Comme je suis particulièrement intéressé par les nanosciences et les nombreuses applications des nanotechnologies, je me suis penché sur le développement de méthodes de fabrication de nanoparticules ultra-petites dont les fonctions peuvent être ajustées avec précision. Récemment, une nouvelle technologie appelée « technologie d’une seule chaîne », c’est-à-dire qui utilise une seule chaîne polymère, est devenue un sujet de recherche de plus en plus motivant pour la communauté scientifique. Cette technologie a l’avantage de dépendre d’une méthode facile de préparation de nanoparticules polymères d’une seule chaîne (SCNPs) et ayant des dimensions typiques de 1,5 à 20 nm. Leurs tailles ultra petites leur confèrent des propriétés spécifiques, ce qui permet de les utiliser comme capteurs, systèmes catalytiques, revêtements à faible viscosité, nanoréacteurs ou pour des applications biomédicales. Grâce aux contributions de nombreux scientifiques durant la dernière décennie, les méthodes de synthèse des SCNPs sont devenues très variées et représentent une technologie désormais mature. Néanmoins, de nombreux problèmes sont à résoudre dans ce domaine, ce qui permettra d’ajouter de nouvelles fonctions ou de les valoriser pour de nouvelles applications. Les polymères sensibles à plusieurs stimuli sont une classe de matériaux intelligents dont les propriétés peuvent être modifiées par l’application d’un stimulus extérieur. Ils sont utilisés extensivement dans les domaines énergétique et biomédical. Comme leurs propriétés physiques et chimiques peuvent être modifiées aisément et efficacement par un contrôle de leur environnement externe, ces polymères sont des candidats pour fabriquer de nouvelles SCNPs. Dans cette thèse, nous nous sommes intéressés au développement de SCNPs ayant de multiples fonctionnalités car cela permet d’ouvrir la voie pour de nouvelles applications. Pour cela, de nombreux polymères sensibles à plusieurs stimuli ont été préparés comme précurseurs à des SCNPs. En concevant spécifiquement ces polymères, il fut possible d’ajouter leurs propriétés de réponse à des stimuli dans les systèmes SCNPs. Le cœur même de cette thèse consiste en trois projets qui utilisèrent trois classes de SCNPs provenant de polymères sensibles aux stimuli. Grâce à leur réponse à plusieurs stimuli, ces SCNPs remplirent de nombreuses fonctions et subirent des modifications soit de leur structure, soit de leur morphologie, soit de leurs propriétés. Et en plus de la variété de fonctions, chaque classe de SCNPs a le potentiel pour de nombreuses applications. Dans la première étude présentée dans cette thèse (chapitre 1), nous avons préparé une classe de SCNPs photodégradables ayant une taille ajustable et inférieure à 10 nm. Il s’agit de polyesters rendus photosensibles par la présence de coumarines à l’intérieur de la chaîne principale (nommés CAPPG) grâce à la copolymérisation de coumarine diol, d’acide adipique et de propylène glycol (PPG). Cette incorporation de coumarines dans la chaîne principale permet au polymère d’être photosensible par deux façons. En effet, les coumarines peuvent se photo-dimériser, lorsqu’elles sont irradiées par des rayonnements UV (> 320 nm) en des cyclobutanes qui peuvent être ouverts à nouveau par d’autres rayonnements UV (254 nm) permettant la restauration des coumarines initiales. Cela a permis la création de SCNPs de tailles inférieures à 10 nm et incluant des propriétés de photodégradation. Cette propriété a été démontrée par une irradiation de 3 h avec des chaînes polymères de 13220 g/mol à 1385 g/mol dans les SCNPs. La taille de ces SNCPs (caractérisée par leur rayon hydrodynamique) peut être modifiée entre 3 nm et 5,3 nm en modifiant le taux de dimérisation des coumarines, ce qui est aisément obtenu en ajustant le temps d’irradiation UV. Les résultats ont démontré que cette méthode permet un contrôle aisé de la taille des SCNPs sans avoir recours à la synthèse de nombreux polymères précurseurs. Finalement, comme le polyester était biodégradable et biocompatible, ces SCNPs peuvent être exploitées pour des applications biomédicales. Dans la deuxième étude effectuée au cours de cette thèse (chapitre 2), nous avons préparé un nouveau type de SCNPs multifonctionnel à partir d’un polymère cristallin liquide. Il s’agit du polyméthacrylate de [2- (7-méthylcoumaryl) oxyéthyle - co - 6-[4-(4’-méthoxyphenylazo) phénoxy] hexyle] (PAzoMACMA). Les groupements latéraux du polymère contiennent, en majorité, des azobenzènes photoisomérisables et, en minorité, des coumarines photodimérisables. Les azobenzènes servent de mésogènes pour la formation de cristaux liquides alors que les coumarines ont été utilisées pour une réticulation photoinduite et intrachaîne. Malgré les dimensions inférieures à 15 nm, le confinement et la réticulation, les phases cristallines liquides (LC) persistèrent même dans les SCNPs. Ces SCNPs cristaux liquides (LC-SCNPs) présentèrent un certain nombre de propriétés intéressantes et particulières. Alors que leurs dispersions dans le THF n’étaient pas fluorescentes, celles dans le chloroforme l’étaient. En plus, ces nanoparticules s’aggloméraient quelque peu dans le chloroforme ce qui induisait des fluorescences différentes entre des SCNPs riches en isomères cis ou riches en isomères trans des azobenzènes. A cause de la photoisomérisation des azobenzènes, ces LC-SCNPs se déformaient sous irradiation comme le font les microparticules ou les colloïdes contenant des azobenzènes. Cependant, la déformation de ces nanoparticules dépend de la longueur d’onde de lumière polarisée. Alors que sous irradiation UV polarisée à 365 nm, l’élongation des SCNPs était perpendiculaire à la polarisation de la lumière incidente, sous irradiation visible polarisée entre 400 et 500 nm, l’étirement se faisait parallèlement à la polarisation. Finalement, un nanocomposite fut préparé par dispersion de LC-SCNPs dans une matrice de polyméthacrylate de méthyle (PMMA). Si celui-ci était étiré mécaniquement, les azobenzènes s’orientaient dans la direction de la déformation induite. Ces propriétés intéressantes des LC-SCNPs que cette étude a permis de dévoiler, suggèrent de nouvelles applications potentielles. Dans la troisième étude de cette thèse (chapitre 3), nous avons préparé une classe de SCNPs sensibles à la présence de CO2 et leurs agrégats micellaires auto-assemblés. D’un côté, des SCNPs ont été préparées à partir d’un polyméthacrylate de {(N, N-diméthylaminoéthyle)-co-4-méthyl-[7-(méthacryloyl)-oxyéthyl-oxy] coumaryle} (PDMAEMA-co-CMA). Lorsqu’elles sont dispersées en solution aqueuse, les nanoparticules individuelles peuvent subir des cycles réversibles d’expansion et de rétrécissement sous une stimulation alternative de CO2 et de N2 qui vont protoner et déprotoner les amines tertiaires. D’un autre côté, des SCNPs de type ‘Janus’ (SCJNPs) ont été préparées à partir d’un copolymère dibloc amphiphile : PS-b-P(DMAEMA-co-CMA) (PS correspond au polystyrène qui est hydrophobe). Ce type de SCJNPs peut s’autoassembler sous forme de micelles en solution aqueuse. Sous stimulation CO2 ou N2, l’expansion ou le rétrécissement à l’intérieur des particules permet de grands changements de volume. En plus, ces particules ont été étudiées comme potentiels nanoréacteurs pour des nanoparticules d’or (AuNPs) que ce soit sous formes SCNPs ou micelles SCJNPs. La vitesse de formation des AuNPs augmente sous bullage de CO2 et décroît sous N2. Cela permet de rendre possible cette réaction contrôlable par ces deux gaz. Qui plus est, utiliser des micelles de SCJNPs dont le volume peut être modifié sur un large intervalle en changeant l’intensité de la stimulation de CO2, permit d’obtenir des AuNPs de taille variable. / Abstract : With interests on nanoscience and nanotechnology for many applications, there is a demand for development of fabrication technology of ultra-small nano-size objects that allow for precise size control and tailored functionality. Recently, a new technology called ‘single-chain technology’, which manipulates a single polymer chain, becomes a rapidly-growing research topic. This technology provides a facile method to prepare polymer single-chain nanoparticles (SCNPs) with a typical size of 1.5-20 nm. Due to the ultra-small size-enabled unique properties, SCNPs have wide range of applications, including sensor, catalytic system, low viscosity coating, nanoreactor and biomedical applications. Through the contributions by many scientists in the past decade, the synthetic methodologies to fabricate SCNPs have been reported using various chemistries and been getting mature. However, there are still several unsolved problems in the field of SCNPs including functions and application. Stimuli-responsive polymers, as a class of smart materials whose properties can be changed by responding to external stimuli, have been widely used in energy and biomedical applications. Since their chemical and physical properties can be changed easily and efficiently via environmental control, stimuli-responsive polymers provide a potential pathway to preparing functional SCNPs. In this thesis, we are focusing on developing functional SCNPs, especially systems with multi-functions, and expanding their applications. To achieve this target, various stimuli-responsive polymers were prepared as polymer precursors and their stimuli-responsive properties were introduced into the SCNP systems by rational design of their chemical structures. The core of this thesis is comprised of three projects which deal with three classes of SCNPs from stimuli-responsive polymers. These stimuli-responsive SCNPs perform multi-functions and undergo certain change either in structure or morphology and properties. In addition, according to their variety of functions, each class of multi-functional SCNPs has diverse potential applications. In the first study presented in the thesis (Chapter 1), we prepared a class of sub-10 nm photodegradable and size-tunable SCNPs based on photo-responsive main-chain coumarin-based polyesters Poly{[7-(hydroxypropoxy)-4-(hydroxymethyl)coumarin adipate]-co- (polypropylene glycol adipate)} (CAPPG) through copolymerization of coumarin diol, adipic acid and polypropylene glycol (PPG). By incorporating coumarin moieties into the chain backbone of a polyester, dual photo-responsive reaction, i.e. photo-dimerization (>320 nm) and photo-induced chain scission (254 nm), occur under two different wavelengths of UV irradiation, enabling the preparation of sub-10 nm SCNPs and their photo-degradation property. The photo-degradability of SCNPs is evidenced under 254 nm UV irradiation for 3 h, which molecular weight of SCNPs decreasing from 13220 g/mol to 1385 g/mol. Moreover, the size of SCNPs can be tunable from 5.3 nm to 3 nm (hydrodynamic diameter) by varying the dimerization degree of coumarin moieties, that is simply controlled by the UV irradiation time. These results demonstrate a facile method to control the size of SCNPs without the need for synthesizing different polymer precursors. Finally, due to the biocompatible and biodegradable nature of polyester as polymer precursor, the SCNPs with photo-degradability and size-tunability have the potential to be exploited for biomedical applications. In the second study realized in this thesis (Chapter 2), we prepared a new type of multi-functional SCNPs from a side-chain liquid crystalline polymer (SCLCP), namely poly{6-[4-(4-methoxyphenylazo) phenoxy]hexylmethacrylate-co-4-methyl-[7-(methacr-yloyl) oxy-ethyl-oxy]coumarin} (PAzoMACMA). The polymer’s side groups comprise photo-isomerizable azobenzene in majority and photo-dimerizable coumarin in minority, with the former as mesogens and the latter for intra-chain photo-crosslinking. Despite the sub-15 nm size, confinement and crosslinking, the liquid crystalline (LC) phases of bulk PAzoMACMA persist in SCLCPs. Such LC-SCNPs exhibit a number of interesting and peculiar properties. While their dispersion in THF is non-fluorescent, when dispersed in chloroform, the nanoparticles appear to agglomerate to certain degree and display significant fluorescence that is different for SCNPs rich in the trans or cis isomer of azobenzene. The azobenzene LC-SCNPs also undergo photo-induced deformation, similar to azobenzene micro- or colloidal particles. However, the elongational deformation of the nanoparticles is dependent upon the linearly polarized excitation wavelength. While under polarized 365 nm UV irradiation the SCNP stretching direction is perpendicular to the light polarization, under polarized 400-500 nm visible light irradiation, the stretching takes place along the light polarization direction. Finally, an all-polymer nanocomposite was prepared by dispersing the LC-SCNPs in poly(methyl methacrylate) (PMMA), and mechanically stretching-induced orientation of azobenzene mesogens developed along the strain direction. The interesting properties of LC-SCNPs unveiled in this study suggest new possibilities for applications including bio-imaging and LC materials. As the third study in this thesis (Chapter 3), we studied a class of CO2-responsive SCNPs and their self-assembled micellar aggregates. On one hand, SCNPs are prepared from a random copolymer of poly{(N,N-dimethylaminoethyl methacrylate)-co-4-methyl-[7-(methacryloyl)oxyethyl-oxy]coumarin} (P(DMAEMA-co-CMA)). When dispersed in aqueous solution, individual nanoparticles can undergo reversible swelling/shrinking under alternating CO2/N2 stimulation as a result of the reversible protonation/deprotonation of tertiary amine groups. On the other hand, tadpole-like single-chain ‘Janus’ nanoparticles (SCJNPs) are prepared using an amphiphilic diblock copolymer of PS-b-P(DMAEMA-co-CMA) (PS is hydrophobic polystyrene). This type of SCJNPs can self-assemble into core-shell micellar aggregates in aqueous solution. Under CO2/N2 stimulation, the collective swelling/shrinking of SCJNPs within the micelle results in large, reversible volume change. In addition, both P(DMAEMA-co-CMA) SCNPs and PS-b-P(DMAEMA-co-CMA) SCJNP micelles are explored as gas-tunable nanoreactors for gold nanoparticles (AuNPs). The rate of AuNP formation increases under CO2 stimulation and decreases upon N2 bubbling, which makes it possible to tune the reaction rate up and down (on/off switching) by using the two gases. Moreover, using the micelles of SCJNPs, whose volume can be controlled over a wide range by adjusting the CO2 stimulation strength, variable-size AuNPs and their aggregates are obtained with continuous redshift of the surface plasmon resonance (SPR) into the long wavelength visible light region.

Indentation hardness, H, has been widely used to characterize the mechanical properties of materials for more than a century because of the following advantages of this technique; (1) it requires small sample and (2) the test is non destructive in nature. Recent technological advances helped in the development of instrumented indentation machines which can record the load, P, vs. displacement, h, data continuously during indentation with excellent load and displacement resolutions. From these, H and the elastic modulus, E, of the indented material can be obtained on the basis of the ‘contact area’ of the indentation at the maximum load. The estimation of true contact area becomes difficult during ‘pile-up’ and ‘sink-in’, commonly observed phenomena while indentation of a low and high strain hardened materials. In order for the better understanding of these phenomena it is important to understand the plastic flow distribution under indenters. It is also important for the prediction of elastic-plastic properties from the P-h data. Recently, there have been considerable theoretical and simulation efforts on this front with a combination of dimensional analysis and finite element simulations. One of the important input parameter for the dimensional analysis is the ‘representative strain’ under the indenter, which is a strong function of the indenter geometry. However there is no comprehensive understanding of the representative strain under the indenter despite several studies till date. One objective of the present thesis is to conduct an experimental analysis of the plastic flow during the sharp indentation. The plastic zone size and shape under conical indenters of different apex angles in a pure and annealed copper were examined by employing the subsurface indentation technique to generate the hardness map. From these isostrain contours are constructed joining the data having similar strain values. The following are the key observations. (1) The plastic strain contours are elliptical in nature, spreading more along the direction of the indenter axis than the lateral direction. (2) The magnitude of the plastic strain in the contact region decreases with increasing the indenter angle. (3) The strain decay in the indentation direction follow a power-law relation with the distance. The estimated representative strains under the indenters, computed as the volume average strain within the elastic-plastic boundary, decreases with increasing indenter angle. We also performed finite element simulations to generate plastic flow distribution under the indenter geometries and compared with the experimental results. The results suggest that the experimental and computed average strains match well. However, the plastic strain contours do not, suggesting that further detailed understanding of the elasto-plastic deformation underneath the sharp indenter is essential before reliable estimates of plastic properties from the P-h curves can be made routinely. The second objective of this thesis is to understand plastic flow in amorphous alloys. It is now well established that plastic deformation in metallic glasses is pressure sensitive, owing to the fundamentally different mechanisms vis-à-vis the dislocation mediated plastic flow in crystalline metals alloys. Early work has shown that the pressure sensitivity of amorphous alloys gets reflected as high constraint factor, C (hardness to yield stress ratio), which sometimes exceed 3.0. In this thesis, we study the temperature dependence of pressure sensitive plastic flow in bulk metallic glasses (BMGs) using C as the proxy for the pressure sensitivity. Experiments on three different BMGs show that C increases with temperature hence the pressure sensitivity. In addition we have carried out finite element simulations to generate P-h curves for different levels of pressure sensitivities and match them with the experimental curves that are obtained at different temperatures. Simulations predict that higher pressure sensitivity index values are required to match the experimental curves at high temperatures confirming that the pressure sensitivity increases with increasing temperature. The fundamental mechanisms responsible for the increase in pressure sensitivity are discussed in detail. Finally we pose a question, is the increase in pressure sensitivity with temperature is common to other amorphous materials such as strong amorphous polymers? In order to answer this question we have chosen PMMA, a strong amorphous polymer. In this study also we have taken C as a proxy to index the pressure sensitivity. Indentation stress-strain curves are constructed at different temperature using spherical indentation experiments. The C values corresponding to different temperatures are determined and plotted as a function of temperature. It is found that C increases with temperature implying that the pressure sensitivity of amorphous polymers also increases with temperature. The micro-mechanisms responsible for the increase in pressure sensitivity are sought.

Broadband Coherent Anti-Stokes Raman scattering (B-CARS) is an advanced Raman spectroscopy technique used to investigate the vibrational properties of materials. B-CARS combines the spectral sensitivity of spontaneous Raman scattering with the enhanced signal intensity of coherent Raman techniques. While B-CARS has been successfully applied in biomedicine for ultra-fast imaging of biological tissue, its potential in solid-state physics remains largely unexplored. This work delves into the challenges and adaptations necessary to apply B-CARS to crystalline materials and shows its potential as a powerful tool for high-speed, hyperspectral investigations. The theoretical part of this work covers inelastic light-matter scattering fundamentals and the signal generation process of B-CARS, with special attention given to the so-called Non-Resonant Background (NRB). This sample-unspecific signal amplifies the B-CARS intensity but also distorts the shape and position of the measured spectral peaks. A reliable NRB correction becomes crucial to retrieve precise spectral parameters containing information on the investigated material's crystallographic structure, defect density, and stress distribution. The first results chapter presents a practical guideline for an optimized workflow of sample preparation, measurement procedure, and data analysis. The influences of sample surfaces, focus positioning, and polarization sensitivity are discussed. The successful NRB removal is achieved by adapting an algorithm initially designed for biomedical purposes. The second chapter involves a transnational Round Robin investigating the same set of materials using different experimental setups. The influences of laser source, detection range, and transmission vs. epi detection are explored to optimize the experimental parameters. This work showcases applications such as high-speed, hyperspectral imaging of ferroelectric domain walls in LiNbO3, demonstrating the potential of B-CARS in the cutting-edge field of domain wall engineering. Additionally, imaging and polarization-sensitive measurements are shown for MoO3 flakes, paving the way for B-CARS investigations of 2D materials. The final chapter presents advanced techniques, such as Three-Color CARS and Time-Delay CARS, applied to crystalline materials. Three-Color CARS is especially promising, as it enhances the signal intensity for low-frequency Raman modes, which are particularly interesting for solid-state physics compared to the usual large-shift modes investigated in biomedical research. Meanwhile, Time-Delay CARS is sensitive to relaxation processes of vibrational and NRB states, enabling experimental NRB removal and lifetime measurements. Additionally, a neural network-based NRB removal method is presented, eliminating the need for a prior NRB spectrum and offering rapid computation. In summary, this work demonstrates the successful implementation of B-CARS for crystalline materials and provides a comprehensive guideline for the optimal experimental setup, workflow, and data processing. The application of B-CARS for imaging bulk crystalline materials, ferroelectric domain walls, and 2D structures shows promising possibilities for future research.

Les propriétés physiques d’un matériau sont intimement liées à sa structure cristalline et dans le cas d’ions dopants aux sites qu’ils occupent. La première partie de ce travail est dédiée au matériau diphosphate de sodium et de zinc Na2ZnP2O7, cristallisé out vitreux et ceci pour les ions dopants Co2+, Ni2+, Mn2+ et Eu3+. Les phases cristallisées ont été obtenues par la méthode Czochralski, les verres par trempe à partir de l’état fondu. Un ensemble de caractéristiques physiques ont été mises en jeu (Raman, infrarouge, RPE, absorption optique, luminescence) pour déterminer les sites occupés par les ions dopants et l’influence sur les propriétés optiques. La deuxième partie de cet travail consiste à une meilleure connaissance des matériaux diélectriques sans plomb appartenant à la famille pérovskite et plus particulièrement à la solution solide NaNbO3-BaTiO3. Des monocistaux ont été obtenus par la méthode des flux et caractérisés en utilisant plusieurs techniques : diffraction X, microanalyse, évolution thermique des domaines ferroélectriques-ferroélastiques, mesures diélectriques, piézoélectriques et pyroélectriques. / Abstract The physical properties of a material are intimately related to its crystalline structure and in the case of doped ions they are dependent on to the sites they occupy. The first part of this work is dedicated to the diphosphate material of sodium and zinc Na2ZnP2O7, glassy and crystallized for doped ions of Co2+, Ni2+, Mn2+ and Eu3+. The crystallized phases were obtained by the Czochralski method whereas the glasses were obtained by quenching from the molten state. A set of physical characteristics were studied (Raman, infrared, EPR, optical absorption, luminescence) to determine the sites occupied by the doped ions and its influence on the optical properties. The second part of this work consists of a better knowledge of lead-free dielectric materials belonging to the family of Perovskite, more particularly to NaNbO3-BaTiO3 solid solution. Single crystals were obtained by the flux method and were characterized using several techniques: X-rays diffraction, microanalysis, thermal evolution of ferroelectric and ferroelastic domains; dielectric, piezoelectric and pyroelectric measurements.