No description available.

No description available.

No description available.

No description available.

Molecular dynamics (MD) simulations enable the prediction of structure and function of biomacromolecules. For example, the activity of small drug molecules can be ranked by molecular docking thereby reducing the number of candidates. To date, several popular empirical force fields are available for MD simulations, such as CHARMM, AMBER, OPLS, and GROMOS. These force fields are commonly used but limited to relatively low accuracy in their predictions as all these force fields compute the atom-atom electrostatic interaction energy using point charges. It is well known that the non-bonded interaction energy, in particular the electrostatic energy, plays a crucial role in determing the structure of proteins, including their docking regions. Therefore, the electrostatic energy is one of the most important parts to be correctly predicted in any force field. As a result, improvements in a force field are critical to obtain more reliable protein simulations. In our work, a novel force field based on Quantum Chemical Topology (QCT) has been actively developed in order to increase the accuracy of the simulations. Currently this force field is referred to as QCTFF. QCTFF calculates the atom-atom electrostatic energy based on the atomic multipole moments instead of point charges. The smallest natural protein, crambin, is used as a pilot system to compare the multipole moment electrostatic interaction energy with the point charge energy. The atom-atom electrostatic interaction energy calculated based on atomic multipole moments is remarkably different from the values calculated by point charges. There are in total fifteen possible atom-atom interactions, including HH, HC, HO, HN, HS, CC, CO, CN, CS, OO, ON, OS, NN, NS, SS, typically occurring in proteins. In order to study the convergence of the multipole expansion behind QCTFF, the minimum internuclear distance in the convergent region for each of the fifteen atom-atom interactions are calculated. The levels of theory used are HF/6-31G(d,p), B3LYP/aug-cc-pVTZ and MP2/aug-cc-pVTZ. This study determines which multipole moment rank is 14 required to obtain the electrostatic energy between two atoms, at a given distance, to a given energy accuracy. In addition, we define an atom type for each element C, H, O, N and S. The element in question is taken as the central atom and a “horizon sphere” centred at its nucleus is systematically enlarged to find out at which radius the atomic multipole moments are no further influenced by atoms entering the growing sphere. All 20 natural amino acids (AAs), from which any protein can be built, are studied in the construction of QCTFF. The minimum energy geometries of each amino acid are optimised and computed at HF/6-31G(d,p), B3LYP/apc-1 and MP2/cc-pVDZ levels of theory. The number of minima of the 20 AAs increased with the size and the flexibility of the side-chain. The multipole moments of the atoms in the five-atom fragment [-NH-Cα-C(=O)]- varied according to the same trends in all three levels of theory. The properties, including bond length and bond angles, of the common seven-atom fragment [-NH-CαH(Cβ)-C(=O)-] (in all but proline and glycine) are investigated to probe the transferability between different AAs. A machine learning method called Kriging is applied to carry out the prediction of the atomic multipole moments in QCTFF. To build the Kriging models, the minima of serine are used to produce a sufficient number of geometries following the distortion procedure of both the normal mode and the redundant internal coordinate approaches. Ultimately, a thousand geometries are used in the training set to build Kriging models and the prediction reliability of the models is tested by the remaining thousand geometries in the test set. The same three levels of theory as before are applied to calculate the wave functions of the distorted geometries using GAUSSIAN03. Moreover, a large number of serine geometries are also extracted from proteins in the Protein Data Bank (PDB). The absolute average error of the total energy predicted by the Kriging models built from these geometries is slightly worse as the range of the geometries is much wider than the geometries produced by the two distortion approaches. Finally, a biologically relevant reaction is studied via the well-known peptide bond hydrolysis enzyme HIV-1 protease. Two types of reaction systems are studied: a small system and a large system. The object of focus is the hydrogen atom in the catalytic water molecule, for which the coordinates change most during the reaction process in both reaction systems. For the small reaction system, the intrinsic reaction coordinate (IRC) method is used to generate 295 different geometries. The Kriging models built from 180 of the 295 geometries can correctly predict both atomic multipole moments and the total interaction energies for the remaining geometries in the test set. For the large reaction system, instead of using the IRC method the motion of the hydrogen atom in question is 15 scanned during the reaction. Finally, 265 geometries are obtained and 180 of these geometries are used to build the Kriging models. The absolute errors of the multipole moments in the test set are all very small.

The aim of this project was to synthesis N-substituted pyrrole monomers with liquid crystal and dendritic side groups, followed by their subsequent polymerisation and mesaurement of the polymer properties. The second objective beyond the synthesis of these novel polymer materials was to investigate how potential liquid crystals and dendritic side groups affected the conjugation and planarit of the polypyrrole backbone, for the purpose of conductivity applications. Four differnt types of molecules were synthesised: 1. Polypyrroles with terminally and laterally attached mesogenic side groups (monomers/polymers 1&2) 2. Polypyrroles with first and second generation dendritic side groups, terminated by alkyl chains (monomers/polymers 3&4) 3. Polyprroles functionalised by first and second generation dendritic moities with hydrophilically-terminated alkyl chains (monomers/polyers 5&6) 4. Polypyrrole hybrid materials: second generation dendrimer with terminal liquid crystal groups (monomers/polymers 7) Although seven polymers were intended to be synthesised, polymers (4) and (6) were unable to be characterised, due to very poor solubility. However polymers (1), (2), (5) and (7), successfully underwent hydrolysis of the terminal and lateral ester side groups to carboxylic acids, making a total of nine novel N-substituted polypyrroles synthesised, within this project. The synthesis began with the generation of the monomer compounds (monomers 1-7). This was done by nucleophilic substitution of the potential side chain liquid crystal or dendritic moieties on a benzyl ring using a Williamson ether synthesis. A flexible hexyl spacer group was sustituted onto the intermediate compounds, followed by the attachment of pyrrole at the heteroaton site. Thin layer chromatography was used to follow the progress of each reaction and the intermediate compounds were subjected to column chromatography and high vacuum distillation for purification. Characterisation techniques such as Fourier-transform infrared spectrophotometry, gas chromatography/mass spectrometry, [sup]1H NMR spectrometry and elemental analysis were used to determine the purity and structural indentity of each monomer compound. Electrochemical (EC) polymerisation of monomers (1) and (3) on conductive indium-tin oxide (ITO) glass were attempted first. The cyclic voltammograms indicated that the polymerisation was successful, and the monomer oxidation and polymer doping and dedoping potentials were determined. However this method of polymer synthesis was abandoned, as the yield of the polymer film was too low and the polymer product was very difficult to remove from the ITO substrate for further analysis. Chemical polymerisation with FeCl[sub]3 in chloroform was successful, as indicated by [sup]1H NMR, KBr IR, and UV-visible spectrometry, and the yields of the polymer products were significantly higher than those of the EC method. Physical properties measurements (electrical conductivity, scanning electron microscopy, hot-stage optical microscopy, solubility testing and differential scanning calorimetry) were carried out, in order to determine the general natures and to indicate any possible applications of the novel polypyrrole materials synthesised. From the spectroscopic data and physical measurements, it was concluded that polymers (1), (2), (3), (5) and (7) and their corresponding carboxylic acids (1a), (2a), (5a) and (7a) had been successfully synthesised, as proton NMR confirmed the disappearance of the 2,5-hydrogens of pyrrole, and in addition conductivity measurements using simple 2-probe and ven der Pauw 4-probe methods on undoped polymers (1a), (2a), and (7a) gave reproducible conductivity values which classified them as being potential semi-conductors (2.2x10[sup]-4, 6.7x10[sup]-5 and 7.8x10[sup]-5 Sm[sup]-1 respectively). UV-visible spectrophotometry indicated that hydrolysis of the laterally attached liquid crystal ester groups to carboxyliv acids significantly reduced the pp* energy gap of polymer (1). (from 3.9eV to 2.48eV). It was suggested that hydrogen bonding in the carboxylic acid might have improved the planarity and conjugation of the polymer backbone and transformed the insulating ester polymers to potential semi-conductors. Differential scanning calorimetry (DSC) and hot-stage optical microscopy (HSM) were used to determine the phase transitions of polymers (1a) and (2a). The DSC traces showed small peaks, indicating liquid crystal phase transitions (nematic phases observed from 105-248[degrees]C, 105-170[degrees]C respectively). These were later confirmed by the use of HSM, as nematic textures were observed in the expected temperature region for nematic liquid crystals. In addition the ethalpy changes for the phase transitions were estimated, and polymers (1a) and (2a) were found to have nematic-isotropic phase transitions within the expected range of values. Scanning electron microscopy was used to examine the morphology of the polymer materials. Generally the N-substituted polypyrroles with dendritic side groups were found to have more porous morphologies, while N-substituted polypyrroles with potentially mesogenic side groups appeared to have more continuous and smoother morphologies. However, polymer (7a) was found to have the most porous morphology and it also add an unexpectedly high conductivity value. Polymer (7a) was the only polypyrrole hybrid material synthesised in this project, and it appeared that combining all three polymeric units (dendrimers, LC and CP) into one polymer system improved the planarity if the polymer backbone, encouraged the formation of a porous structure which facilitated p-type doping of the polymer by iodine vapour.

The magnetic properties of the first row transition metal oxides are wide and varied and have been studied extensively since the 1930’s. Observations that the magnetic properties of these material types change with the dimension of the sample have stimulated many theoretical and experimental studies of the systems involved. As sample sizes decrease towards the nanoscale long range crystallographic order is no longer possible. However, the application of mesoporous silica samples as hard exo-templates to direct the formation of mesoporous metal oxides has provided a new opportunity to explore the influence of scale of crystallographic order further. These types of samples have pore systems running through the material on the mesoscale (diameter between 2nm to 50nm) with pore walls truly in the nanoscale region (7nm to 9nm thick) crystallographically ordered over large scale distances. The work presented in this thesis presents magnetic and crystallographic studies of a variety of the first row transition metal oxides from chromium to nickel in three dimensional mesoporous forms predominantly using SQUID magnetometry and neutron powder diffraction. Rietveld refinements of diffraction data from hematite and eskolaite (®-Fe2O3 and Cr2O3) show that the samples have space groups identical to their bulk counterparts, however slight differences in lattice parameters are observed. Refinement of magnetic properties has also been performed and compared to magnetic property measurements. Of particular interest are results from a mesoporous hematite which show suppression of a well defined first-order magnetic phase transition (the Morin transition). This suppression has been studied extensively with neutron powder diffraction and preliminary inelastic neutron spectroscopic measurements. Comparisons with hematite nanoparticles which also show the suppression of the Morin transition can be drawn. Parametric neutron powder diffraction studies on Co3O4 and NiO samples shows that the Néel ordering temperatures are lowered as the mesoporous structure is imposed. This too was observed in eskolaite. Other studies have been carried out on mesoporous alpha-MnO2 (magnetometry) and nanoscale Li1+xMn2–xO4 (X-ray photo electron spectroscopy) with comparisons to their bulk counterparts and finally nanoparticulate hausmannite Mn3O4 (magnetometry and muon spin relaxation) which exhibits spin-glass type behaviour.

Spin, charge and orbital ordering in various crystalline compounds have been studied under extreme conditions. The main techniques used were synchrotron X-ray and neutron powder diffraction. High-pressure conditions were obtained by using a diamond anvil cell and the Paris-Edinburgh cell. Changes in the valence state of BiNiO3 perovskite under pressure have been investigated by a neutron powder diffraction study and bond valence sum (BVS) calculations. At ambient pressure, BiNiO3 has the unusual charge distribution Bi3+0.5Bi5+0.5Ni2+O3 with ordering of Bi3+ and Bi5+ charges on the A sites of a highly distorted perovskite structure. High pressure neutron diffraction measurements show that the pressure-induced melting of the charge disproportionated state leads to a simultaneous charge transfer from Ni to Bi, so that the high pressure phase is metallic Bi3+Ni3+O3. This exceptional charge transfer between A and B site cations coupled to electronic instabilities at both sites gives rise to a remarkable variety of ground states. Furthermore, Rietveld analysis of low temperature neutron powder diffraction data shows that the structure of BiNiO3 remains triclinic (space grp 1P) throughout the temperature range 5 to 300 K. BVS calculations confirm that the charge distribution is Bi3+0.5Bi5+0.5Ni2+O3 down to 5 K. The magnetic cell is identical to that of the triclinic superstructure and a G-type antiferromagnetic model gives a good fit to the magnetic intensities, with an ordered Ni2+ moment of 1.76(3) μB at 5 K. However, BiNiO3 is ferrimagnetic due to the inexact cancellation of opposing, inequivalent moments in the low symmetry cell. The effect of high pressure on the structural properties of (EDT-TTF-CONH2)6[Re6Se8(CN)6], a conducting, molecular, mixed-valence, π-conjugated radical, cation salt has been examined using synchrotron X-ray diffraction and a diamond anvil cell set-up. It has previously been shown that this compound undergoes a low temperature phase transition from a rhombohedral (space group 3R) to a triclinic (space group1P) structure at ~150 K. This transition is caused by a charge ordering. A LeBail profile fitting of powder diffraction data revealed a change in compressibility at 0.7 GPa indicative of a phase transition. This was confirmed by single crystal data which showed that the structure remains rhombohedral )3(R, up i to 0.4 GPa but is triclinic )1(P at 0.8, 1.2 and 1.8 GPa. Hence, high pressure, as well as low temperature, can drive the charge ordering in (EDT-TTF-CONH2)6 [Re6Se8(CN)6]. The transition pressure is between 0.5-0.7 GPa at 300 K. The crystal and magnetic structures of the orbitally ordered perovskite KCrF3 have been determined from neutron powder diffraction measurement at temperatures from 3.5 to 300 K. A phase transition from a tetragonal to a monoclinic structure occurs at 250 K but the orbital ordering is sustained. Long range antiferromagnetic order of the A-type occurs below TN = 46 K and the refined magnetic moment for the Cr2+ sites was found to be 4.39(7) μB.

No description available.

This work investigates two main areas of interest. The first is the synthesis and study of a variety of Lewis acid-Lewis base adducts of general formula X(_3)B-PR(_3) (X = CI, Br; R= alkyl or aryl). The (^11)B-P and (^10)B-P NMR coupling phenomena are investigated. The partial dissociation of some of these compounds in solution is probed by (^31)P and (^11)B NMR. The crystal structures of many of these compounds are presented and analysed for correlations between the properties of the phosphines and B P bond length. It was found that only the Tolman cone angle gave a good correlation to B P bond length. A compound containing a novel B2O2 ring was synthesised during these studies, and theoretical calculations performed to gain an insight into the planarity of this unusual 4- membered ring. The first crystallographically characterised example of the tri(o-toly)phosphonium cation is also presented. The related compounds X(_3)B pyridine, X(_3)B-2-picoline and X(_3)B 4-picohne (X = CI, Br) are investigated by various analytical methods. Crystal structures of these B-N adducts are presented. It is found that the structure of CI3B 4-picoline exhibits an interesting phase transition from orthorhombic Pnma at higher temperatures to monoclinic space group P2i/n below ca. 170K.The second area of work involves crystal structure analyses of a variety of co-crystals of his(phenylethynyl)benzene compounds, of general type R-Ph-CC-Ph-CC-Ph-R, with R = H, F, MeO, MeS, CH3 and CF3 and a varying fluorinadon pattern on the phenyl rings. The effect of these para substituents on the arene-peifluoroarene interaction was investigated. The effect of each substituent was found to be subtle. In most cases, the co-crystallisation induced an arene-perfluoroarene interaction that was not present in the individual components. Starting materials for the bis(phenylethynyl)benzene compounds often include halogenated tolans of the form Ph-CC-Ph-X (X = CI, Br, I), with either the halogenated or the non-halogenated ring fluorinated. Crystal structures of PhF(_5)-CC-Ph-X (X = I, Br, CI), Ph-CC-PhF4-I and Ph-CC-PhF4-CC-SiMe3 are presented and the arene- perfluoroarene interaction investigated. It is found that the PhPg-CC-Ph-X (X = I and Br) compounds show no evidence of the interaction. The dominating intermolecular force is a Type n halogen -halogen short contact. In contrast, the structure of the chloro analogue is dominated by the arene-perfluoroarene interaction, with no halogen -halogen short contacts. Neither intermolecular force is present in the Ph-CC- PhF(_4)-I or Ph-CC-PhF(_4)-CC-SiMe(_3) structures.