Les recherches effectuées au cours de cette thèse s'inscrivent dans le cadre du projet FUI Parex-IT (PARement EXtérieur pour l'Isolation Thermique). Le but de ce projet est de développer un enduit d'isolation thermique par l'extérieur à fort pouvoir isolant pour la réhabilitation. Plus particulièrement, ces travaux concernent la caractérisation mécanique de l'accroche de cet enduit avec son support structurel. Celui-ci est appliqué par projection directe d'une couche de 4 cm d'épaisseur. En considérant un mur enduit, soumis à des sollicitations horizontales telles que le vent ou le séisme, ce type de mise en place implique une transmission d'efforts de la structure vers l'enduit isolant par l'interface entre les deux matériaux. La résistance de cette interface est donc considérée ici comme la zone critique pour une bonne pérennité du complexe. Parallèlement, une étude équivalente a été menée sur un enduit de finition classique, à titre comparatif. Deux échelles ont été prises en compte pour l'étude du comportement de cette interface. Une première à l'échelle locale dont l'objectif est de déterminer un critère de rupture de cette interface. Celle-ci a donc été soumise, expérimentalement, à différents états de contraintes entraînant sa rupture. Cependant, les différences importantes de propriétés mécaniques entre cet enduit et son support sont susceptibles, d'après la bibliographie, de provoquer des concentrations de contraintes du point de vue expérimental. Une analyse numérique a donc été réalisée afin de quantifier les écarts obtenus entre les critères expérimentaux et numériques, pour les deux matériaux testés. A l'échelle d'un mur, l'analyse s'est portée sur une maçonnerie soumise à des sollicitations de cisaillement dans son plan. L'objectif étant tout d'abord d'étudier le comportement d'un mur enduit, et de le comparer à une maçonnerie nue, de référence. Dans un deuxième temps, la pertinence des critères des interfaces support-enduits obtenus à l'échelle locale a été étudiée d'après une étude numérique. Les analyses par corrélation d'images numériques ont, de manière générale, permis de comparer les comportements expérimentaux et numériques pour faciliter la validation des modèles proposés. / This thesis takes part of the FUI project PAREX-IT (PARement EXtérieur pour l'Isolation Thermique). Its main purpose is to develop an efficient insulating coating from the outside, for refurbishment. More particularly, this work investigates the mechanical bond behaviour between this solution and its structural substrate. This thermal insulation is 4 cm thick and pneumatically placed. Considering a coated wall, solicited by wind or seismic loads, this configuration implies a load transfer from the structure to the insulating coating, through the interface between those two materials. The interfacial strength is though considered here as the critical area for a good sustainability of this complex. Meanwhile an equivalent study has been made on a more classical coating, by comparison. Two different scales have been studied. First, a local scale study to determine a failure criterion of this interface, which have been solicited in different states of stresses provoking its failure. However, the important differences of mechanical properties between the tested coatings and the mechanical support can, according to bibliography, provoke stress concentration. A numerical analysis has been conducted to quantify their influence on the parameters of the experimental criterion, for the two tested coatings. At a wall scale, a masonry submitted to shear loads was studied. The objective was to compare the behaviour between a coated wall and an uncoated one, considered as reference. In a second time, a numerical study was performed to confirm the criteria obtained at the local scale. Image correlation has been used to compare the experimental and numerical behaviours to facilitate modelling validation.

In order to reduce postharvest losses of exported Pome fruit and increase export revenue, export quality pears require a reduction in stem-end shrivelling and an extension in shelf-life, regardless of the presence or absence of refrigerated storage. A kafirin coating may fulfil these requirements during export and at the export destination, in retail and at fresh fruit markets. A two-phased approach was followed. During Phase 1, the physiological and biochemical behaviour of ’Packham’s Triumph’ pears were studied under ideal refrigerated (-0.5°C), temperature-abused (10°C) and typical ripening (20°C) conditions. These storage conditions were selected to simulate potential conditions during the export process. Phase 2 involved the development and application of a kafirin-based coating to increase the shelf-life of pears. In Phase 1, two experiments were conducted concurrently on freshly harvested, uncoated pears. In Experiment 1, pears were stored at –0.5, 10 and 20°C and 95 to 98% RH for 42, 42 and 21 d respectively. An increase in storage temperature increased the metabolic activity of the pears and the rate of quality deterioration. Very few quality changes occurred in pears during storage at -0.5°C. Pears stored at 20°C ripened and became senescent in approximately half the time taken by pears at 10°C. However, fully ripe ‘Packham’s’ pears from 10 and 20°C exhibited similar final colour and firmness values. Stem-end shrivelling was exacerbated by storage at 20°C after only 4 days but not observed during storage at -0.5 or 10°C. In Experiment 2, ‘Packham’s Triumph’ pears were stored at -0.5 and 10°C (95 to 98% RH) for 42 and 35 days, respectively before being ripened at 20°C for 7 days. Storage of pears at 10°C prior to ripening accelerated softening and yellowing in the pears, when compared to pears from -0.5°C storage. Storage duration prior to ripening at 20°C also resulted in pears of increasing softness and yellowness by the end of 7 days at 20°C. The effect of storage duration at -0.5°C was less severe on the ripening rate and intensity of softening and yellowing than storage at 10°C. Thus, storage at -0.5°C extended pear shelf-life and resulted in pears of better quality after ripening than storage of pears at 10°C. In Phase 2, pears from Controlled Atmosphere (CA) storage were coated with a kafirin-based coating and stored at 20°C (35 to 45% RH) for 24 days. The ripening rate and the physiological behaviour and physico-chemical changes of pears used in Phase 2 were probably accelerated by 18 weeks under CA conditions and one week under RA conditions prior to the start of the shelf-life study. The kafirin coating did not retard ripening, which was probably already induced during storage before coating, but senescence in the coated pears was delayed by approximately 6 days. The rate of respiration, ethylene production, flesh softening and especially yellowing, was delayed by the coating. Coated and uncoated pears exhibited no growth of coliforms or lactic acid bacteria. Overall, coated pears had lower levels of aerobic mesophiles and yeast and mould growth than uncoated pears. Unfortunately, pear surface-shrivelling was intensified by the coating, probably due to the dehydrating action of the ethanol in the coating solution during dipping. However, the kafirin coating was able to extend pear shelf-life by delaying senescence and microbiological growth. The coating formulation may require a higher concentration of kafirin to increase its hydrophobicity and reduce pear shrivelling. The kafirin coating has possible potential to markedly extend the quality and shelf-life of ‘Packham’s Triumph’ pears, provided that the pears are coated after minimal RA storage when pears are in the pre-climacteric phase. Copyright 2007, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. Please cite as follows: Buchner, S 2007, Coating of pears (Var. ‘Packhams Triumph’) with kafirin protein and its effect on postharvest physiology and shelf-life, MSc(Agric) dissertation, University of Pretoria, Pretoria, viewed yymmdd < http://upetd.up.ac.za/thesis/available/etd-02242012-113144/ > E549/gm / Dissertation (MSc(Agric))--University of Pretoria, 2012. / Food Science / unrestricted

Bond coats used in thermal barrier coatings (TBCs) for gas-turbine engine applications are studied in this thesis, with a focus on oxidation behaviour, surface rumpling and stress evolution. Bond coats made of γ/γ’ Ni-Al-Pt alloys have been widely used in TBCs and it has been found that addition of platinum greatly improves the oxidation resistance of the coatings. The mechanisms behind this benefit, however, are not well understood. For this reason, the oxidation behaviour of four γ/γ’ Ni-20Al-xPt (x= 0, 5, 10 and 15 at. %) alloys at 1150 °C is studied and compared in terms of oxide spallation, oxide microstructure and growth, residual stress in the oxide scale and oxide/alloy interface morphology. The progressive increase of platinum addition into the alloys results in (1) greater resistance to oxide spallation, (2) reduction in oxidation of nickel, (3) lower stresses in the α-Al2O3 scale and (4) more planar oxide/alloy interfaces. It is found that the selective oxidation of aluminium promoted by platinum plays a central role in the evolution of the oxidation behaviour of the alloys. Surface rumpling of a NiCoCrAlY bond coat deposited on a Ni-base superalloy during cyclic oxidation at 1150 °C is studied. The extent of rumpling is found to depend on thermal history, coating thickness and exposure atmosphere. While the coating surface progressively roughens with cyclic oxidation, bulk NiCoCrAlY alloys with the same nominal composition show a much less tendency to rumple under the same thermal cycling condition. The coatings, especially the thin ones, experience substantial degradation (e.g. β to γ phase transformation and exhaustion of yttrium) induced by oxidation and coating/substrate interdiffusion during thermal exposure. The observations together suggest that rumpling is driven by the lateral growth of the thermally grown oxide (TGO) and the coating deforms in compliance with the TGO. While the dependence of rumpling development on experimental conditions is generally in agreement with the prediction of the existing model, it is suggested that the degradation of the NiCoCrAlY coating and its dependence on coating thickness need to be taken into consideration when predicting the rumpling development of NiCoCrAlY coatings. The residual stresses in a NiCoCrAlY bond coat deposited on a Ni-base superalloy are studied by X-ray diffraction using the sin2Ψ technique. The stresses at room temperature are found to be tensile; they first increase and then decrease with oxidation time. The stress develops and builds up upon cooling, predominantly within the temperature range from 1150 °C to 600 °C. Due to the limited penetration depth into the bond coat, the X-ray only probes the stress in a thin surface layer consisting of a single γ-phase formed through aluminium depletion during oxidation. Above 600 °C, the volume fraction of the β-phase in the bond coat increases with decreasing temperature. The mechanisms of stress generation in the coating are examined and discussed based on experiments designed to isolate the contribution of possible stress generation factors. It is found that the measured bond coat stresses are mainly induced by the volume change of the bond coat associated with the precipitation of the β-phase upon cooling.

Modern automobiles are painted with basecoat, technology which is either metallic, solid colour or pearlescent. This requires protection from chemicals, scratching, weathering and UV light by applying a protective top coat (clearcoat) over the basecoat. For the clearcoat to cure into a hard protective shell it undergoes an irreversible crosslinking process. This usually takes place over the first four to five hours, depending on the formulation and weather conditions. The speed of crosslinking can be enhanced by temperature. Pot life is important as it can affect the overall quality of the painted surface. If crosslinking occurs too quickly, before the clearcoat is applied onto the surface, the clearcoat cannot be used to produce a good quality finish. The “expired” mixture is thus discarded. If used, the quality of the finished product cannot be guaranteed to last, and the paintwork may have to be redone. This often means removing the underlying paint and primer as well with the clearcoat film. Besides the time lost, the discarded clearcoat mixture often lands in the landfill, polluting ground water and the environment. It is thus important from the point of view of both the environment preservation and waste management, that as much clearcoat as possible is used without being wasted. It was proven in an earlier study (BSc Hon Formulation Science Treatise, 2011) that adding eugenol to a clearcoat mixture after crosslinking had started could reduce its viscosity, which is an indicator of crosslinking progress. Crosslinking subsequently resumed at a lower rate than in traditional blends. If stored away from oxygen and high temperatures, this blend could maintain optimum viscosity indefinitely. In this follow up study an optimum formulation was developed using D - optimal experimental design. It sought to extend the pot life to avoid waste to spray painters while saving the environment from pollution. The formulation that gave the desired viscosity after five hours of pot life was adopted. It was hypothesised that the optimum formulated clearcoat mixture would have a longer pot life than its traditional counterparts, and that it would perform just as well as the traditional clearcoat mixtures. To study the rate of crosslinking (disappearance of functional groups and appearance of the urethane bond), FTIR spectrometry was performed on a mixture produced from the optimized formula in comparison to that of a traditional mixture (the control). The rate of disappearance of functional groups was found to be lower in the eugenol mixture than in the control mixture. After six hours, eugenol was added into the control mixture, and this seemed to reduce the viscosity with the re-emergence of functional groups in the mixture. After 24 hours of crosslinking, an FTIR scan was done on the solid sample and this revealed that the eugenol mixture had crosslinked fully, with no detectable functional groups in the sample.

Yttria stabilized zirconia (YSZ) coatings were produced from a YSZ suspension in acetylacetone (ACAC) using electrophoretic deposition (EPD) and then consolidated via the natural drying and isothermal sintering with the constraint of the metal substrates. Before EPD, the operational pH of the suspension was adjusted by addition of acetic acid or organic bases. The effect of suspension pH on the deposition of EPD coatings was studied with respect to the suspension stability, coating density and microstructure both for a mono-sized system and micro-nano binary systems. The constrained drying process of the deposits was examined via the measurement of the critical cracking thickness (CCT). The sinterability of coatings was evaluated by micro-hardness and microstructure. For a mono-sized (0.26μm) suspension, results showed that the zeta potential had a high positive value on both sides of the isoelectric point (IEP). This probably resulted from the adsorption of base molecules triethanolamine (TEA), detected by fourier transform infrared spectroscopy. Three alkalis with different molecular structure were compared and the effect of their molecule length on the interparticle repulsion was discussed. Accordingly, the double layer thickness of the particles can be estimated. Based on this, particle interactions were estimated for different pH suspensions. The reduced particle coagulation increased the packing density of the EPD coatings from 38 % at pH 7.4 to 53 % at pH 8.4. Therefore, subsequent sintering of coatings was promoted. After sintering at 1200 °C, coatings made in pH 8.4 suspensions obtained a much higher hardness and had fewer big pores than coatings fabricated in pH 7.4 suspensions. The CCT of the latter is slightly higher than the former which might be ascribed to its particle network structure. In a binary suspension composed of the coarse (1μm) and fine (with average size of 100 nm or 10 nm, content varied in 0-30 wt. % to the powder mixture) YSZ powders, interactions between different species can be tuned by the zeta potential of individual component. Binary particles can be well dispersed at pH 4 when both of the coarse and fine powders reached their highest zeta potentials. Heterocoagulation occurred between them to form a haloing structure with fine powders covered on the coarse particle surfaces when they exhibited zeta potentials of the opposite sign at pH 8.6. Particle interactions were estimated and the microstructures of the binary coatings were examined to discuss how the different fine particle sizes influenced the particle packing after EPD. At pH 4, there existed a “stability window” for the 10 nm fines at 10 wt. % whereas no noticeable the border of the window can be observed for 100 nm fines within the measuring range. 10 nm and 100 nm fine powders gave similar overall densities of binary EPD coatings which were independent of the fine powder content. For heterocoagulation coatings made at pH 8.6, although the adsorption of fine particles reduce the agglomeration of coarse powder, the low zeta potential of the halos led to a loose structure of the “skeleton” ( the packing of the coarse powder) in the final binary coatings. 10 nm fine powders was observed to give a higher CCT and denser particle packing than 100 nm fine powders especially in a pre-saturated heterocoagulated binary coatings at 20 wt. % fine powder content. In order to further improve the sintering of the EPD coatings at low temperature sintering, a layer of CuO was applied on the coarse powder surface. With the addition of 30 wt. % fine powders, the hardness of EPD coatings after sintering 2 hours at 1150°C increased from 6 to 61 Vickers. With the presence of CuO, the hardness values were enhanced by 2.5-4.25 times. The density measurements indicated that the CuO layer not only served as a sintering aid, the CuO layer also helped with the binary particle packing particularly in the heterocoagulation condition because of the stronger particle interactions between the fine powders and CuO modified coarse powders. It seems that CuO had no significant impact on the cracking resistance of the binary coatings during drying, however t-m phase transformation was observed during sintering possibly due to the liquid phase induce by CuO.

Recent commercial developments have created a need for alternative materials and methods for imparting oil/grease resistance to paper and/or paperboard used in packaging. The performance of a novel grease resistant functional coating comprised of polyvinyl alcohol (PVA), sodium tetraborate pentahydrate (borate) and acetonedicarboxylic acid (ACDA) and the application of said coating by means of flexographic press is presented herein. Application criteria is developed, testing procedures described, and performance assessment of the developed coating materials are made. SEM images along with contact angle data suggest that coating performance is probably attributable to decreased mean pore size in conjunction with a slightly increased surface contact angle facilitated by crosslinking of PVA molecules by both borate ions and ACDA.

As an alternative to the present toxic chromate-based coating system now in use, the Mg-rich primer technology has been designed to protect Al alloys (in particular Al 2024 T3) and developed in analogy to Zn-rich primers for steel substrate. As an expansion of this concept, metal-rich primer systems based on Mg alloy particles as pigments were studied. Five different Mg alloy pigments, AM60, AZ91B, LNR91, AM503 and AZG, were characterized by using the same epoxy-polyamide polymer as binder, a same dispersion additive and the same solvent. Different Mg alloy-rich primers were formulated by varying the Mg alloy particles and their pigment volume concentrations (PVC). The electrochemical performance of each Mg alloy-rich primer after the cyclic exposure in Prohesion chamber was investigated by electrochemical impedance Spectroscopy (EIS). The results indicated that all the Mg alloy-rich primers could provide cathodic protection for AA 2024 T3 substrates. However, the Mg alloys as pigments in metal-rich primers seemed to exhibit the different anti-corrosion protection performances, such as the barrier properties, due to the different properties of these pigments. In these investigations, multiple samples of each system were studied and statistical methods were used in analyzing the EIS data. From these results, the recommendation for improved EIS data analysis was made. CPVC studies were carried out on the Mg alloy-rich primers by using three Mg alloy pigments, AM60, AZ91B and LNR91. A modified model for predicting CPVC is proposed, and the results showed much better agreement between the CPVC values obtained from the experimental and mathematical methods. Using the data from the AM60 alloy pigment system, an estimate of experimental coarseness was done on a coating system, the first time such an estimate has been performed. By combining various surface analysis techniques, such as scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS) and confocal Raman microscopy, the oxidation products formed after exposure were identified. It was found that variation of Al content in Mg alloy could significantly affect the pH of the microenvironment in the primer films and result in the formation of various oxidation products. / Air Force Office of Scientific Research (Grant No. 49620-02-1-0398)

As an alternative to the present toxic chromate-based coating system now in use, the Mg-rich primer technology has been designed to protect A1 alloys (in particular A1 2024 T3) and developed in analogy to Zn-rich primers for steel substrate. As an expansion of this concept, metal-rich primer systems based on Mg alloy particles as pigments were studied. Five different Mg alloy pigments. AM60, A719B, LNR91, AM503 and AZG, were characterized by using the same epoxy-polyamide polymer as binder, a same dispersion additive and the same solvent. Different Mg alloy-rich primers were formulated by varying the Mg alloy particles and their pigment volume concentrations (PVC). The electrochemical performance of each Mg alloy-rich primer alter the cyclic exposure in Prohesion chamber was investigated by electrochemical impedance Spectroscopy (EIS). The results indicated that all the Mg alloy-rich primers could provide cathodic protection for AA 2024 T3 substrates. However, the Mg alloys as pigments in metal-rich primers seemed to exhibit the different anti-corrosion protection performances, such as the barrier properties, due to the different properties of these pigments. In these investigations, multiple samples of each system were studied and statistical methods were used in analyzing the EIS data. From these results. the recommendation for improved EIS data analysis was made. CPVC studies were carried out on the Mg alloy-rich primers by using three Mg alloy pigments, AM60, A2918 and LNR91. A modified model for predicting CPVC is proposed, and the results showed much better agreement between the CPVC values obtained from the experimental and mathematical methods. Using the data from the AM60 alloy pigment system, an estimate of experimental coarseness was done on a coating system, the first time such an estimate has been performed. By combining various surface analysis techniques, such as scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS) and confocal Raman microscopy, the oxidation products formed alter exposure were identified. It was found that variation of A1 content in Mg alloy could significantly affect the pH of the microenvironment in the primer films and result in the formation of various oxidation products. / Air Force Office of Scientific Research (AFOSR) (Grant No. 49620-02-1-0398)

Coating of organophobic silver nano-particles with thiols was applied to facilitate their dispersing in aprotic organic solvents and solutions of polyacrylic acid and polylactide. Impact of chemical nature of thiols and dipole moment of the solvents on dispersibility of the modified silver particles was studied by UV-Vis spectroscopy (surface plasmon resonance) and scanning electron microscopy. It was shown that thiols' grafting to the surface of metal nanoparticles is an effective route to improve and stabilize dispersions in organic systems.

Scanning electrochemical microscopy is a method that incorporates an ultramicroelectrode capable of facilitating electrochemical reactions paired with an XYZ positioning system capable of micron-level movements. This study investigates the corrosion behavior of structural transition joint clad material that contains steel, pure aluminum, and an aluminum alloy blast welded into a single joint. This study will characterize the corrosion response of the structural transition joint and identify the galvanic activity measured between its layers. High-temperature coatings in this study are designed to be used effectively up to 1400?F. In this study, we characterized several commercial high-temperature coatings exposed to different levels of heat. General trends of decreasing barrier performance were observed with the exception when these coatings are exposed to their rated temperature limit of 1400?F, at which the barrier increased slightly, indicated by their low-frequency impedance modulus. The cause is a combination of sintering and oxide formation.