Preview

PDF, English Download (41MB) | Terms of use

Abstract

The inherent chemical and morphological richness and complexity of Organic Semiconductors make the detailed theoretical description almost impossible. This leads to the necessity of simulation-based approaches that should be as detailed and computationally efficient as possible, which turns out to be a far from trivial task. While over the decades, different models of different levels of detail and computational effort have been proposed, there is still ample room for improvement. In this thesis a semi-empirical Tight Binding model of organic semiconductors was developed that is based on user-specified morphologies like Molecular Dynamics morphologies and is combined with kinetic Monte Carlo simulations to obtain a tool that captures different correlations and includes partially delocalized charge carriers. In this thesis, kinetic Monte Carlo simulations are used to describe how anisotropic localization lengths can lead to increased thermoelectric powerfactors, breaking the usual trade-off between Seebeck coefficient and conductivity. The developed numerical model is also used to study the validity of the effective temperature model of the field dependence of the conductivity that assumes a one-to-one relation between thermal- and field-activated charge transport. It is found that the effective temperature model is not always self-consistent and breaks down for inhomogeneous systems. Additionally, other parameters influencing the field dependence are not correctly incorporated in the effective temperature model. The numerical model was also used to explain the experimentally observed superlinear increase of conductivity at high charge carrier concentrations. Delocalization of charges at the Fermi level is found to explain this observation. Lastly, the tight binding model was used to explain the renormalization of the Density of States at intermediate doping levels that is necessary to explain the experimentally observed roll-off in conductivity for intermediate doping in the conductivity dependence of the Seebeck coefficient which is the relevant regime for optimized thermoelectric devices.