“Polymer-fullerene bulk heterojunctions : morphology and its implications on the performance and stability of photovoltaic devices.”
Over the last decade, great improvements have been made within the field of organic BHJ solar cells. It has become clear that the active layer morphology plays a key role in achieving competitive devices. Within this work, the active layer morphology of several polymer:PCBM systems is correlated with the performance and stability of the corresponding solar cells. The morphology of polymer:fullerene blends has to be tuned to obtain state-of-the-art photovoltaic performance. But, an important general conclusion of this work is that the polymer:fullerene blended films are not stable. An increase of temperature can induce and accelerate PCBM diffusion and phase separation. In this work, also some solutions for this problem are investigated.
Chapter 2 discussed the techniques used in this thesis for studying the morphology of the active layers. With BFTEM, the bulk morphology of a sample can be displayed. SAED gives information about the ordering within the samples. Also UV-Vis can be applied to calculate the amount of P3XT fibers present in the active layer.
In chapter 3, it was shown that P3XT fibers, prepared in solution, can be used to prepare photovoltaic devices that do not need a post-production annealing treatment. The P3XT fibers were PAT derivatives with side chains ranging from 4 to 9 carbon atoms. A drawback of this method was the need to use different solvents for each material to obtain the fibers in solution. For some of these solvents, the solubility of PCBM was not optimal and this clearly had an effect on the active layer morphology. When PCBM had a high solubility in the solvent, good intermixing of PCBM and polymer was obtained. For low PCBM solubility, the PCBM assembled in large (> 500nm) chunks. A method, based on solution heating, was presented, which allowed to control the amount of P3XT fibers in the P3XT:PCBM blend. The photovoltaic properties and the morphology of the various P3XT materials applied in solar cells, have been investigated. Jsc largely depended on the morphology of the P3XT:PCBM blends. When good intermixing of P3XT and PCBM was obtained, Jsc reached an optimum. Voc was correlated with the P3XT fiber content. Upon increasing the solution temperature, and in this way decreasing the fiber content, Voc increased. It was demonstrated before that the aggregation of P3HT into fibers increases the polymer’s oxidation potential. Since Voc is related to the HOMO level of the donor material, the increase in Voc for the P3XT:PCBM solar cells was attributed to a decrease in fibrillar P3XT upon solution heating. The efficiencies for the different P3XT (X = 4 to 9) ranged from 0.6% to 3.1%; the best performance was obtained with P36T. This efficiency did not increase linearly with fiber content, but formed an optimum at average fiber contents, also observed by van Bavel et al. for state-of-the-art P3HT:PCBM solar cells.
Though efficiencies of organic BHJ solar cells are gradually increasing towards values necessary for large scale consumption, their stability is still poor compared to their inorganic counterparts. Chapter 4 discussed the thermal stability of MDMO-PPV:PCBM solar cells. It was seen that upon thermal annealing, a thermally accelerated demixing of MDMO-PPV and PCBM takes place. The reduced interfacial area, resulting from this process, leads to less efficient exiton dissociation (which happens at the polymer-PCBM interface). This is reflected in the photovoltaic performance where a strong decrease in Jsc is observed. An Arrhenius model is used to describe the degradation kinetics at the various ageing temperatures and to extrapolate the lifetime of the solar cells to other temperatures. An activation energy of 0.87 eV was found.
In chapter 5, the stability of several polymer:PCBM systems was compared. First the stability of the popular P3HT was studied. Though this material initially behaves positive during an annealing treatment (P3HT crystallizes for short annealing times leading to improved charge transport and better photovoltaic performance), its long time stability turns out to be quite poor. For long annealing at elevated temperatures, a demixing of P3HT and PCBM occurs. This can be correlated again to a decrease in Jsc. Small changes in the preparation method of the solar cells (i.e. using less PCBM or applying a slow drying treatment) can improve the initial performance of the solar cells, because of the instant crystallization of P3HT, but do not lead to enhanced thermal stability. Introducing functionalized side chains in the polymer structure did prove to enhance the thermal stability. An implementation of only 10% of these side chains turned out to be enough to hamper the free movement of PCBM molecules in the polymer matrix and in this way stabilize the performance of the device. Also using a polymer with a high glass transition temperature (above the operating temperatures of the device) induced a more stable morphology coinciding with a more stable performance.