As the demand for global renewable energy grows, so does the demand for more efficient energy conversion machines and better wind resource assessment. The need to convert as much energy as possible with little cost remains the biggest challenge. In the wind energy sector, the quantity of the resource “wind” is not hard to locate, as with current ground and space based remote sensing technologies, and climate reanalysis techniques, the mapping of average wind speeds across the globe is achievable. The difficulty lies in identifying the “quality” of the wind resource. “Quality” is the measure of the time variant properties of the wind, and time scale here does not represent seasonal, monthly, or the daily variability, but rather the changes within hours, minutes, seconds, and sub‐second periods. Wind possesses a highly unpredictable, and non‐universal character, which is referred to as turbulence. These intermittencies in the wind speed create variable mechanical loads on the structure of wind turbines leading to fatigue, and ultimately failure. Identifying site specific qualities of the wind resource is very crucial in the design and selection process of the wind turbine. Physical theories explaining wind turbulence phenomena over flat terrain have been critiqued and tested by observations, and in general, have achieved reasonable success in explaining surface layer wind dynamics that can be applied universally. This universality, and the extrapolation of flat terrain theories to complex terrain applications, breaks down most of the time due to the newly recognized spatial and temporal spectrum of interaction modes, mechanically and thermodynamically, with the surrounding complex terrain. In terrain as found in New Zealand, most of the wind farm development is carried out over complex terrain, with ridge top and mountainous installations. In this study, an experimental campaign was carried out over a coastal ridge top, proposed for wind farming, to investigate mean and turbulent wind flow features significant for wind turbine selection and placement across the ridge. The steep sloped faces of the ridge, high wind speeds and its proximity to the sea made this location ideal for a benchmark investigation site. Ultra‐sonic ii anemometers, a sodar (sound detection and ranging) wind profiler, and high resolution LES (large eddy simulation) numerical modelling were all utilized separately and in an interconnected way to provide a comprehensive analysis of the wind dynamics over the ridge top. The three principal components of the investigation were: the effect of the upstream topography and the thermal circulation associated with the proximity to the sea on the observed and modelled wind shear vertical profile; the role that the near upwind terrain plays in shaping the turbulence energy spectrum and influencing the predicted spectrum, ultimately affecting isotropy in the flow field and turbulence length scales; turbulence advection from far topography, and the role that far upwind terrain plays in altering the wind turbulence in a measurement area or at a single point. Results showed that the thermal wind circulations and upstream steep topography could dictate the wind shear profile, and consequently have a large impact on wind turbine height selection and placement. The sodar proved to be a very useful tool in identifying vertical shear zones associated with effects of steep upstream terrain, vertical mixing of horizontal momentum, and thermal circulation from the local sea breeze. In complex terrain, the added multi‐directional perturbations from the underlying roughness redistribute the statistical variations (measured by variances) in the three spatial dimensions. Isotropy, based on measured variances, was attained for both sites on the ridge. Isotropy also held true for the energy spectrum via Fourier analysis of the high temporal resolution data, but not for both sites. In general, local isotropy can be attained in cases of higher wind speeds and increased terrain relief. Measured spectral ratios did not converge to the limit suggested by the local isotropy hypothesis. These results identify contradictions in assessing the turbulence isotropy in both real space (statistically through variances) and Fourier space (through power spectrum analysis), which suggests caution in deriving or interpreting turbulence information for wind turbine design and selection. iii 2D‐LES experiments showed that turbulent kinetic energy (TKE) can attain long range memory of underlying terrain, which can then react accordingly with upcoming terrain. Under the high wind speed scenarios, which are suitable for wind farming, and over relatively complex terrain, the flow retained some aspects of terrain information at least 30H (H is the terrain height) upstream and downstream of the terrain. In general, as the turbulence field travels over new terrain it tends to increase in intensity downstream of that feature. The newly modified TKE field acquires geometric features from the underlying terrain; mainly these features register as amplifications in the wave structure of the field at wavelengths comparable to the height of the underlying terrain. The 2D‐LES sensitivity experiments identified key areas of high mean wind speed and turbulence in relation to terrain effects, all of which should be taken into consideration when thinking of locating a wind farm in such areas.

Traditionally, agricultural nitrous oxide (N₂O) emission of New Zealand has been measured using chambers or lysimeters, and micrometeorological flux measurement experiments have been very few. Since micrometerological flux measurement systems have the advantage of measuring spatially integrated flux values for longer time periods compared to measurements made using chambers, development and verification of such a system was needed for New Zealand's agro-meteorological conditions. In this study, efficacy of such a combined flux gradient (FG) - eddy covariance (EC) micrometeorological flux measurement system is verified by continuously measuring N₂O fluxes from some control and mitigated agricultural plots of New Zealand. The control fields had natural N₂O emission, whereas, the mitigated plots were treated with chemicals to reduce N₂O emission. In this combined FG-EC method, the turbulent eddy diffusivities were estimated using the Monin-Obukhov (M-O) similarity theory based parameterization (where diffusion velocity `dhp' was used) and a thermal approach (where eddy diffusivity `kht' was used) from the EC measurements. These transfer coefficients (kht and dhp) along with the measured N₂O concentration differences were then fitted to the traditional FG equation to compute final flux values. As the primary objective of this study, measured fluxes from two different seasons and from two approaches were compared for consistency and then verified against published results. Under this wider objective of verification of the FG-EC micrometeorological method of N₂O flux estimation, this research thesis addresses three key issues: (i) assessment of error propagation in the measured flux through the eddy diffusivity - to understand the random error dynamics of the system and to estimate precision of the overall method, (ii) quantification and separation of N₂O source area emission rates from adjacent plots - to identify the contribution of an individual plot to the measured flux when multi-plot fluxes were measured from sources with different biogenic characters, and (iii) quantification of the effect of animal grazing and mitigation on the measured flux and actual emission rate of N₂O - to assess robustness of the FG-EC micrometeorological system. As a fourth objective of this study, (iv) new scaling properties of a turbulence surface layer model of a convective atmosphere is investigated as an alternative to the standard M-O similarity theory, as significant questioning of the M-O theory has been reported in some recent publications. Results from the verification experiment showed that the daily measured flux values obtained from this combined micrometeorlogical system for control plots varied between 0-191.9 and 0-491.8 gN₂O-N.ha⁻¹.day⁻¹ for autumn and spring experiments, respectively, for the parameterization method. Similarly, the daily mean flux values were found to be 10.9 ± 0.98 and 11.7 ± 0.57 gN₂O-N.ha⁻¹.day⁻¹ for the autumn and spring seasons, respectively. All these values were found to be of the same order of previously reported values in the literature and found to verifying that this FG-EC system works well under a range of meteorological conditions within a defined error range. Therefore, when the propagated random error was computed in the final flux value using kht and dhp, the mean relative error in kht was found to be higher than the mean relative error in dhp, irrespective of stability. From a Monte-Carlo type simulation of the random error, it was found that the maximum error can be up to 80% for kht irrespective of stability, and 49% and 35% for dhp respectively for stable (1/L ≥ 0, where L is Obukhov length) and unstable (1/L < 0) atmosphere. Errors in the concentration differences were estimated based on the minimum resolvable estimates from the gas analyzer and the associated random errors were found to be 6% and 8% for unstable and stable conditions. Finally, the total mean random error in the N2Oflux values was found to be approximately of the order of 9% and 12% for the parameterization method for unstable and stable conditions, respectively, and 16.5% for the thermal method, irrespective of stability. Objective (ii) of this research was addressed by developing a `footprint fraction' based inverse footprint method. Results of the footprint analysis method were assessed, first, by comparing footprint fractions obtained from both an analytical footprint model and a `forward' simulation of a backward Lagrangian stochastic (bLs) model; and second, by comparing the source area emission rates of a control plot obtained from the footprint analysis method and from the `backward' simulation of the bLs model. It was observed that the analytical footprint fractions were realistic as they compared well with the values obtain from the bLs model. The actual emission rates were found to be on average 2.1% higher than the measured flux values for the control plots. On average 4.3% of the measured fluxes were found to be contributed by source areas outside of the field domain. Again, the proposed footprint method of emission rate estimation was found to work well under a wider range of atmospheric stability, as the inverse footprint model and bLs model based emission rates were found to correlate well (0.70 and 0.61 for autumn and spring, respectively) with a 99% statistical significance. Similarly when the effect of grazing on the N₂O fluxes was considered, a 90% enhancement in the flux values was observed after grazing, followed by a decreasing trend in fluxes. However, contrary to existing knowledge of mitigation of N₂O flux by an inhibitor, this study found no statistically significant effect of mitigation in the pastoral emission of N₂O. Error accumulation, lesser soil N₂O production potential and/or inefficiency of the FG-EC method was conjectured to be reason/s for such discrepancy and some alternative convective boundary layer turbulence scaling was tested. Separate field measurement data, including the vertical profile measurements of the convective boundary layer and sonic anemometer measurements within the surface layer were used for this purpose. The spectral analysis of the vertical wind component, temperature and heat flux revealed that this new model of the convective boundary layer, which explains atmospheric boundary layer turbulence in terms of some nonlocal parameters, is more suitable than the traditional Monin-Obukhov similarity theory based model of atmospheric turbulence where the atmospheric flow properties are local. Therefore, it can be concluded that this new model of turbulence might provide the framework for a newer model of flux estimation in future. Overall, the FG-EC model of N₂O flux estimation method seems to work well within a certain error range. However, more field applications of this FG-EC method are needed for different agro-meteorological conditions of New Zealand before this method is accepted as a standard method of flux estimation, particularly, inefficiency in detecting the effect of mitigation should be tested. Development of an alternative flux gradient model which includes nonlocal atmospheric surface parameters might also be considered as a future research objective.

Wireless Surface Acoustic Wave (SAW) devices were fabricated and tested using planar Lithium Niobate (LiNbO₃) as substrate. The working frequencies were in the 180 MHz and 360 MHz range. Using a network analyser, the devices were interrogated with a wireless range of more than 2 metres. Trials with Electron Beam Lithography (EBL) to fabricate SAW devices working in the 2450 MHz with a calculated feature size of 350 nm are discussed. Charging problems became evident as LiNbO₃ is a strong piezoelectric and pyroelectric material. Various attempts were undertaken to neutralise the charging problems. Further investigation revealed that sputtered Zinc Oxide (ZnO) is a suitable material for attaching SAW devices on irregularly shaped material. DC sputtering was used and several parameters have been optimised to achieve the desired piezoelectric effect. ZnO was sputtered using a magnetron sputtering system with a 75 mm Zn target and a DC sputter power of 250 Watts. Several trials were performed and an optimised material has been prepared under the following conditions: 9 sccm of Oxygen and 6 sccm of Argon were introduced during the process which resulted in a process pressure of 1.2x10⁻² mbar. The coatings have been characterised using Rutherford Backscattering, X-ray diffraction, SEM imaging, and Atomic force microscopy. SAW devices were fabricated and tested on 600 nm thick sputtered ZnO on a Si substrate with a working frequency of 430 MHz. The phase velocity has been calculated as 4300m/s. Non-planar samples have been coated with 500 nm of sputtered ZnO and SAW structures have been fabricated on using EBL. The design frequency is 2450 MHz, with a calculated feature size of 1 µm. The surface roughness however prevented a successful lift-off. AFM imaging confirmed a surface roughness in the order of 20 nm. Ways to improve manufacturability on these samples have been identified.

A variety of synthetic bone graft materials such as the polymer poly(lactic-co-glycolic acid) (PLGA) have been investigated as alternatives to current tissue based bone graft materials. In this study, efforts have been made to improve the tissue-PLGA interface by immobilizing bone morphogenetic protein-2 (BMP-2) in an oriented manner on scaffolds using covalently immobilized heparin. The results demonstrated a four-fold increase in covalently immobilized heparin compared to non-specific heparin attachment. Furthermore, the scaffolds with covalently attached heparin retained approximately three-fold more BMP-2 than did either scaffolds with no heparin attached or scaffolds with non-specific heparin attachment. The activity of scaffolds with BMP-2 immobilized in various manners was examined using an alkaline phosphatase assay on C3H10T1/2-seeded scaffolds. These results indicated approximately twice the amount of activity with scaffolds that had BMP-2 immobilized with covalently attached heparin than on scaffolds with adsorption of BMP-2 and a three-fold increase in activity when compared to scaffolds that had non-specific heparin attachment as the mechanism for BMP-2 immobilization. These results demonstrated that PLGA with covalently linked heparin has potential to immobilize BMP-2 in a specific orientation that is favorable for cell-receptor binding, leading to the more efficient use of the bone-growth factor.

Diamond is a unique material with excellent properties. As a result of the development within the area of CVD synthesis, doping and surface functionalization, diamond has become a strong candidate for use in electrochemical, electronic and biomedical applications. In this thesis, theoretical calculations have been used with the purpose to investigate various properties of the diamond surfaces. The effect of doping elements (N and B) on the stability of different surface terminations with X (where X = H, OH, Oontop or Obridge) has been investigated for a diamond (100) surface. As a result, the adsorption energy for all termination types was shown to decrease from the situation with a non-doped diamond surface, to the scenario with a N- (or B-doped) diamond thin film.. This result was found to correlate well with the changes of the calculated Csurface-X bond lengths. Furthermore, the spin density has been calculated and used to show the local distribution of the unpaired electron, which is the consequence of the introduction of dopants into the diamond slab. As a result, the spin density was found to be localized in the vicinity to the dopants for H- (or OH-) terminated diamond (100) surfaces. On the other hand, a delocalised spin density over the Oadsorbate and Csurface layer for Oontop- and Obridge-terminated surfaces, has also been observed. Moreover, the results of the pDOS calculations indicate the electron donating ability of N, and the hole donating ability of B. The Fermi level was shifted towards the lower conduction band edge for N-doped diamond, and towards the upper edge of the valence band edge for B-doped diamond. Hence, N-doped diamond will render n-type conductivity, and B-doped diamond will show p-type conductivity. In addition, an interesting observation was made for Oontop –terminated diamond surfaces. Localized electron conductivity, involving only this type of termination situation,, was also observed for N- (or B-) doped and completely  Oontop-terminated diamond surfaces. With the purpose of applying diamond substrates in the formation of epitaxial graphene, the annealing process of an ideal diamond (111) surface has also been simulated in the present work. It was thereby shown that high temperatures (over 2000 K) will be required for the epitaxial formation of graphene ontop of the diamond (111) surface. However, in the presence of hydrogen radicals (by saturating the radical sites in the system), the required temperature was observed to decrease to 1000 K. In addition to these MD simulations, by using an interlayer iron ontop of the diamond (111) surface, the adhesion energies between the graphene and the Fe//diamond slab, as well as the adhesion energy between the graphene//Fe layer and the diamond (111) surface, have been calculated. Thereby, the interaction between the graphene and Fe layer was obtained to be very weak, and of an electrostatic type. On the other hand, the interaction between the Fe interlayer and the diamond substrate was calculated as a moderately strong covalent bond. Moreover, the changes in these interactions, correlating to the changes in the pDOS spectra of graphene, Fe and diamond, gave a tendency of one-dimensional quantum size effect, depending on the thickness of Fe interlayer.

No description available.

Surface-induced osseointegration and antibacterial capability are very important criteria for the clinical success of titanium implants. To enhance these two criteria, an architectural hybrid system is constructed onto Ti-6Al-4V with a rough surface. First, thermal oxidation (TO), treatment with hydrogen peroxide (H2O2) and a mix of TO and H2O2 (Mixed) are used to modify the surface topography and chemistry of Ti-6Al-4V disks. Surface characterizations by the use of microscopes and spectroscopes indicate that TO can induce more favorable topography, roughness, wettability and hydroxyl group concentration on Ti-6Al-4V surfaces. Therefore, an alginate/chitosan LBL film that incorporates antibacterial nano-silver is bridged onto thermally oxidized Ti-6Al-4V alloy by mussel-inspired dopamine. The microscopies and spectrometers confirm that the hybrid system is successfully fabricated onto the Ti-6Al-4V surface while the sub-micron topography induced by TO is maintained. Bone marrow stem cell (BMSC) adhesion, proliferation and differentiation are up-regulated by the synergy of sub-micron surface produced by TO and alginate/chitosan LBL film. The incorporation of nano-silver into the hybrid system is demonstrated to inhibit the growth of Escherichia coli and Staphylococcus aureus, but not jeopardize the enhanced BMSC activities. Taken together, this thesis presents a promising strategy to fabricate novel Ti-6Al-4V implants with enhanced osseointegration and antibacterial capability.

The overall objective of this PhD thesis research is to characterize, understand and ultimately control phase-separated structures in mixed films consisting of a perfluorinated fatty acid and a photopolymerizable surfactant. In these systems, film morphology, mechanical properties and spectroscopic properties are inter-related and this thesis explores these relationships. In this context the interaction between perfluorotetradecanoic acid (C13F27COOH, referred to as PF in this dissertation), and 10,12-pentacosadynoic acid (CH3(CH2)11−C≡C−C≡C−(CH2)8COOH, referred to as PCDA in this dissertation) has been studied in monolayers using a combination of surface and spectroscopic characterization techniques. To investigate the inter-relationship of the properties described above, film behavior under a variety of conditions, including behavior at different interfaces (solid-liquid, air-liquid), different film compositions and under different conditions of photoillumination and mechanical stress were explored. Thermodynamic and morphological studies of mixed monolayer surfactant films of PF and the photo-polymerizable diacetylene molecule, PCDA, were carried out. The films were prepared at the air-water interface and transferred onto solid supports such as a glass slides via Langmuir-Blodgett (LB) deposition technique. The presence of the perfluoroacid helped to stabilize the diacetylene surfactant monolayer in comparison with the diacetylene alone, allowing film transfer onto solid substrates without needing to add cations to the sub-phase or photo-polymerize the components prior to deposition. Addition of the perfluorocarbon to PCDA resulted in films with the photopolymer strands oriented perpendicular to the direction of the film compression in a Langmuir trough. This is in contrast with film structures formed from pure PCDA. Formation of these features could be explained by a two-step process that happened sequentially: first, the compression of monolayer with trough barriers while trying to maintain the surface pressure constant induces stress on the film surface; second, additional film buckling which was enhanced by the strong cohesion between PF and PCDA. Film compression data, supported by in situ fluorescence spectrophotometry, Brewster angle microscope imaging and atomic force microscope images of deposited films, supported this mechanism. Factors that controlled the orientation of the photopolymer fibers were also investigated. Fibers were found to consist of multiple strands, with each strand having a different orientation. Our investigation also revealed there was a preferred orientation for fibers in the film as a whole. The angle of approximately 60o to the direction of film compression during deposition from a Langmuir trough has been calculated with the help of dual-view, polarized fluorescence microscopy. This orientation was attributed to the mechanical stress exerted by the trough compression barriers coupled with rotation of the polymer fibers during film draining. The combination of Atomic Force Microscope (AFM) and fluorescence microscopy (FM) provided a thorough and comprehensive mapping of fundamental properties of mixed monolayer system, and enabled a quantitative determination of the degree of selectivity of the polymerization process.

Label free detection techniques such as surface plasmon resonance, carbon nanotubes, nanowires, and interferometry have been progressing rapidly for biosensing applications. Surface plasmon resonance is considered one of the most promising label free optical techniques. The use of nanohole arrays in a metal film allows for extraordinary transmission and has been motivated by their application as biosensors. Nanohole arrays present several advantages like smaller foot print, dense integration, lower limits of detection, and collinear optical detection. This thesis presents the design parameters for the optimization of sensitivity and resolution of nanohole arrays for refractive index sensing. A systematic study is provided of the influence of the nanohole array periodicity, diameter, and gold thickness. Focused ion beam was used to fabricate the nanohole arrays. A microfluidic device with a set of embedded nanohole arrays was developed and used to measure the sensing characteristics. The results are encouraging for potential future biosensing tests. / Graduate

The statistical prediction of local sea surface winds at a number of locations over the global ocean (Northeast Pacific, Northwest Atlantic and Pacific, tropical Pacific and Atlantic) is investigated using a surface wind statistical downscaling model based on multiple linear regression. The predictands (mean and standard deviation of both vector wind components and wind speed) calculated from ocean buoy observations on daily, weekly and monthly temporal scales are regressed on upper level predictor fields (derived from zonal wind, meridional wind, wind speed, and air temperature) from reanalysis products. The predictor fields are subject to a combined Empirical Orthogonal Function (EOF) analysis before entering the regression model. It is found that in general the mean vector wind components are more predictable than mean wind speed in the North Pacific and Atlantic, while in the tropical Pacific and Atlantic the difference in predictive skill between mean vector wind components and wind speed is not substantial. The predictability of wind speed relative to vector wind components is interpreted by an idealized Gaussian model of wind speed probability density function, which indicates that the wind speed is more sensitive to the standard deviations (which generally are not well predicted) than to the means of vector wind component in the midlatitude region and vice versa in the tropical region. This sensitivity of wind speed statistics to those of vector wind components can be characterized by a simple scalar quantity theta=arctan(mu/sigma) (in which mu is the magnitude of average vector wind and sigma is the isotropic standard deviation of the vector winds). The quantity theta is found to be dependent on season, geographic location and averaging timescale of wind statistics. While the idealized probability model does a good job of characterizing month-to-month variations in the mean wind speed based on those of the vector wind statistics, month-to-month variations in the standard deviation of speed are not well modelled. A series of Monte Carlo experiments demonstrates that the inconsistency in the characterization of wind speed standard deviation is the result of differences of sampling variability between the vector wind and wind speed statistics. / Graduate