Cette étude a été motivée par les nombreux avantages du procédé de micromoulage, qui couple la croissance électrolytique et la localisation du film avec un moule en résine épaisse. Ce procédé permet en effet la réalisation de micro-objets, dont les dimensions sont uniquement dépendantes de la résolution des techniques de lithographie employées pour définir les moules en résine. Le micromoulage permet donc de réaliser des microstructures métalliques et est compatible avec la technologie MEMS. Nous avons mis en évidence l’influence de différents paramètres expérimentaux dans le cadre de l’étude de la croissance électrolytique du samarium-cobalt en solution aqueuse dans une cellule de Hull. Cette étude nous a permis de déterminer plusieurs points de fonctionnement conduisant à des teneurs en samarium et des épaisseurs élevées : jusqu'à 50 % de samarium et plusieurs microns d’épaisseur. En outre, un certain nombre d’hypothèse ont été émises, qui lient le procédé d’élaboration et le mécanisme de croissance. Nous avons aussi réussi a montré qu’il est possible de réaliser des micromotifs de plusieurs microns d’épaisseur contenant un rapport Sm/(Sm+Co) relativement élevé (10 %) et une faible contamination en oxygène (8 %). / This study was motivated by the advantags of micromolding process, which couples the electrolytic growth and location of the film with a thick resin mold. This method makes it possible the achievement of micro-objects, the dimensions of which are only dependent on the resolution of the lithographic techniques used to define the resin molds. The micromolding can therefore produce metal microstructures and is compatible with MEMS technology. We have highlighted the influence of various experimental parameters in the context of the study of the electrolytic growth of samarium-cobalt in aqueous solution in a Hull cell. this study allowed us to identified several operating points resulting in samarium contents and high thicknesses up to 50% of samarium and several microns thick. In addition, a number of hypotheses have been put forward, which link the process of development and growth mechanism. We also successfully showed that it is possible to create micropatterns of several microns thick report containing Sm / (Sm + Co) relatively high (10%) and low oxygen contamination (8%).

<p>Cilj ovog rada bio je da se ustanovi da li dolazi do elektrohemijskog taloženja&nbsp; niobijuma i aluminijuma iz hloroaluminatnih rastopa obogaćenih niobijumom na podlogama od staklastog&nbsp; ugljenika, platine i zlata. Nadalje, ako dolazi do rečenog&nbsp; taloženja, da li pri tome dolazi i do formiranja legura niobijuma sa izabranim podlogama kao i da li dolazi do formiranja legura niobijuma sa zajednički taloženim aluminijumom. I na kraju,&nbsp; cilj je&nbsp; bio da se navedeni fenomeni ostvare na &scaron;to nižim temperaturama.</p><p>Nb&nbsp; i&nbsp; Al&nbsp; su elektrohemijski taloženi na staklastom ugljeniku, platini i zlatu pri 200&nbsp; &deg;C, pod atmosferom argona, iz rastopa napravljenog od ekvimolarne sme&scaron;e AlCl₃ + NaCl kome&nbsp; je dodat niobijum.</p><p>U istraživanjima su kori&scaron;ćene: linearna&nbsp; i&nbsp; ciklička voltametrija, metoda potenciodinamičke polarizacione krive, metoda &bdquo;otvorenog kola&ldquo;,&nbsp; hronoamperometrija,&nbsp; SEM,&nbsp; EDS, AFM&nbsp; i XRD analiza.</p><p>Elektrohemijsko taloženje Nb i Al iz primenjenih rastopa pri potpotencijalima uočeno je na podlozi od zlata. Taloženje niobijuma pri potpotencijalima na zlatu je novost, a taloženje Al na zlatu pri&nbsp; potpotencijalima poznato je od ranije. Elektrotaloženje&nbsp; Al&nbsp; i&nbsp; Nb&nbsp; iz primenjenih rastopa pri potpotencijalu na zlatu prethodi procesima elektrotaloženja niobijuma i aluminijuma pri natpotencijalima.&nbsp; Rezultat ovih procesa su Nb/Au i Al/Au legure nastale zajedničkim taloženjem i interdifuzijom istaloženog Nb, odnosno&nbsp; Al i podloge&nbsp; -&nbsp; zlata (AlAu,&nbsp; AlAu₂ , AlAu₄, AlNb₂ , AlNb₃ , Au₂Nb, Au₂Nb₃ , AuNb₃ , AlNb₃).</p><p>Elektrohemijsko taloženje Nb iz primenjenih rastopa pri potpotencijalima uočeno je na platini i prethodi procesima elektrotaloženja&nbsp; Nb&nbsp; i&nbsp; Al. Rezultat su formirane Nb/Pt legure nastale interdifuzijom&nbsp; istaloženog Nb i podloge&nbsp; -&nbsp; platine (Nb₃Pt). Pri natpotencijalima aluminijuma na&nbsp; platini dolazi do taloženja metalnog niobijuma, metalnog aluminijuma i zajedničkog taloženja Nb i Al. Kao posledica&nbsp; ovih procesa dolazi i do formiranja legura (AlNb₃, Al₃Pt₅ ,&nbsp; Nb₃Pt).</p><p>Elektrohemijsko taloženje niobijuma iz primenjenih rastopa na izabranim podlogama odigrava se na potencijalima koji su pozitivniji od potencijala&nbsp; elektrotaloženja aluminijuma.</p><p>Sve dobijene legure formirane su pri temperaturi od 200 &deg;C &scaron;to je nekoliko stotina stepeni niže od odgovarajućih temperatura potrebnih da se formiraju termičkim putem.&nbsp;</p> / <p>The aim of this study was to establish electrodeposition of niobium and aluminium from chloroaluminate melt (AlCl₃ + NaCl) enriched with niobium onto gold, platinum and glassy carbon substrates. Furthermore, if such deposition is shown to occur,&nbsp; to&nbsp; assess if it can lead to formation ofniobium alloys with chosen substrates or&nbsp; formation of Nb/Al alloys as a result of codeposition. And finally, if the deposition is indeed confirmed,&nbsp; to&nbsp; establish the lowest operational temperature at which it is possible.</p><p>Niobium and aluminium were electrodeposited onto glassy carbon, gold and platinum at&nbsp; 200 &deg;C, under an argon atmophere&nbsp; from the melt made of&nbsp; equimolar mixture AlCl₃ + NaCl with niobium being added.</p><p>Research methods used were:&nbsp; cyclic voltammetry, potentiodynamic polarization curves, &quot;open circuit&rdquo;, chronoamperometry,&nbsp; SEM,&nbsp; EDS, AFM and&nbsp; XRD&nbsp; analysis. Underpotential deposition of&nbsp; Nb&nbsp; and Al&nbsp; from the melt onto gold substrate was observed. Of the two, niobium deposition is a novelty, while alumininum deposition has been previously studied.&nbsp; Underpotential deposition of&nbsp; NB and Al onto gold from given melt precedes the overpotential deposition of aluminium and niobium onto gold. The result of these processes are Nb/Au and Al/Au alloys formed by codeposition of the two metals&nbsp; and&nbsp; interdiffusion between the Nb and Al deposits and the substrate -gold (AlAu, AlAu₂, AlAu₄, AlNb₂, AlNb₃,&nbsp; Au₂Nb, Au₂Nb₃, AuNb₃, AlNb₃). Underpotential deposition of niobium onto platinum from used melts was recorded and it precedes the&nbsp; processes of overpotential deposition of niobium and aluminium. The result is formation of Nb/Pt alloys as a consequence of&nbsp; interdiffusion between the substrate and the deposited Nb (Nb₃Pt). The potentials more negative than the aluminium deposition potential in the used system lead to deposition of niobium and aluminium&nbsp; individually and jointly. Thus, a number of alloys are formed (AlNb₃, Al₃Pt₅, Nb₃Pt).&nbsp; Niobium electrodeposition onto chosen substrates from the melt applied takes place at potentials which are more positive than the potential of electrodeposition of aluminium. Formation of&nbsp; the&nbsp; obtained alloys occurred at a temperature of 200 &deg;C which is several hundred degrees lower than the temperatures needed for their formation by thermal methods.</p>

De nombreux traitements de surface (sablage, attaques acides…) ont été mis au point sur les implants dentaires afin de favoriser leur ostéointégration. Par ailleurs, depuis plusieurs années des revêtements à base de phosphates de calcium sont également développés dans le même but.L’objectif principal de la thèse est de mettre au point un procédé de revêtement à basse température afin de déposer à la surface de l’implant en titane une couche mince de phosphate de calcium, de structure et de composition analogue au minéral osseux en vue de favoriser l’ostéointégration. Il est aussi souhaité que ce revêtement présente des propriétés antibactériennes afin de lutter contre les infections post-opératoires. Dans un premier temps, un traitement de surface de l’implant en titane composé d’une étape de sablage et d’une attaque acide a été développé. Il permet d’obtenir une rugosité moyenne de surface comprise entre 1,4 et1,8 µm ainsi qu’une texture microporeuse et une mouillabilité de surface améliorée. Puis, les procédés de revêtement d’électrodéposition et d’immersions successives ont été étudiés. L’étapede centrifugation implémentée dans le procédé d’immersions successives s’est révélée cruciale et un revêtement d’environ 2 µm d’épaisseur composé d’apatite biomimétique a été obtenu. La composition et l’épaisseur du revêtement élaboré par électrodéposition est fortement influencée par la durée du dépôt. Ainsi une durée d’électrodéposition de 1 mn menée à un potentiel de -1,6V/ECS permet d’obtenir un revêtement d’environ 1,5 µm d’épaisseur composé d’une couche de phosphate octocalcique et de cristaux de brushite. Un test de vissage/dévissage dans une mâchoire artificielle a démontré la tenue mécanique des revêtements obtenus selon les deux procédés. Enfin, des ions antibactériens tels que l’argent, le cuivre ou le zinc ont été incorporésaux revêtements. Il a été démontré que des taux importants d’incorporation allant jusqu’à 40 %molaire par rapport au calcium peuvent être atteints pour le cuivre et le zinc. Des tests biologiques permettant d’évaluer l’effet de ces revêtements sur l’activité biologique de cellules mésenchymateuses humaines ainsi que sur la formation d’un biofilm (modèle de péri-implantite)en vue de la prévention des infections post-opératoires ont conduit à des résultats prometteurs pour le développement de tels revêtements bioactifs. Ces travaux de thèse s’inscrivent dans le cadre du projet BIOACTISURF, soutenu par la Région Midi-Pyrénées, et réalisé en collaboration avec le Laboratoire de Génie Chimique (LGC) ainsi qu’un partenaire industriel / Numerous surface treatments have been developed in order to improve osseointegration of dentalimplants (sandblasting, acid etching…). Moreover, various strategies involving calcium phosphate coatings have emerged for the same target for a few decades. The main purpose of this work is todevelop a low temperature process allowing the deposition of a thin calcium phosphate layer at the titanium implant surface. The composition and structure of this calcium phosphate coating have to be close to the bone mineral to enable osseointegration improvement. Moreover, thecoating should have antibacterial properties in order to prevent post-operative infections. A surfacetreatment protocol composed of sandblasting and acid etching was firstly developed creating anaverage roughness of 1.4 – 1.8 µm with micropits and improved wettability. Secondly, two processes were studied to produce the calcium phosphate coating: the alternate soaking processand the electrodeposition. It was demonstrated that the centrifugation step implemented in thealternate soaking process is crucial and a coating of about 2 µm thick composed of biomimeticapatite was obtained. Among all the operating parameters of the electrodeposition process, time ofelectrodeposition has the major impact on the composition and the thickness of the coating. An electrodeposition of 1 mn at -1.6 V/SCE leads to a 1.5 µm thick coating composed of a layer ofoctacalcium phosphate and dicalcium phosphate dihydrate crystals. A screw/unscrew test demonstrated the mechanical stability of the coatings obtained by both processes. Finally,antibacterial ions such as silver, copper and zinc were incorporated in the coatings. Highincorporation rate up to 40 mol.% compared to calcium were determined for copper and zinc.Biological tests were conducted to qualify the effect of these coatings on the biological activity ofhuman mesenchymal cells and on the formation of a biofilm (peri-implantitis model) to preventpost-operative infections. They led to promising results for the development of such bioactivecoatings. This work is part of the BIOACTISURF project, supported by the Midi-Pyrénées Region,and carried out in collaboration with the Laboratory of Chemical Engineering (LGC) and an industrial partner

A novel method based on pulse current electrodeposition (PCE) employing four different waveforms was developed and utilized for fabricating membrane-electrode assemblies (MEAs) with low platinum loading for use in low-temperature proton exchange membrane fuel cells. It was found that both peak deposition current density and duty cycle control the nucleation rate and the growth of platinum crystallites. Based on the combination of parameters used in this study, the optimum conditions for PCE were found to be a peak deposition current density of 400 mA cm-2, a duty cycle of 4%, and a pulse generated and delivered in the microsecond range utilizing a ramp-down waveform. MEAs prepared by PCE using the ramp-down waveform show performance comparable with commercial MEAs that employ ten times the loading of platinum catalyst. The thickness of the pulse electrodeposited catalyst layer is about 5-7 µm, which is ten times thinner than that of commercial state-of-the-art electrodes. MEAs prepared by PCE outperformed commercial MEAs when subjected to a series of steady-state and transient lifetime tests. In steady-state lifetime tests, the average cell voltage over a 3000-h period at a constant current density of 619 mA cm-2 for the in-house and the state-of-the-art MEAs were 564 mV and 505 mV, respectively. In addition, the influence of substrate and carbon powder type, hydrophobic polymer content in the gas diffusion layer, microporous layer loading, and the through-plane gas permeability of different gas diffusion layers on fuel cell performance were investigated and optimized. Finally, two mathematical models based on the microhardness model developed by Molina et al. [J. Molina, B. A. Hoyos, Electrochim. Acta, 54 (2009) 1784-1790] and Milchev [A. Milchev, “Electrocrystallization: Fundamentals of Nucleation And Growth” 2002, Kluwer Academic Publishers, 189-215] were refined and further developed, one based on pure diffusion control and another based on joint diffusion, ohmic and charge transfer control developed by Milchev [A. Milchev, J. Electroanal. Chem., 312 (1991) 267-275 & A. Milchev, Electrochim. Acta, 37 (12) (1992) 2229-2232]. Experimental results validated the above models and a strong correlation between the microhardness and the particle size of the deposited layer was established.

A novel method based on pulse current electrodeposition (PCE) employing four different waveforms was developed and utilized for fabricating membrane-electrode assemblies (MEAs) with low platinum loading for use in low-temperature proton exchange membrane fuel cells. It was found that both peak deposition current density and duty cycle control the nucleation rate and the growth of platinum crystallites. Based on the combination of parameters used in this study, the optimum conditions for PCE were found to be a peak deposition current density of 400 mA cm-2, a duty cycle of 4%, and a pulse generated and delivered in the microsecond range utilizing a ramp-down waveform. MEAs prepared by PCE using the ramp-down waveform show performance comparable with commercial MEAs that employ ten times the loading of platinum catalyst. The thickness of the pulse electrodeposited catalyst layer is about 5-7 µm, which is ten times thinner than that of commercial state-of-the-art electrodes. MEAs prepared by PCE outperformed commercial MEAs when subjected to a series of steady-state and transient lifetime tests. In steady-state lifetime tests, the average cell voltage over a 3000-h period at a constant current density of 619 mA cm-2 for the in-house and the state-of-the-art MEAs were 564 mV and 505 mV, respectively. In addition, the influence of substrate and carbon powder type, hydrophobic polymer content in the gas diffusion layer, microporous layer loading, and the through-plane gas permeability of different gas diffusion layers on fuel cell performance were investigated and optimized. Finally, two mathematical models based on the microhardness model developed by Molina et al. [J. Molina, B. A. Hoyos, Electrochim. Acta, 54 (2009) 1784-1790] and Milchev [A. Milchev, “Electrocrystallization: Fundamentals of Nucleation And Growth” 2002, Kluwer Academic Publishers, 189-215] were refined and further developed, one based on pure diffusion control and another based on joint diffusion, ohmic and charge transfer control developed by Milchev [A. Milchev, J. Electroanal. Chem., 312 (1991) 267-275 & A. Milchev, Electrochim. Acta, 37 (12) (1992) 2229-2232]. Experimental results validated the above models and a strong correlation between the microhardness and the particle size of the deposited layer was established.

This dissertation presents an investigation of miniaturized magnetic induction and permanent magnet (PM) machines, intended for use in a microengine. Similar to a macroscale turbogenerator, a microengine comprises a small, gas-fueled turbine engine for converting chemical fuel energy into mechanical power and an integrated electrical generator for converting mechanical power to electrical power. The microengine system is proposed as a revolutionary, high power-density source for portable electronics. In this research, miniaturized magnetic induction machines and PM machines were designed, fabricated, and characterized. Both types of machines used axially directed magnetic fields and were nominally 10 mm in diameter and 1.5-2.3 mm in thickness. Innovative microfabrication techniques were developed to demonstrate the feasibility of integrating magnetic machines within a bulk-micromachined, silicon-based microengine system. Two-phase, eight-pole induction machines were constructed within silicon substrates using Cu coils in a laminated, slotted ferromagnetic NiFe or CoFeNi stator core. Silicon etching, wafer bonding, and electrodeposition were used to form all of the magnetic machine components. The induction machines were characterized in motoring mode using tethered rotors and demonstrated motoring torques of up to 2.5 uN-m. Also, three-phase, eight-pole, surface wound PM machines were built using a hybrid microfabrication/assembly approach. The stators were fabricated by electroplating Cu coils on ferromagnetic NiFeMo (Moly Permalloy) substrates. The rotors were formed by assembling a magnetically patterned SmCo PM with a FeCoV (Hiperco 50) back iron. The PM machines were tested as generators with free-spinning rotors, powered by an air-driven spindle, and demonstrated 2.6 W of mechanical-to-electrical power conversion with continuous DC power generation of 1.1 W at 120 krpm rotor speed. The primary contributions of this work are (1) the demonstration of microfabricated magnetic machines integrated within bulk-micromachined silicon and (2) the demonstration of multi-watt power conversion from a microfabricated PM generator. These achievements represent progress in the ongoing development of silicon-based microengines, but in addition, the fabrication technologies and device structures may find application in other microsystems.

Bipolar electrochemistry deals with the exposure of an isolated conducting substrate that has no direct connection with a power supply except via an electric field. Therefore it can be considered as a "wireless technique". The polarization of the substrate with respect to the surrounding medium generates a potential difference between its opposite ends which can support localized electrochemical oxidation reduction reactions and break the surface symmetry of the substrate. The method was applied in the present thesis to materials science and biological systems. In the frame of designing asymmetric particles, also called "Janus" particles, bipolar electrochemistry was adapted for the bulk preparation of these objects. Conductive substrates with different nature, sizes and shapes have been modified with various materials such as metals, ionic and inorganic compounds using this approach. Moreover, a control over the deposit topology could be achieved for substrates at different length scales. Bipolar electrodeposition is also a good tool for investigating the generation of different metal morphologies. Further developments in the bipolar setup allowed us to use the technology for microstructuration of conductive objects. Furthermore the concept has shown to be very useful in the field of the induced motion of particles. The asymmetric objects that have been prepared by bipolar electrodeposition were employed as microswimmers which could show both translational and rotational motion. The application of electric fields in the bipolar setup can be used for the direct generation of motion of isotropic objects through bubble generation. A levitation motion of objects combined with light emission was possible using this concept. Finally, bipolar electrochemistry was also used for studying the intrinsic conductivity of biological molecules (DNA), which is of great importance in the nanotechnology.

Thin film solar cells of Copper Indium Telluride and Cadmium Sulfide junctions were fabricated on plain ITO glass slides and also on those coated with intrinsic Tin Oxide. CdS was deposited through chemical bath deposition and CIT by electrodeposition. Both compounds were subjected to annealing at temperatures between 350°C and 500°C which produced more uniform film thicknesses and larger grain sizes. The CIT/ CdS junction was characterized after performing XRD and spectral absorption of individual compounds. Studies were also made on CdS / CdTe solar cells with respect to effect of annealing temperatures on open circuit voltages. NP acid etch, the most important process to make the surface of CdTe tellurium rich, was also studied in terms of open circuit voltages. Thermally evaporated CdS of four different thicknesses was deposited on Tin Oxide coated ITO and inferences were drawn as to what thickness of CdS yields better results.

Austenitic stainless steels are commonly used in nuclear reactors and have been considered as potential structural materials in fusion reactors due to their excellent corrosion resistance, good creep and fatigue resistance at elevated temperatures, but their relatively low yield strength and poor radiation tolerance hinder their applications in high dose radiation environments. High angle grain boundaries have long been postulated as sinks for radiation-induced defects, such as bubbles, voids, and dislocation loops. Here we provide experimental evidence that high angle grain boundaries can effectively remove radiation-induced defects. The equal channel angular pressing (ECAP) technique was used to produce ultrafine grained Fe-Cr-Ni alloy. Mechanical properties of the alloy were studied at elevated temperature by tensile tests and in situ neutron scattering measurements. Enhanced dynamic recovery process at elevated temperature due to dislocation climb lowers the strain hardening rate and ductility of ultrafine grained Fe-Cr-Ni alloy. Thermal stability of the ultrafine grained Fe-Cr-Ni alloy was examined by ex situ annealing and in situ heating within a transmission electron microscope. Abnormal grain growth at 827 K (600°C) is attributed to deformation-induced martensite, located at the triple junctions of grains. Helium ion irradiation studies on Fe-Cr-Ni alloy show that the density of He bubbles, dislocation loops, as well as irradiation hardening are reduced by grain refinement. In addition, we provide direct evidence, via in situ Kr ion irradiation within a transmission electron microscope, that high angle grain boundaries in nanocrystalline Ni can effectively absorb irradiation-induced dislocation loops and segments. The density and size of dislocation loops in irradiated nanocrystalline Ni were merely half of those in irradiated coarse grained Ni. The results imply that irradiation tolerance in bulk metals can be effectively enhanced by microstructure refinement.

Dans les dispositifs actuels de sauvegarde de données, les bits d'informations sont stockées sous la forme de paroi de domaines dans une couche mince, voire des media "patternés". Le support reste donc 2D. De nos jours, la densité de stockage tend vers une valeur maximale qu'il est difficile de dépasser pour des raisons fondamentales et technologiques. Ainsi, récemment des efforts ont été réalisés pour développer des dispositifs 3D qui allient la polyvalence de la mémoire RAM solide avec un coût comparable à celui des disques durs actuels.Un nouveau concept théorique particulièrement intéressant pour une mémoire magnétique en 3D a été proposé en 2004 par S. Parkin et al.. Cette mémoire de type registre à décalage est constituée d'un réseau de nanofils magnétiques verticaux avec une section transversale cylindrique ou bien rectangulaire. Dans ce nouveau type de mémoire, les bits sont codés sous forme d'une série de parois de domaine. Cette dernière peut être déplacée vers une tête de lecture intégrée par des impulsions de courant polarisé en spin de quelques nanosecondes.Les parois de domaines magnétiques dans des nanofils cylindriques ont suscité l'intérêt de la communauté scientifique en raison de leur application possible dans un dispositif fonctionnel ainsi qu'en raison de nouvelles propriétés intéressantes qui résultent du confinement géométrique des parois. A ce jour, seules quelques études expérimentales sur de telles parois de domaines existent. Elles ont mis en évidence la difficulté de maîtriser la propagation de parois dues à des forts effets de piégeage. Jusqu'à présent, l'origine microscopique de ce piégeage n'a été que partiellement comprise. On s'attend à ce qu’indépendamment de la qualité géométrique du fil, la microstructure du matériau puisse jouer un rôle non négligeable.Dans le cadre du projet européen FP7 m3D, l'objectif de mon travail de thèse a été d'étudier la propagation des parois de domaine dans des nanofils cylindriques avec des modulations de diamètre. L'énergie de ces parois de domaine augmentant avec le diamètre du fil, on s'attend à ce que des excroissances (ou des constrictions) agissent comme des barrières d'énergie artificielles (respectivement puits). Par conséquent, une propagation de paroi de domaine contrôlée via la géométrie du fil semble possible.La première partie de mon travail concerne l'optimisation des matériaux. Des fils d'un alliage de NiCo (diamètre de 100-200nm et longueur de plusieurs dizaines de micromètres) avec deux géométries distinctes ont été fabriqués par électrodéposition en collaboration avec le groupe du Prof. J. Bachmann à l' Université d'Erlangen. Pour chaque géométrie, j'ai exploré l'effet de la composition de l'alliage ainsi que d'un recuit sur la microstructure du matériau. Par la suite, la propagation des parois de domaine dans des nanofils individuels a été étudiée sous l'influence d'un champ magnétique quasi-statique ou d'une impulsion de champ magnétique avec une durée d'impulsion de l'ordre de la nanoseconde. Dans la dernière partie de ma thèse, j'ai effectué des simulations micromagnétiques complémentaires pour étudier l'effet de la géométrie des modulations sur le piégeage de ces parois de domaine magnétiques. / In all current data storage devices, the information bits are stored in form of domain walls in a thin film or in patterned media on a two-dimensional surface . Within the next decade, further increase of the storage density in these devices is expected to come to a halt due to several fundamental and technological issues. Thus there have recently been efforts to develop three-dimensional devices combining the versatility of solid state RAM with the cost efficiency of common hard disk drives.A particularly interesting theoretical concept for a three-dimensional magnetic memory has been proposed in 2004 by S. Parkin et al. . Their racetrack memory consists of a vertical array of magnetic nanowires with either cylindrical or rectangular cross section. The bits are encoded in a series of up to 100 domain walls per wire. Using nanosecond spin polarized current pulses these walls are shifted past an integrated read head.Magnetic domain walls in cylindrical nanowires have raised the interest of the scientific community due to their possible application in a functional device as well as due to exciting new properties which arise from the geometric confinement. Up to date, only a few pioneering experimental studies on such domain walls exist. They indicate strong pinning effects preventing a deterministic domain wall propagation. So far the microscopic origin of this pinning has only partially been understood. It is expected however that beside the wire geometry the material microstructure may play a considerable role.Situated within the framework of the European FP 7 project m3D, the objective of my work has been to investigate the domain wall propagation in cylindrical nanowires with diameter modulations by means of magnetic force microscopy and micromagnetic simulation. As the domain wall energy increases with the wire diameter, protrusions (resp. notches) are expected to act as an artificial energy barrier (resp. well). Consequently, a deterministic domain wall propagation controlled via the wire geometry seems possible.A first part of my work concerns material optimization. For this, NiCo alloy wires (100-200nm diameter and multiple tens of micrometers in length) with two distinct geometries have been fabricated by template assisted electrodeposition (Chemist collaborators at Univ. Erlangen, Prof. J.Bachmann). I have then explored the impact of the alloy composition as well as of possible post-fabrication annealing on the material microstructure. Subsequently, domain wall propagation in individual nanowires has been investigated under the influence of either a quasistatic magnetic field or a nanosecond magnetic field pulse. In addition I have performed complementary micromagnetic simulations to study the effect of the modulation geometry on the domain wall pinning.