Carrier, spin, charge current, and spin current injection by one- and two-color optical schemes are investigated within 30-band k·p theory. Parameters of the band model are optimized to give full-Brillouin zone band structures for GaAs and Ge that give accurate Γ-point effective masses and gyromagnetic factors and give access to the L valley, and to the E₁ and E₁+Δ₁ critical points in the linear optical absorption. Calculations of one- and two-photon carrier and spin injection and two-color current injection are performed for excitation energies in the range of 0—4 eV in GaAs and 0—3.5 eV in Ge. Significant spin and spin current injection occurs with 30% spin polarization in GaAs and Ge at photon energy matching the E₁ critical point. Further, the anisotropy and disparity of the current injection between parallel and perpendicular linearly-polarized beam configurations are calculated. For light propagating along a <111> crystal axis, anisotropic contributions in coherent current control and two-photon spin injection give rise to normal current components and in-plane spin components. In Ge, contributions from the holes to spin, electrical current, and spin current injection are investigated. Optical orientation results in 83% spin-polarized holes at the band edge. The effects of carrier dynamics in Ge are treated within a rate-equation model. The detection of spin dynamics in a pump-probe setup is considered, and the Fermi-factor approach is justified for electrons but not for holes. Carrier and current injection are further investigated in single-layer and bilayer graphene within the tight-binding model. In single-layer graphene, the linear-circular dichroism in two-photon absorption yields an absorption coefficient that is twice as large for circularly polarized light compared to linearly polarized light. Coherent current injection is largest for co-circularly polarized beams and zero for cross-circularly polarized beams. For linearly polarized beams, the magnitude of the injected current is independent of beam polarizations. In contrast, the injected current in bilayer graphene shows disparity between parallel and perpendicular configurations of the beams. The resulting angular dependence of the current is a macroscopic, measurable consequence of interlayer coupling in the bilayer.

L'AGS fournit un faisceau de proton polarisé à RHIC. Le faisceau est accéléré dans l'AGS de Gγ = 4.5 à Gγ = 45.5 et la transmission de la polarisation est critique pour le programme de spin de RHIC. Au cours des dernières années, divers systèmes ont été mis en œuvre pour améliorer la transmission de la polarisation dans l'AGS. Ces améliorations consistent essentiellement en l'introduction de deux serpents Siberien partiels et du system de saut de nombre d'onde. Cependant, la transmission de la polarisation n'atteint pas encore 100 % durant le cycle d'accélération de l'AGS. L'efficacité actelle de la transmission de la polarisation est estimée à environ 85 % dans les conditions de fonctionnement typiques. Comprendre les sources de dépolarisation dans l'AGS est essentiel pour améliorer les performances en protons polarisés de la machine. La dynamique complexe de faisceau et de spin, notamment en présence des aimaint spécialisés appelé serpent Sibériens, justifient le fort intérÃa t pour des méthodes de simulation originales. Le code Zgoubi, capable de résoudre l'équation du mouvement et de l'évolution du spin directement à partir d'une carte de champs, est utilisé pour modéliser l'AGS. Un modèle de l'AGS utilisant le code Zgoubi a pout cette reaison été développé et interfacé avec le système actuel par une simple commande: l'AgsFromSnapRampCmd. L'interfa ̧age avec le système de contrôle de la machine permet la modélisation rapide en utilisant les paramètres de réels la machine. Ces développements ont permis de reproduire fidèlement l'optique de l'AGS le long du cycle d'accéleration. Des développements supplémentaires sur le code Zgoubi, ainsi que sur des outils de post-traitement et de pré-traitement, ont fourni au code la possibilité de suivre les faisceaux sur de nombreux tours, ce qui s'avère être fondamental pour une représentation realiste du cycle d'accélération complet de la machine. Des simulations de faisceaux sur de nombreux tours dans l'AGS, en utilisant des conditions réalistes de faisceau et de machine, ont fourni une unique vision des les mécanismes sous-jacents de l'évolution de l'émittance et de la polarisation du faisceau au cours du cycle d'accélération. Des programmes de post-traitement ont été développés pour permettre la représentation des quantités pertinentes des données simulées par Zgoubi.Les simulations se sont avérées particulièrement utiles pour mieux comprendre les pertes de polarisation à travers résonances horizontales intrinsèques de spin. Le modèle Zgoubi ainsi que les outils développés ont également été utilisées pour certaines applications directes. Par exemple, les simulations d'expériences de faisceau ont permis l'estimation précise des gains de polarisation attendus en fonction des changements apportés. En particulier, des simulations d'expériences impliquant le système de saut des nombres d'onde ont fournis des estimations précises de la polarisation gagné et permis le choix des conditions optimales de la machine. / The AGS provides a polarized proton beam to RHIC. The beam is accelerated in the AGS from Gγ = 4.5 to Gγ = 45.5 and the polarization transmission is critical to the RHIC spin program. In the recent years, various systems were implemented to improve the AGS polarization transmission. These upgrades include the double partial snakes configuration and the tune jumps system. However, 100 % polarization transmission through the AGS acceleration cycle is not yet reached. The current efficiency of the polarization transmission is estimated to be around 85 % in typical running conditions. Understanding the sources of depolarization in the AGS is critical to improve the AGS polarized proton performances. The complexity of beam and spin dynamics, which is in part due to the specialized Siberian snake magnets, drove a strong interest for original methods of simulations. For that, the Zgoubi code, capable of direct particle and spin tracking through field maps, was here used to model the AGS. A model of the AGS using the Zgoubi code was developed and interfaced with the current system through a simple command: the AgsFromSnapRampCmd. Interfacing with the machine control system allows for fast modelization using actual machine parameters. Those developments allowed the model to realistically reproduce the optics of the AGS along the acceleration ramp. Additional developments on the Zgoubi code, as well as on post-processing and pre-processing tools, granted long term multiturn beam tracking capabilities: the tracking of realistic beams along the complete AGS acceleration cycle. Beam multiturn tracking simulations in the AGS, using realistic beam and machine parameters, provided a unique insight into the mechanisms behind the evolution of the beam emittance and polarization during the acceleration cycle. Post-processing softwares were developed to allow the representation of the relevant quantities from the Zgoubi simulations data. The Zgoubi simulations proved particularly useful to better understand the polarization losses through horizontal intrinsic spin resonances The Zgoubi model as well as the tools developed were also used for some direct applications. For instance, some beam experiment simulations allowed an accurate estimation of the expected polarization gains from machine changes. In particular, the simulations that involved involved the tune jumps system provided an accurate estimation of polarization gains and the optimum settings that would improve the performance of the AGS.

In my thesis I will discuss some aspects of collective dynamics of excitons and exciton-polaritons in nanoscale heterostructures. In the first Chapter I will make a brief introduction to the modern semiconductor physics and willdescribe the general elements and notions which will be used further. Other four chapters would be devoted to four works in which I participated, notably, in Chapter 2 I will speak about the coherent interactions between phonons and exciton orexciton-polariton condensates, in Chapter 3 I will discuss the quantum dots lasing and its amplification by an acoustic pulse. Chapter 4 and 5 will be devoted respectively to the polariton multistability and to the condensates of indirect excitons.

Magnetic thin films and patterned nanostructures have been studied with respect to their magnetic properties using SQUID-magnetometry, magnetic force microscopy, electrical measurements, and micromagnetic calculations. Properties of vortex domain walls, trapped in Permalloy nanowires with artificial constrictions, were investigated experimentally and by numerical calculations. In particular, the geometrical extent and strength of the pinning potential were evaluated. In these wires, long-range vortex domain wall displacement induced by spin polarized alternating currents was obtained numerically at reduced threshold current densities as compared with the direct current case. Due to the asymmetry of the energy potential, the long-range displacement direction is determined by the vortex chirality. Strained FeCo/Pt superlattices with strong perpendicular anisotropy were investigated experimentally. The strain was controlled by varying the thickness of each alternating layer with monolayer precision and was found to have a dominating effect on the total anisotropy. Epitaxial films of the diluted magnetic semiconductor (Ga,Mn)As were studied with focus on how the ferromagnetic transition temperature could be controlled by post-growth annealing. The ferromagnetic transition temperature was enhanced by approximately 85% for a Mn-doping concentration of 6% under certain conditions. A method to manipulate micrometer sized magnetic particles on patterned arrays of elliptical Permalloy microstructures was studied. Controlled motion and separation of the magnetic particles were obtained using applied rotating magnetic fields. The domain structure of the elliptical elements was studied numerically.

L'AGS fournit un faisceau de proton polarisé à RHIC. Le faisceau est accéléré dans l'AGS de Gγ = 4.5 à Gγ = 45.5 et la transmission de la polarisation est critique pour le programme de spin de RHIC. Au cours des dernières années, divers systèmes ont été mis en œuvre pour améliorer la transmission de la polarisation dans l'AGS. Ces améliorations consistent essentiellement en l'introduction de deux serpents Siberien partiels et du system de saut de nombre d'onde. Cependant, la transmission de la polarisation n'atteint pas encore 100 % durant le cycle d'accélération de l'AGS. L'efficacité actelle de la transmission de la polarisation est estimée à environ 85 % dans les conditions de fonctionnement typiques. Comprendre les sources de dépolarisation dans l'AGS est essentiel pour améliorer les performances en protons polarisés de la machine. La dynamique complexe de faisceau et de spin, notamment en présence des aimaint spécialisés appelé serpent Sibériens, justifient le fort intérÃa t pour des méthodes de simulation originales. Le code Zgoubi, capable de résoudre l'équation du mouvement et de l'évolution du spin directement à partir d'une carte de champs, est utilisé pour modéliser l'AGS. Un modèle de l'AGS utilisant le code Zgoubi a pout cette reaison été développé et interfacé avec le système actuel par une simple commande: l'AgsFromSnapRampCmd. L'interfa ̧age avec le système de contrôle de la machine permet la modélisation rapide en utilisant les paramètres de réels la machine. Ces développements ont permis de reproduire fidèlement l'optique de l'AGS le long du cycle d'accéleration. Des développements supplémentaires sur le code Zgoubi, ainsi que sur des outils de post-traitement et de pré-traitement, ont fourni au code la possibilité de suivre les faisceaux sur de nombreux tours, ce qui s'avère être fondamental pour une représentation realiste du cycle d'accélération complet de la machine. Des simulations de faisceaux sur de nombreux tours dans l'AGS, en utilisant des conditions réalistes de faisceau et de machine, ont fourni une unique vision des les mécanismes sous-jacents de l'évolution de l'émittance et de la polarisation du faisceau au cours du cycle d'accélération. Des programmes de post-traitement ont été développés pour permettre la représentation des quantités pertinentes des données simulées par Zgoubi.Les simulations se sont avérées particulièrement utiles pour mieux comprendre les pertes de polarisation à travers résonances horizontales intrinsèques de spin. Le modèle Zgoubi ainsi que les outils développés ont également été utilisées pour certaines applications directes. Par exemple, les simulations d'expériences de faisceau ont permis l'estimation précise des gains de polarisation attendus en fonction des changements apportés. En particulier, des simulations d'expériences impliquant le système de saut des nombres d'onde ont fournis des estimations précises de la polarisation gagné et permis le choix des conditions optimales de la machine. / The AGS provides a polarized proton beam to RHIC. The beam is accelerated in the AGS from Gγ = 4.5 to Gγ = 45.5 and the polarization transmission is critical to the RHIC spin program. In the recent years, various systems were implemented to improve the AGS polarization transmission. These upgrades include the double partial snakes configuration and the tune jumps system. However, 100 % polarization transmission through the AGS acceleration cycle is not yet reached. The current efficiency of the polarization transmission is estimated to be around 85 % in typical running conditions. Understanding the sources of depolarization in the AGS is critical to improve the AGS polarized proton performances. The complexity of beam and spin dynamics, which is in part due to the specialized Siberian snake magnets, drove a strong interest for original methods of simulations. For that, the Zgoubi code, capable of direct particle and spin tracking through field maps, was here used to model the AGS. A model of the AGS using the Zgoubi code was developed and interfaced with the current system through a simple command: the AgsFromSnapRampCmd. Interfacing with the machine control system allows for fast modelization using actual machine parameters. Those developments allowed the model to realistically reproduce the optics of the AGS along the acceleration ramp. Additional developments on the Zgoubi code, as well as on post-processing and pre-processing tools, granted long term multiturn beam tracking capabilities: the tracking of realistic beams along the complete AGS acceleration cycle. Beam multiturn tracking simulations in the AGS, using realistic beam and machine parameters, provided a unique insight into the mechanisms behind the evolution of the beam emittance and polarization during the acceleration cycle. Post-processing softwares were developed to allow the representation of the relevant quantities from the Zgoubi simulations data. The Zgoubi simulations proved particularly useful to better understand the polarization losses through horizontal intrinsic spin resonances The Zgoubi model as well as the tools developed were also used for some direct applications. For instance, some beam experiment simulations allowed an accurate estimation of the expected polarization gains from machine changes. In particular, the simulations that involved involved the tune jumps system provided an accurate estimation of polarization gains and the optimum settings that would improve the performance of the AGS.

Dans cette thèse, on s’intéresse à l'étude théorique et à la simulation numérique de la dynamique de charges et de spins dans des nano-structures métalliques. Ces dernières années la physique des nano-structures métalliques à connu un intérêt croissant, aussi bien d'un point de vue de la physique fondamental que d'un point de vue des applications technologiques. Il est donc essentiel d'avoir des modèles théoriques nous permettant de décrire correctement ce type d'objets. Cette thèse comporte deux études distinctes. Dans un premier temps on utilise un modèle semi-classique dans l'espace des phases afin d'étudier la dynamique de charges et de spins dans des films ferromagnétiques(Nickel). On décrit dans le même modèle le magnétisme itinérant et le magnétisme localisé. On montre qu'il est possible, en excitant le système avec un laser pulsé femtoseconde dans le domaine du visible, de créer un courant de spin oscillant dans la direction normal du film sur des temps ultrarapides(femtoseconde). Dans un second temps on s’intéresse à la dynamique de charge d'électrons confinés dans des nano-particules d'Or ou bien encore par des potentiels anisotropes. On montre que de telles systèmes sont des candidats intéressant pour faire de la génération d'harmoniques. / In this thesis we focus on the theoritical description and on the numerical simulation of the charge and spin dynamics in metallic nano-structures. The physics of metallic nano-structures has stimulated a huge amount of scientific interest in the last two decades, both for fundamental research and for potential technological applications. The thesis is divided in two parts. In the first part we use a semiclassical phase-space model to study the ultrafast charge and spin dynamics in thin ferromagnetic films (Nickel). Both itinerant and localized magnetism are taken into account. It is shown that an oscillating spin current can be generated in the film via the application of a femtosecond laser pulse in the visible range. In the second part we focus on the charge dynamics of electrons confined in metallic nano-particles (Gold) or anisotropic wells. We show that such systems can be used for high harmonic generation.

For every phenomenon found in inorganic materials, organic counterparts have been found in the last 50 years. The discovery of metallic conductivity in the inorganic conjugated polymer (SN)x was a forerunner to the discovery of metallic conductivity in Polyacetylene [1]. It was soon followed by the development of Polypyrrole and Polythiophene, and by the rediscovery of Polyaniline as a conducting polymer [2]. In polymers like Polyacetylene and Polythiophene, doping is by a redox reaction where the incorporation of electron withdrawing groups creates charge carriers in the polymer backbone. In contrast to these polymers, the main doping mechanism in Polyaniline is protonation, that is the attachment of a proton (donated by an acid) to specific sites (imine and amine groups) in the polymer. The protonated groups are also the sites where water and oxygen interact with the charge carriers on the polymer chain. A wide variety of quasi-particle states (excitons, bipolarons, separated polarons and polaron lattice forms) exist in Polyaniline, in its different states of oxidation and protonation. All of them have different transport and optical signatures. Out of these, only the polaron lattice gives rise to a half-filled conduction band, and therefore a metallic state [3]. This fascinating interplay of protonation and metallic features in Polyaniline, combined with its easy processibility, has made Polyaniline an attractive conducting polymer. Therefore the main focus of this thesis is on the role of the dopant on the electronic and optical properties of doped Polyaniline. The first chapter describes the main features of Polyaniline and its doping by protonation. The second chapter describes the experimental and simulation methods used in this thesis. Steady improvements in processing have led to reduced disorder in the samples, and have given rise to stronger metallic features like metallic (Drude-like) reflectivity in the infrared frequencies, and a positive temperature coefficient of the logarithmic derivative of the conductivity. High molecular weight Polyaniline doped with sulfonic acid dopants by surfactant-counterion processing, like Polyaniline doped with AMPSA (2-acrylamido-2-methyl-1-propanesulfonic acid) [4] and cast from dichloroacetic acid (DCA), shows all the metallic features indicative of an intrinsic metallic state [5]. In this thesis, the third chapter describes the spin-charge dynamics of Polyaniline doped with AMPSA (PANIAMPSA) through X-band Electron Spin Resonance studies [6]. Electron Spin Resonance (ESR) is an important technique to probe the spin-charge dynamics of conducting polymers [7, 8]. The X-band ESR spectra of PANI doped with AMPSA showed the presence of two lines (one broad and one narrow) at all temperatures and doping levels, indicative of two types of spin carriers. Three interesting features were observed in our study: a large linewidth ( ~100 Gauss), a maximum of ESR linewidth at ~ 25 K, and a surprising independence of linewidth on water/O2 . The temperature dependence of both linewidths suggests that the broad line is due to the delocalised charge carriers in well-ordered regions, and that the narrow line is due to localised spins in the disordered regions in the sample. Although the XRD spectra showed minimal crystallinity, the ESR and SQUID susceptibility had a strong Pauli contribution, indicative of an intrinsic metallic state. A similarity of the temperature dependence of linewidths of PANI-AMPSA with MWNT-s and HOPG graphite suggested that some quasi-2-D (Q2D) ordering is present in PANI-AMPSA. From Semi-empirical molecular modelling studies, a plausible hydrogen bonding pattern is suggested that can give rise to the Q2D graphene-like arrangement of the PANI polymer chains. This ordering is due to hydrogen bonding between the acrylamido group of the dopant and the amine fragment of the Polyaniline backbone. Hydrogen bonds are not just structural linkers between adjacent chains; they can have subtle effects on electronic states of the polymer backbone due to charge transfer/withdrawal by the hydrogen bond from the delocalised β-electron system of the backbone. The same Q2D model is used to explain the water/oxygen independence of linewidth in PANI-AMPSA. The temperature dependence of linewidth of both lines has been explained in terms of the QTDG (Quasi Two Dimensional Graphite) model, where a strong exchange interaction is presumed to arise between the 2D delocalised charge carriers and the localised spins, leading to a low-temperature peak in the the linewidth. Water is known to significantly enhance the conductive properties of doped Polyaniline [9]. A detailed DFT (Density Functional Theory) modelling study of the influence of water in doped Polyanilines is presented, which clearly indicated that water enhances the charge transfer between the counterion and the polymer backbone. The torsion angles between the adjacent phenyl rings of the emeraldine base decrease when the imine nitrogens are protonated by inorganic acids like HCl and HBr, and hydration of the acid counterions further decrease the torsion angles. In contrast, the torsion angles of the AMPSA protonated Polyaniline are already low (comparable to the hydrated cases), and the charge transferred by AMPSA is also enhanced. Visualisation of the molecular structure of the PANI-AMPSA complex suggested that water molecules may play a minimal role in the electronic properties of AMPSA doped Polyaniline. We suggest the Q2D ordering as the reason for the temperature dependence of the linewidth, the lack of oxygen and water dependence of the linewidth, as well as the enhanced metallic properties in PANI-AMPSA, as compared to other doped Polyanilines. The electronic states of Polyaniline are modified by both redox processes and protonation. This gives rise to a wide variety of optical states, which can be easily accessed by both applied potential and pH [10]. Therefore Polyaniline displays strong electrochromism across the visible, near-IR (NIR), IR and even microwave spectral regions. This feature has wide applications in electrochromic devices. However, a fundamental understanding of the phenomena behind this electrochromism, the charge carrier(s) responsible, and the relation of nanoscopic morphology and electrochemical properties to the electrochromism, is still not clear. In the fourth chapter, we have analysed extensive data from electrochromic devices [11]. Clear assignments are that certain population states contribute predominantly to certain spectral regions (e.g. bipolaron states to the IR, the valence band to the visible and other mid-gap states to the microwave). Among more specific findings, a prominent 7µm (0.16 eV) peak in MIR devices is ascribed to bipolarons, while a low-energy transition at 0.054 eV is ascribed to inter/intra-chain transitions. Each of these transitions is tracked with respect to changes in applied potential, as well as correlated with device morphology and construction. Our analysis of UV-Vis-MIR-FIR-microwave results along with detailed SEM data clearly relates performance in different wavelength regions to morphology. Preliminary kinetics analysis show that the diffusion rates in these devices could be improved further. These findings point to the potential design of very broad-band electrochromic systems encompassing the visible through microwave regions. Polyaniline in its insulating states can be considered as a series of linked oligoanilines. These oligoaniline states can either be considered as a model for describing the properties of the polymer, or can be interesting systems themselves in the light of single-molecular electronic devices [12]. Both applied potential and pH can change the electronic states of these systems. The ability of pH to modify the oxidation states in these systems (and induce electronic transport), and the influence of water on these properties can be a model for biological systems too. While a wealth of information on oligoanilines has been generated from experiments, computational modelling of these systems is less reported. Among many computational methods that have been developed for calculation of optical absorption spectra of molecules, Time Dependent Density Functional Theory (TDDFT) is the method with the widest use. TDDFT obtains the excitation energies of a molecule from the linear response of the electronic density to a external perturbing field [13]. Solvent effects, which are known to affect the excitation energies, are included through the SCRF/PCM (Self-consistent Reaction Field/ Polarizable Continuum Model). PCM is a method that treats the solvent molecules as a continuum, and self-consistently evaluates their electronic distribution around the solute. In the fifth chapter, a systematic study of the optical properties of neutral oligoaniline, in three oxidation states, is performed by varying the chain length and linearity of the backbone. The intrinsic accuracy in the excitation energies obtainable by the combined TDDFT/PCM formalism has enabled us to suggest effective oligomer lengths for the optical transitions in Polyaniline; these are 4 rings for emeraldine base, 4–8 rings for leucoemeraldine base and 4 rings for pernigraniline. The sensitivity of the 2.0 eV exciton peak in emeraldine base to the chemical environment is also apparent from this work. The Valence Density of States (VDOS) and vibrational frequencies, that have been obtained in course of these simulations, have been quantitatively analysed and are a useful addition to understanding the optical properties of neutral Polyanilines. A summary of the results of the dopant and water dependence on the electronic and structural properties of protonated oligoanilines was presented in the third chapter; the appendix describes the methodology in detail. It is worthwhile to emphasize that doped Polyaniline is a system where protonation, hydration and extended β-conjugation all occur together synergistically, and a good overall description of this system is necessary. Modelling the doped state of Polyaniline is a bit more difficult, due to spin polarisation. Ideally, conducting Polyaniline should be modelled in the solid state, with neighbouring chains, counterions and water molecules. Water is known to reversibly increase the macroscopic conductivity and ESR linewidth of doped Polyaniline. In the sixth chapter of this thesis, optical spectra of the bipolaron, separated polaron, and the polaron lattice forms of doped Polyaniline, explicitly including the counterions (Cl, Br, AMPSA) are obtained by the TDDFT method. All the polaronic lattice forms show a dominant absorption at 1.0–1.2 eV, with no absorptions in the range 1.4–2.0 eV. The inclusion of water molecules to solvate the counterions is shown to only weakly modify the optical properties in the polaron lattice form. In the case of polarons on a twisted chain, the 1.0 eV peak is shifted to 1.5 eV. For bipolarons, there is an absorption at 1.3–1.5 eV, along with another peak at 1.8 eV. Comparing with experimental spectra we suggest that the 1.5, 2.8 eV set belongs to a polaron lattice form wherein the chains are twisted. However, individually the 1.5 eV peak may equally come from bipolarons or separated polarons. The peak at 1.8 eV may either be ascribed to a bipolaron form (in which case there should be a 1.5 eV peak too), or to an isolated polaron. The isolated polaron may also show a peak at 2 eV and 3.5 eV that is clearly from a residual emeraldine base electronic state. The steady evolution of the (a) 2 eV exciton peak in emeraldine base to a (b) 1.6– 1.8 eV peak (isolated polarons) to a (c) 1.5 eV peak in the bipolaron form to (d) 1.3 eV peak in the separated polaron form to (e) a 1.0–1.2 eV peak in the fully doped metallic polaron lattice form is clear. This steady evolution observed from TDDFT simulations may help in clarifying the experimental assignments, especially in electrochemical studies on Polyaniline. Simulations including the water molecules were performed to study the experimentally observed dramatic changes on hydration in Polyaniline. However hydration of ions is a dynamic process and static geometries may not provide a fully realistic description. Combined ab initio Molecular Dynamics (AIMD) and TDDFT calculations may be necessary to realistically model the transport properties of doped Polyaniline. This chapter tries to lay a foundation for such work. The main results obtained in this thesis are summarized in the conclusion. To conclude, this thesis is on the electronic and optical properties of Polyaniline. An ESR study on AMPSA doped Polyaniline indicated a unique 2D nanoscopic morphology, and this structure was validated by molecular modelling. The detailed analyses on electrochromic devices led us to perform TDDFT simulations of neutral and doped Polyanilines. These simulations have resulted in clear UV-VIS-IR assignments in all forms of Polyaniline.

A novel experimental setup has been realized to measure weak magnetic moments which can be modulated at radio frequencies (~1–5 MHz). Using an optimized radio-frequency (RF) pickup coil and lock-in amplifier, an experimental sensitivity of 10 -15 Am2 corresponding to 10 -18 emu has been demonstrated with a one second time constant. The detection limit at room temperature is 9.3 10 -16 Am2/√Hz limited by Johnson noise of the coil. In order to demonstrate the sensitivity of this technique it was used to electrically detect the polarized spins in semiconductors in zero applied magnetic fields. For example in GaAs, the magnetic moment due to a small number (~ 7 x 108) of spin polarized electrons generated by polarization modulated optical radiation was detected. Spin polarization was generated by optical injection using circularly polarized light which is modulated rapidly using an electro-optic cell. The modulated spin polarization generates a weak time-varying magnetic field which is detected by the sensitive radio-frequency coil. Using a radio-frequency lock-in amplifier, clear signals were obtained for bulk GaAs and Ge samples from which an optical spin orientation efficiency of ~ 10–20% could be determined for Ge at 1342 nm excitation wavelength at 127 K. In the presence of a small external magnetic field, the signal decayed according to the Hanle Effect, from which a spin lifetime of 4.6 ± 1.0 ns for electrons in bulk Ge at 127 K was extracted. The spin dynamics in n-Ge was further explored and the temperature dependence of the spin lifetime was plotted for a temperature range of about 90 K to 180 K. The temperature dependence of the optical pumping efficiency was also measured though no quantitative conclusions could be derived. The signals observed for semi-insulating GaAs, n-GaAs, GaSb and CdTe which are direct gap semiconductors are much larger than expected (almost two orders of magnitude). An attempt was made to explain this unexpected behavior of these direct gap semiconductors using the spin hall effect.

Cette thèse est consacrée à de nouveaux phénomènes en physique liées au spin et à la topologie des quasi-particules lumière-matière dans des hétérostructures. Elle est divisée en quatre parties. Chapitre 1 donne un fond nécessaire et introduit les propriétés fondamentales des polaritons et des excitons indirects dans des puits quantiques couplés. Chapitre 2 est concentré sur la dynamique de spin et sur formation de défauts topologiques dans des systèmes aux excitons indirects. Les 2 derniers chapitres considèrent les structures basées sur les microcavités. Chapitre 3 est consacré à la dynamique de spin des polaritons dans des oscillateurs paramétriques optiques. Finalement, chapitre 4 étudie les réseaux des microcavités en forme des piliers et introduit l’isolant topologique polaritonique. / The present thesis manuscript is devoted to new phenomena in physics of light-matter quasiparticles in heterostructures, related to spin and topology. It is divided into four parts. Chapter 1 gives a necessary background, introducing basic properties of microcavity polaritons and indirect excitons in coupled quantum wells. Chapter 2 is focused on spin dynamics and topological defects formation in indirect exciton many-body systems. The last 2 chapters are related to microcavity-based structures. Chapter 3 is devoted to polariton spin dynamics in optical parametric oscillators. Finally, Chapter 4 studies pillar microcavity lattices and introduces the polariton topological insulator.

The spin transport and dynamics of optically injected spin polarized carri- ers are studied with a high spatial and/or time resolution in semiconductor GaAs-based heterostructures in multiple transport regimes. An unexpectedly long-scale and high-speed spin diffusion transport is observed in a long-lived electron sub-system induced optically at an undoped single GaAs/AlGaAs heterointerface. A diffusion and drift-dominated spin transport is investi- gated using an electrical spin-detection via the inverse spin Hall effect in doped GaAs-based systems at room and low temperatures. It is shown that the inverse spin Hall signal and the spin transport parameters can be con- trolled by a direct application of an electric field or by expanding a depleted zone of a planar pn-junction.