In the past few decades, devices such as computers have become smaller, and their performance has improved beyond comparison. Semiconductors and interconnectors are used in almost all devices today and are made of thin films. Chemical vapor deposition (CVD) and atomic layer deposition (ALD) are common methods for depositing conformal films, where the film grows by being exposed to precursors either continuously as in CVD or by pulses as is pulsed CVD and ALD. It has been reported that Cu amidinate and guanidinate are precursors for depositing metallic Cu (Copper), but they tend to decompose. Precursors that are thermally stable and can be used to deposit a thin film of Cu are (1,3-di-tert-butyltriazenides) copper(I) (Cu-triazenides). This precursor has been synthesized and developed by the Pedersen Group in the Department of Physics, Chemistry, and Biology (IFM) at Linköpings University. This thesis investigates if Cu-triazenides are suitable as vapor deposition precursors using pulsed CVD and if Cu-triazenides are suitable as an ALD precursor with H2O as the second precursor. The main objective of this thesis is to focus on films deposited with ALD by finding the ALD window that gives a growth per cycle and making a saturation curve for each precursor. Si(100) and glass were used as the substrate, and SEM-EDX (Scanning electron microscopy- Energy- dispersive X-ray spectroscopy), XRD (X-ray diffraction), and XRR (X-ray reflectivity) analyzed the chemical, physical and, optical properties of the films. The Cu-triazenides were suitable as single-source precursors, as films were deposited with pulsed CVD on both Si(100) and glass. ALD deposited no film on Si(100) between the temperature 175–300 °C, and the analysis methods did not provide information on the thickness and composition of the films deposited on the glass.

Ultra-thin films of pure elements are important in microelectronics due to their wide range of applications. Atomic Layer Deposition (ALD) has drawn increasing attention as the thin films deposition technique for applications in microelectronics, due to its ability to deposit thin films with high conformality with atomic level control of the thickness of the film. However, due to the limited number of suitable precursor/co-reactant pairs available, only a few pure elements have been deposited successfully by ALD to date. The current study involves the synthesis and identification of potentially suitable precursor and co-reactant molecules for ALD of elemental Zn and Ge, neither of which have previously been achieved. MeZnOiPr, Zn(OiPr)2, and ZnEt2 were investigated as Zn precursors while GeCl2(Dioxane), Ge{N(SiMe3)2}2, and Ge(OCH2CH2NMe2)2 were investigated as Ge precursors. Co-reactants of interest were, 4,4,5,5-tetramethyl-1,3,2-dioxaborolane (HBpin), PhSiH3, [H2Al(tBuNCH2CH2NMe2)] (LAlH2), BH3(NMe3), and AlH3(Quinuclidine). Ligand-exchange reactions between precursors and co-reactants were expected to produce unstable zinc or germanium hydride species, which would then reductively eliminate to produce the pure element. Solution reactivity studies were employed to identify potential precursor/co-reactant pairs. Solution reactions of Zn precursors with the selected co-reactants indicated that unstable ZnH2 is produced during the reactions, and will dissociate into its elements (Zn and H2) at room temperature. These solution reactivity studies revealed that, HBpin and LAlH2 were more reactive as co-reactants than BH3(NMe3), AlH3(Quinuclidine), and PhSiH3. Additionally, MeZnOiPr and ZnEt2 exhibited the highest reactivity as precursors, although the lower reactivity of Zn(OiPr)2 may simply be due to low solubility. Solution reactions of Ge precursors produced a polymeric mono-germanium hydride species (GeH)x, which will only dissociate into its elements upon heating at elevated temperatures. While LAlH2 indicated high reactivity with all Ge precursors, it was difficult to arrange co-reactants in order of reactivity as most reactions immediately produced insoluble (GeH)x upon mixing reagents at room temperature. Ge(OCH2CH2NMe2)2 found to be the most reactive precursor out of all Ge precursors investigated. / Thesis / Master of Science (MSc)

Manipulating grain boundaries to pursue favorable physical or chemical properties is essential in materials design. As a prominent candidate for direct heat-to-electricity conversion applications, the performance of thermoelectric (TE) materials is strongly affected by the chemical compositions and physical properties of grain boundaries. As a layer-by-layer deposition technique, atomic layer deposition (ALD) is recognized as a unique method for depositing highly uniform films in a controlled manner. This gives us fresh insight into applying ALD approaches on a powder surface to realize a uniform coating on each particle with a specific thickness and composition of ALD layers. Powder ALD also provides a new way to construct complex layer structures, such as multiple layers, with precision layer composition control. In this thesis, a strategy of interface modification based on powder ALD is introduced in various TE materials (Bi, CuNi, and Zn4Sb3) to accurately control and modify the phase boundaries by oxide layer coating. For elemental bismuth, as the first discovered TE material, ultrathin layers of Al2O3, TiO2, and ZnO are typically deposited on powders via 1–20 cycles. All of the oxide layers significantly alter the microstructure and suppress grain growth. The hierarchical interface modifications aid the formation of an energy barrier by the oxide layer, resulting in a substantial increase in the Seebeck coefficient that is superior to that of most pure polycrystalline metals. Conversely, taking advantage of strong electron and phonon scattering, an exceptionally large decrease in thermal conductivity is obtained. A maximum figure of merit (zT) of 0.15 at 393 K and an average zT of 0.14 at 300–453 K were achieved in 5 cycles of Al2O3-coated Bi. Additionally, newly developed Sb2O5 thin films produced from SbCl5 and H2O2 were formed on the surfaces of Bi powders. Because of the high Kapitza resistance generated by Sb2O5 layers on Bi particles, a substantial decrease in total thermal conductivity from 7.8 to 5.7 W/m·K was obtained with just 5 cycles of Sb2O5 layer deposition and a 16% reduction in lattice thermal conductivity. Because of strong phonon scattering, the maximum zT values increased by approximately 12% and were relocated to 423 K. For CuNi, first, single-type ZnO and Al2O3 layers were deposited on the surface of CuNi powder, and their effect on the TE performance of the bulk was thoroughly investigated. The enhancement of the Seebeck coefficient, caused by the energy filtering effect, compensates for the electrical conductivity deterioration due to the low electrical conductivity of the oxide layers. Furthermore, the oxide layers may significantly increase the phonon scattering. Therefore, to reduce the resistance of phase boundaries, a multiple-layer structure was constructed by inserting Al2O3 into ZnO. Atom probe tomography shows that the Al atoms diffused into ZnO and realized the doping effect after pressing. Al diffusion has great potential to increase the electrical conductivity of coating layers. In comparison to pure CuNi, zT increased by 128% as a result of the decrease in resistance and stronger phonon scattering at the phase boundaries. The oxide layer coating not only yields a significant enhancement in the TE performance but also behaves as an energy barrier to suppress the migration of Zn ions in Zn4Sb3. With increasing ZnO layer cycle numbers, the layer thickness can be precisely tuned, and Zn migration can be effectively blocked with an oxide barrier. In the 100 cycle ZnO-coated sample, there was little deterioration of the power factor due to increasing resistivity. However, the decrease in total thermal conductivity results in similar zT values compared with pure Zn4Sb3, indicating that the TE performance of the 100 cycle ZnO layer-coated sample did not degrade. Additionally, 100 cycles of ZnO layers result in significantly enhanced thermal stability and effectively block Zn atom movement after 10 thermal cycling tests. The study demonstrates that ALD-based interface modification is a versatile method for decoupling TE parameters and precisely modifying phase boundaries, which is practical for other TE materials.:Abstract Kurzfassung Contents Chapter 1 Introduction Chapter 2 Background and motivation 2.1 Fundamental knowledge of thermoelectricity 2.1.1 Thermoelectric effects 2.1.2 Thermoelectric parameters 2.2 Interface/surface modification of thermoelectric materials 2.2.1 Principles 2.2.2 Discontinuous interface modification 2.2.3 Continuous interface modification 2.3 Background of atomic layer deposition 2.3.1 An ALD process example 2.3.2 Growth characteristics 2.3.3 Powder ALD 2.4 Powder ALD in thermoelectricity 2. 5 State-of-art in Bi, CuNi alloys, and Zn4Sb3 2.5.1 Bi and CuNi 2.5.2 β-Zn4Sb3 Chapter 3 Experimental techniques 3.1 Material syntheses and preparations 3.2 Material characterizations 3.2.1 ICP-OES 3.2.2 APT 3.2.3 LSR and LFA 3.2.4 GIXRD 3.2.5 XPS Chapter 4 Effect of powder ALD interface modification on the thermoelectric properties of Bismuth 4.1 Introduction 4.2 The influence of Al2O3, TiO2, and ZnO layers on TE properties 4.2.1 Characterizations of Al2O3, ZnO, and TiO2 ALD thin films 4.2.2 Microstructural characterizations of bulks 4.2.3 Effect of ALD surface modification on the TE properties of Bi 4.3 The influence of newly developed Sb2O5 layers on TE properties 4.3.3 New developed Sb2O5 ALD films 4.3.2 Microstructural characterizations of bulks 4.3.3 Effect of ALD surface modification on the TE properties of Bi Chapter 5 Precision interface engineering of CuNi alloys by multilayers of powder ALD 5.1 Introduction 5.2 Analysis of CuNi powders coated with ZnO and Al2O3 5.3 The effect of single-kind oxides on TE performance 5.4 The effect of multilayers on TE performance 5.5 Summary Chapter 6 Blocking ion migration in Zn4Sb3 by powder ALD 6.1 Introduction 6.2 Zn ion migration analysis in pure Zn4Sb3 6.3 The powder ALD effect on microstructure and thermoelectric properties 6.3 Stability testing on ZnO-coated samples 6.4 Summary Chapter 7 Summary and outlook 7.1 Summary 7.2 Outlook Appendix Appendix A: XRD patterns of Al2O3, TiO2 and ZnO-coated Bi Appendix B: TE properties of Al2O3 and ZnO-coated CuNi alloys References Abbreviations and symbols Acknowledgments List of publications List of awards List of attending conferences

No description available.

Afin de répondre au besoin de la société actuelle qui utilise toujours plus de moyens de transport et de dispositifs portables, les modes de production, de conversion et de stockage de l'énergie, sont en train de connaître de véritables mutations. Afin de créer des systèmes capables de générer une énergie maîtrisée et renouvelable, les nanosciences et nanotechnologies sont des domaines de premier plan. Le travail présenté dans ce manuscrit décrit la fabrication de structures, de taille nanométrique, organisées à grande échelle. La fonctionnalisation se fait par synthèse de films ou de particules par ALD. Des systèmes MIM sont synthétisés sur des structures ordonnées d'alumine poreuse. Les matériaux déposés en couches minces sont TiN, Al2O3 et HfO2. L'objectif est de fabriquer des nano-condensateurs à hautes performances pouvant être utilisés pour des applications de stockage de données, de mémoire ou pour le stockage d'énergie dans des petits dispositifs comme la technologie RFID.Dans une deuxième partie, des catalyseurs métalliques Pd/Ni sont déposés sur des membranes d'alumine pour l'électro-oxydation de l'acide formique. De la même manière, des nano-tubes de TiO2 fabriqués par oxydation anodique sont fonctionnalisés par des nano-particules de Pd pour l'électrooxydation de l'éthanol. Ces deux études systèmes peuvent conduire à leur utilisation comme catalyseurs au niveau de l'anode des piles à combustible liquide à combustion directe. Enfin, la dernière partie de ce travail consiste au dépôt par voie électrochimique de Cu2O, sur des nano-tubes de TiO2 qui servent de support. La jonction p/n ainsi fabriquée pourra servir pour la photo-conversion de l'énergie solaire. / In order to meet the growing needs in today's society that requests more transportation and portable devices, energy production, conversion and storage systems are now experiencing real changes. To fabricate systems able to generate a controlled and renewable energy, nanoscience and nanotechnology are leading research fields. The work presented in this manuscript describes the fabrication of nanosized, large-scale organized structures. These nanostructures have been functionalized through film and particle synthesis using a chemical vapor deposition method: the Atomic Layer Deposition (ALD).Thus, metal/insulator/metal (MIM) systems are fabricated on highly-ordered high-aspect ratio porous alumina. The materials that have been deposited are TiN, Al2O3 and HfO2. The aim is to produce high performance nanocapacitors that can be used for data storage (DRAM) application or for energy storage in small devices such as RFID.In a second part, metallic catalysts such as Pd/Ni have been deposited on alumina membranes for formic acid electro-oxidation. Similarly, TiO2 nanotubes have been fabricated by anodic oxidation and they have been functionalized with Pd nanoparticles for ethanol electro-oxidation. Both studied systems can be used as anode catalysts in direct liquid fuel cells.Finally, the last part of this work has consisted in the study of Cu2O, as a p-type semiconductor, that has been electrochemically deposited onto TiO2 nanotubes (n-type semiconductor). The resulting p/n junction can be interesting for solar energy photoconversion.

Le disulfure de molybdène, MoS2, est un composé lamellaire de la famille des dichalcogénures de métaux de transition utilisé depuis près d'un siècle comme lubrifiant solide et catalyseur d'hydrotraitement. Depuis la découverte en 2010 de ses propriétés de photoluminescence et de conduction (semiconducteur possédant un gap direct) lorsqu'il est isolé à l'état d'une seule monocouche, ce nouveau matériau 2D a suscité un intérêt croissant au sein de la communauté scientifique et permis d'envisager de nombreuses applications dans le domaine de l'énergie ou pour la réalisation de composants électroniques. Au-delà du disulfure de molybdène, cette découverte s'étend également à d'autres dichalcogénures (WS2, NbS2, MoSe2, WSe2,…) dont la combinaison des propriétés avec celles d'autres matériaux 2D déjà connus (graphène, h-BN,…) offre encore d'avantage de possibilités. Aujourd'hui, la réalisation de nombreux prototypes en laboratoire, principalement assemblés à partir de monocouches exfoliées, a pu démontrer le potentiel applicatif de ces matériaux, justifiant la nécessité de mettre au point des méthodes de synthèse qui permettront l'élaboration de dichalcogénures 2D à une échelle industrielle.Dans ce contexte, où semble actuellement être privilégié le développement de procédés de CVD à très haute température nécessitant des temps de croissance élevés et l'utilisation de substrats épitaxiés, nous avons décidé d'évaluer le potentiel d'une approche à basse température par des méthodes de dépôt en phase vapeur sur silice amorphe. Ce travail nous a ainsi permis d'identifier plusieurs couples de précurseurs pouvant se prêter au dépôt par CVD ou par ALD de couches minces amorphes de sulfure de molybdène ou de tungstène à moins de 250°C, puis de démontrer leur capacité à se réorganiser en monocouches de MoS2 et WS2 cristallines par un simple recuit thermique sous atmosphère inerte / MoS2, a transition metal dichalcogenide (TMD) possessing a mica-like layered structure, has been widely used over the past century as solid lubricant and hydrotreating catalyst. Since 2010, the discovery of new semiconducting (direct gap) and photoluminescence properties emerging in monolayer MoS2 has attracted much interest, with a wide range of potentialities for next-generation electronics or energy storage devices. Beyond MoS2, this discovery also concerns other TMDs (WS2, NbS2, MoSe2, WSe2,…), displaying a wide variety of electronic and optical properties, and whose combination with other 2D materials (graphene, BN,…) offers outstanding opportunities. While exfoliated materials have provided a convenient way to demonstrate the feasibility of proof-of-concept-devices, the development of reliable synthesis methods allowing the industrial production of monolayer TMDs has now to be investigated.In this booming research field, currently dominated by high-temperature CVD processes which are time-consuming and often require the use of epitaxial substrates, we investigated the potentiality of a low-temperature chemical vapor deposition approach on amorphous SiO2 substrates. This work allowed us to identify suitable precursors for the CVD or ALD of ultrathin amorphous molybdenum or tungsten sulfide deposits below 250°C, and to point out their ability to self-reorganize into crystalline MoS2 and WS2 monolayers upon thermal annealing

Thermische TaN ALD mit den Präkursoren TBTDET und TBTEMT, NH3 als zweiten Reaktanten und Ar als inertes Spülgas ist untersucht worden. Als Messverfahren zur Bewertung ist zeitlich aufgelöste in-situ spektroskopische Ellipsometrie mit einer Datenerfassungsrate von 0,86 Datenpunkte/s, sowie in-vacuo XPS und AFM verwendet worden. Es konnten sehr glatte homogene geschlossene TaN-Dünnschichten mit einem Ta:N-Verhältnis von 0,6, -Verunreinigungen von ca. 5 at.% (TBTDET) bzw. 9 at.% (TBTEMT) und einem GPC von ca. 0,6 nm/Zyklus im linearen Wachstumsbereich hergestellt werden. Eine O3-Vorbehandlung einer SiO2-Oberfläche beschleunigt die initiale Phase der TaN-Abscheidung. Die abgeschiedenen TaN-Schichten zeigen sich sehr reaktiv auf O2. / Thermal ALD with the precursors TBTDET and TBTEMT, NH3 as the second reactant and Ar as inert purging gas was studied. For measuring purposes time-resolved in-situ spectroscopic ellipsometry with an data acquisition rate of 0,86 data points/s, in-vacuo XPS and AFM was used. It was possible to deposit very smmoth homogenous closed TaN thin films with a Ta:N rate of about 0,6, contaminations of 5 at.% (TBTDET) and 9 at.% (TBTEMT), respectively, and a GPC of about 0,6 nm/Zyklus. An O3 pretreatment of a SiO2 surface accelerated the initial phase of the TaN atomic layer deposition (ALD) deposition. These TaN-Schichten were very reactiv against O2.

L'utilisation de sources naturelles illimitées telles que l'énergie solaire, éolienne ou hydraulique est en plein essor. Cependant leurs productions énergétiques sont fortement liées aux conditions climatiques et sont donc intermittentes. Ces systèmes nécessitent donc d'être associés à du stockage d'énergie, afin de lisser la production avant injection sur le réseau. Pour toutes ces raisons, les batteries Li-ion doivent intégrer de nouveaux matériaux d'électrode permettant d'obtenir une grande puissance et une haute densité d'énergie, tout en conservant une durée de vie élevée et une sécurité d'utilisation.Dans ce contexte, notre travail a consisté à préparer des matériaux à hauts potentiels, le LiCoPO4 (LCP) et le LiNi0.5Mn1.5O4 (LNMO). Ces derniers s'inscrivent parfaitement dans le contexte de développement de matériaux à haute densité d'énergie puisqu'ils possèdent des potentiels d'oxydation de 4,8 et 4,7 V vs Li +/Li et des densités d'énergies massiques théoriques de 802 et 691 Wh.kg-1 par rapport au lithium, respectivement.Ces matériaux ont d'abord été synthétisés sous forme de couches minces afin d'obtenir des électrodes modèles pour étudier de manière fondamentale les propriétés de transport des matériaux et ses interactions en présence d'un électrolyte liquide notamment les phénomènes à l'interface électrode/électrolyte. La compréhension des matériaux acquise au cours de ce premier axe a permis de transposer et d'adapter ces techniques de caractérisation aux systèmes plus complexes que sont les électrodes composites.Les propriétés de ces matériaux vis-à-vis de l'insertion et la désinsertion du lithium ont ensuite été testées et caractérisées en cellules électrochimiques. De nombreux paramètres cinétiques et thermodynamiques ont été extrait grâce à plusieurs techniques électrochimiques telles que la titration intermittente (GITT), la spectroscopie d'impédance (PEIS et GEIS), le cyclage galvano-statique et les tests de puissance. Nous proposons d'ailleurs une méthode simple, à partir de ces tests de puissance, pour déterminer le coefficient de diffusion du lithium. Enfin, nous tentons de répondre à plusieurs questions qui demeurent en suspens concernant la cyclabilité et la perte de capacité de ces matériaux à haut potentiel au cours de cyclage long et nous proposons une technique très simple permettant d'améliorer de façon étonnamment efficace la cyclabilité d'électrodes composites de LNMO. / The use of unlimited natural sources such as solar, wind or hydraulic power is booming. However, their energy production is dependant of climatic conditions and is therefore intermittent. These systems are usually associated with energy storage, in order to smooth the production before injection on the network. For all these reasons, Li-ion batteries need to incorporate new electrode materials to achieve high power and high energy density while maintaining a long life and safe use.In this context, our work consisted in preparing high potential materials, LiCoPO4 (LCP) and LiNi0.5Mn1.5O4 (LNMO). The latter are perfectly integered in the context of development of materials with high energy density since they have an oxidation potential of 4.8 and 4.7 V vs Li + / Li and theoretical mass energy densities of 802 and 691 Wh.kg-1 over lithium, respectively.These materials were first synthesized in the form of thin thin films to obtain model electrodes to study the kinetics and thermodynamics properties of materials transport and interactions with the presence of a liquid electrolyte including phenomena at the electrode interface / electrolyte. The understanding acquired during this first axis allowed us to transpose and adapt these characterization techniques to more complex systems : composite electrodes.The properties of these materials with respect to lithium insertion and desinsertion have been tested and characterized in electrochemicals cells. Kinetic and thermodynamic parameters have been extracted using several electrochemical techniques such as intermittent titration (GITT), impedance spectroscopy (PEIS and GEIS), galvano-static cycling and power tests. We propose a simple method, based on power tests, to determine the diffusion coefficient of lithium. Finally, we attempt to answer several questions that remain unsolved about the capacity loss of high potential materials during long cycling, and we propose a very simple technique for improving the cycling of composites electrodes of LNMO.

Le développement des mémoires dynamiques (DRAM) à haute performance basées sur la structure métal-isolateur-métal (MIM) nécessite de remplacer la couche de dioxyde de silicium par des matériaux diélectriques à haute permittivité diélectrique. L'utilisation de ces isolants dits high-k permet de réduire la taille du dispositif DRAM tout en conservant une densité de capacité élevée et un faible courant de fuite pour diminuer la fréquence de rafraichissement. Parmi les nombreux matériaux high k, le dioxyde de titane (TiO2) est l'un des candidats les plus prometteurs en raison de sa constant diélectrique relativement élevée pouvant atteindre 170 dans le TiO2 cristallisé en phase rutile. De plus, il est possible d’obtenir cette phase à basse température par le procédé ALD (< 250 °C) si le dépôt est réalisé sur un substrat RuO2 (phase rutile) grâce à une très faible différence de paramètres de maille entre les deux matériaux. L'objectif principal de cette thèse est d'étudier les mécanismes des réactions chimiques qui se produisant à l'interface RuO2/TiO2 lors du dépôt et leur influence sur les propriétés structurales et diélectriques du film TiO2, en particulier l'influence des espèces oxydantes, le plasma O2 et le H2O. L’influence des électrodes supérieure et inferieures sur les propriétés électriques et structurales de TiO2 a également été étudiée. Ensuite, la constante diélectrique, la conductivité ac et la tangente de perte des structures MIM à base d’oxyde de titane dopé aluminium ont été étudiés dans une gamme de fréquences large bande, de 1 Hz à 2 GHz. Enfin, la réalisation des MIM tridimensionnelles (3D) utilisant un substrat de silicium structuré en réseaux des trous coniques denses a été démontrée. Les structures MIM 3D réalisées ont permis d’augmenter sensiblement la densité de capacité tout en gardant de bonnes performances en termes de courant de fuite. / The development of high performance dynamic random access memory (DRAM) based on metal-insulator-metal (MIM) structure made it necessary to replace the conventional silicon dioxide layer by dielectric materials with high dielectric constants. The use of these so-called high-k insulators allows aggressive scaling of DRAM devices while keeping high capacitance density and, more importantly, low leakage current. Among the numerous high k dielectrics, titanium dioxide (TiO2) is one of the most attractive candidate due to its rather high dielectric constant (k). Rutile TiO2 is the interesting phase due to its high dielectric constant and the possibility to deposit this phase at low temperature by ALD (< 250 °C) by using RuO2 substrate thanks to a very small lattice mismatch between the two materials. The main objective of this thesis is to investigate the surface chemical reactions mechanisms at the RuO2/TiO2 interface and their influence on the ALD TiO2 film properties, especially the influence of oxidizing species, namely, H2O or O2 plasma. The influence of bottom and top electrode on electrical and structural proprieties of TiO2 MIM structure was also studied. Then, the dielectric constant, the ac conductivity and the loss tangent of aluminum doped titanium oxide are measured through a wide band frequency range, from 1 Hz to 2 GHz. Finally, the feasibility of three-dimensional (3D) MIM structures was studied by using dense array of truncated conical holes etched in a silicon substrate. The 3D MIM capacitors showed a large increase in the capacitance density while retaining very good electrical properties especially a leakage current comparable to planar MIM devices.

L'approche développée est une méthode "bottum-up" pour la réalisation d'électrodes nanostructurées pour des dispositifs de stockage et de production de l'énergie. Tout d'abord, la nanostructuration de surface est effectuée par voie électrochimique, une technique peu coûteuse, permettant un contrôle précis de la géométrie et facile à mettre en oeuvre sur de grandes surfaces. Ces nanostructures sont ensuite fonctionnalisées par la technique de dépôt de couche atomique ("Atomic Layer Deposition ", ALD). Cette technique est particulièrement intéressante pour la fonctionnalisation de surface car elle permet de déposer des matériaux de haute qualité sur de grandes surfaces avec un contrôle de l'épaisseur à l'échelle atomique. La combinaison de la nanostructuration par voie électrochimique et de la fonctionnalisation de surface par ALD est donc particulièrement intéressante pour les applications visées : la photocatalyse, l'électrocatalyse et les batteries au lithium. L'élaboration d'une électrode hybride à partir de nanotubes de TiO2 et de SnO2 pour les batteries Li-ion est décrite. La seconde étude concerne l'élaboration d'une anode pour des piles à combustibles à éthanol, composée de nanotubes de TiO2, de SnO2 et de Pd. La dernière partie concerne la fabrication d'une photoélectrode nanotubulaire ordonnée pour la production de dihydrogène. Les membranes d'alumine servent de matrice pour le d epôt de NiO et de Sb2S3 par ALD. / The approach developed here is the so-called bottum-up method to design nanostructured electrodes that can be use for energy storage and production. First, the surfaces are structured at the nanoscale using electrochemical methods. This method is not expensive, it allows a total control of the geometry and it is easy to implement on large surfaces. Then, the nanostructures are functionalized by Atomic Layer Deposition (ALD). This technique is particularly useful for surface functionalization as it allows to deposit high quality materials on large surfaces with a thickness control at the atomic scale. The combination of the electrochemical structuring with the surface functionalization by ALD is particularly attractive for the targeted applications : photocatalysis, electrocatalysis and lithium batteries. The development of a hybrid anode consists of SnO2 coated TiO2 nanotubes for Li-ion batteries is described. The second study is the development of anodes for direct ethanol fuel cells consists of the nanotubes of TiO2, SnO2 and Pd. The last part focused on the production of ordered nanotubular photoelectrode for the production of hydrogen. The alumina membranes are used as a template for the deposition of NiO and Sb2S3 by ALD.