To manufacture a high precision component, the aircraft industry in particular, employs computer numerical control machining to produce the aircraft components. However, it has often been found that problems of distortion occur when thin and slender components are manufactured this way. It is believed that the distortion arises due to the re-distribution of the residual stress field that was present in the original component in addition to the effects of cutting forces and temperatures that were induced during the machining process. At present, the process to produce a component with acceptable tolerance limits is one of trial and error. This often consumes a lot of time and produces considerable material wastage. The finite element method has been used to develop a methodology to model the actual machining process that produced the aircraft component. The finite element model that has been developed so far was able to demonstrate the effects of redistribution of the residual stresses on the distortion of the final machined component. It was found that the final component distorted differently when the materials were simply removed from a solid billet with different material removal sequences. A preliminary investigation on the effects of cutting temperature and forces has also been made. The investigation showed that such effects could not be totally ignored as they significantly affect the distortion of the final component. In addition to modelling the machining process, a methodology has also been developed to introduce and simulate the clamps and machine worktable that are part of the machining process. Although there is a considerable amount of work relating to modelling machining processes, there has not been any previous attempt to address the problem as a whole as described in this thesis. It is hope that the work undertaken here would be able to provide a stepping-stone to such attempts in future. However, further experimental data would be required to complement a more refined model in the attempt to accurately predict the distortion in machined components. It is believed that it would be possible to then extend the developed methodology to minimise the distortion by changing the machining sequence and the shape of the material to be removed. This would reduce the timescales involved to correct the problems of distortion and eliminate material wastage.

No description available.

No description available.

Thesis (Ph. D.)--University of Illinois, 1922. / Vita. "Reprinted from the University of Illinois studies in language and literature, volume 13, number 3-4." "This ... is a linguistic and historical study solely, intended to supplement from the side of written or inscriptional records the works on archaeology."--P. 9. Bibliography (p. 201-202).

Submitted in fulfillment of the requirements of the degree of Master of Engineering: Chemical Engineering, Durban University of Technology, Durban, South Africa, 2015. / The implementation of new water regulations from the local government has been a motivation for most industries to treat the effluent before disposal or reuse within the plant, in order to save costs and avoid sanctions. Tongaat-Huletts sugar refinery has therefore invested in this collaborative research with the Durban University of Technology in order to investigate new technologies for wastewater treatment and water recovery using an organic coagulant called chitosan. Chitosan is a natural non-toxic polymer extracted from the exoskeleton of crustaceans. Chitosan has gained extensive attention as a coagulant in the treatment of wastewaters from various industries. However, no attention has been given to the coagulation of effluents from the sugar industry using this polymer. In this work, chitosan coagulant (CCo) was prepared by dissolution of known amounts of chitosan powder in aqueous acid at 50℃. The solution was diluted to desired concentrations using distilled water at room temperature. The removal of impurities using chitosan was investigated for two effluent streams from the sugar refinery, namely the final effluent (FE) and the resin effluent (RE) by applying the one-factor-at-a-time (OFAT) method. The optimum chitosan loading was found to be 138 mg/l for the RE and 7.41 mg/l for the FE, beyond which the efficiency of the coagulant decreased. The coagulation of FE removed 97% of the total suspended solids (TSS), 61% colour and 35% chemical oxygen demand (COD). The treatment of RE resulted in the removal of 68% TSS, 30% colour and 15% COD due to its high content of impurities. Therefore, RE was not considered for statistical studies. The Box-Behnken (BBD) design, which is a statistical response surface methodology (RSM) model was used to study the simultaneous effect of pH, coagulant loading and settling time on the removal of the COD, TSS and colour, with the help of an overlay plot for the FE. The optimum values from the overlay plot were 92% for TSS, 83% for colour and 29% for COD. The model equations generated by the BBD for individual responses involved all the manipulated variables contrary to the OFAT which only considered one manipulated parameter per response. Moreover, the BBD allowed the simultaneous analysis of all the parameters and the identification of interactions which occur when the effect of one factor is dependent on the level of another. The most important interaction for the removal of TSS was the combination of the variation in pH and coagulant dosage. The COD removal was mostly affected by the interaction between the coagulant loading and the settling time. The colour removal increased with the simultaneous increase of the pH and the settling time. A comparative study between the wastewaters from the sugar industry, the brewery industry and milk processing industry revealed that the performance of the chitosan was also affected by the amount of total dissolved solids (TDS) in the wastewater. A model was developed relating the TSS, COD and TDS from all these wastewaters, and was used to predict the TSS removal for the effluent from the olive oil mills and the wastewater from the winery. Chitosan can be considered as a good alternative to inorganic and synthetic coagulants for the pre-treatment of the FE due to its ability to efficiently remove the levels of TSS and colour. Furthermore, the production of chitosan from crustacean shells is a good method of reducing pollution from the fishery industry. Chitosan can be produced locally at low cost due to both the abundance of crustacean shells in the coastal regions of South Africa and the simplicity of its preparation process. It is recommended that a mathematical model be developed to accurately predict the influence of chitosan on all types of effluent. Such a model will provide an indication of the performance of the chitosan and guide experimenters. It is further recommended that the effect of the use of organic coagulants on the destabilization of dissolved solids in wastewater be given greater attention.

Citrus essential oils (EO) are potential antimicrobials, first described as such in 1949 by Piacentini. Due to their acceptability in terms of fragrance and flavour characteristics, they lend themselves to use both in food and in clinical practice. Lemon (Citrus limon) or sweet orange (Citrus sinensis) or bergamot (Citrus bergamia) essential oils and their components (limonene, linalool, citral, hersipidin and neoericitrin) and vapours at different temperatures and pHs were tested for their anti-microbial activity against vancomycin-sensitive Enterococcus faecium and Enterococcus faecalis. Lemon essential oil alone or blended and orange EO alone were not as effective as citral or linalool alone or bergamot blends with inhibition diameters of > 9cm. The ranges established for differences in inhibition of growth were; 5-15°C, 20-25°C, 30-45°C and 50°C (p ≤ 0.005) and pHs 4.5-6.5, pH 7.5, and pHs 8.5 —10.5 (p ≤ 0.001). A blend of 1:1 (v/v) orange/bergamot EO was the most effective with MICs at 25°C and pH 5.5 of 0.25% - 0.5% (v/v) and an MID of 50mg/i at 50°C at pH 7.5, under these conditions viable counts were reduced by 5.5 - 10 log10 cfu/ml (p ≤ 0.001). The mechanisms by which a blend of orange/bergamot EO and its vapour bring about their antimicrobial effect were assessed. Transmission electron microscope (TEM) images established morphological changes as well as suggesting EO blend uptake into the cell. The permeability of the cell increased by x2 after being subjected to the EO and by x40 after exposure to the vapour. There were decreases of 1.5 in intracellular pH, 20 a.u. in membrane potential and 18 pmol/mg protein of intracellular ATP. The application of the EO blend impregnated into wipes reduced surface microbial load of vancomycin resistant and vancomycin susceptible E. faecium and E. faecalis by up to 4 log10 compared with control wipes (p ≤ 0.00 1). The use of the EO blend vapour on surfaces resulted in a 2.5 log10 reduction over 24 hours, whilst the vapours (15mg/L air) showed a microbial reduction of up to 4 log10 on cucumber skin and lettuce leaf after 45 seconds exposure via a heat diffuser at 25°C (p ≤ 0.001). Sensory tests demonstrated that there were no changes to the taste of the foodstuff after treatment with the EO blend vapour compared with untreated controls. Investigations into the applications of the EO blend showed no significant difference in effect of EO blend between vancomycin resistant and vancomycin susceptible strains. The results of this study suggest that the use of a citrus essential blend both in oil and vapour form could be a potential alternative to chemical based antimicrobial

This thesis seeks for a better understanding of the sensory properties of hydrocolloid thickened foods during oral processing through studying both flow and lubrication behaviours. In addition, during oral processing, saliva plays an important part through mixing with samples, and it is therefore the mixture of foods and saliva that is perceived. However, the role of saliva in sensory perception is not fully elucidated. This research also features a preliminary study on both flow and lubrication properties of saliva in presence of the 5 basic tastants and also how lubrication properties of hydrocolloids are changed when mixing with saliva. Two groups of five samples were designed to have either similar viscosity at a shear rate of 50 s-1 or 105 s-1 by varying the concentrations of xanthan and dextran with the aim to find out which shear rate(s) is related to mouthfeel perceptions. Samples had the same levels of sucrose and banana flavour (isoamyl acetate) added to them and the flavour release and in mouth perceptions measured. The flow behaviour of samples were further characterised in small amplitude dynamic oscillatory shear and stretch flow. A trained sensory panel generated and evaluated mouthfeel, aroma and taste attributes of these solutions. Sensory results indicated that both low and high shear viscosity were related to mouthfeel perceptions. Models including both low and high shear viscosity values predicted the ‘Thickness’ perceptions better than the models including a single shear viscosity. Stickiness and mouthcoating perceptions were better predicted through models including both low shear viscosity and extensional viscosity. Mouthfeel perceptions were also found to be related to complex viscosity at angular frequency of 100 rad.s-1. In terms of sweetness perception, it was affected by the low shear viscosity. However, for samples having similar low shear viscosity, higher scores of overall sweetness were given to samples that were less shear thinning. The high shear viscosity of hydrocolloid samples determines the lubrication properties. Samples with higher viscosity at high shear rate were found to have lower friction in mixed regime but higher in hydrodynamic regime. The mouthfeel perceptions were found to be correlated with friction coefficient at speed of 40-100 mm/s and flavour and aroma were negatively correlated with friction coefficient at speed of 10-30 mm/s. The flow and lubrication behaviour of saliva is changed significantly when stimulated by five basic tastes. The presence of saliva mixed with hydrocolloid samples reduced the friction by up to two orders in boundary and mixed regime but did not affect the friction in hydrodynamic regime which is more related to mouthfeel perceptions.

High Gravity (HG) and Very High Gravity (VHG) fermentations are increasingly attractive within the brewing industry as a means of energy-saving and to optimise process efficiency. However, the use of highly concentrated worts is concomitant with a number of biological stress factors and in particular elevated osmotic pressure, which can impact on yeast quality and fermentation performance. In order to eliminate or reduce such negative effects, yeast cells often respond to their environment by adapting their central carbon metabolism and by making osmotic adjustments. The aim of this research was to investigate the impact of wort gravity on carbon flux, key performance indicators and to examine the effect of external osmolality (as a measure of osmotic pressure) on cell physiology. The fermentation performance of lager and ale brewing yeasts in standard (13 °P), HG (18 °P) and VHG (24 °P) worts was assessed with respect to the uptake of wort sugars, and the production of key carbon metabolites. Estimation of carbon partitioning revealed that products including trehalose, glycogen, higher alcohols and esters had only minor effects on carbon distribution, whereas the production of yeast biomass acted as a major trade-off with ethanol production. Moreover, parallel analysis of the fermentation environment indicated that osmolality increased during fermentations, particularly at high gravities, with the largest contribution directly related to ethanol production. Consequently, yeast cells were subjected to a series of increasing osmolality levels induced by sorbitol, designed to replicate high gravity conditions. These conditions were shown to have a negative impact on yeast viability and vitality, although cell genome integrity was unaffected. In addition, cells responded to osmotic pressure by modifying membrane components leading to a change in fluidity, and by promoting glycerol production. It is anticipated that the data presented here will provide a greater understanding of the response of yeast to HG and VHG conditions, potentially leading to process optimisation in the future.

Potato granules from different sources were found, on extrusion, to produce potato snacks of variable quality. In some instances strip formation was unsatisfactory, in other instances blistering of the snack occurred on frying. In total, about 20-25 batches of potato granules were examined and classified in relation to these two phenomena. The amylose/amylopectin ratios of these samples of potato granules were determined by the semi-micro potentiometric iodine titration technique, but it was found that there was no significant differences in the amylose/amylopectin ratio in relation to the extrusion behaviour of the granules. The determination of the amount of free starch present outside the potato granules also did not show any clear differences between the satisfactory and unsatisfactory potato granules, though the extract from the unsatisfactory potato granules tended to indicate that they contained more lipid than the satisfactory ones. The unsatisfactory potato granules yielded a higher amount of total extractable soluble starch than their satisfactory counterpart, for the same variety of potato granules. The amount of soluble starch increased on extrusion and it was also found to be related to the breaking strength of the extrudate and their blistering behaviour. A stronger strip which gave a higher amount of soluble starch possessed a greater tendency to blister. Addition of PSA additives to potato granules prior to extrusion increased the soluble starch content, but the degree of blistering was suppressed. Gel permeation chromatography experiments showed that the starches extracted from inside the potato granules consisted of mostly high molecular weight macromolecules (very similar to that of amylopectin) and a smaller quantity of low molecular weight macromolecules similar to amylose. The soluble starch extracted from the outside of potato granules also consisted of a greater amount of high molecular weight macromolecules whereas, after extrusion, a greater quantity of smaller molecular weight macromolecules was found to be present which presumably had been expressed from the granules and was assisting in forming the network, binding the potato granules together to form a coherent strip. Determination of the total phosphate and glucose-6-phosphate contents of both potato granules and extracted starches did not show any correlation between the satisfactory and unsatisfactory potato granules. The quantities of phosphate present were small (0.8%). Studies were conducted to elucidate the differences in the cell wall material between satisfactory and unsatisfactory potato granules. Results showed only minimal differences in both the amount of cell wall material and the composition of neutral sugars in the cell wall extract of different potato granule samples. Investigations of the macromolecular order in the granules were pursued along several lines. By x-ray diffraction it was found that the manufactureof potato granules by the add-back process produced changes in the molecular order of the starch component transforming the B-type x-ray pattern of native potato starch to, in the potato granules, the A-type typically found in cereal starches. The differences in molecular order of the starch component between the satisfactory and unsatisfactory potato granules were also investigated using infra-red spectroscopy and differential scanning calorimetry besides x-ray diffractometry. However, there was no simple correlation with either the crystal type or relative crystallinity of potato granules and their extrusion behaviour. Infra-red studies also did not reveal any differences in the spectra nor the absorbance values at wavelengths of 935, 855 and 760 cm-1. D.S.C. and x-ray results, followed subsequently by lipid analysis, established that the unsatisfactory behaviour of certain batches of potato granules was due to the presence of excess GMS (>0.3%), which prevented the formation of a coherent strip, whereas satisfactory granules had a normal amount of GMS (0.3%) which was not detected by either D.S.C. or x-ray techniques. The excess GMS of unsatisfactory granules when monitored as the unassociated material (by the M L peak), was found to decrease with storage time. Extrusion of satisfactory potato granules, when examined by D.S.C. was accompanied by an increase in the amylose-lipid complex (V-amylose) as indicated by the M A-L peak and this was further enhanced on the addition of PSA additives. However, the formation of the V-amylose complex was not very evident in the x-ray patterns. X-ray crystallinity studies indicated that the extrudates exhibited a lower order and it was presumed that there was less x-ray order in the V- than in the A-form. The crystallinity of potato granule samples were found to be affected by their moisture content, the highest crystallinity was achieved with samples having approximately 38% moisture. NMR experiments in conjunction with heating experiments which had been designed to ascertain the amount of bound water before and after heating the granules to 90°C, indicated that although the bound water content before heating did not show any clear differences between satisfactory and unsatisfactory samples, there was a tendency for the unsatisfactory granules to show a slight decrease in bound water after heating. Experiments using varying amounts of water revealed that the spin-spin relaxation time (T2) increased with the water content of both satisfactory and unsatisfactory granules. The increase was greater for the satisfactory granules at 65% and 55% moisture content and lower for 45% and 35% moisture content, compared to the unsatisfactory granules. A mathematical model has been constructed which assumed that blistering was the consequence of an over-strong intergranular network which had a low permeability for water vapour. The experimental physical constraints of time, temperature and water content when introduced into the model showed that the blistering phenomenon could readily be explained. It was finally concluded that unsatisfactory strip formation was a consequence of the presence of excess GMS while blistering occurred when extrusion yielded a higher intergranular polysaccharide network (as evidenced by the amount of the total soluble starch).

No description available.