabrication : le premier "top-down" est basé sur la nanogravure directe par faisceau d’ions focalisés (FIB)de couches bi-dimonsionnelles de SiGe. Ce procédé permet de contrôler la taille des NFs, les déformations, et leur localisation précise. Il permet de fabriquer des réseaux de larges piliers. Les NFs réalisés par cette technique sont peu denses et de diamètre important. Le second procédé est de type "Bottom-Up" ; il s’appuie sur la croissance VLS à partir de catalyseurs métalliques (AuSi). Les NFs réalisés ont étudiés à l’échelle locale afin de mesurer la taille moyenne de contrainte ainsi que leur effet sur le confinement quantique et sur la structure de bande des NFs. / SiGe/Si core/shell nanowires (NWs) and nanodots (NDs) are promising candidates for the future generation of optoelectronic devices. It was demonstrated that the SiGe/Si heterostructure composition, interface geometry, size and aspect ratios can be used to tune the electronic properties of the nanowires. Compared to pure Si or Ge nanowires, the core-shell structures and exhibit extended number of potential configurations to modulate the band gap by the intrinsic strain. Moreover, the epitaxial strain and the band-offsets produce a better conductance and higher mobility of charge carriers. Recent calculations reported that by varying the core-shell aspect ratio could induce an indirect to direct band gap transition. One of the best configurations giving direct allowed transitions consists of a thin Si core embedded within wide Ge shell. The Germanium condensation technique is able to provide high Ge content (> 50%) shell with Si core whom thickness of core and shell can be accurately tuned. The aim of this work is to develop two types of synthesis processes: the first "top-down" will be based on direct nanoetching by focused ion beam (FIB) of 2D SiGe layer. This process allows the control of the size of NWs, and their precise location. The NWs achieved by this technique are not very dense and have a large diameter. The second processes called "bottom-up"; are based on the VLS growth of NWs from metal catalysts (AuSi). Grown NWs have been studied locally in order to measure the mean size and the strain and their effects on the quantum confinement and band structure of NWs.

Recently, efforts have been made for self-powering the batteries and portable electronic devices by piezoelectric nanogenerators. The piezoelectric nanogenerators can work as a power source for nano-systems and also as an active sensor. The piezoelectric nanogenerator is a device that converts random mechanical energy into electrical energy by utilizing the semiconducting and piezoelectric properties. Also, the mechanical energy is always available in and around us for powering these nano devices. The aim of the present thesis work is to explore 1D and 2D ZnO nanostructures (nanorods and nanosheets) on metal alloy substrates for the development of piezoelectric nanogenerators in energy harvesting and sensors applications. Hydrothermal synthesis method was adopted for the growth of ZnO nanostructures. The nanogenerators were fabricated by using the optimized synthesis parameters and subsequently studied their performance for power generation and as an active speed sensor. These 1D and 2D nanostructures based nanogenerators have opened up a new window for the energy harvesting applications and sensors development. The thesis is divided into following six chapters. Chapter 1: This chapter gives a general introduction about energy harvesting devices such as nanogenerators, available energy sources, mechanical energy harvesting, ZnO material and the details on hydrothermal synthesis process. A brief literature survey on different applications of piezoelectric nanogenerators is also included. Chapter 2: A novel flexible metal alloy (Phynox) and its properties along with its applications are discussed in this chapter. Details on the synthesis of 1D ZnO nanorods on Phynox alloy substrate by hydrothermal method are presented. Further, the optimization of parameters such as growth temperature, seed layer annealing and substrate temperature effects on the synthesis of ZnO nanorods are discussed in detail. As-synthesized ZnO nanorods have been characterized using XRD, FE-SEM, TEM and XPS. Chapter 3: It reports on the fabrication of piezoelectric nanogenerator on Phynox alloy substrate as power generating device by harvesting the mechanical energy. Initially, the performance of the nanogenerator for power generation due to finger tip impacts was studied and subsequently its switching polarity test was also carried out. Output voltage measurements were carried out using the in-house developed experimental setup. Stability test was also carried out to see the robustness of the nanogenerator. Finally, the output voltage response of the nanogenerator was studied for its use as an active speed sensor. Chapter 4: Synthesis of Al doped 2D ZnO nanorsheets on Aluminum alloy (AA-6061) substrate by hydrothermal method is reported in this chapter. The optimized parameters such as growth temperature and growth time effects on the synthesis of ZnO nanosheets are discussed. As-synthesized ZnO nanosheets were characterized using XRD, FE-SEM, TEM and XPS. The Al doping in ZnO is confirmed by EDXS and XPS analysis. Chapter 5: Cost effective fabrication of Al doped 2D ZnO nanosheets based nanogenerator for direct current (DC) power generation is reported in this chapter. The performance of the nanogenerator for DC power generation due to finger tip impacts was studied and subsequently its switching polarity test was also carried out. Output voltage measurements were carried out using the in-house developed experimental setup. Stability test was also carried out to see the robustness of the nanogenerator. Finally, the DC output voltage response of the nanogenerator was studied for its use as an active speed sensor. Chapter 6: The first section summarizes the significant features of the work presented in this thesis. In the second section the scope for carrying out the further work is given.

Le contexte du réchauffement climatique entraîne un développement des technologies CSP (Concentrated Solar Power). La réduction des coûts de production de ces technologies passe par une amélioration de la durabilité et de l'efficacité des composants des centrales solaires. Les températures de fonctionnement élevées du CSP (250-600°C) nécessitent d'employer des absorbeurs spectralement sélectifs afin de limiter les pertes par radiation. Cette thèse propose une architecture originale d'absorbeurs sélectifs stables à haute température sous air en combinant un dépôt de TiAlN en couches minces avec un réseau de diffraction. L'ajout d'une texturation de surface augmente l'absorption solaire du dépôt par un effet d'absorption et de gradient des indices optiques effectifs conduisant à une augmentation du rendement photothermique de l'absorbeur. Dans ce mémoire, la modélisation du comportement optique des absorbeurs texturés, les méthodes de fabrication du réseau de diffraction ainsi que le dépôt des couches minces par PVD et PECVD seront abordés et les mesures expérimentales seront comparées aux modélisations. L'analyse des absorbeurs texturés fabriqués confirme un gain de rendement photothermique pouvant atteindre +3% ainsi qu'une stabilité thermique remarquable à 500°C sous air jusqu'à 300 h de recuit / The global warming context reinforces the development of CSP technologies. Cost reduction of CSP requires the improvement of component durability and efficiency. The solar absorbers should be spectrally selective since the high working temperatures of CSP plants increase the radiative thermal losses. This thesis proposes an original, spectrally selective absorber structure combining TiAlN based coatings and diffractive gratings. The surface texturing provided by the diffractive gratings improves the solar absorption of the thin coating by an effective optical index gradation effect leading toincreased photothermal efficiency. In this thesis, the modeling of the textured absorber’s optical behavior, fabrication methods of diffractive gratings, as well as layer deposition by PVD/PECVD will be discussed. Experimental measurements will be compared to the theoretical modelling. The experimental analysis of textured absorbers confirms the increase of photothermal efficiency by almost 3%, as well as a good thermal stability at 500°C in air for 300 hours of annealing

Les nanocristaux colloïdaux sont des fluorophores semi-conducteurs de taille nanométrique. Fluorescents à température ambiante et synthétisés par voie chimique, ces nanoémetteurs représentent d'excellents candidats pour divers domaines d'application tels que l'éclairage, le marquage biologique ou le photovoltaïque.Mon travail expérimental, s'inscrit dans le développement de ces nanoémetteurs via deux approches différentes : leur couplage avec des nanostructures plasmoniques à 4K et la mise en place de régime d'émission collective.Dans un premier temps, nous avons étudié le couplage de nancoristaux individuels avec des films d'or plan dans le but de réduire les pertes par effet Joule. Tout d'abord, nous nous sommes intéressés à l'influence de la température sur ce couplage. A partir d'expériences de photoluminescence et de simulations numériques effectuées par Gérard Colas des Francs du LCIB, nous avons pu mesurer et simuler une forte diminuation de l'accélération du temps caractéristique d'émission des nanocristaux lorsque la température passe de 300K à 4K. Cette diminution est directement liée à une importante diminution des pertes ohmiques des couches d'or. L'efficacité quantique des nanocristaux est ainsi augementée d'un facteur 3. Ensuite, nous avons étudié l'influence de la cristallinité de l'or sur ce couplage. A nouveau une forte réduction de l'accélération de l'émission des nanocristaux a été mesurée sur or cristallin en comparaison avec des couches d'or amorphe. Ces mesures laissent de nouveau présager une réduction des pertes par effet Joule ainsi qu'une augmentation d'au moins un facteur deux de l'efficacité quantique des nanocristaux.Dans un second temps, nous avons effectué les premières caractérisations d'agrégats de nanocristaux enrobés dans une coquille de silice. A température ambiante, nous avons mis en évidence la présence d'interaction de type FRET entre les nanocristaux émettant dans le bleu et les nanocristaux émettant dans le rouge. Cette interaction permet ainsi une accélération de l'émission globale des agrégats. A 4K, nous avons observé une modification de la dynamique d'émission des agrégats avec l'apparition de deux échelles de temps différentes. Pour les temps courts, la dynamique d'émission est accélérée et est régie par la recombinaison de l'exciton. Pour les temps longs, la dynamique d'émission est régie par une loi de puissance traduisant ainsi l'apparition de temps caractéristiques d'émission extrêmement longs. / Colloidal semiconductor nanocrystals are fluorescent semiconductors with a nanometric size. Bright at room temperature and chemically synthesized, nanocrystals are interesting candidates for differents applications as lighting, biological labeling or photovoltaic.My experimental work, is part of the development of these emitters by two differents approaches : coupling with plasmonic structures at 4,K and formation of collective emission.First, we studied the coupling between single nanocrystals and a flat gold film in order to decrease the optical losses. To begin we studied the influence of the temperature. With some photoluminescence measurements and some simulations, we show significant decrease of the enhancement of the photoluminescence decay rate at 4,K. This reduction is linked to the decrease of optical losses. Then, we studied the influence of crystallinity of gold. We show again an important reduction of enhancement of the photoluminescence decay rate with crystalline gold compared to amorphous layer.Second, we investigated the optical properties of compact nanocrystal clusters encapsulated in a silica shell. At room temperature, we observed an enhancement of the photoluminescence decay rate through Förster resonance energy transfer (FRET). At 4K, we measured an important variation of the emission dynamic with emergence of two times scales. At short time scale, emission is accelerated and governed by the exciton recombination. At long time scale, the decay is governed by power law showing the emergence of long-lived states.