Abstract
The work presented in this thesis encompasses an investigation into a number of porphyrin species. All of this work has a central focus on analysis of the porphyrin electronic structure and within the context of photovoltaics. The first exemplar of structures encompasses a series of Zn(II) tetraphenyl porphyrins which have asymmetrically substituted meso groups. The dihedral angle between the porphyrin macrocycle and the meso binding functionality has been shown not to infer any insulating properties towards delocalisation of the excited state. It is proposed that the electron withdrawing ability of the carboxylate unit is more than sufficient to overcome the poor π-overlap between the aromatic units. The inclusion of redox active functionalities at the distal meso site is shown to affect the energy and nature of the oxidised porphyrin species. It does so without almost any effect to the excited state properties of the structure. Both of these findings are important within the context of electroactive porphyrin chromophores.
The second major family of porphyrins studied consists of functionalisation at the various β positions. The use of electron withdrawing and donating groups (cyano and thio ether respectively) has the expected effect on the electronic structure of the porphyrin macrocycle. All visible electronic transitions are red shifted compared to the parent species, with a strong enhancement in the intensity of the Q bands observed. This is explained by a loss in degeneracy of the eg orbital set and a further splitting of the a1u/a2u MOs. Likewise the electronic properties have a substantial effect on the geometry of these porphyrin species. Further functionalisation of these species with a conjugated linker at another β position has a major effect on the electronic structure. The previously egx MO mixes with substituent based orbitals and splits into two new MOs with mixed porphyrin/substituent character.
The ground and excited state electronic properties of a set of Ru(II) porphyrins was also investigated. A similar electronic effect of conjugated β substituents to that presented above was also observed. Extensive DFT calculations have shown that π-withdrawing axial ligands act to lower the energy of the metal based dπ MOs below that of the a1u/a2u orbital set. This is also manifested in the excited state where calculated T1structures can be characterised as either π,π* (σ-donating axial ligands) or dπ,π* (π-withdrawing axial ligands). The β substituted species have the same general characteristics of their parent macrocycles but their excited states are substantially delocalised across the conjugated unit and consequently are of lower energy. Extended resonance Raman experiments on the β substituted structures showed that the FC state of the Soret band is substantially delocalised across the substituent, in contrast to a rather localised Q state.
Chapter 6 is involved with the investigation of a series of free-base and Ni(II) N-confused porphyrins (NCPs). DFT calculations have shown that the free-base NCPs show significant deviation in the energy and nature of the frontier molecular orbitals. The Ni(II) NCPs are not as well perturbed and presumably the Ni(II) ion acts as a scaffold to the conjugated macrocycle preventing major perturbation of the FMOs. An extensive resonance Raman study (the first of this porphyrin family) has established that structural differences to the parent porphyrin macrocycles are restricted to the external bonds of the porphyrin core. Despite the analogous loss in symmetry with chlorin macrocycles the NCPs displayed only minor variations in the energy of key marker bands, giving credence to the theory that the electronic structure of the ring is more important in defining vibrational energy compared to symmetry-species mode enhancement. These materials have been used in ternary blend bulk heterojunction solar cells and a model accounting for energy considerations of exciton formation was used to explain variations in device performance. The results of this have shown that there is a positive correlation between the calculated ionization potential of the porphyrin component and the Voc of the device. With a related negative correlation between electron affinity and Isc also determined.
The final chapter describes a theoretical construct which is concerned with establishment of a Marcus-Hush model to account for the electron injection rate from an absorbed dye into TiO2 within the DSSC device. This model has been developed in conjunction with some limited experimental data and shows that complete energetic coupling of the excited state of the dye, ES* and the TiO2 conduction band, ECB is achieved at an energy ES*>ECB+2λ. It has also been demonstrated that only a weak electronic coupling between the dye and the semiconductor is required as the large number of accepting states ensures that the electron transfer will always be activationless and the process is entropically driven. Both of these findings are in direct contrast with synthetic design parameters (all based on empirical evidence) found in the literature and go some way in establishing a more complete model of desirable dye characteristics within these devices.
A new method to determine the complete potential energy surface of large, complex molecules which exhibit structural changes with differing electronic configuration has also been developed. This is achieved by taking a projection of the total PES and condensing it to a single dimension. This allows for a fast, computationally efficient method to determining PES of large molecules, something that up this point was not easy to achieve. A case study of Re(CO)3Cl(2,2’-bipyridine) showed that the calculated potentials were largely harmonic in nature. This occurs since the structural distortions between geometric minima are multidimensional, overlaying a number of orthogonal anharmonic potentials will result in a harmonic surface. Surprisingly the degree of anharmoncity of a PES depends on the electronic configuration of that particular surface. This can be accounted for by considering how well the bond order of an electronic configuration is coupled to the structural distortion across a particular PES.