A Computational Study of Substituted 2,2'-bithiophene as Building Blocks for Organic Solar Cells

The ability of 2,2’-bithiophene to act as an electron donor in an organic solar cell can be influenced through the addition of substituents, the nature of which can result in differing impacts on the molecule’s excitation energy. The ground and excited state geometries of trans and cis 2,2-bithiophene were investigated using the CC2 and ADC(2) ab inito methods. Adiabatic excitation energies were calculated from these geometries and compared to resonance-enhanced multiphoton ionization spectroscopy data in the literature, good agreement to which determined the basis set to be used when performing subsequent calculations on the substituted molecule. OH, formyl, nitro, nitroso, OC(H)O, OC(CH3)O, phenyl and dimethylamino substituents were chosen with the aim of influencing the excitation energy through both differing electronic and steric effects. Substitutions were made singly and doubly in five combinations of position and number of substituents. Ground and excited state geometries were calculated, and it was found that the CC2 and ADC(2) methods were unable to yield results for the formyl, nitro and nitroso substituents due to the multi-reference character of their ground states. The effect of the substituents on the bond lengths and dihedral angles of the thiophene rings compared to unsubstituted 2,2’-bithiophene was analysed before adiabatic excitation energies were calculated and the diethylamino and phenyl substituents found to cause the greatest decrease in excitation energy, almost regardless of substitution position. A final series of calculations were performed on unsubstituted, OH and OC(H)O substituted oligomers of three and four repeat units and the reciprocal of the repeat units against adiabatic excitation energy plotted in order to extrapolate to the excitation energy of a theoretical substituted polythiophene of infinite length. The utility of performing this type of extrapolation with this few repeat units, due to constraints imposed by computational cost, was reviewed.