The dye-sensitized solar cell (DSC) is a promising efficient low-cost molecular photovoltaic device. One of the key components in DSCs is the dye, as it is responsible for the capture of sunlight. State-of-the-art DSC devices, based on ruthenium dyes, show record efficiencies of 10-12 %. During the last decade, metal-free organic dyes have been extensively explored as sensitizers for DSC application. The use of organic dyes is particularly attractive as it enables easy structural modifications, due to fairly short synthetic routes and reduced material cost. Novel dye should in addition to the light-harvesting properties also be compatible with the DSC components. In this thesis, a series of new organic dyes are investigated, both when integrated in the DSC device and as individual components. The evaluation methods consisted of different electrochemical and photoelectrochemical techniques. Whereas the light-harvesting properties of the dyes were fairly easily improved, the behavior of the dye integrated in the DSC showed less predictable photovoltaic results. The dye series studied in Papers II and IV revealed that their dye energetics limited vital electron-transfer processes, the dye regeneration (Paper II) and injection quantum yield (Paper IV). Further, in Papers III-VI, it was observed that different dye structures seemed to alter the interfacial electron recombination with the electrolyte. In addition to the dye structure sterics, some organic dyes appear to enhance the interfacial recombination, possibly due to specific dye-redox acceptor interaction (Paper V). The impact of dye sterical modifications versus the use of coadsorbent was explored in Paper VI. The dye layer properties in the presence and absence of various coadsorbents were further investigated in Paper VII. The core of this thesis is the identification of the processes and properties limiting the performance of the DSC device, aiming at an overall understanding of the compatibility between the DSC components and novel organic dyes. / QC 20100730

Les satellites sont vulnérables aux particules de forte énergie piégées dans les ceintures de Van Allen. Afin d’en assurer la protection, il est nécessaire de prédire avec précision la dynamique des électrons au sein de la magnétosphère. Dans un premier temps nous proposons une méthode originale de résolution analytique de l’équation de Fokker-Planck réduite qui modélise le transport et les pertes des électrons de la magnétosphère interne. La résolution repose sur une technique de décomposition spectrale. Si la solution analytique s’avère utile pour mettre en exergue certaines propriétés physiques des ceintures de radiation, elle est également pertinente pour valider le code numérique de résolution de l’équation de Fokker-Planck réduite, développé durant la thèse. Ce dernier nous amène à généraliser l’étude précédente en illustrant l’évolution des flux d’électrons pour diverses énergies et positions. Nous prouvons notamment que la structure des ceintures de radiation ainsi que leur temps d’évolution ne dépendent que de quelques facteurs bien choisis. Dans la perspective de reproduire un événement particulier de retour au calme après un orage magnétique, mesuré par les satellites de la NASA dédiés aux ceintures de radiation, nous sommes en mesure de simuler la précipitation des électrons dans l’atmosphère terrestre causée par les interactions avec les ondes électromagnétiques de la magnétosphère. L’utilisation de conditions bâties sur des données empiriques et spécifiques à la période en question nous permet de corroborer les flux observés. Enfin, l’influence du champ magnétique terrestre sur la dynamique des ceintures de radiation est étudiée sous divers aspects. Nous nous concentrons sur la ceinture externe pour comprendre comment les asymétries du champ magnétique, considérablement façonnées par l’activité solaire, affectent notre manière de concilier théorie et observations. Nous explorons également l’importance de certains processus diffusifs nouveaux et cachés, qui émergent à cause de l’irrégularité naturelle du champ magnétique au plus proche voisinage de la Terre. / Satellites are vulnerable to high-energy particles trapped in the Van Allen belts. To ensure their protection, it is necessary to predict properly the electron dynamics in the magnetosphere. We first propose an original method to find the analytical solution of the reduced Fokker-Planck equation that models the transport and loss of electrons in the inner magnetosphere. The resolution relies on an eigenfunction expansion approach. If the analytical solution is proven to be useful at uncovering some of the physical properties of the radiation belts, it is also relevant to validate the numerical code that solves the reduced Fokker-Planck equation, which has been developed during the PhD. The latter code is used to generalize the previous study in illustrating the evolution of the electron fluxes for various energies and locations. We demonstrate that the structure of the radiation belts as well as their dynamical timescales only depend on a few well-chosen parameters. In the perspective of reproducing a specific storm-recovery event reported by the NASA Van Allen Probes, we are able to simulate the electron scattering in the Earth’s atmosphere due to the interaction with magnetospheric electromagnetic waves. The consideration of data-driven and event-specific conditions enables us to corroborate the observed fluxes. Finally, various influences of the Earth’s magnetic field on the dynamics of the radiation belts are investigated. We focus on the outer belt to see how the magnetic field asymmetries, which are strongly shaped by solar activity, affect the way of conciliating theory and observations. We also explore the importance of new hidden diffusive processes that emerge due to the natural irregularity of the magnetic field in the closest vicinity of the Earth.

The solid-state mesoscopic solar cells (sMSCs) have been developed as a promising alternative technology to the conventional photovoltaics. However, the device performance suffers from the low hole-mobilities and the incomplete pore filling of the hole transport materials (HTMs) into the mesoporous electrodes. A variety of HTMs and different preparation methods have been studied to overcome these limitations. There are two types of sMSCs included in this doctoral thesis, namely solid-state dye-sensitized solar cells (sDSCs) and organometallic halide perovskite based solar cells. Two different types of HTMs, namely the small molecule organic HTM spiro-OMeTAD and the conjugated polymer HTM P3HT, were compared in sDSCs. The photo-induced absorption spectroscopy (PIA) spectra and spectroelectrochemical data suggested that the dye-dye hole conduction occurs in the absence of HTM and appears to be of significant importance to the contribution of hole transport. The PIA measurements and transient absorption spectroscopy (TAS) indicated that the oxidized dye was efficiently regenerated by a small molecule organic HTM TPAA due to its excellent pore filling. The conducting polymer P3HT was employed as a co-HTM to transfer the holes away from TPAA to prohibit the charge carrier recombination and to improve the hole transport. An alternative small molecule organic HTM, MeO-TPD, was found to outperform spiro-OMeTAD in sDSCs due to its more efficient pore filling and higher hole-mobility. Moreover, an initial light soaking treatment was observed to significantly improve the device performance due to a mechanism of Li+ ion migration towards the TiO2 surface. In order to overcome the infiltration difficulty of conducting polymer HTMs, a state-of-the-art method to perform in-situ photoelectrochemical polymerization (PEP) in an aqueous micellar solution of bis-EDOT monomer was developed as an environmental-friendly alternative pathway with scale-up potential for constructing efficient sDSCs with polymer HTMs. Three different types of HTMs, namely DEH, spiro-OMeTAD and P3HT, were used to investigate the influence of HTMs on the charge recombination in CH3NH3PbI3 perovskite based sMSCs. The photovoltage decay measurements indicate that the electron lifetime (τn) of these devices decreases by one order of magnitude in the sequence τspiro-OMeTAD > τP3HT > τDEH.