Evaluating the structure-property relationships of pressure-responsive materials

Abstract

The field of material science is continuously working on identifying and designing materials for highly-desired applications, such as for gas separation, energy storage, or as part of electronic or sensory devices. This includes a sub-class of materials whose functionality is pressure-responsive, opening up a new plethora of new design possibilities. The capabilities of candidate materials are, however, often evaluated without a thorough understanding of the material structure and how it enables certain properties to be exhibited. The structure-property relationship is central to fully appreciate the behaviour of these materials and facilitate more targeted material development. The overarching theme of this thesis is to provide greater understanding of the structure property relationships of a selection of pressure responsive materials, via the use of high pressure X-ray diffraction and various complementary computational techniques. Prior to the experimental chapters, Chapter One provides an overview of the importance of the structure-property relationship in materials alongside two exemplary investigations from the literature and Chapter Two focuses on the theory and defining characteristics of the investigative techniques used in this thesis to outline their suitability and complementary nature to explore the structural and property responses of materials.

Chapter Three focuses on the extensive investigation of four square-planar Pt(II) complexes, specifically two 1,2 dionedioximato complexes and two Magnus salts, which form one-dimensional stacks of metal centres in the solid state and have been identified for their piezoresistivity capabilities through the contraction of the short Pt···Pt separation distances. The full structural response of these four compounds to pressure have been evaluated experimentally for the first time, with platinum bis(1,2 benzoquinonedioximato) demonstrating by far the strongest anisotropic pressure-volume response. Solid-state hybrid density functional theory calculations revealed this response was enabled by strong Pt···Pt interaction in the valence band and a combination of Pt···Pt and inter-ligand interactions in the conduction band, the latter of which had not been identified in previous studies. These interactions thus result in a small ambient-pressure electronic band gap (0.5 eV), which contracted quickly to form the metallic state by 1 GPa. The other Pt complexes did not exhibit such strong responses due to a lack of the desired strong interlayer interactions in one or both of the frontier bands. The results discussed herein have provided a more detailed set of design criteria for future piezoresistive material development.

Chapter Four continues in a similar manner to Chapter Three, in terms of the touted application of the material under investigation and the techniques utilised, but instead focuses on two polymorphs of an Au(I) dithiolene Mott insulator. In this material, the unpaired electron site is delocalised across the dithiolene components of the ligand and thus communication between neighbouring sites stems mainly from S···S interactions, rather than Au···Au interactions. Both polymorphs were determined to have similar ambient pressure band gaps (ca. 0.50 eV) thereby slightly overestimating the literature-stated experimental band gap (0.22 eV). The P21/c polymorph showed a stronger pressure-volume response and faster band gap compression, the latter due to greater delocalisation of the Hubbard sub-bands originating from short S···S interactions (< 4 Å) in two directions of the crystal structure. On the other hand, the newly-discovered P21 polymorph displayed a poor band gap response due to only possessing uni-directional S···S interactions, although the results obtained were hindered by poor crystal quality. These results however further highlight the relationship between crystal packing and the material properties, providing a word of caution to future research in this field that the possibility of polymorphism must be considered carefully.

Chapter Five builds upon the current understanding of the small heterocyclic radical molecule 1,3,5-triathia-2,4-6-triazapentalenyl (TTTA) which is known to exhibit magnetic bistability via a strongly hysteretic paramagnetic diamagnetic switch with an associated structural phase transition between P21/c and P1 ̅ polymorphs. Both polymorphs are dominated visually by π-stacking columns, with these columns linked by short lateral N···S interactions. This new study has revealed the full structural response of the P21/c polymorph under pressure, and suggests that the literature-reported gradual suppression of paramagnetism in this material as a response to pressure is due to the decreasing intermolecular separation in the regular π-stacking chains instead of a full structural phase transition as seen under variable temperature. Furthermore, semi-empirical calculations were utilised to calculate the strength of the myriad of intermolecular interactions in both polymorphs, which revealed that the strongest inter-column interactions are comparable to the slipped π stacking interactions (ca. 30 kJ mol-1), verifying their importance in the construction of the two polymorphs and as the source of the wide magnetic hysteresis.

Chapter Six switches focus to a small-pore Sc-based metal-organic framework which has been identified from various crystallographic studies to undergo a Fddd→C2/c structural phase transition upon sufficient adsorption of CO2, as a pure gas or as part of a gas mixture, and shows strong selective uptake for CO2 over CH4 which is a highly desirable trait. To provide clarity to the results obtained from X-ray diffraction experiments, classical grand canonical Monte-Carlo simulations have been employed on various fixed structures of the framework obtained under a range of experimental conditions to investigate the adsorption site hierarchies of both CO2 and CH4 and quantify the strength of the guest-framework interactions. This study has shown that the selectivity for CO2 uptake in this material stems from stronger CO2-framework interactions (between -40 and -20 kJ mol-1) compared to those calculated between CH4 and framework (no stronger than -20 kJ mol-1). Furthermore, a shift to stronger CO2 adsorption site energies was observed during the phase transition, whereas no such change in energy was seen in the simulations involving CH4 adsorption, thus clarifying the different structural responses observed experimentally. This study has highlighted the strengths of utilising classical-based simulations for analysing these type of systems as well how changes in adsorption site environment are highly correlated with subtle changes in the framework structure.