Chapter 1 of my PhD thesis "Organic Solar Cell Architectures" comprises the motivation (limited oil resources, environmental impact of fossil fuel combustion, status of present inorganic solar cell technology) and outline of the thesis.

Chapter 2 is aimed to introduce researchers who are planning to work on organic solar cells into this very specialised but also interdisciplinary field. The more general properties of organic semiconductors can be found in many excellent text books and other references and are only summarized briefly in the first part whereas the second part gives a more comprehensive insight into the important characteristic solar cell parameters and links between them. In fact, this Chapter contains a unique compilation and summary of "organic solar cell relevant knowledge" that is consistent with the experience, understanding and view of the author.

However, since a full theoretical understanding of organic solar cells is still not possible we have tried to improve understanding of device physics by drawing analogies to inorganic cells using the equivalent circuit diagram and the ``traditional" interpretation of current voltage characteristics.

The subsequent Chapters deal with the four known device architectures: the single layer (Chapter 3), double layer (Chapter 4), blend (Chapter 5) and the laminated device (Chapter 6).

Each of them begins with a survey of characteristic parameters of already reported devices - including the results of this thesis - pointing out specific advantages and encountered problems.

Chapter 7 concerns single layer devices comprising a liquid crystalline semiconductor. The outstanding properties of discotic liquid crystals justify the discussion of this device in a Chapter separated from the single layer device Chapter. It also comprises a survey of the interesting transport (charge carrier mobility) properties of liquid crystalline semiconductors as well as the mesogenic characterization of a series of discotic molecules from which one has been used to fabricate a single layer solar cell.

How solar cell efficiencies can be determined in a reasonable yet practical way either by setting up a solar simulator or numerical simulation is discussed in Chapter 8.

Chapter 9 concludes with a summary of the characteristic parameters comprising all four solar cell architecture, an overall assessment, some suggestions for future investigations and a comprehensive bibliography.

Details of sample preparation and measurements as well as a list of publications by the author and a brief CV can be found in the Appendices in Chapter 10.

The object of this thesis was the investigation of various types of organic semiconductors (preferably with low bandgaps) in different solar cell architectures. However, the following findings may be of particular interest for both experts and newcomers in the field:

• We have introduced a new device architecture that combines advantages of double layer and blend devices and opens exciting new possibilities in device design such as selective doping. Two laminated devices are discussed in Chapter 6.

• We have shown for the first time that dye/dye interfaces can be used for photogeneration of charges in solar cells.

• In Chapter 7 we have shown that dyes with liquid crystalline properties can be used as active semiconducting components in solar cells. Our results together with the recent literature indicate that heating into the liquid crystalline phase is not necessary.

• All devices discussed in the Chapters on double layer, blends and laminated structures show spectral responses covering at least the wavelength range of visible light. Two devices even had a clear photo-response down to a wavelength of 1000nm.

• The single layer device comprising PTV shows a very strong monoton dependency of the EQE on the wavelength so that the device can be used as simple colour - or even - wavelength detector covering the entire visible range.

• We have studied effects of film thickness and have found that those devices that have the thinnest films (10-30nm, which is near the estimated exciton diffusion length) give the highest currents. A method to estimate the optimal thickness has been introduced. However, we have also found that the shunt resistor grows 30 to 100 times faster than the series resistor with increasing film thickness favouring thicker films for larger fill factors.

• In Chapter 8 we describe how the standard solar spectrum can be simulated with a relatively simple setup that can be built in most laboratories for a fraction of the cost of commercial simulators. We also discuss a method solely based on intensity dependent photocurrent/voltage measurements that can be used to estimate the AM1.5 efficiency of solar cells.

  Note: The term "we" is used throughout the text in the thesis to underline the fact that every single result of the author's work as represented in this thesis was only possible because of the provision of equippment, materials and scientific input from others. This is also reflected in the long list of people mentioned in the acknowledgment and consistent with the fact that modern research relies on collaboration and teamwork.

• Download thesis here