School of Chemistry - Theses

Permanent URI for this collection

Search Results

Now showing 1 - 10 of 133
  • Item
    Thumbnail Image
    Fabrication of PVDF–TiO2 electrospun membranes incorporating with carbon nitride for solar fuel production and organic pollutant photodecomposition
    Tan, Jeannie Ziang Yie ( 2016)
    Semiconductor–mediated photocatalysis for the decomposition of pollutants and production of industrially important species, i.e., methane by photoreduction of CO2 (g) is an emerging technology. However, problems, including low quantum efficiency, visible light inactivity and the difficulty to deploy and recover the photocatalyst, have to be mitigated. In order to enhance the photocatalytic activity of titanium dioxide, the sensitization of TiO2 with visible light active carbon nitrides (CNx) was proposed. Nonetheless, as an important step in the fabrication of a photocatalytic device, the integration of photocatalytic nanoparticles into a solid matrix, such as an electrospun fibrous membrane, forms a research objective in this thesis. A low temperature synthesis route to fabricate TiO2 nanoparticles with different crystal phase compositions was developed. The Ti–precursor concentration (9–45 mM) and the presence of Cl– during hydrothermal treatment influenced the TiO2 crystal phase composition. Overall, anatase–rich TiO2 samples showed higher photocatalytic decomposition activity than rutile–rich samples. However, all samples and a commercial TiO2 reference produced only trace amounts of methane during CO2 photoreduction. A polyvinylidene fluoride (PVDF)–TiO2 nanocomposite was fabricated by electrospinning followed by a low temperature hydrothermal treatment to induce the in situ growth of TiO2 nanoparticles on the electrospun PVDF nanofibres. The crystal phase composition of TiO2 was tuned by manipulating the concentration of the Ti–precursor (0.030–0.125 M) and acidity (pH <0–6.5) in the hydrothermal solution. The surface accessibility, crystal phase composition and the presence of Ti3+ within the nanocomposite significantly influenced the photocatalytic activity for CO2 reduction and organic pollutant decomposition. The maximum production of methane was 19.8 µmol per gram of photocatalyst per hour (quantum efficiency for the photomethanation reaction, Q. E.CH4 : 0.44 %) under UV irradiation. The visible light absorption of the PVDF–TiO2 nanocomposite was enhanced by the addition of CNx. A facile, low temperature wet–chemical synthesis was developed for CNx. The synthesized CNx possessed C=O functional groups that resulted in a negatively charged surface across pH 3–9, and led to an enhanced adsorption capacity and organic pollutant photodegradation under visible light irradiation. CNx also showed a relatively high capacity for heavy metal ion adsorption. Unfortunately, the CNx particles were too large for successful incorporation into the PVDF–TiO2 nanofibres. As an alternative, graphitic–CNx quantum dots (g–CNQDs) were synthesized by microwave heating, and were introduced into the PVDF–TiO2 nanofibres during electrospinning. The g–CNQDs were evenly distributed along the nanofibres, and significantly extended the photoresponse of the nanocomposite into the visible range. Methane production from CO2 photoreduction increased with the amount of g–CNQDs incorporated into the nanocomposite, with a maximum production of 39.8 µmol of methane per gram of photocatalyst per hour (Q.E.CH4 = 0.58%) under simulated sunlight irradiation.
  • Item
    Thumbnail Image
    Fabrication of nanostructured titania electron transport materials for high-performance perovskite solar cells
    Wu, Wuqiang ( 2016)
    Excessive energy consumption and aggravated environmental pollution has called for the utilization of clean energy for efficient and effective energy conversion and storage. Semiconducting titania thin films with fascinating optical, electrical and electrochemical properties have been intensively investigated as promising anode materials for thin film optoelectronic and photovoltaic devices. However, facile solution-processed techniques for the preparation of high-quality titania-based thin films have not been extensively explored. The focus of this thesis is to synthesise uniform thin films of various titania nanostructures on transparent conducting oxide substrates via a simple hydrothermal route in a cost-effective and highly controlled manner, and to apply them as electron transport layers (ETLs) in perovskite solar cells (PSCs). In the first research chapter, diverse TiO2 ETLs with different dimensionalities (i.e. 0D nanoparticles (TNP), 1D nanowires (TNW) and 2D nanosheets (TNS)) were hydrothermally grown on the transparent conducting oxide substrate by varying the organic solvent used. PSCs constructed using 1D TNW or 2D TNS yield enhanced photovoltaic performance compared to the 0D TNP counterparts. On one hand, the porous TNW or TNS network is beneficial to the infiltration of perovskite precursor. On the other hand, the electron transport and charge extraction at the TNW/perovskite or TNS/perovskite interfaces was facilitated, and thus reduced interfacial recombination loss. Employing a bilayered ETL film consisting of a self-assembled TiO2 blocking layer and a subsequent TNW active layer, produced PSC devices with a power conversion efficiency (PCE) exceeding 16%. Then, efforts were focused on optimizing the electric properties and exploring low-temperature processability of the TNW structure. In Chapter 3, a glucose-assisted self-assembly solvothermal protocol is developed to prepare electron-rich TNW thin films that exhibit enhanced electron mobility. PCEs up to 17.75% were attained for PSCs based on an integrated TiO2/CH3NH3PbI3 planar heterojunction and CH3NH3PbI3/spiro-MeOTAD bulk heterojunction structure, which is efficient in light harvesting and charge collection. In Chapter 4, an amorphous titania nanowire thin film was prepared via hydrolysis of potassium titanium oxide oxalate in an aqueous solution at 90 °C. High photovoltaic performance is obtained owing to the porous nanowire network that allows for efficient perovskite infiltration, the unique 1D geometry and excellent substrate coverage for efficient electron transport, as well as suppressed charge recombination. The 3D branched nanowire array architecture has emerged as a promising ETL design to achieve high-performance photovoltaic devices owing to their high internal surface area and well-interconnected networks for sufficient perovskite infiltration and superior charge transport characteristics. In Chapters 5 and 6, a series of branched TiO2 nanowires with homogeneous or heterogeneous compositions have been constructed. The 3D branched nanowires could efficiently capture light, block the holes and transport the photoinjected electrons. Moreover, the enlarged surface area was beneficial for the perovskite precursor infiltration leading to efficient charge extraction. In addition, the heterojunction branched TiO2 nanowires with appropriate band alignment between the constituent trunks and branches is beneficial for an energy cascade favouring charge extraction, as well as beneficial surface passivation effect for suppressed charge recombination. The research contained in this thesis provides approaches for the construction of novel semiconducting metal oxide electrodes for application in high performance optoelectronic devices achieved by incorporating nanostructured design and ETL/perovskite interfacial engineering. The findings from this work could also be implemented in other energy conversion and storage research, such as water splitting, Li-ion batteries and photocatalysis.
  • Item
    Thumbnail Image
    Molecular characterisation of the multicopper oxidases CueO and ceruloplasmin: key proteins in copper and iron metabolism
    Cortes Castrillon, Laura ( 2016)
    Copper and iron are both essential metals in most biological systems. The ability of these two metals to cycle between different oxidation states within the frameworks of biomolecules is widely utilized in many biological processes. However, these redox properties make them potentially toxic. Consequently, biological systems have evolved sophisticated components and mechanisms to safely metabolise them. Versatile multicopper oxidases (MCOs) have evolved for such action. This thesis describes a series of studies aimed at a molecular understanding of the biological functions of two of these oxidases: the first, CueO, is a cuprous oxidase essential for handling excess copper in bacteria; the second, ceruloplasmin (Cp), is a ferroxidase essential for ron metabolism in humans. The MCOs are a large family of metallo-enzymes present in all three domains of life. They couple the one-electron oxidation of substrate to the four-electron reduction of O2 to H2O. They feature at least four Cu atoms classified into three sites: type 1 (T1), type 2 (T2), and the binuclear type 3 (T3). The T1 site catalyzes substrate oxidation while the T2/T3 trinuclear cluster (TNC) reduces O2 to water. Since the substrate oxidation occurs at the T1 Cu site, the substrate specificities are determined primarily by the nature of a substrate docking/oxidation (SDO) site associated with the T1 site. Metallo-oxidases are a sub-set of MCOs that specifically bind and catalyze the oxidation of low-valent transition metal ions including Cu(I), Fe(II) and Mn(II). CueO is a soluble periplasmic cuprous oxidase that confers copper tolerance in E.coli under aerobic conditions. In addition to the classic machinery (T1, T2, T3) common to MCO function, it features a Met-rich insert (residues 355-402; 14 Met; 5 His) that includes a helix that physically blocks access to the catalytic T1 centre. This insert provides at least three additional Met-rich Cu(I) binding sites, designated as Cu5, Cu6 and Cu7, that are tailored to facilitate Cu(I) binding and oxidation. Protein variants targeting these metal sites were generated and the enzyme activities were compared to that of the wild type form. The results demonstrate that these three copper sites play related but distinct roles in the robust cuprous oxidase activity of CueO. Cp is required for iron metabolism in mammals and is one of the most complex of MCOs. It receives Fe(II) delivered by the export pump ferroportin and oxidizes it to Fe(III) that is delivered subsequently to plasma metallo-chaperone transferrin. Study of Cp at the atomic and molecular levels has been hampered for over 15 years due to lack of a suitable model ferrous substrate. The chromophoric complex FeIIHx(Tar)2 (HxTar, 4-(2-thiazolylazo)resorcinol; x = 0-2) was developed in this work as a novel and robust substrate that mimics the Fe(II) delivery function of ferroportin. It allows a direct ferroxidase activity assay for Cp and related enzymes in real time via an intense absorbance at 720 nm in solution. The enzyme formally converts substrate FeIIHx(Tar)2 to its ferric form [FeIIIHx(Tar)2]+ by coupling to the reduction of O2 to H2O. The formation constants of both ferrous and ferric forms at pH 7.0 were determined (log β_2^' = 13.6 and 21.6, respectively), as well as average dissociation constant of the SDO site(s) in Cp (log K_D^'= –7.2). This thermodynamic data is consistent with structural evidenc for an iron transfer route within the enzyme. This new assay has a detection limit of 10 nM Cp and allows direct detection of Cp activity in serum. Cp undergoes various post-translation modifications (PTMs) in vivo to satisfy its cellular functions and in response to various disease states. A PTM screening for Cp samples from serum was performed and allowed identification of several new phosphorylation sites and residues that are most susceptible to the oxidative stress associated with neurological diseases. In combination with the new activity assay, a proteomic approach was employed for the molecular characterisation of Cp in sera from a small sample of Parkinson’s and Alzheimer’s disease patients, buffered by healthy controls. However, no significant differences were observed in protein expression, copper content and enzyme activity. The work covered by this thesis provided not only a set of versatile research tools for study of metallo-oxidases in general, but also valuable new insights into the reaction mechanisms of these enzymes. This study may guide further development towards a better understanding of the diverse functions of this class of enzymes.
  • Item
    Thumbnail Image
    Interactions between inverse bicontinuous cubic phase materials and encapsulated biomolecules
    Meikle, Thomas G. ( 2016)
    The inverse bicontinuous lipidic cubic phase provides a robust, thermodynamically stable membrane mimetic, with the ability to encapsulate a wide range of biomolecules, including amino acids, peptides and proteins. The unique structural architecture and desirable properties of cubic phase lipids have led to their use in wide range of applications, including as a crystallization matrix for membrane proteins, as well as a hosting environment for therapeutic compounds, enabling the design of drug delivery materials. Within the cubic phase-biomolecule system, a complex structural relationship exists between the chemical structure of the lipid, the overall mesostructure of the system, and the structure and properties of encapsulated compounds. At present this relationship is poorly defined, slowing the further development and implementation of these materials. To investigate the structural relationship between lipidic bicontinuous cubic phase bilayers and encapsulated biomolecules, a number of different cubic phase-biomolecule systems have been examined. The encapsulated biomolecules range in complexity from free amino acids to large integral membrane proteins, and give this work relevance to a number of different end use applications, including the delivery and release of small molecule drugs and peptides, as well as the in meso crystallization of integral membrane proteins. We have examined the effects of encapsulating L-histidine and L-phenylalanine within a range of different cubic phase lipids on the mesostructure of the cubic phase. Subsequently, the translational diffusion of these compounds was measured using PFG-NMR, and a linear relationship was discovered with the diameter of the nanoscale water channels. It was demonstrated that a simple mathematical model can be used to predict in vitro release rates of small molecules based solely upon NMR determined in meso diffusion coefficients. Encapsulation of the transmembrane, anti-microbial peptide gramicidin A’ in the lipidic bicontinuous cubic phase revealed a number of interesting structural changes, both in the bilayer and the peptide itself. These changes were found to be strongly correlated with bilayer parameters such as lateral pressure profile and hydrophobic mismatch, providing further insight into the structural considerations of the system. Studies of peptide encapsulation were expanded to include a broader range of antimicrobial peptide structures, including melittin and alamethicin, which along with gramicidin A’ were encapsulated within a number of different cubic phase nanoparticle formulations. Thorough characterization and analysis of each system revealed that as in the bulk phase, the structural changes in cubosome systems were dependent on factors such as hydrophobic mismatch and lateral pressure of the bilayer, as well as peptide structure and charge. The host lipid identity was also found to significantly modulate peptide uptake. In later work, we examine the encapsulation of the integral membrane protein intimin in the lipidic cubic phase, in a study which elucidates the mesophase properties which lead to successful protein crystallization. The structural parameters of the mesophase were tracked throughout the crystallization process, providing insight into the mesoscale changes occurring during crystal nucleation. Cubic phase bilayers comprised of a range of lipids were also characterized and a strong correlation was found between properties such as protein and lipid diffusion, and successful crystallization. This body of work constitutes a significant contribution to our understanding of the interactions between encapsulated biomolecules and the cubic phase bilayer. The structural trends and considerations highlighted have implications for the further implementation of these materials, including in the rational design and formulation of drug delivery materials and in meso crystallization experiments. Information of a fundamental interest is provided through broader observations regarding the interaction of biomolecules with lipid membranes.
  • Item
    Thumbnail Image
    Predicting the thermodynamic properties of liquids from computer simulations
    LIU, MAOYUAN ( 2016)
    Computer simulations provide information on atomic structures that can be challenging to obtain experimentally. This thesis demonstrates the use of computer simulations to predict macroscopic thermodynamic quantities that are experimentally verifiable. This thesis also combines different computational free energy methods to evaluate the realism of structural explanations for liquid phenomena. One computational free energy method is the `alchemical' particle insertion method, which can provide benchmark thermodynamics for simulation models. Theories which claim to explain liquid phenomena must then be consistent with both simulation benchmarks and experimental measurements. Three scientifically and industrially important liquid systems are studied in detail to establish the generality of this approach: molten sodium chloride, molten calcium aluminosilicates, and water. The study of molten sodium chloride establishes the robustness of the `alchemical' particle insertion method. The study of molten calcium aluminosilicates establishes that structural ordering entropy methods can identify structural features which control macroscopic thermodynamics. The study of the hydrophobic effect in water evaluates possible structural explanations against `alchemical' simulation benchmarks, and rejects several popular theories for the hydrophobic effect. We propose a parameter-free theory of dipole correlations, the first structural explanation for the hydrophobic effect consistent with all simulation and experimental evidence.
  • Item
    Thumbnail Image
    Synthesis and electroluminescent applications of II-VI semiconductor quantum dots
    Kirkwood, Nicholas ( 2016)
    This thesis studies next-generation light-emitting-diodes or LEDs as a means to create sustainable, energy-efficient lighting. It is shown that high-quality semiconductor nanocrystals can be used to create tunable, efficient and photostable LEDs. These new devices can be fabricated using wet-chemical technologies that drastically reduce the energy cost of production.
  • Item
    Thumbnail Image
    Substrates, substrate mimics and inhibitors of bacterial and fungal mannanases and transmannosidases
    Belz, Tyson ( 2016)
    Glycoside hydrolases are a diverse group of enzymes that have been classified into more than 130 sequenced based families. Glycoside hydrolase family 76 (GH76) is poorly characterized and is populated exclusively by bacterial and fungal members. The only known biochemical function for members of this family has been described for bacterial enzymes, termed α-1,6-mannanases, which hydrolytically cleave α-1,6-mannans, such as those found in the fungal cell wall. Separately, it has been hypothesized that certain fungal members, termed DCW1 and DFG5, are transmannosidases, with the ability to cleave an α-mannosidic linkage within the glycosylphosphatidylinositol (GPI) glycan of cell wall proteins, and then catalyze the formation of a new glycosidic linkage to the cell wall glycan, resulting in covalent attachment of mannoproteins to the fungal cell wall. The first part of this thesis details part of a large collaborative effort dedicated to establishing a detailed chemical mechanism for the foundation member of glycoside hydrolase family 76, Aman6 α-1,6-mannanase from Bacillus circulans. In an effort to obtain non-hydrolyzable substrate mimics, a series of S-linked α-1,6-oligomannosides were synthesized using alternative [2+2+2] approaches that involve assembly from either the reducing end or the non-reducing end. In the first approach, involving construction from the reducing end, coupling of a disaccharide thioacetate with a 6’-iodo reducing end disaccharide, followed by activation of the resulting tetrasaccharide to a 6’’’-iodide, and iterative coupling with the same disaccharide thioacetate afforded an S-linked hexasaccharide, as well as the intermediate di- and tetrasaccharides. In the second approach, involving construction from the non-reducing end, coupling of the above disaccharide thioacetate with an anomeric S-trityl protected 6’-iodo disaccharide, afforded an S-trityl tetrasaccharide, which was converted was converted to a tetrasaccharide thioacetate, which was in turn coupled with the same anomeric S-trityl protected 6’-iodo disaccharide to afford the hexasaccharide, which was finally elaborated to the methyl thioglycoside. X-ray crystallographic studies with the catalytic domain of Aman6 (BcGH76) and the synthetic S-linked oligomannosides provided evidence only for binding of sugar residues in the -3/-2 subsites, and no active site spanning complexes could be obtained for the wildtype enzyme. However, mechanistically informative complexes were obtained with α-1,6-mannosylisofagomine (ManIFG) and an active-site spanning complex with α-1,6-mannopentaose and a catalytically disabled mutant. In tandem with combined quantum mechanics/molecular mechanics calculations, a conformational itinerary for the enzyme-catalyzed reaction was proposed. The second part of this thesis details efforts to develop a non-hydrolyzable trisaccharide homologue of ManIFG that contained only S-linkages. Building upon the methodology and key intermediates developed in the previous part, S-linked di- and trisaccharide isofagomines were readily assembled. Crystallographic studies with BcGH76 revealed the disaccharide to bind unexpectedly in the -3/-2 subsites; the trisaccharide bound in a more orthodox fashon in the -3/-2/-1 subsities, but even so bound without proper engagement with the catalytic machinery, and in particular the catalytic nucleophile. We conclude that the introduction of one or more S-linkages substantially perturbs the structure of the S-linked oligomannosides such that they are poor mimics of substrate. The final chapter of this thesis discloses the synthesis of a GPI anchor fragment. Two tetrasaccharides, α-Man-1,2-α-Man-1,6-α-Man-1,4-α-GlcN-OMe, and the corresponding acetamide, were prepared using a novel procedure inspired by previous work of F. Kong. The route is notable for use of only ester, acetal and amide protecting groups, and unlike previous work, the tetrasaccharides were deprotected. It is anticipated that these tetrasaccharides may prove of use in obtaining direct experimental evidence in support of mannoprotein cross-linking into th fungal cell wall catalyzed by the family GH76 enzymes DCW1/DFG5.
  • Item
    Thumbnail Image
    A new method for amide bond formation
    Pourvali, Aysa ( 2016)
    The synthesis of amide bonds is one the most important issues in organic chemistry. The reaction of thioamides and silver carboxylates to generate imides, and ultimately amide bonds, was investigated. This system was shown to be effective for the coupling of a range of N-protected amino acids with amino acid derived thioamides, leading to the preparation of dipeptide imides. Selective cleavage of the imide compounds by methanolysis or hydrolysis was shown to be a viable general method for amide bond formation. However, the described method was not suitable for the generation of polypeptides by coupling peptide acids with peptide N-thioacetamides. Nevertheless, the synthesis of polypeptides could be achieved by coupling of a peptide containing a free C-terminal carboxylic acid to an amino ester thioamide in the presence of silver(I), which after imide hydrolysis and ester deprotection gives the homologated peptide. This cycle is repeatable to synthesize longer polypeptides. The application of our method to synthesise polypeptides was exemplified by the synthesis of the pentapeptide thymopentin. The synthesis was performed in the N→C direction, without any epimerization. Initial investigations towards the synthesis of cyclopeptide alkaloids in the presence of Ag2CO3 were also undertaken.
  • Item
    Thumbnail Image
    Structured photocatalysts for solar energy storage with improved boundary layer mass transfer
    Parris, David Hayshiv ( 2016)
    As solar energy contributes more of our energy needs, energy storage becomes critical. There are two main strategies for solar energy storage; in batteries or as chemical energy. These two strategies fit differently into future renewable energy infrastructures. Batteries suit small, distributed photovoltaic installations such as rooftop solar panels. They are less suited to industrial scale installations not attached to the grid. Just as green plants store energy whilst the sun shines, and use it overnight, we need systems that efficiently store the solar energy as chemical energy in forms that can readily be converted back to electrical energy as required. Methods for capturing and storing solar energy using photocatalytic as against photovoltaic cells include water-splitting, conversion of carbon dioxide to methane, methanol or hydrocarbons, and direct recharging of redox flow batteries. All of these have the same requirements as leaves – light capture, the need to flow liquids through the cells with a minimum of pumping energy, and molecules that store the energy in a useful way. The capture and storage of solar energy is an active field. Focus in this work was on methods of energy storage other than conventional batteries, and on work using variations of the Z-Scheme approach to light capture as it allows a greater proportion of the available spectrum to be captured. For water splitting, closely coupled schemes, where hydrogen and oxygen are generated simultaneously, were considered too dangerous to scale up and so were largely ignored. The Z-Scheme can be split into an oxidation reactor and a reduction reactor, with a recirculating shuttle molecule. In water splitting these would generate oxygen and hydrogen gas respectively. This allows separate photocatalysts to be used for the two reactions. In this thesis, activated tungsten oxide was chosen as the photo-oxidation catalyst due to its high selectivity for converting ferric ions to oxygen and ferrous ions, and anatase was selected as the basis of the photoreduction catalyst because of its low cost, availability, and the large amount of work carried out controlling surface area, porosity and chemistry of anatase photocatalysts. The photocatalysts were screened as slurries of dispersed beads and particles. This allowed rapid and simple photocatalyst preparation, but did not allow continuous reactor operation due to the need to separate the catalyst from the reactant solution. Catalysts were characterised by nitrogen adsorption for surface area and porosity, scanning and transmission electron microscopy for particle size, crystal size and morphology. Spectrometry was used to look at the absorption and scattering of light by the catalysts, and to monitor progress of reaction colorimetrically. Colorimetric actinometry was used to calibrate the light source, a mercury-xenon arc lamp. Methylene blue decolourization was used to follow the reduction reaction, and ferrous and ferric bipyridyl complexes were used to follow the oxidation reaction. Results from slurry reactors proved difficult to analyse. Calculations were carried out that showed that slurries did not behave as simply as assumed in most of the literature. In particular there was evidence of mass transfer limitation from the bulk liquid to the photocatalyst surface, even in well-stirred reactors, and variability in light capture and light loss by transmission and back scattering introduced significant errors. A 65 mm diameter heterogeneous reactor was designed and built to study continuous reaction. The initial design used thin film photocatalysts mounted on fused quartz, produced by evaporative casting. Data were logged continuously using an in-line cuvette connected to a CCD spectrophotometer. This reactor system was effective, with high reproducibility. The main source of errors was bubbles passing through the cuvette. Using this reactor, reaction rates were shown to be limited by boundary layer diffusion between the flowing liquid and the catalyst, not diffusion within the catalyst layer as had been expected. The reaction appeared to follow Langmuir-Hinshelwood kinetics closely, where the diffusion limitation depends in part on the binding of the substrate to the photocatalyst surface. So diffusion had a much greater impact on the bleaching of the weakly binding methylene blue over anatase than on the strongly binding ferric ions over tungsten oxide. To overcome boundary layer diffusion it is necessary to develop catalysts structured to optimise both light capture and mass transfer. To retain the advantages of thin film casting, engineering approaches to reducing boundary layer diffusion were developed. Designs tested included the “Leaf” reactor and a static mixer. Both were produced using 3D printing. Little or no improvement over the standard thin film reactors was observed, probably because the resolution of the printer was not fine enough (a target of less than 180 μm was estimated from thin film results, with the effective printer resolution being greater than 1,500 μm). A more successful approach was the development of structured catalysts based on supporting the active components on a felted quartz mat. Unlike glass fibre supported photocatalysts reported in the literature, the quartz mat fibres were laid down in a more random and open structure that did not bias liquid flow. Supported photocatalysts were prepared using evaporative casting, and growing thin films onto the fibres using sol-gel or aqueous methods. High reaction rates per unit area were obtained using structured catalysts, although some diffusion limitation remained at practical flowrates. The results obtained clearly demonstrate that the next generation of photocatalytic cells must be designed and engineered to further reduce or eliminate diffusion limitation while maintaining low flow resistance and hence energy lost to pumping. This should combine finite element modelling, improved methods of coating substrates with photocatalysts, and optimisation of catalyst activity by controlling surface area, porosity and light capture to ensure efficient use of as much of the solar spectrum as possible. The most cost effective reactions for energy storage, transport and recovery have yet to be identified.
  • Item
    Thumbnail Image
    High accuracy measurements of x-ray absorption spectra and application in conformation and structural analysis of Ni(II) complexes and ferrocene
    ISLAM , M TAUHIDUL ( 2016)
    X-ray Absorption Spectroscopy (XAS) is uniquely sensitive to local and element specific structure applying to samples as diverse as solution, gas, solid, crystal or films. However, standard XAS applied particularly in disordered environment (i.e. solution) does not currently use error analysis from experimental systematics, raising concerns about the reliability of refined structures. This work develops experimental and analytical methodologies, which allow new high accuracy measurements of XAS spectra from dilute samples using synchrotron radiation. A new technique - the hybrid technique - is developed and applied to two nickel(II) complexes including ([bis(N-n-propyl- salicylaldiminato)] nickel(II), (n-pr, Ni) and [bis(N-i-propyl- salicylaldiminato)] nickel(II), (i-pr, Ni) complexes, and two sandwich compounds including Ferrocene(Fc) and decamethylferrocene (dmFc). Applying the technique, XAS spectra of the complexes are measured on both relative and absolute scales to high accuracy (a 1%-5% accuracy for the nickel(II) complexes; 0.2% to 2% accuracy for Fc and dmFc compounds ) by the characterisation and correction of experimental systematics. These results are the first ever report of high accuracy measurements from dilute systems. The solid-state structures of the (i-pr, Ni) and (n-pr, Ni) complexes are well defined from the crystallographic measurements. Their structural forms in solution have not been investigated. With the use of high-accuracy XAS spectra, application of a statistically robust XAFS refinement of the isomers confirms that the (i-pr, Ni) complex belongs to a distorted tetrahedral geometry, and the (n-pr, Ni) complex represents a distorted square planar geometry. The general structural forms of the complexes are consistent with crystallographic data, but the key bond-lengths and bite angles varied from their crystalline structures. The analysis is the first time demonstration of XAFS analysis of dilute isomers propagating experimental uncertainties for meaningful chi^2_r determination and hypothesis testing. The structural analysis of the nickel(II) complexes has provided us with necessary understanding for an application of the statistically robust method to the stereochemistry of high symmetry molecule Fc. The structural form of Fc is however experimentally unclear, whether Fc is dominated by a single conformer or a mixture of the two conformers, or the structures depend on medium. Previous XAFS approaches failed to distinguish between the conformations of Fc. This work provides exciting evidence on this topic. In addition to XAFS approaches, previous studies of vibrational IR spectroscopy of Fc were inadequate to resolve the issue. This work therefore aimed an extensive study to obtain a clear answer. An application of statistically-validated XAFS analysis indicated the dominance of the D5h form in Fc structure. Consistently, the quantitative analysis of the measured high-resolution and temperature series (300 K - 7 K) IR spectra in gas-phase and in condensed-phase confirms the dominance of the D5h in Fc. This work also provides a first experimental measure of the barrier height (6.8 +/- 0.1 kJ/mol) for the solution state species from the modelling of vibrational-levels in observed IR spectra.