| The thesis at hand was motivated by the somewhat surprising observation that in certain cases, properties of monomer building blocks can be inherited to larger copolymers that contain them, i.e., in these cases the property of a complex chemical system is fully (or at least predominantly) determined by that of a much simpler subsystem. We set out to investigate and rationalize this phenomenon in a systematic fashion and identify factors that affect the occurrence of inheritance. This study is part of an ongoing effort to advance our understanding of structure-property relationships in copolymer-based organic semiconductors, which is a prerequisite for rational design and inverse engineering of next-generation materials.;The basic approach of this work is to employ data mining and materials informatics to a suitable data set to extract the desired insights. As an initial case study, we focused on the inheritance of principal energy levels, i.e., those of the highest occupied and lowest unoccupied molecular orbitals (HOMO and LUMO, respectively), as they play a major role in the performance of organic semiconductors. We utilized the first-principles quantum chemical data set of the Harvard Clean Energy Project as a clean and well-defined foundation for this study. The work presented in this thesis includes the calculation of missing monomer data, the development and application of chemical pattern recognition and substructure searches to identify the monomer make-up of given polymers, the comparison of monomer and polymer property data, the application of a number of statistical means to analyze the resulting mapping, and an interpretation of the findings that emerge from our study based on orbital interaction theory. |
