A number of industrial sectors suffer from the production of large amounts of low quality wasted thermal energy. Thermoelectric materials offer the potential to capture and repurpose this energy, by converting heat into electricity, which is generally much easier to transform and utilise. The current state of the art for thermoelectric materials are generally inorganic systems, such as metals and semimetals. Molecules offer an attractive alternative to these materials as they are made from more abundant elements, can be assembled into larger scale materials with relatively low-cost, and present an unrivalled level of structural control, which could offer new routes towards thermoelectric optimisation. It should be said however, that molecular thermoelectricity is much less explored, and although a number of interesting works have been performed many unanswered questions remain. The work presented within this thesis focuses on a number of new potential routes through which molecular thermoelectricity could be enhanced. This primarily focussed on; i) the incorporation of redox-active centres into a molecular wire, ii) the utilisation of quantum interference, iii) the construction of multi-component materials, and iv) the design of systems for phonon-suppression.
Initially the work focussed on the design and synthesis of ferrocene-containing molecular wires, constructed around the 1,1’-dialkynylferrocene motif. Here it was shown to be relatively straightforward to construct a range of molecules containing different anchor groups, which displayed interesting electronic tunability, as affected by the identity of terminal aryl groups contained within each system. It was found that pyridine- and thioanisole-terminated ferrocenes were generally poor candidates for studies by STM-BJ, due to rotation around the Fe-Cp axis, leading to the formation of a ‘hairpin’ geometry. However, this rotation lent itself to the formation of SAMs which displayed unexpected binding of only one of two anchors which were both pointed towards the surface.
Following this, a route was devised to access previously elusive dialkynyl-ferrocene systems, terminated with a thioacetate, through use of a cyanoethyl sulphur protecting group. This route was extended to the construction of a large family of molecules which contained a range of diverse functional groups, and once again, this series demonstrated an array of interesting optoelectronic properties. Preliminary studies of these molecules suggest that they form thicker, much more conductive SAMs than their pyridine-terminated counterparts. However, the mechanism behind this requires further elucidation to discern whether this relates to the formation of an ‘open’ 1,1’-dialkynylferrocene geometry.
A range of organic wires were also explored in this work. For this, a number of anthracenes containing different anchor group configurations were synthesised and subsequently deposited on a surface. Here it was found that the thermoelectric properties of these systems were strongly controlled by both their anchor groups, and their connectivity around a central aromatic core, hinting at the presence of a room-temperature CQI effect. Going further still, multi-component SAMs were constructed via a coordinative interaction between a pyridyl-terminated anthracenyl SAM and a zinc-centred porphyrin. These multi-component materials exhibited small changes in conductance, as well as boosted Seebeck coefficients, which were boosted further when measured with a graphene-coated metal probe.
Finally, the synthesis of molecules for ‘phonon-suppression’ was investigated. Two strategies for this were designed based around the construction of a typical OPE-3 type wire, containing distal substituents. The first of these focussed on the synthesis of a family of molecules with a point change in the mass of a substituent, and the second focussed on the introduction of a reactive moiety, which could allow multiple molecular wires to be linked to one-another on a surface. Both ideas were realised synthetically, however are still awaiting experimental confirmation by thermoelectric characterisation.
In short, the work reported herein presents an extension of the synthesis of a range of molecular wires, and also expands our knowledge of a number of routes which could be applied in the optimisation of molecular thermoelectricity.