Le développement de la technologie des cellules photovoltaïques organiques nécessite des compétences diverses liées à l’ingénierie moléculaire, à l’ingénierie des interfaces, au contrôle et à la caractérisation de la morphologie des films, à l’optimisation de la structure du dispositif et à la compréhension de la photo-physique des matériaux utilisés. Dans ce contexte, le travail présenté dans cette thèse contribue à la compréhension des propriétés photo-physiques des matériaux organiques π-conjugués et propose des outils de caractérisations optiques pour le suivi de la morphologie de ces matériaux. Dans un premier temps, une méthodologie rigoureuse de détermination des indices optiques des films organiques par ellipsométrie a été proposée. Les modèles utilisés en ellipsométrie ont ainsi été choisis en tenant compte des propriétés physiques des matériaux organiques π-conjugués ce qui a permis de remonter à la structure électronique de dérivés de fullèrenes (PC60BM et PC70BM). Dans un second temps, nous avons associé des données ellipsométriques à des mesures complémentaires d’absorbance et de photoluminescence dans le cas de deux copolymères (PTB7 et PTB7-Th) en films et en solutions afin d’isoler les interactions inter et intra-chaînes. Nous avons démontré que la photo-physique de ces copolymères diffère de celle du P3HT. Nous avons montré que même en solution dans le chlorobenzène, le PTB7 et le PTB7-Th s'agrègent fortement. Ces agrégats, de type H, se cassent plus facilement dans les solutions de chlorobenzène à base de PTB7-Th que dans celles à base de PTB7. / The development of organic photovoltaic cell technology requires various skills related to the molecular engineering, interface engineering, controlling and characterizing the morphology of the films, device structure optimization and understanding of photophysics of the materials. In this context, the work presented in this thesis contributes to the understanding of the photophysical properties of π-conjugated organic materials and propose optical characterizations tools for probing the morphology of these materials. First, a rigorous methodology for determining refractive indices of organic films by ellipsometry has been proposed. The models used in ellipsometry have been chosen by taking into account the physical properties of π-conjugated organic materials which allow the determination of the electronic structure of fullerene derivatives (PC60BM and PC70BM). Secondly, we associated ellipsometric data to complementary measurements of absorbance and photoluminescence in the case of two copolymers (PTB7 and PTB7-Th) in films and solutions in order to isolate inter and intra-chain interactions. We have demonstrated that the photophysics of these copolymers differs from the P3HT. We have shown that even in solution in chlorobenzene, the PTB7 PTB7-Th aggregate strongly. These aggregates, H-type, break more easily in the chlorobenzene solutions based of PTB7-Th as in those based on PTB7.

Ce travail de thèse a été consacré à l’étude de la stabilité dans différents milieux des matériaux constitutifs de la couche active des cellules solaires organiques. Les deux objectifs principaux étaient premièrement d’approfondir la compréhension sur certains mécanismes complexes intervenant dans la couche active, et deuxièmement d’étudier les processus de dégradation intervenant spécifiquement aux interfaces de la couche active au sein de la cellule. La première partie de ce mémoire est consacrée à l’étude de la dégradation photochimique et thermique de la couche active des cellules solaires en faisant varier le matériau polymère qui la constitue. La deuxième partie est dédiée au rôle d’un troisième composant de la couche active que peut être la présence d’additifs résiduels provenant de la mise en forme, ou d’un additif stabilisant ajouté à dessein. La troisième partie est consacrée au processus de délamination susceptible d’intervenir à l’interface couche active / couche transporteuse de trous. Enfin, une dernière partie concerne l’étude de l’alignement des niveaux énergétiques entre la molécule de C60 et divers substrats transporteurs d’électrons. L’influence de la lumière et de la température sur les propriétés d’interface couche active / couche transporteuse d’électrons est également reportée. / This word was devoted to the stability in various conditions of materials used in the active layer of organic solar cells. The main goals of this work were first to provide deeper understanding about complex mechanisms occurring in the active layer and second to investigate interfacial degradation pathways involving the active layer. A first part was dedicated to the photo and thermal stability of the polymer blend materials which constitute the active layer of the solar cells. A second section focused on the role of the third component of the active layer which can be undesired residual additives coming from the processing or the desired insertion of a stabilizer additive. A third part concerned the delamination issue which takes place at the active layer / hole transporting layer interface. Finally, a last section was devoted to the energy level alignment between the C60 molecules and various electron transporting substrates. The photo and thermal stability of the active layer / electron transporting layer interface was also studied in this section.

The performance of non-fullerene acceptor (NFA)- based organic solar cells (OSC) has shown continuous increase in recent years, reaching power-conversion efficiencies up to 17% through the design and synthesis of efficient acceptor materials. Recent research is directed towards achieving higher efficiency of OSC, which is limited by the open-circuit voltage (Voc) which is lower than the Voc values achieved in inorganic or perovskites solar cells with comparable bandgaps. In this work, voltage losses in NFA based OSC were calculated by investigating charge-transfer state energy (ECT) using electroluminescence spectroscopy and sensitive external quantum efficiency in three polymer:non-fullerene bulk heterojunction solar cells. PCE10:ITIC device acquired the highest ECT with a Voc of 0.82V, and a a power conversion efficiency (PCE) of 7.91%. While PCE10:O-IDTBR obtained the highest Voc of 1.03V, a PCE of 8.02% compared to PCE10:O-IDTBCN solar cell that has a lower Voc of 0.73V with a PCE of 7.98%. Both radiative and non-radiative voltage losses were calculated. In this thesis, the high open circuit voltage of PCE10:O-IDTBR is explained by the low non-radiative voltage losses compared to PCE10:O-IDTBCN and PCE10:ITIC devices.

For its inherent advantages, such as lightweight, low cost, flexibility, and opportunity to cover large surface areas, organic solar cells have attracted more and more attention in both academia and industry. However, the efficiency of organic solar cell is still much lower than silicon solar cells, but steadily rising as it now stands above 8%. The architecture of bulk heterojunction solar cells can improve the performance of organic solar cell a lot, but these improvements are highly dependent on the morphology of photoactive layer. Therefore, by controlling the morphology of photoactive layer, most commonly composed of a P3HT donor polymer and PCBM small molecule, the performance of organic solar cells could be optimized. The use of solvent additives in the solution formulation is particularly interesting, because it is a low cost method of controlling the phase separation of the photoactive layer and possibly removing the need for subsequent thermal and solvent vapor annealing. However, the role of the solvent additive remains not well understood and much debate remains on the mechanisms by which it impacts phase separation. In the first part of this thesis, we investigate the role of the solvent additive on the individual components (solvent, donor and acceptor) of the solution and the photoactive layer both in the bulk solution, during solution-processing and in the post-processing solid state of the film. In the second part of this thesis, we investigate the role of the additive on the blended solution state and resulting thin film phase separation. Finally, we propose a new method of controlling phase separation based on the insight into the role of the solvent additive. In the first part, we used an additive [octandiethiol (OT)] in the solvent to help the aggregation of P3HT in the solution. From the UV-vis experiments, the crystallinity of P3HT in the solutions increased while it decreased in thin films with steady increase of additive concentration. This method could be used for one step, annealing-free fabrication of organic solar cell with high performance. The solution can potentially be used to prepare ink for the large scale roll-to-roll ink-jet printing of P3HT thin films. Secondly, from the experiments it is found that differences in the evaporation rate and solubility of the components of the photoactive layer may be part of the reason for morphological changes. With lower evaporation rate than the host solvent, the additive concentration in the solution keeps increasing with time during the final stages of spin coating. In addition, the phase separation is increased with the increase of additive concentration, as demonstrated by AFM and TEM. By controlling the additive concentration, it is possible to control the phase separation of photoactive layer in pristine device. It is also found that the additive can change the wetting ability of the solvent to produce films with high surface coverage. With this information in hand, we modified the solution process of BHJ layers. A layer of crystals was deposited from the OT-containing solution by postponing the start of the spin coating for several minutes (delay time) after the solution is dropped on the surface of substrate. We found this to be a very effective method of increasing the phase separation and crystallinity of the photoactive materials. This effect was not possible when using oDCB solvent without any additive.

In organic photovoltaics (OPVs) the key processes relevant to device operation such as exciton dissociation and free carriers recombination occur at the donor-acceptor (D-A) interface. OPV devices require the bulk heterojunction (BHJ) architecture to function efficiently. In these BHJs, D-A interfaces are arranged in three dimensions, which makes molecular arrangements at these interfaces ill defined and hard to characterize. In addition, molecular materials used in OPVs are inherently disordered and may exhibit variable degrees of structural order in the same BHJ. Yet, D-A molecular arrangements and structure are crucial because they shape the energy landscape and photovoltaic (PV) performance in OPVs. Studies that use well-defined model systems to look in details at the interfacial molecular structure in OPVs and link it to interfacial energy landscape and device operation are critically lacking. We have used in situ photoelectron spectroscopy and ex situ x-ray scattering to study D-A interfaces in tailored bilayers and BHJs based on small molecule donors. We show preferential miscibility at the D-A interface depending on molecular conformation in zinc phthalocyanine (ZnPc)/ C60 bilayers and we derive implications for exciton dissociation. Using sexithiophene (6T), a crystalline donor, we show that the energy landscape at the D-A interface varies markedly depending on the molecular composition of the BHJ. Both the ionization energies of sexithiophene and C60 shift by over ~0.4 eV while the energy of the charge transfer state shifts by ~0.5 eV depending on composition. Such shifts create a downward energy landscape that helps interfacial excitons to overcome their binding energies. Finally, we demonstrate that when both disordered and ordered phases of D coexist at the interface, low-lying energy states form in ordered phases and significantly limit the Voc in devices. Overall our work underlines the importance of the aggregation and conformation states of molecular materials at and near the D-A interface in determining the operation and performance of OPV devices. This work shows that the role of D-A interfaces in complex BHJ devices can be unraveled through careful experimental design and by in depth characterization of planar heterojunction bilayer devices recreating model interfaces.

Organic photovoltaics (OPVs) have received tremendous attention in recent years due to their numerous attractive attributes such as, the potential for high power conversion efficiency (PCE), mechanical flexibility, and the potential for large-scale manufacturing via low-cost techniques. To date, the record PCE values for bulk-heterojunction (BHJ) OPVs exceed 18% for single-junction cells thanks to the rapid development of donors and acceptors materials for active layer. However, the progress of hole-transporting layer (HTL) systems, which is a key device component to reduce the additional performance losses of OPVs, has been limited with only a handful of materials available like PEDOT:PSS and MoOX. In this thesis, I introduce serval materials to unitize as hole-selective contact in high-performance OPVs. Firstly, the application of liquid-exfoliated two-dimensional transition metal disulfides (TMDs) is demonstrated as the HTLs in OPVs. The solution processing of few-layer WS2 suspensions was directly spun onto transparent indium-tin-oxide (ITO) electrodes yield solar cells with superior power conversion efficiency (PCE), improved fill-factor (FF), enhanced short-circuit current (JSC), and lower series resistance than devices based on PEDOT:PSS. Based on PM6:Y6:PC71BM BHJ layer, the cells with WS2 HTL exhibit the highest PCE of 17% thanks to the favorable photonic structure and reduced bimolecular recombination losses in WS2-based cells. Next, the self-assembled monolayer (SAM) namely 2PACz is utilized as hole-selective contact directly onto the ITO anode. The 2PACz modifies the work function of ITO while simultaneously affecting the BHJ layer’s morphology deposited atop. This ITO-2PACz anode is utilized in OPV with PM6:BTP-eC9:PC71BM, showing a remarkable PCE of 18.0%. The enhanced performance is attributed to reduced contact-resistance, lower bimolecular recombination losses, and improved charge transport within the BHJ layer. Lastly, the previously 2PACz SAM was functionalized with bromide functional groups, namely Br-2PACz, which is investigated as hole-extracting interlayers in OPVs. The highest occupied molecular orbital (HOMO) energy of Br-2PACz was measured at -6.01 eV, and significant changes the work function of ITO electrodes upon chemical functionalization. OPV cells based on PM6:BTP-eC9:PC71BM using ITO/Br-2PACz anodes exhibit a maximum PCE of 18.4%, outperforming devices with ITO/PEDOT:PSS (17.5%), resulting from lower interface resistance, improved hole transport, and longer carrier lifetimes.

The evolution of the internet of things (IoT) opens the market opportunity for organic photovoltaic cells, especially for indoor applications where the lifetime of the organic cells is longer than outdoor. For example, IoT requires off-grid energy sources for many devices with low power consumption. In this work, new materials were tested as candidate components in the active layer of printed organic photovoltaics by fabrication of devices. The initial electrical performance of these devices and their stability over time were investigated by measurements of the current-voltage characteristics. Three selected active layers were further investigated with atomic force microscopy (AFM) measurements. The current-voltage measurements showed that the addition of a solvent additive to the active layer ink affects the initial electrical performance as well as the stability of the devices. The AFM measurements showed that the surface topography of the active layer was affected by the sort of solvent additive that was used. Three new electron acceptor material and two solvent additives showed promising electrical performance and stability.

Organic solar cells are one of the most promising candidates for future solar power generation. They are thin and lightweight with several additional advantages such as scalability, environmental sustainability and low cost for processing and installation. However, the low charge carrier mobility of the absorbing material for organic solar cells requires thin absorber layers, limiting photon harvesting and the overall power conversion efficiency. Several attempts, e.g., periodically patterned structures and scattering layers have been tried to enhance the absorption of thin-film solar cells as light trapping elements. However, much effort is required to introduce light trapping structures to conventional rigid metal oxide electrodes and glass substrate. For instance, almost 13 hours are required to fabricate micro structures of 1 m2 area on glass, in contrast, 1 minute on PET using a same laser set-up and an additional scattering layers are demanded for providing light trapping effects to solar cells. In the last years, flexibility is emerging as the one of the major advantages of organic solar cells. To realize flexibility of solar cells, the classically used glass substrates and ITO electrodes are too brittle. Therefore, polymer materials are promising candidates to replace them as flexible electrodes and substrates. In this thesis, the highly transparent conducting polymer, PEDOT:PSS and PET equipped with an AlOx encapsulation layer are used as electrode and substrate, respectively. Besides the flexibility, additional light trapping elements, e.g. scattering particles, nano- and microstructures can be easily applied to the polymer materials since they have the potential for easier shaping and processing. In this study, we apply different light trapping and in-coupling approaches to organic solar cells. First, PET substrates are structured with a direct laser interference patterning system, which is a powerful and scalable one-step technique for patterning polymers. Almost 80 % of the light is diffracted by these patterned PET substrates and thereby the light path in the absorption layer is increased. Optical display films, commercially developed to be used as back light units of liquid crystal displays are also examined as light trapping substrates and exhibit similar enhancement as patterned PET. Moreover, since PEDOT:PSS is prepared by a solution-based process, TiO2 nanoparticles are added as light scattering elements to the PEDOT:PSS electrodes. Consequently, those electrodes provide a dual function as electrical contact and light trapping element. Finally, 2- or 3-dimensional nanostructures are printed by a nano-imprinting technique onto the surface of PEDOT:PSS with PDMS stamps. By controlling the temperature and the time of PEDOT:PSS during an annealing step, nanostructures are transferred from PDMS masks to PEDOT:PSS. To evaluate the effects of light trapping for all above mentioned approaches, flexible organic solar cells are produced by vacuum evaporation using blends of DCV5T-Me and C60 as absorber layer. The substrates are optically characterized using UV-vis spectrometer and goniometer measurements. The topography of the samples is measured by atomic force microscopy, scanning microscopy and optical microscopy. Bending tests with various radii are performed to test the flexibility of the substrates. In summary, light trapping effects are successfully implemented in the electrodes and substrates for OPVs, giving efficiency improvements of up to 16 %. The light trapping mechanisms in our approaches are extensively discussed in this thesis. / Organische Photovoltaik ist einer der vielversprechendsten Kandidaten für die zukünftige Solarstromgewinnung auf flexiblen Substraten. Um diese Flexibilität zu ermöglichen, sind herkömliche Glassubstrate mit ITO-Elektroden zu spröde. Ein vielversprechender Kandidat, um sowohl flexible Elektroden als auch flexible Substrate herzustellen, sind Polymere, da diese sehr biegsam und leicht zu verarbeiten sind. Deshalb wird in dieser Arbeit das hoch transparente, leitfähige Polymer PEDOT:PSS als Elektrode und PET (mit einer AlOx Verkapselungsschicht) als Substrat untersucht. Aufgrund der guten Prozessierbarkeit der Polymere konnten wir zusätzlich zu den eigentlichen Funktionen des Substrates und der Elektrode noch den Mechanismus des Lichteinfangs hinzufügen. Zusätzlich zu ihrer Flexibilität haben organische Solarzellen noch weitere Vorteile: sie sind dünn, leicht, skalierbar und verursachen vergleichsweise geringe Kosten für Herstellung und Installation. Ein Nachteil organischer Solarzellen ist die vergleichsweise geringe Ladungsträgerbeweglichkeit der Absorbermaterialien, welche oft die Schichtdicke der Absorbermaterialien begrenzt. Dies hat weniger absorbierte Photonen, weniger Stromdichte und somit einen geringeren Wirkungsgrad zur Folge. In den letzten Jahren wurden periodisch strukturierte Substrate und streuende Schichten als Lichteinfangelemente eingesetzt, um den Wirkungsgrad organischer Solarzellen mit dünnen Absorberschichten zu erhöhen. Gestaltungsregeln für solche Lichteinfangelemente sind noch weitestgehend unbekannt. Im Rahmen dieser Arbeit strukturieren wir PET Substrate mit einem direkten Laserinterferenzsystem, welches ein leistungsfähiges, skalierbares Einschrittverfahren zur Polymerstrukturierung ist. Da PEDOT:PSS aus der Lösung prozessiert wird, können wir weiterhin Nanopartikel hinzufügen, die der Elektrode zusätzlich noch lichtstreuende Eigenschaften geben. Außerdem können 2- bzw. 3-dimensionale Nanostrukturen leicht mithilfe einer Stempeltechnik eingeprägt werden. Um die Effekte des Lichteinfangs, welcher durch die oben genannten Methoden erzeugt wird, zu untersuchen, werden flexible organische Solarzellen mittels Vakuumverdampfung prozessiert. DCV5T-Me und C60 bilden dabei die photoaktive Schicht. Somit werden die Licht fangenden Eigenschaften dieser flexiblen Solarzellen ausgenutzt und ausführlich in der Arbeit diskutiert.

Philosophiae Doctor - PhD / The commercial need to improve the performance of low-cost organic solar cells has led to the idea for this research. The study discusses the synthesis of one dimensional TiO2 and ZnO nanowire arrays synthesised using a hydrothermal autoclave method and their application in bulk heterojunction inverted organic solar cells. Previous literature has shown that the precise manipulation, positioning and assembly of 1D nanostructures remain one of the greatest challenges in the field of nanotechnology, with much of the difficulty arising primarily from the lack of size and scale of the materials as well as the inability to visualise the nanostructures. In particular, one dimensional metal-oxides such as TiO2, ZnO and Fe2O3 have emerged as attractive alternatives to traditional semiconductor structures such as Si and GaAs as they are simple and inexpensive to manufacture, with research showing that application of ZnO nano-cones yield efficiencies of 8.4%, which is very attractive given the scope that exists in optimising the metal-oxide architecture. Much is still to be learned from the precise structural features of these materials and their influence on device performance. In this regard, this work largely focuses on this aspect of metal-oxide nanowires prior their application in organic solar cells.

Renewable energy is required for a sustainable future and one way to meet this is with organic solar cells (OSCs). The OSC can be easily manufactured at a low cost, be lightweight and be used on flexible surfaces. If the efficiency in high illumination intensities and stability in harsh environments increase for OSCs, they can com- pete with the other technologies even in outdoor conditions. Another advantage of OSCs is their good performance under low-light and indoor conditions. This is utilized by Epishine, a Swedish company based in Linköping working with small, thin and flexible organic printed solar cells optimized for indoor applications. The goal of this thesis is to determine how Epishine’s solar cells for low-light indoor usage work in more challenging conditions and to identify which are the factors that are detrimental for the lifetime of the cells. The result showed that all modules had a similar initial electrical performance which indicates that the modules have high reproducibility and degradation in darkness is negligible (since the initial measurements were made at different times). The tests showed that the temperature affected the modules. The test in the oven showed a little less than half the degradation compared to tests under the solar simulator, although both tests were subjected to the same temperature. The hu- midity test and the test exposed to LED-light showed almost no degradation. For the levels exposed to the sun or simulated sunlight, the decrease of the short circuit current density shows a burn-in time, which is typical for organic solar cells. After the first couple of hours, the decrease slows down to a more linear behaviour. All modules that were exposed to bright light also showed some recovery effect for short circuit current density and efficiency after they have been kept in the dark. It would be interesting to investigate the behaviour of the modules after even more exposure and look into how the recovery effect works.