Morphology control and device optimization for efficient organic solar cells

Abstract

Renewable energy is paramount for a sustainable global future. Solar cells convert solar light directly into electricity and are therefore of great interest in meeting the world’s energy demand. Currently crystalline silicon solar cells dominate the market. Solution processed organic solar cells can potentially be made on large scale using fast and easy roll-to-roll processing, which will make them an attractive alternative for crystalline silicon solar cells that are fabricated using a time- and energy consuming production process. The photoactive layer in an organic solar cell consists of a combination of an electron donating and an electron accepting material. Light is absorbed, creating an exciton that is split into charges at the interface of these materials. The charges are transported through the layer and collected at the electrodes. Efficient exciton generation, dissociation, and charge transport are key to obtain the highest efficiencies from these active layers. All these are affected by the morphology of the layer. The morphology of the active layer depends critically on different material parameters, such as solubility, miscibility, and tendency to crystallize and it can be influenced and optimized by changing the processing conditions. In this thesis control over the morphology of the photoactive layer is described for different material combinations. Higher efficiencies can be obtained when two complementary photoactive layers are combined in a tandem solar cell. The tandem construction alleviates losses that arise from thermalization of charge carriers and the transmission of light by absorbing high energy photons in a wide bandgap front cell and low energy photons in a small bandgap back cell. Device optimization of two different layer combinations into a tandem architecture is also described in this thesis. Hybrid organic solar cells combine a conjugated polymer with an inorganic semiconductor to potentially make use of the light-absorbing properties and the processability of the polymer, and high electron mobility and morphological stability of the inorganic semiconductor. These materials are very different in nature and therefore morphology control in a hybrid organic solar cell is a challenge. In Chapter 2 the synthesis of titanium dioxide nanocrystals capped with oleic acid ligands and their mixing with poly (3-hexylthiophene) (P3HT) into an active layer is described. The effect of the oleic acid capping on the solar cell performance was investigated by partial removal or replacement by less insulating ligands. Higher solar cell efficiencies were obtained when the amount of oleic acid on titanium dioxide was reduced, mainly as a result of an increase in photocurrent. This indicates that the oleic acid ligands act as a barrier for charge separation and charge transport in hybrid solar cells. Full removal of the ligands was not possible as they are needed to provide solubility to the titanium dioxide. Maximum solar cell efficiencies remained very low. Chapter 3 describes an investigation into bilayer organic solar cells of P3HT and a C60 fullerene derivative ([60]PCBM). Photoactive layers of P3HT and [60]PCBM were sequentially deposited from orthogonal solvents. The solar cells made from these layers were studied before and after thermal annealing and compared to mixed P3HT:[60]PCBM bulk heterojunction solar cells produced from a single solvent and bilayers of P3HT with vacuum evaporated C60 on top. Comparison of the spectral shape and magnitude of the experimental and theoretically modeled external quantum efficiencies showed that P3HT/[60]PCBM stacks made via orthogonal solution processing do not lead to bilayers with a sharp interface. A sharp interface was only obtained for P3HT/C60 bilayers. Thermal annealing of the diffuse-interface P3HT/[60]PCBM bilayers led to increased mixing, and with that increased efficiencies. Remarkably annealing of the solution processed bilayers did not result in the same mixed bulk heterojunction morphology that was obtained when P3HT and [60]PCBM were cast simultaneously from single solution and the efficiency in solar cells was higher. In small molecule organic solar cells different parameters can be used to control the morphology of the active layer. The effect of the side chain position was studied using diketopyrrolopyrrole (DPP)-based small molecules in Chapter 4. Efficiencies up to 3.3% were obtained when the DPPs were blended with a C70 derivative ([70]PCBM) in a bulk heterojunction solar cell. A variety of characterization techniques was used to study the behavior of these molecules in pristine films and mixed films with [70]PCBM, such as absorption, photoluminescence, electron and X-ray diffraction, and electron microscopy. The side chain position affects the molecular packing, varying from H-type to J-type, as well as the kinetics and ordering of DPP crystallinity in the mixed films. These differences affect the morphology of the active layer blends and in the end efficiency and behavior of solar cells made with these DPP-based molecules. Finally, also the use of these DPP small molecules in solution processed tandem solar cells was tested, to investigate the possibilities of processing layers of these small molecules from a low viscous, fast drying solution in a tandem device. The results showed that working tandem solar cells can successfully be prepared using these DPP-based small molecule active layers. In Chapter 5 a novel small bandgap DPP-based dendritic small molecule was investigated in active layer blends with [60]PCBM and [70]PCBM fullerene derivatives. The morphology that was obtained upon spin casting the blend strongly depends on the fullerene used. This morphology could be influenced by the addition of very small amounts of a processing additive to the spin casting mixture. While the short circuit current increased when using the [70]PCBM, the fill factor was significantly lower due to the differences in morphology evolution. In the end this resulted in similar efficiencies that could be obtained using either of the fullerene derivatives. Next an effort was made to produce a small molecule tandem solar cell with this small bandgap DPP-based dendritic molecule and a wide bandgap thiophene dendrimer. In the conventional architecture the integrity of the intermediate contact was not good enough to obtain a working tandem solar cell. An inverted architecture was needed to successfully process all the necessary layers on top of each other from orthogonal solvents. The efficiency of the inverted tandem solar cell was higher than that of the individual inverted single junction solar cells. In the last chapter polymer tandem solar cells with a power conversion efficiency of 7.0% were obtained using efficient complementary absorbing small and wide bandgap layers. The small bandgap layer consisted of a DPP-based polymer and [60]PCBM, while the wide bandgap material was the very efficient PCDTBT polymer with [70]PCBM. The morphology of the individual layers was optimized using solvent additives and thickness series were made. With the solar cell performance of the small and wide bandgap layers and optical modeling of the complete device stack the best tandem configuration was calculated. With 7.0% the tandem polymer solar cell performed 20% better than each of the best single junction solar cells that gave efficiencies of 5.3% and 5.8%.