Magnetism of carbon-based materials is a challenging area for both fundamental research and possible applications. We present studies of low-dimensional carbon-based magnetic systems (fullerene-diluted molecular magnets, carbon nanotubes, graphite fluoride, and nanoporous carbon) by means of SQUID magnetometer, X-ray diffraction and vibrational spectroscopy, the latter techniques used as complementary instruments to find a correlation between the magnetic behaviour and the structure of the samples.In the first part of the thesis, characteristic features of the magnetization process in aligned films of carbon nanotubes with low concentration of iron are discussed. It is shown that the magnetism of such structures is influenced by quantum effects, and the anisotropy behaviour is opposite to what is observed in heavily doped nanotubes.In the second part, Mn12-based single molecular magnets with various carboxylic ligands and their 1:1 fullerene-diluted complexes are studied. We prove that magnetic properties of such systems strongly depend on the environment, and, in principle, it is possible to design a magnet with desirable properties. One of the studied compounds demonstrated a record blocking temperature for a single molecular magnet. Both fullerene-diluted complexes demonstrated “magnetization training” effect in alternating magnetic fields and the ability to preserve magnetic moment.The third and the fourth parts of the thesis are dedicated to the analysis of various contributions to the magnetic susceptibility of metal-free carbon-based systems – intercalated compounds of graphite fluorides and nanoporous oxygen-eroded graphite. The magnetic properties of these systems are strongly dependent on structure, and can be delicately tuned by altering the π-electron system of graphite, i. e. by degree of fluorination of intercalated compounds and by introduction of boron impurity to the host matrix of nanoporous graphite. / Magnetism av kolbaserade material är ett utmanande område för både grundforskning och möjliga tillämpningar. Vi presenterar studier med låg-dimensionella kolbaserade magnetiska system (fulleren-utspädda molekylära magneter, kolnanorör, grafit fluorid och nanoporösa kol) med hjälp av SQUID magnetometer, röntgendiffraktion och vibrerande spektroskopi, de senare tekniker som används som komplement instrument för att finna sambandet mellan den magnetiska uppträdande och strukturen hos proven. I den första delen av avhandlingen är egenheter från magnetisering processen i linje filmer av kolnanorör med låg koncentration av järn diskuteras. Det visas att magnetism av sådana strukturer påverkas av kvantmekaniska effekter och anisotropin beteende är motsatsen till vad som observerats i kraftigt dopade nanorör. I den tvåa delen är Mn12-baserade enda-molekyl magneter med olika karboxylsyror ligander och deras 1:1 fulleren-utspädda komplex studeras. Vi visar att magnetiska egenskaperna hos sådana system beror i hög grad på miljön, och i princip är det möjligt att utforma en magnet med önskvärda egenskaper. En av de studerade föreningarna visade en post blockeringstemperaturen för en enda molekylär magnet. Både fulleren-utspädda komplex visade "magnetisering utbildning" effekt i alternerande magnetfält och möjligheten att bevara magnetiskt moment. Den tredje och fjärde delarna av avhandlingen är avsedda för inneboende magnetism av analys av olika bidrag till magnetisk susceptibilitet av metall-fritt kol-baserade system -inskjutna föreningar grafit fluorider och nanoporösa O2-eroderade grafit. Magnetiska egenskaperna hos dessa system är starkt beroende av strukturen, och kan fint avstämmas genom att man ändrar π-elektronsystem av grafit, i. e. med graden av fluorering av inskjutna föreningar och genom införandet av bor föroreningar till värd matris av nanoporösa grafit.

Increased demand of power, limited fuel resources and environmental concerns have recently prompted a huge thrust on research areas of alternative energy and photovoltaics have been hailed as energy source for future. Particularly, third generation solar cell configurations like dye-sensitized solar cells and quantum dot Schottky barrier solar cells have drawn more attention because of their ease of processability, cheap cost with decent performance, low payback time and portability. Quantum dots are very attractive materials as sensitizers because of their size dependent bandgap tunability, increased oscillator strength and hence higher absorption coefficient, possibility of multiple exciton generation and photochemical robustness. Hence syntheses of quantum dot based hybrid nanostructures have received huge attention among researchers for using it quantum dot sensitized solar cell configuration. This dissertation can be divided in two parts. In the first part two different methods have been reported to prepare interconnected mesoporous nanostructures of wide band gap semiconductors like TiO2 and ZnO which is very important in providing high surface area for absorption or attachment of the sensitizers. In the second part, methods have been developed to establish direct contacts between quantum dots and wide bandgap substrates without molecular linkers which are expected to increase the electron injection rate from quantum dots to TiO2/ZnO. The entire thesis based on the results and findings obtained from the present investigation is organised as follows: Chapter-I provides a general introduction on the working principle of different type of solar cells and then gives a detailed description of the structure and electronic process of dye sensitised solar cells. Then, benefits of quantum dots as sensitizer over dye molecules has been discussed followed by the modification needed in case of quantum dot sensitized solar cells. Chapter-II deals with the materials and methods which essentially gives the information about the materials used for the synthesis and the techniques utilized to characterize the materials chosen for the investigation. Chapter-III describes a hybrid sol-gel combustion technique to synthesize large quantities of highly crystalline and phase-pure anatase powder in a single step. Titanium isopropoxide reacts with oleic acid to form a viscous liquid (oxocarboxoalkoxide) which undergoes non-hydrolytic polycondensation to form TiO2 during combustion. Oleylamine takes part in formation of reverse micelle which expands during combustion giving rise to porous interconnected membrane like microstructure of pore size ~5 nm, BET surface area of ~100 m2/g and porosity of ~ 48%. More importantly, this porous powder having a pre-existing network can be used to form thicker film by doctor blade technique from its paste and at the same time is expected to have better transport properties due to its less particulate nature. Chapter-IV presents a general method to prepare mesoporous structure from rod-like morphologies by partial sintering of a green pellet. Material having inherent anisotropy in their crystal structure tends to grow in a particular direction rather undergoing equiaxial growth. For instance, hexagonal ZnO and tetragonal rutile usually grow as rod-shaped particles. A loose compact of these nanorods give nanoporous morphology upon heating. Advantage of this method is the tunability of pore size by tuning the aspect ratio of the nanorods. Preparation of porous TiO2, ZnO and hydroxyapatite has been demonstrated from their corresponding nanorods. Chapter-V deals with a solvothermal based technique that has been developed for in-situ deposition of nanoparticles on any plane or curved surfaces conformally. This has been demonstrated for nanoparticles of FeCo, Au, Co, CdS on substrates like glass, mica, Si, NaCl, Al2O3 M-plane and also conformal coating of Au nanoparticles on polystyrene latex spheres. CdSe on rutile nanorods, ZnO nanorods and CNTs are promising hybrid nanostructures for third generation photovoltaics and their successful preparation has been detailed in the chapter. The mechanism proposed is based on dominant attractive sphere-plate interaction under high temperature and high autogeneous pressure condition which at reduced density and surface tension of the solvent reduces the dispersibility of the nanoparticle and allow effective spreading of the nanoparticles on the substrate. This method is also advantageous for coating of complicated geometry like inner walls of porous structures. Chapter-VI presents a method to coat chalcogenide nanoparticles on mesoporous TiO2 without any molecular linker which can enhance the electron injection rate from the chalcogenide quantum dots to TiO2. CdS, PbS can be easily synthesized through aqueous chemistry. For deposition of these sulfides, the ion layer gas absorption and reaction (IGLAR) method was modified to form uniform dense nanoparticles on anatase and ZnO surfaces. Nitrate salts of corresponding metal ions are dried directly on the semiconductor surface and instead of exposing it to H2S gas, it was treated with a concentrated sulfide solution. This introduces two competitive process i) dissolution of nitrate salt ii) formation of the metal sulfide. This dissolution step was absent when treated with H2S gas (IGLAR) and hence lead to a continuous coating. We have successfully produced CdS-TiO2 and PbS-TiO2 composites using this approach. Photoelectrochemical measurements on CdSTiO2 composites show an overall efficiency of 2.8% which is among the highest values obtained for this system demonstrating the applicability of the method to engineer interfaces to achieve high efficiency solar cells. Chapter-VI explores the combination of strategies of nanocrystal conversion chemistry with previously described sol-gel combustion technique to create dense and uniformly coated QD sensitized TiO2 electrode without compromising heat-treatment routines which is essential for better adhesion and to enhance performance with reduced leakage. Intimate biphasic oxide mixtures of PbO and CdO with TiO2 are first synthesized by nonhydrolytic solgel process with is followed by combustion to produce porous morphology. This powder can be coated as electrode and can sustain high temperature heat treatment routines and finally can be selectively converted to sulfides with Na2S treatment as TiO2 is immune to sulfidation under this condition. Materials at different stages are characterised by XRD, TEM, EDS, UV-Vis and XPS.

Nanoporous carbons have become a material of interest in many applications such as electrochemical double layer capacitors (supercapacitors). Supercapacitors are being studied for their potential in storing electrical energy storage from intermittent sources and in use as power sources that can be charged rapidly. However, a lack of understanding of the charge storage mechanism within a supercapacitor makes it difficult to optimize them. Two components of this challenge are the difficulties in experimentally characterizing the sub-nanoporous structure of carbon electrode materials and the electrical performance of the supercapacitors. This work provides a means to accurately characterize the porous structure of sub-nanoporus carbon materials and identifies the current limitations in characterizing the electrical performance of a supercapacitor cell. Future work may focus on the relationship between the sub-nano porous structure of the carbon electrode and the capacitance of supercapacitors, and on the elucidation of charge storage mechanisms.

Nanoporous carbons have become a material of interest in many applications such as electrochemical double layer capacitors (supercapacitors). Supercapacitors are being studied for their potential in storing electrical energy storage from intermittent sources and in use as power sources that can be charged rapidly. However, a lack of understanding of the charge storage mechanism within a supercapacitor makes it difficult to optimize them. Two components of this challenge are the difficulties in experimentally characterizing the sub-nanoporous structure of carbon electrode materials and the electrical performance of the supercapacitors. This work provides a means to accurately characterize the porous structure of sub-nanoporus carbon materials and identifies the current limitations in characterizing the electrical performance of a supercapacitor cell. Future work may focus on the relationship between the sub-nano porous structure of the carbon electrode and the capacitance of supercapacitors, and on the elucidation of charge storage mechanisms.

Made available in DSpace on 2014-12-17T14:07:09Z (GMT). No. of bitstreams: 1 JoseFTJ_TESE.pdf: 4834280 bytes, checksum: 1a618560296186081bd6c34117d226be (MD5) Previous issue date: 2012-07-04 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / Thin commercial aluminum electrolytic and passed through reactions was obtained with anodic alumina membranes nanopores. These materials have applications in areas recognized electronic, biomedical, chemical and biological weapons, especially in obtaining nanostructures using these membranes as a substrate or template for processing nanowires, nanodots and nanofibers for applications noble. Previous studies showed that the membranes that have undergone heat treatment temperature to 1300? C underwent changes in morphology, crystal structure and optical properties. This aim, this thesis, a study of the heat treatment of porous anodic alumina membranes, in order to obtain and to characterize the behavior changes structures during the crystallization process of the membranes, at temperatures ranging between 300 and 1700? C. It was therefore necessary to mount a system formed by a tubular furnace resistive alumina tube and controlled environment, applying flux with special blend of Ag-87% and 13% N2, in which argon had the role of carrying out the oxygen nitrogen system and induce the closing of the pores during the densification of the membrane. The duration of heat treatment ranged from 60 to 15 minutes, at temperatures from 300 to 1700? C respectively. With the heat treatment occurred: a drastic reduction of porosity, grain growth and increased translucency of the membrane. For the characterization of the membranes were analyzed properties: Physical - thermogravimetric, X-ray diffraction, BET surface area; morphological - SEM, EDS through compositional and, optical absorbance, and transmittance in the UV-VIS, and FTIR. The results using the SEM showed that crystallization has occurred, densification and significant changes in membrane structure, as well as obtaining microtube, the BET analysis showed a decrease in specific surface area of the membranes has to 44.381 m2.g-1 to less than 1.8 m2.g-1 and in the analysis of transmittance and absorbance was found a value of 16.5% in the range of 800 nm, characteristic of the near infrared and FTIR have confirmed the molecular groups of the material. Thus, one can say that the membranes were mixed characteristics and properties which qualify for use in gas filtration system, as well as applications in the range of optical wavelength of the infra-red, and as a substrate of nanomaterials. This requires the continuation and deepening of additional study / L?minas delgadas de alum?nio comercial passaram por rea??es eletrol?ticas e obtiveram-se membranas de alumina an?dica com nanoporos. Estes materiais t?m reconhecidas aplica??es nas ?reas eletr?nicas, biom?dicas, qu?micas e biol?gicas, principalmente, na obten??o de nanoestruturas utilizando estas membranas como substrato ou molde para processamento de nanofios, nanopontos e nanofibras para aplica??es nobres. Estudos anteriores apontaram que as membranas que passaram por tratamentos t?rmicos at? a temperatura de 1300? C, sofreram altera??es na morfologia, na estrutura cristalina e nas propriedades ?pticas. O objetivo deste trabalho foi o estudo do tratamento t?rmico de membranas de alumina an?dica porosas, com o intuito de obter e caracterizar as altera??es de comportamento das estruturas, durante o processo de cristaliza??o das membranas, em temperaturas que variaram entre 300 e 1700? C. Assim, foi necess?rio montar um sistema formado por um forno resistivo tubular e tubo de alumina com ambiente controlado, aplicando fluxo com mistura especial de Ar-87% e N2-13%, no qual o arg?nio teve o papel de carrear o oxig?nio para fora do sistema e o nitrog?nio de induzir o fechamento dos poros, durante a densifica??o das membranas. A dura??o dos tratamentos t?rmicos variou de 60 a 15 minutos, para as temperaturas de 300 at? 1700? C respectivamente. Com o tratamento t?rmico ocorreu redu??o dr?stica da porosidade, crescimento do gr?o e aumento da translucidez da membrana. Para a caracteriza??o das membranas, foram feitas an?lises das propriedades: f?sica - termogravim?trica; difra??o de raios-X, ?rea superficial BET; morfol?gica - MEV, composicional atrav?s do EDS; e, ?ptica - transmit?ncia e absorb?ncia no UV-VIS e FTIR. Os resultados por meio do MEV mostraram que ocorreu cristaliza??o, densifica??o e mudan?as significativas na estrutura das membranas, bem como, a obten??o de microtubo; a an?lise de BET constatou uma diminui??o na ?rea superficial espec?fica das membranas que passou de 44,381m2.g-1, para menos de 1,8m2.g-1; na an?lise de transmit?ncia e absorb?ncia foi encontrado um valor de 16,5 % na faixa de 800nm, caracter?stico do infravermelho pr?ximo e no FTIR foram confirmadas os grupos moleculares do material. Assim, pode-se afirmar que as membranas apresentaram caracter?sticas mistas e propriedades que as qualificam para o uso em sistema de filtra??o de gases, bem como, de aplica??es ?ticas na faixa do comprimento de onda do Infravermelho, e como substrato de nanomateriais. Isto requer a continuidade e aprofundamento em estudos complementares

Les nanofils de silicium présentent un fort potentiel d'intégration, et leur utilisation dans des dispositifs électroniques tels que des cellules solaires photovoltaïques ne peut se faire que si leur élaboration et leurs propriétés structurales sont maitrisées. Nous présentons dans cette thèse une méthode de fabrication de matrices de nanofils de silicium par croissance catalysée par l'or ou le cuivre en dépôt chimique en phase vapeur et faisant appel à des matrices de guidage de la croissance en alumine nanoporeuse. Cette technique permet notamment la croissance d'assemblées de nanofils ultra-denses (1.10^{10} nanofils/cm²) sur substrat non préférentiel ou d'hétérostructures comme des nanofils de germanium sur substrat de silicium. Grâce à la diffraction des rayons X nous montrons ensuite que les nanofils produits sont de très bonne qualité structurale malgré leur substrat non préférentiel et la présence d'une légère déformation de leur maille cristalline. Le contrôle de la déformation cristalline de nanofils de germanium est par ailleurs démontré en encapsulant les nanofils dans une coquille de nitrure de silicium. De nouveaux éléments de réflexion sont également rapportés concernant la contamination des nanofils de silicium par le catalyseur de leur croissance. Enfin l'intégration des nanofils de silicium dans des dispositifs solaires photovoltaïques est démontré en faisant appel à des jonctions PN radiales entre le coeur et la coquille des nanofils. / Silicon nanowires are promising objects but their integration in electronic devices such as photvoltaic solar cells relies on the ability to control their production and tailor their structural properties. In this thesis we present a method to produce nanowire matrices using a gold or copper catalysed growth process by chemical vapor deposition and using a nanoporous alumina growth template. This method enables the fabrication of ultra-dense nanowire arrays (1.10^{10} nanowires/cm²) on non preferential substrate or heterostructures such as germanium nanowires on silicon substrate. Using X-ray diffraction we also show that the structural quality of the template grown nanowires is very good in spite of their non preferential substrate and the presence of a small cristalline lattice strain. The control of germanium nanowires strain is also demonstrated by embeding them in a silicon nitride shell. Besides, new results are presented concerning the catalyst contamination of silicon nanowires. Silicon nanowires integration in photovoltaic devices is eventually demonstrated using a radial geometry for the PN junction between the core and the shell of the nanowires.

Anodisation is a surface treatment process, commonly used to form a protective oxide coating on the surface of metals like aluminium. Anodised coatings, being grown out of the base metal have excellent interface strength but are porous and brittle. Porosity of the coating reduces the hardness and the brittle nature of the oxide induces cracking. In practice, the pores are typically filled with organic dye and sealed. Under certain controlled electrochemical conditions, anodisation results in a highly ordered hexagonal porous structure in pure aluminium. In this work, we explore the possibility of using this ordered porous alumina to form a novel metal nanocomposite as a tribological coating. By optimizing the nonporous structure and tuning the electrodeposition process, we uniformly filled the ordered pores with copper. We have measured the hardness of the resulting ordered and aligned nanocomposite. We explore the possibility of using this composite coating for tribological applications by carrying out some preliminary reciprocating wear test. Ordered porous alumina layer is formed by a two-step anodisation process. By optimizing the anodisation conditions, we control the thickness of the coating and the pore size. The interface of the porous structure and aluminium substrate is defined by a non-conducting dense barrier oxide layer. However, to deposit metal into the pores, a conducting path should be established through the barrier layer. One possibility is to etch out the bottom of the pores at the cost of the interface strength and losing out on the main advantage of anodised coatings. To be able to fill metal without this sacrifice, we utilised the dendritic structure in the barrier layer formed by a step-wise reduction of voltage towards the end of anodisation process. Optimisation of this dendritic structure led to uniform deposition of metal into pores, achieved by pulsed electrodeposition. In pulse lectrodeposition, a positive pulse is applied to remove accumulated charge near to the bottom of pores, followed by a negative pulse to deposit metal and a delay to allow diffusion of ions. By optimising the pulse shape and duration, we have achieved uniform growth of metal into pores. Further, monitoring the deposition current helped us to identify and control different phases of growth of the nanowire. The properties of the porous alumina and the nanocomposite were measured by nanoindentation. The deformation characteristics were obtained by observing the indents in a FE-SEM. We find that dendritic modification of interface has very little effect on the hardness of the porous alumina layer. We also found that the porous alumina deformed either by compaction or by forming circumferential and radial cracks. When copper is filled in the nano pores, the hardness increased by 50% and no circumferential cracks were found up to the load of 10 mN for a film thickness of about 1 µm. Coefficient of friction of the coating reciprocated against steel in dry condition is found to be around 0.4. Minimal wear was observed from the SEM images of wear track. In summary, a novel nanocomposite coating with ordered porous alumina as matrix embedded with aligned metal nano rods has been developed. This was achieved by optimally modifying the barrier layer without sacrificing the interfacial strength. Uniform coating has been achieved over an area of 10 mm x 10 mm. The coating is found to have high hardness and high wear resistance.

L'objectif de la thèse présentée ici est l'optimisation de matériaux carbonésnanoporeux au moyen de la “conception de matériaux virtuels”. En ce qui concerne cette échelle de travail (~ 10nm), la Nanotomographie FIB-SEM est la seule technique d'imagerie donnant accès à une information sur la géométrie tridimensionnelle. Cependant, pour l'optimisation du comportement, l'espace des pores doit être reconstruit à partir des données tirées des images obtenues. Jusqu'à présent ce problème n'était pas résolu. Pour pouvoir le maîtriser, on a développé une simulation d'images FIB-SEM. Les images FIB-SEM simulées peuvent être utilisées pour la vérification et la validation des algorithmes de segmentation. En utilisant les données d'image simulées, un nouvel algorithme pour la reconstruction de l'espace des pores à partir des données FIB-SEM a été développé.Deux études de cas avec des carbones nanoporeux utilisés pour le stockage d'énergie sont présentées, en utilisant les nouvelles techniques pour la caractérisation et l'optimisation des électrodes Li-ion de type EDLC'S (« electric double-layer capacitors », soit supercondensateurs). L'espace des pores reconstruit est modélisé géométriquement à l'aide de la géométrie stochastique. Enfin, on a simulé les propriétés électriques des matériaux enutilisant des structures modélisées et simulées. / The aim of the work presented here is to optimize nanoporous carbon materials by means of 'virtual material design'. On this length scale (~ 10nm) Focused Ion Beam – Scanning Electron Microscopy Nanotomography (FIB-SEM) is the only imaging technique providing three dimensional geometric information. Yet, for the optimization, the pore space of the materials must be reconstructed from the resulting image data, which was a generally unsolved problem so far.To overcome this problem, a simulation method for FIB-SEM images was developed. The resulting synthetic FIB-SEM images could then be used to test and validate segmentation algorithms. Using simulated image data, a new algorithm for the morphological segmentation of the highly porous structures from FIB-SEM data was developed, enabling the reconstruction of the three dimensional pore space from FIB-SEM images.Two case studies with nanoporous carbons used for energy storage are presented, using the new techniques for the characterization and optimization of electrodes of Li-ion batteries and electric double layer capacitors (EDLC's), respectively. The reconstructed pore space is modeled geometrically by means of stochastic geometry. Finally, the electrical properties of the materials were simulated using both imaged real and modeled structures.

L’objectif de ce mémoire est de présenter une étude aussi complète que possible de l’effet Stark dans les toupies asymétriques de type X2Y4 possédant le groupe de symétrie D2h et de l’appliquer au problème de la spectroscopie de l’éthylène piégé dans une zéolithe. Pour la première fois, un formalisme tensoriel adapté à l’étude de l’effet Stark pour la chaine de groupes O(3) D2h a été développé. Parallèlement au développement du modèle théorique, des programmes informatiques ont été mis au point sous la forme d’une nouvelle génération de logiciel appelé D2hTDS-ST utilisant une base plus simple, permettant une économie en temps de calcul. Sur la base des outils théoriques et informatiques ainsi qu’à partir des spectres expérimentaux enregistrés au laboratoire, nous avons fait une première estimation de la valeur du champ électrique effectif moyen produit par la silicalite-1. Les valeurs du champ électrique obtenues avec notre méthode sont cohérentes avec les valeurs obtenues par les calculs ab initio PBE1PBE6-31++G(2d,2p). / The objective of this report is to present a study as complete as possible of the Stark effect in the asymmetric top molecules of X2Y4 type possessing the D2h symmetry group and to apply it to the problem of the spectroscopy of the ethylene molecule trapped in a zeolite. For the first time, a tensorial formalism adapted to the study of the Stark effect for the O(3) D2h group chain was developed. In parallel, a computer package was worked out as a new generation of software called D2hTDS-ST using a simpler basis, allowing an economy in the calculation time. On the basis of the theoretical and computing tools as well as from the experimental spectra, recorded in the laboratory, we made a first estimation of the average effective electric field produced by silicalite-1. The values of the electric field obtained with our method are coherent with the values obtained by the ab initio PBE1PBE6-31++G(2d,2p) calculations.

L’objectif principal de la thèse a été d’étudier les propriétés de résistance des matériaux nanoporeux par des approches théoriques et numériques. Dans le contexte des méthodes d’homogénéisation, des critères de résistance macroscopiques ont été établis par des approches analytiques à l’homogénéisation non-linéaire et à l’analyse limite. Les critères de résistance ainsi obtenus permettent de tenir en compte les effets de taille, tout en améliorant les formulations déjà existantes.
 En outre, dans le but de servir de référence pour la calibration et/ou la validation des modèles analytiques, des simulations numériques basées sur la dynamique moléculaire ont été conduites, en se référant à des monocristaux d’aluminium contenant des nanopores sphériques, sous des conditions multiaxiales de vitesse de déformation. Par rapport aux simulations actuellement disponibles dans la littérature, les résultats obtenus ont clairement établi que les surfaces de résistance sont significativement influencées par les trois invariants isotropes de contrainte.
 Finalement, dans le but de mettre à profit les indications fournies par les simulations numériques, le cas d’un matériau nanoporeux constitué d’une matrice ductile sensible aux trois invariants isotropes a été étudié par une approche par l’analyse limite, en prenant en compte une formulation modifiée du critère de résistance de bigoni. La solution exacte du problème a été établie dans le cas d’un chargement isotrope. A partir des résultats ainsi obtenus, une approche d’analyse limite cinématique a été mise en place, et permet de fournir des estimations des propriétés de résistance macroscopiques sous chargements axisymétriques. / The main objectif of the thesis consisted in investigating strength properties of nanoporous materials by means of theoretical and numerical approaches. In the framework of homogenization methods, novel macroscopic strength criteria have been established via a non-linear homogenization procedure and a kinematic limit-analysis approach. Resulting yield functions allowed to take into account void-size effects on nanoporous materials strength properties, thereby resulting in a strong enhancement of available estimates. Furthermore, aiming to funish effective benchmarking evidence for the calibration and/or the assessment of theoretical models, molecular-dynamics based computations have been carried out on in-silico single crystals embedding spherical nanovoids, simulation domains undergoing multiaxial strain-rate boundary conditions. With respect to available numerical studies, proposed results clearly showed the influence of all the three isotropic stress invariants on computed material strength surfaces. Finally, with the aim to account for physical indications coming from numerical simulations, a ductile nanoporous material with a general isotropic plastic matrix has been investigated via a limit analysis approach, by referring to a modified version of the bigoni strength criterion. The limit state of a hollow-sphere model undergoing isotropic loadings has been exactly determined. Correspondigly, a novel strength criterion has been analytically established in the case of axysimmetric boundary conditions.