Thesis (Ph.D.)--Hong Kong University of Science and Technology, 2007. / Includes bibliographical references (leaves 156-167). Also available in electronic version.

Proefschrift Rijksuniversiteit Limburg, Maastricht. / Met lit. opg. en een samenvatting in het Nederlands.

Laser cladding is a process that is used to improve the properties of a metal surface. The properties in question may include hardness, wear-, corrosion- and/or fatigue-resistance. The process involves fusing a thin layer of additional metal to the original surface, using a laser as the heat source. Unfortunately, residual stresses are generated due to the rapid and highly localised thermal expansion and contraction that occur during the heating-melting-solidification-cooling cycle. These residual stresses can have a detrimental effect on the final performance of the clad component, especially with respect to corrosion resistance. Detrimental tensile residual stresses can be mitigated through the use of post-processing techniques such as laser shock peening (LSP). LSP is a process that uses a pulsed laser to generate intense spots of recoil pressure on a surface, thereby introducing compressive residual stresses. Post weld heat treatment (PWHT) is another process that could be also used in laser cladding in order to relieve tensile residual stresses. In this work, laser cladding was carried out by depositing a clad layer of AISI grade 316L stainless steel on to either a S275 steel substrate or an AISI grade 316L stainless steel substrate, using different process parameters. The hardness and residual stresses in the overlay and substrate were assessed for each laser clad sample before and after being treated with LSP and PWHT. The corrosion rate and microstructure were also assessed in each case. The novelty of this work is two-fold. Firstly, to the author's knowledge, it is the first study that attempts to link process parameters to both the residual stresses and the corrosion performance of austenitic stainless steel overlays deposited by laser cladding. The second novel aspect is based on the application of both LSP and PWHT to the deposited overlay in order to investigate whether an improvement in the mechanical properties and the corrosion resistance can be realised. In this study, tensile residual stresses were generated in the clad layers. However, the magnitude of the residual stresses did not appear to be particularly sensitive to the deposition parameters. Indeed, it was found that the number of layers that is deposited is more important than the choice of process parameters. LSP was effective in reducing the tensile residual stresses and in fact it introduced compressive stresses to all the samples that were treated. In contrast, PWHT only led to satisfactory stress relief when the AISI grade 316L stainless steel was deposited on to a matching substrate material. This was related to the fact that a difference between the thermal expansion coefficients of the overlay and substrate led to the development of significant tensile residual stresses on cooling down after PWHT. The corrosion tests on the clad coupons led to the development of pits and cracks. However, after LSP only pits were found, without any sign of cracking, for the test durations that were investigated owing to the fact that compressive stresses were generated. Similar results were found after PWHT for the clad samples in which the overlay material matched the substrate material. However, signs of cracking were observed after PWHT in samples where AISI grade 316L stainless steel was deposited on to an S275 steel substrate due to tensile residual stresses remaining within the overlay. This result suggests that there may be little benefit in carrying out PWHT for components in which grade 316L stainless steel is deposited on to a steel substrate. In contrast, there appear to be clear benefits associated with carrying out LSP in order to mitigate the residual stresses and retard the onset of cracking.

This thesis is concerned with the design and development of measuring devices for the characterisation of acoustic transients with high temporal and spatial resolution. Three new techniques are demonstrated characterising acoustic transients generated by Nd-YAG laser (1060nm, 30ns, 55mJ) assisted breakdown of water and air. The first technique demonstrates the use of a high power semiconductor laser in a high speed multiple exposure imaging system. This system developed is capable of illuminating an event with up to 10 pulses of light at a maximum repetition rate of 5MHz, with a timing accuracy of ≈5ns. Each semiconductor laser light pulse has a FWHM duration of 50ns, peak power of 30W, and a wavelength of 860nm. Images of individual acoustic transients are displayed on the same CCD camera frame, and it was found that this is best achieved using a dark field imaging technique such as Schlieren imaging.

A study of the response of submerged mast-like structures to shock loads has been carried out. An analytical model has been developed which uses a lumped parameter system whose equations of motion are solved by the use of finite difference time integration techniques. The effects of the stiffness and damping inherent in the supporting restraints are successfully included into the analysis by assuming that they provide linear rotational stiffness and a viscous type of damping. An analysis of hydrodynamic interaction effects has been developed in order to investigate how the frequency response of the structure is affected by submergence. Changes in both the frequency and the damping characteristics are predicted as a function of the depth of submergence. Methods of incorporation of these hydrodynamic forces into the time domain solution of the lumped mass system are investigated and are found to be most accurate for slender structures. In order to develop a fuller understanding of the hydrodynamic response, two experimental investigations have been carried out. The first is laboratory based and uses carefully controlled free vibration tests to excite a number of modal frequencies of a rigidly clamped vertical mast in both submerged and partially submerged conditions. Results show that the frequency changes are closely predicted by the hydrodynamic analysis developed in this thesis and also show that hydrodynamic damping is a linear function of amplitude of deflection for the first natural response mode over the mast's elastic range of response. To investigate the shock response of submerged structures, a second test rig is used to subject a slender mast to a shock load radiating from an underwater explosion. The structural significance of the higher modal frequencies is evaluated through the use of Fourier Analysis and digital filtering techniques. The instrumentation performance including both damped and undamped transducers is assessed in order to make recommendations for future shock trials. The experimental investigation has shown that both the direct pressure loading and the movement of the supports are important factors in the mast response and that their combination produces important high frequency response modes.

>Magister Scientiae - MSc / Due to economic and environmental concerns associated with use of fossil fuels, humanity is seeking alternative fuels. Ethanol is one of the alternative fuels produced commercially. Current ethanol production technologies using first generation ethanol processes is criticised for depleting the food supply and escalating food prices. Biomass is a target feedstock for use in bioethanol production and would resolve the criticism associated with the current bioethanol industry. Bacterial strains such as Geobacillus thermoglucosidasius NCIMB 11955 can be used to produce ethanol from biomass because they assimilate hexose and pentose sugars, a property that is lacking in first generation ethanol producing microbes (Saccharomyces cerevisiae and Zymomonas mobilis) (Riyanti and Rogers, 2009). Due to the low ethanol tolerance (4 % v/v (maximum)) of G. thermoglucosidasius, use of this species for bioethanol production is not economically viable. GroES and GroEL genes are involved in stress tolerance in bacteria: activation of these genes has been observed in stress induced bacteria (Rasouly and Ron, 2009). In this study the ethanol tolerance of G. thermoglucosidasius NCIMB 11955 was characterised by culturing at 45 ºC and 55 ºC in the presence of ethanol. A greater ethanol tolerance was observed at the sub-optimal growth temperature of 45 ºC. Escherichia coli metabolic systems are well understood. Aiming to improve the ethanol tolerance G. thermoglucosidasius NCIMB 11955, the GroES and GroEL genes of the organism were cloned in an expression vector and expressed in E.coli before testing their ability to confer an increased tolerance to ethanol. Proteomic analysis of the recombinant E. coli strain showed that GroES was over-expressed while GroEL was not. After over expression of GroES, the optical density of cultures was periodically measured. Over-expression of the G. thermoglucosidasius NCIMB 11955 GroES gene improved the ethanol tolerance of E. coli Rosetta pLySs growing in 4% (v/v) ethanol.

The co-chaperone murine stress-inducible protein 1 (mSTI1), a heat shock protein 70 (Hsp70)/ heat shock protein 90 (Hsp90) organizing protein (Hop) homologue, mediates the assembly of the Hsp70/Hsp90 chaperone heterocomplex. mSTI1 is phosphorylated in vitro by cell cycle kinases, proximal to a putative nuclear localization signal (NLS), substantiating a predicted CKII-cdc2-NLS (CcN) motif at position 189-239. Stable transfectants of NIH 3T3 fibroblasts that expressed mSTI1-EGFP, NLSmSTI1-EGFP and EGFP, were prepared. Fluorescence microscopy revealed mSTI1 was cytoplasmically localized, and that this localization was not affected by the fusion of mSTI1 with the EGFP moiety. NLSmSTI1-EGFP was targeted to the nucleus compared to EGFP, suggesting that the NLSmSTI1 was a functional NLS. The localization of mSTI1 was determined under normal and heat shock conditions, inhibition of nuclear export (leptomycin B), inhibition of CKII 5,6-dichlorobenzimidazole riboside, DRB), inhibition of cdc2 kinase (olomoucine), and G1/S phase arrest (hydroxyurea). mSTI1-EGFP and mSTI1 were excluded from the nucleus in the majority of resting cells, but accumulated in the nucleus following leptomycin B treatment, implying that mSTI1 possibly undergoes a functional import process, and export via the chromosomal region maintenance 1 (CRM-1)-mediated export pathway. Hydroxyurea and olomoucine (but not DRB or heat shock) treatment increased the proportion of cells in which mSTI1-EGFP exhibited cytoplasmic and nuclear localization. 2D gel electrophoresis detected three endogenous mSTI1 isoforms, which changed following hydroxyurea treatment. Furthermore, point inactivation and mimicking of phosphorylatable residues in mSTI1 altered the translocation of the protein and the isoform composition. Modification of mSTI1 at S189 and T198 decreased the number of isoforms of mSTI1-EGFP, suggesting that the protein is modified at these sites in vivo. The removal of the in vitro cdc2 kinase site at T198 promoted a nuclear localization during G1/S phase arrest. Therefore active cdc2 kinase, but not CKII, may be required for cytoplasmic localization of mSTI1. The CKII site appears to have no regulatory role under heat shock conditions or during the cell cycle. In vitro phosphorylation studies on untagged mSTI1 further supported the prediction that S189 is the only site recognised by CKII. The cdc2 kinase site at T198, however, although the major site, was not the only site phosphorylated in vitro. However, mSTI1 and cdc2 kinase did not interact in a detectable stable complex. Bioinformatic analysis of mSTI1 revealed NLS residues were conserved in STI1 proteins, and the NLS and TPR2A motifs were in close proximity. This may have mechanistic implications for the formation of the Hsp90-mSTI1 heterocomplex. The cytoplasmic or nuclear localization of mSTI1 is predicted to be the result of a dynamic equilibrium between nuclear import and nuclear export, the fulcrum of which may be shifted under different cell cycle conditions. These data provide the first evidence of regulated nuclear import/export of a major Hsp70/Hsp90 co-chaperone, and the regulation of this nuclear import by cell cycle status and cell cycle kinases.

Introduction: Picornaviruses are a family of RNA viruses which are economically and clinically significant. Like many other viruses, picornaviruses utilise host cell machinery to facilitate their replication and assembly, including heat shock proteins (Hsps). The aim of this research was to investigate the role of Hsp40, Hsp70 and Hsp90 during picornavirus infection using the cardiovirus, Theiler’s murine encephalomyelitis virus (TMEV), as a study model. Methodology: Picornavirus VP1 capsid proteins were analysed by multiple sequence alignment and multiple structural comparisons. Protein domain architecture was used to analyse Hsp90 cellular and viral client proteins. Effects of Hsp90 inhibitors, novobiocin and geldanamycin, on TMEV growth in BHK-21 cells was observed over a 48hr period. Localisation of Hsp40, Hsp90 and Hsp70 in TMEV-infected BHK-21 cells was investigated by indirect immunofluorescence and confocal microscopy. Results and Discussion: VP1 proteins of picornaviruses are highly divergent within the family at the amino acid level, which might be linked to the protein’s function in determining virus tropism and antibody neutralisation. An eight-stranded anti-parallel beta-barrel structure was found conserved in the VP1 protein structures which might be linked to the highly conserved picornavirus capsid assembly process. Absence of a common protein domain between Hsp90 viral and cellular client proteins that might be functionally connected to Hsp90, suggests that Hsp90 most likely recognises surface features rather than sequence motifs/patterns. The Hsp90 inhibitors, novobiocin and geldanamycin, had a negative effect on virus growth as virus-induced cytopathic effect was not observed in treated cell after 48hrs. TMEV 2C protein was detected by Western analysis in infected cell lysates treated with geldanamycin but not novobiocin, suggesting novobiocin affects the translation or processing of TMEV 2C. Immunofluorescence analysis of TMEV-infected cells showed a relocalisation of Hsp40 into the nucleus during infection. Overlap of Hsp40 and TMEV P1 was observed in the perinuclear region, suggesting colocalisation between these proteins. Hsp70 converged around the replication complex during infection but did not overlap with TMEV 2C. Hsp90 concentrated in the region of the replication complex where it overlapped with TMEV 2C and this redistribution was found to be dependent on the stage of infection. The overlap between Hsp90 and TMEV 2C signals observed, suggested colocalisation between the two proteins. Conclusion: This study identified Hsp90, Hsp70 and Hsp40 as possible host factors required in TMEV replication.

Photoelectric measurements were made of the shock-excited spectrum of a mixture of helium and argon. The electron density behind the shock wave and the temperatures of the plasma components were deduced from the spectroscopic measurements, assuming thermal equilibrium conditions in the shock plasma. The two temperatures were in fairly good agreement, supporting the equilibrium assumption. On the other hand, the temperature and electron density differ significantly from values expected for a one-dimensional shock wave. / Science, Faculty of / Physics and Astronomy, Department of / Graduate

The plasmas produced in electromagnetic shock tubes have previously been studied in this laboratory and elsewhere. In general the temperatures and electron densities deduced from time-resolved spectra emitted by the plasma do not agree with the values calculated from shock theory. Photographs taken with a Kerr cell shutter revealed that luminous discharge gases with a very irregular front were driven down the tube and that no separate shock front could be seen ahead. The plasma behind the luminous front consisted of a mixture of rest gas and a considerable amount (~50%) of impurity from the driving discharge. In the work reported here, further attempts were made to produce shock heated plasmas. Various electrode configurations were tried but no improvement was observed. Some measure of success was attained with an electrodeless driver on the shock tube. Kerr cell photographs showed that with argon in the tube a shock wave appeared to be formed ahead of the discharge plasma. The shock speed was much slower than the speed of the advancing luminous front in the tubes with electrodes. However, no shock wave could be observed with helium. With argon in the electrodeless tube radiation could be observed from the gas ahead of the shock wave. Time resolved spectroscopic measurements on this radiation allowed rough determination of electron density and of the population of excited states of argon atoms and ions ahead of the shock front. This "preheating" of the gas is presumably due to ultraviolet light emitted from the discharge and the shock plasma. The values of electron density and temperature expected behind the shock front were calculated from shock theory, taking into account the preheating of the gas. The expected values agreed well with the electron density and temperature determined from spectroscopic measurements on the shock plasma. The study of the precursor radiation was continued in a shock tube with electrodes. In this tube the driving discharge was more luminous and the excitation and ionization of helium and argon ahead of the luminous front could be more readily observed than with the electrodeless tube. The number densities of helium atoms in various excited states were determined from the time resolved line intensities before and after the passage of the luminous front. The ratios of atoms in different levels differ from the expected ratios for thermal equilibrium conditions, both ahead of the luminous front and behind it. An estimate was made of the time required for the attainment of equilibrium by electron impact. The calculation indicates that ahead of the luminous front there is not sufficient time to attain equilibrium. On the other hand, for the high electron density found behind the luminous front, the equilibrium distribution is expected to be reached in times shorter than the observation times, in disagreement with the behaviour observed. / Science, Faculty of / Physics and Astronomy, Department of / Graduate