Oyster meats are currently sorted by hand using volume as the sorting parameter. Hand grading is inaccurate, time consuming and costly. Previous research on physical properties of oyster meats showed a high correlation between projected area of oyster meats and their volume thus allowing the use of projected area measurements as a sorting criterion. A machine vision based oyster meat sorting machine was developed to mechanize the sorting process. The machine consists of a dark conveyor belt transporting singulated oysters through a grading station and then along a row of fast acting water jet valves which separates the stream of oysters into 3 classes. The vision system consists of a monochrome television camera, flash light illumination to "freeze" the images, a digitizer/transmitter and a Personal Computer as an image processing unit. Software synchronizes the flash light and digitization of images and calculates projected area of each meat using the planimeter method. The grading results are sent to a valve control board which actuates the spray valves. The sorting rate is 37 oyster meats/min with a sorting accuracy of 87.5%. A description of the design work, adjustment and l calibration procedures and a final sorting test is included. / Master of Science

The increasing level and dispersion of toxic chemicals in the environment has stimulated a need for accurate methods capable of detecting and quantifying the activity of toxicants. The primary objective of this study was to determine, through in vitro tests, the potential of three purified enzymes: α-chymotrypsin, acid phosphatase, and carbonic anhydrase for use in the early assessment of toxicants at environmentally significant levels. Activities of α-chymotrypsin and acid phosphatase were measured spectrophotometrically, while carbonic anhydrase activities were determined through monitoring a pH change. The chemical agents investigated included several heavy metals, common herbicides and pesticides, and various environmentally significant anions. In addition, several techniques were explored to amplify enzyme response to chemical agents. The results of the study demonstrated that α-chymotrypsin did not significantly respond to cadmium, nickel, or 2,4-D, and consequently, does not appear to be useful in indicating potential toxicity problems associated with these agents. The acid phosphatase enzyme system appears to be useful in signaling the presence of low levels of certain anions (fluoride at 0.45 mg/ℓ and nitrate at 7.3 mg/ℓ), but does not appear to have potential for detecting toxic activity due to cadmium, nickel, or 2,4-D. Enzyme inhibition induced by fluoride at 37°C was not altered by changing the assay temperature to 50°C. The results of the experiments with carbonic anhydrase show that the enzyme does not appear to be affected by cadmium or nickel ions. However, enzyme activity was inhibited by fluoride (4.5 mg/ℓ), sulfide (0.5 mg/ℓ), and nitrate (73 mg/ℓ). Enzyme inhibition was also induced by 10 mg/ℓ of atrazine, malathion, or carbaryl, and 150 to 500 mg/ℓ of 2,4-D. Inhibitory effects induced by sulfanilamide appeared to be slightly enhanced by the addition of Cd²⁺, Ni²⁺, or Zn²⁺ cations. These findings, although preliminary, suggest that carbonic anhydrase demonstrates potential for signaling the presence of anions, and appears to be useful in indicating potential toxicity problems due to pesticides and herbicides. / Master of Science

Concentration-versus-depth profiles have been calculated for Ti and C in Ti-implanted 52100 steel. A computer formalism was developed to account for diffusion and mixing processes, as well as sputtering and lattice dilation. A Gaussian distribution of Ti was assumed to be incorporated at each time interval. The effects of sputtering and lattice dilation were then included by means of an appropriate coordinate transformation. C was assumed to be gettered from the vacuum system in a one-to-one ratio with the surface Ti concentration up to a saturation point. Both Ti and C were allowed to diffuse. A series of experimental (Auger) concentration-versus-depth profiles of Ti implanted steel were analyzed using the above-mentioned assumptions. A best fit procedure for these curves yielded information on the values of the sputtering yield, range and straggling, as well as the mixing processes that occur during the implantation. The effective diffusivity of Ti was found to be 6x10⁻¹⁵ cm²/sec, a value that is consistent with the cascade mixing mechanism. The effective diffusivity of C was found 6x10⁻¹⁵ cm²/sec, and the sputtering yield by Ti atoms was best fit by a value of about 2. The observed range and straggling values were in very good agreement with the values predicted by existing theories, so long as the lattice was allowed to dilate. / M.S.

Development of a methodology of applying symbolic and computational models of conjugate geometry to several problems in design and manufacturing areas forms the basis of this study. The philosophy of this work is the development of generalized symbolic models for two-dimensional and three-dimensional conjugate geometry applications. The generalized models have been implemented using what seems to be the best tool for these kind of applications - a symbolic manipulation system. The unique feature of this research is reflected in the fruitful combination of the elegance of conjugate geometry theory and the inherent versatility of a symbolic manipulation system. The generalized conjugate geometry algorithms were programmed and run using MACSYMA. Typical cases in design of mechanisms have been studied using these symbolic programs. The manufacture of helically swept surfaces is of special interest to this work. Helically swept surfaces have been designed and three schemes of manufacturing these surfaces are presented here. Examples of these three schemes of manufacturing helically swept surfaces have been carried out using the symbolic program in MACSYMA. The results of all the examples have been presented both analytically and graphically. / Master of Science

MScEng / Thesis (MScEng (Mechanical and Mechatronic Engineering))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: Accurate dimensional measurement devices are critical for international industrial competitiveness for South Africa. An overview of all the necessary components of a laser tracking system using a multilateration technique for very accurate dimensional metrology is presented. A prototype laser tracker station was built to further investigate this type of system. The prototype successfully tracks a target within a volume of at least 200 200 200 mm3, approximately 300 mm away from the tracker. This system includes the mechanical design of a prototype tracker station, electronic implementation of ampli cation and motor control circuits, a tracking control algorithm, microcontroller programming and interfacing, as well as a user interface. Kinematic modelling along with Monte Carlo analyses nd the main error source of such a tracker as the beam steering mechanism gimbal axes misalignment. Multilateration is also motivated by the results found by the analysis. Furthermore, an initial sequential multilateration algorithm is developed and tested. The results of these tests are promising and motivate the use of multilateration over a single beam laser tracking system. / AFRIKAANSE OPSOMMING: Dit is van kritieke belang dat Suid-Afrika akkurate dimensionele metingstoestelle ontwikkel vir internasionale industriële medinging. 'n Oorsig van al die nodige komponente vir 'n Laser-Volgsisteem, wat slegs van multilaterasie gebruik maak om baie akkurate drie dimensionele metings te kan neem, word in hierdie projek voorgestel. 'n Prototipe Laser-Volgsisteem-stasie word gebou om so 'n sisteem verder te ondersoek. Die prototipe slaag wel daarin om 'n teiken, binne 'n volume van 200 200 200 mm3 op 'n afstand van omtrent 300 mm te volg. Die sisteem sluit die meganiese ontwerp van die sodanige stasie, elektroniese seinversterking, motorbeheer, 'n volgingsbeheer algoritme, mikroverwerker programeering en intergrasie, asook 'n gebruikerskoppelvlak program in. Kinematiese modelering, tesame met Monte Carlo simulasies, toon aan dat die hoof oorsaak van metingsfoute by so 'n stasie by die rotasie-asse van die laserstraal-stuurmeganisme, wat nie haaks is nie, lê. Die multilaterasie metode word ook verder ondersteun deur dié modelering. 'n Algoritme wat sekwensiële multilateratsie toepas word boonop ontwikkel en getoets. Die resultate van die toetse dui daarop dat die algoritme funksioneer en dat daar voordele daarin kan wees om so 'n metode in plaas van 'n Enkelstraal-Volgsisteem te gebruik.

For decades, duplex filters have been put to use virtually unmodified. Technologies, handling and use of materials show enormous potential for improvement. Filter element emoval/replacement is performed according to a complex process sequence. With the newly developed Duplex Filter, the market demands concerning simple filter element removal/replacement, as well as weight and pressure loss reduction are fully met.

Thesis (MEng) -- Stellenbosch University, 2014. / ENGLISH ABSTRACT: This project investigates aspects of a novel inherently safe nuclear power steam supply system as safety is of paramount importance. The system envisaged has unique features namely: a) a two-phase flow flash-tube type natural circulating primary loop (also the secondary radioactive particle containment); b) a twophase flow thermosyphon heat pipe type heat exchanger secondary loop is used to transfer heat from the primary loop to the steam generators, thereby physically separating the two flow streams from one another; c) a natural convection air cooled condenser for the removal of the reactor’s residual heat; d) a unique core using TRISO type fuel (acting as the primary radioactive particle containment) with life of at least 8.9 years; e) a steel containment vessel acting as a tertiary radioactive product containment; f) a concrete containing structure with air vents to allow air to pass over the main steel containment vessel for cooling purposes in the case of an emergency, and for the removal of parasitic heat during operation. In particular the primary and secondary loops of the proposed system are investigated. This is done by design, construction and testing of a small scale experimental set-up of the primary and secondary loops as well as the development of theoretical models for the two loops. A literature survey focusing on nuclear technology, thermosyphon loops, natural circulating loop instabilities, heat pipes, and two-phase flow modelling is presented to give a brief overview of the technologies as well as tools used in the work undertaken. Observations of the inside flow behaviour of the primary loop experimental set-up were made possible by windows providing many insights into the inner workings, such as plume formation and geysering. The transient response of the secondary heat pipe loop start-up is also investigated. A thermal resistance theoretical model was developed for the secondary loop using heat transfer formulae from theory as well as experimentally semiempirical correlated formula. Different states of operation of the secondary loop were observed during testing with the theoretical model of the condensing regime correlating well, two-phase regime correlating acceptably and liquid regime correlating poorly to experimental results and thus were modelled using an experimentally determined overall heat transfer coefficient. The secondary loop model of the liquid regime is coupled with the primary loop theoretical model to predict the system’s performance. A homogeneous, one-dimensional, simple theoretical model for the primary loop was derived and computer simulated. The results did not compare well with experimental results for single phase flow and failed to capture the onset of two-phase flow. The assumptions of one dimensional model with a unidirectional flow, a hydrostatic pressure problem, a constant volumetric flow rate and the inability of the implementation of the code to handle expansion are noted as some of the flaws in the theoretical model. The following recommendations are made: a more advanced design of the pressuriser should be incorporated into the experiment; the secondary loop’s theoretical model should be characterised under a broader set of operating conditions; the computer program can be used as the basis for further research and implementation of alternative solution algorithms and models. / AFRIKKANSE OPSOMMING: Hierdie projek ondersoek aspekte van ’n ongewone, essensieel veilige kernkrag stoomtoevoer-stelsel, omdat veiligheid van kardinale belang is. Die stelsel wat voorgestel is, het unieke eienskappe, naamlik: a) ’n twee-fasevloei flits-buistipe natuurlik sirkulerende primêre lus (wat ook die sekondêre inperking van radioaktiewe materiaal bevat); b) ’n twee-fasevloei termo-heweleffek sekondêre lus hitte-pyp hitte-uitruiler word gebruik om die hitte vanaf die primêre lus oor te dra na die stoomkragopwekkers en daardeur word die twee strome se vloei fisies geskei van mekaar; c) ’n natuurlike konveksie lugverkoelde kondensor word gebruik vir die verwydering van die reaktors se oortollige hitte; d) ’n unieke kern gebruik TRISO-tipe brandstof (wat as die primêre inperking van radioaktiewe materiaal optree) met ’n lewe van minstens 8.9 jaar; e) ’n inperkingshouer van staal wat optree as ’n tersiêre radioaktiewe produkhouer; f) ’n betonstruktuur met lugventilasie om toe te laat dat lug oor die hoof staalhouer vloei vir verkoeling in ’n noodgeval, en vir die verwydering van parasitiese hitte tydens werking. Hoofsaaklik word die primêre en sekondêre lusse van die voorgestelde stelsel ondersoek. Dit word gedoen deur die ontwerp, konstruksie en die toets van ’n eksperimentele opstelling van die primêre en sekondêre lusse op klein skaal, sowel as die ontwikkeling van teoretiese modelle vir die twee lusse. ’n Literatuurstudie wat fokus op kerntegnologie, termo-heweleffeklusse, natuurlik sirkulerende lus instabiliteit, hitte-pype, en twee-fase vloeimodellering word aangebied om ’n kort oorsig te gee van die tegnologie, sowel as gereedskap gebruik in die werk wat onderneem is. Om die interne vloeigedrag van die primêre lus se eksperimentele opstelling waar te neem, word daar gebruik gemaak van vensters wat dien as ’n manier om die innerlike werking van die proses soos pluimvorming en die kook van die water in die warmwaterkolom te toon. Die oorgangsreaksie van die sekondêre hittepyplus aanvangs is ook ondersoek. ’n Teoretiese termiese weerstandmodel is ontwikkel vir die sekondêre lus met behulp van hitte-oordragformules waarvoor hitte-oordragteorie gebruik is, wat met eksperimentele semi-empiriese formules gekorreleer is. Verskillende toestande van die sekondêre lus se werking is waargeneem gedurende die toetse. Die teoretiese model het goed met die kondensasiestaat gekorreleer, terwyl by die twee-fasewerkswyse aanvaarbare korrellasies aangetref is en die uiteindelike vloeitoestand swakker gekorrelleer het met eksperimentele resultate en dus gemodelleer is met behulp van die NTU-effektiwiteitsmetode. Die sekondêre lusmodel van die vloeistoftoestand is gekoppel met die primêre lus teoretiese model om die werking van die stelsels te voorspel. ’n Homogene een-dimensionele eenvoudige teoretiese model van die primêre lus is afgelei en ’n rekenaar simulasie is uitgevoer. Die resultate vergelyk nie goed met die eksperimentele resultate vir enkelfasevloei en kon nie die aanvang van twee-fasevloei beskryf nie. Die aannemings van ’n een-dimensionele model met eenrigting vloei, ’n hidrostatiese druk probleem, ’n konstant volumetries vloeitempo en die onvermoë van die implementering van die kode om uitbreiding te hanteer is bekend as ’n paar van die foute in die teoretiese model. Die volgende aanbevelings word gemaak: ’n meer gevorderde ontwerp van drukreëlaar moet in die eksperiment ingesluit word; die sekondêre lus se teoretiese model moet gekenmerk word onder ’n wyer stel bedryfsomstandighede, en die rekenaar program kan gebruik word as die basis vir verdere navorsing en die implementering van alternatiewe algoritmes en modelle.

Thesis (MScAgric)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: The distinctive varietal flavour of wines is a combination of absolute and relative concentrations of chemical compounds. Volatile compounds are responsible for the odour of wine and non-volatiles cause the sensation of flavour. Accompanying these senses, a third, tactile, sense of ‘mouth-feel’ is recognizable. This forms the complete organoleptic quality of wine. Several hundred different compounds are simultaneously responsible for the odour release in wine, and since there is no real character impact compound, the aroma of wine can be described as a delicate balance of all these compounds. One of the most important groups of volatiles is the monoterpenes, which play a role in both aroma and flavour. This is especially significant for the Muscat varieties, but these flavour compounds are also present in other non-muscat grape varieties, where they supplement the varietal aroma. Monoterpenes occur in wine as free, volatile and odorous molecules, as well as flavourless non-volatile glycosidic complexes. The latter slowly releases monoterpenes by acidic hydrolysis, but the impact on varietal aroma is considered insufficient for wines that are consumed young. It is therefore important to supplement the release mechanism, in order to enhance the varietal aroma of the wine. The enzymatic hydrolysis mechanism functions in two successive steps: firstly, depending on the precursor, the glycosidic linkage is cleaved by α-L-arabinofuranosidase, α-L-rhamnosidase, β-D-xylosidase or β-D-apiosidase. The second step involves the liberation of the monoterpene alcohol by a β-glucosidase. This enzymatic hydrolysis does not influence the intrinsic aromatic characteristics of the wine, as opposed to acid hydrolysis. Pectolytic enzymes play an important role in cell elongation, softening of tissue and decomposition of plant material. These enzymes are used to improve juice yields, release colour and flavour compounds from grape skins, as well as improve clarification and filterability. Pectolytic enzymes work synergistically to break down pectins in wine. Protopectinase produce water-soluble and highly polymerised pectin substances from protopectin, it acts on non-methylated galacturonic acid units. Pectin methylesterase split methyl ester groups from the polygalacturonic chain. Polygalacturonase break down the glycosidic links between galacturonic acid units. Pectin and pectate lyases have a β-eliminative attack on the chain and it results in the formation of a double bond between C4 and C5 in the terminal residues. From the above it can be seen that enzymes play a pivotal role in the winemaking process. Unfortunately, in winemaking a lot of factors can influence the effects of enzymes. One possible factor in the wine medium is the presence of acidprotease, from yeast and/or fungal origin. This type of enzyme utilizes other enzymes as substrates and renders them useless. Pure enzyme preparations were used to study the interactions of a yeast acid-protease and a report activity (β-glucosidase) in vitro. A bottled wine and a buffer were used as in vitro conditions. Enzyme assays were performed to determine the relative activity over a number of days. The results indicated that even though both enzymes showed activity in both the media, the yeast protease did not have any significantly affect on the report activity. Subsequently wine was made from Sauvignon blanc grapes, with varying enzyme preparation additions. Enzyme assays were performed during the fermentation; and chemical, as well as sensory analysis were done on the stabilized wine. The results confirmed that the yeast protease did not have any significant affect on the report activity in these conditions. The protease’s inability to affect the report activity seems unlikely due to the fact that it is active at a low pH range and has been suggested as the only protease to survive the fermentation process. It seems possible that a winerelated factor, possibly ethanol, is responsible. Thus it seems that yeast protease does not threaten the use of commercial enzymes in the winemaking process in any significant way. Future work would entail more detailed enzyme studies of interactions between protease, both from yeast and fungal origin, and other report activities in specified conditions. The degradation capability could be directed towards unwanted enzyme activities that cause oxidation and browning of the must. The characterization of interactions between protease and β-glucosidase activities may hold key to producing wines with enhanced aroma and colour potential, as well as the elimination of unwanted enzyme activities. / AFRIKAANSE OPSOMMING: Die herkenbare kultivar karakter van wyn is ‘n kombinasie van absolute en relatiewe konsentrasies van verskeie chemiese komponente. Vlugtige komponente is verantwoordelik vir die geur, of aroma, van wyn en die nie-vlugtige komponente veroorsaak die sensasie van smaak. ‘n Derde, fisiese sensasie, die ‘mondgevoel’, is ook herkenbaar. Dit vorm die omvattende organoleptiese kwaliteit van die wyn. ‘n Paar honderd verskillende komponente is gelyktydig verantwoordelik vir die aroma vrystelling in wyn en omdat daar geen werklike karakter ‘impak’ komponent is nie, kan die aroma van wyn beskryf word as ‘n delikate balans van al die betrokke komponente. Een van die mees belangrike groepe vlugtige komponente is die monoterpene wat ‘n rol speel in beide aroma en smaak. Dit is veral belangrik by Muskaat kultivars, maar hierdie aroma komponente is ook teenwoordig in niemuskaat druif kultivars, waar hulle bydra tot die kultivar karakter en aroma. Monoterpene kom in wyn voor as vry, vlugtige en aromatiese molekules en in geurlose, nie-vlugtige glikosidies-gebonde komplekse. Die gebonde vorm word stadig vrygestel deur ‘n suurhidrolise, maar dit word as onvoldoende beskou vir wyne wat vroeg gedrink word. Dit is dus belangrik dat die vrystelling van geurstowwe verhoog word om die kultivar karakter van die wyn te versterk. Die ensiematiese hidrolise proses behels twee opeenvolgende stappe: eerstens, afhangende van die aard van die voorloper, word die glikosidiese verbinding deur α-L-arabinofuranosidase, α-Lramnosidase, β-D-xilosidase, of β-D-apiosidase gebreek. In die tweede stap word die monoterpeen-alkohol deur β-glukosidase vrygestel. Hierdie ensiematiese afbraak proses verander nie die intrinsieke aromatiese kenmerke van die wyn, soos met suurhidrolise die geval is nie. Pektolitiese ensieme speel ‘n fundamentele rol in selverlenging, sagwording en afbraak van plant materiaal. Hierdie ensieme word gebruik om sap opbrengs te verhoog, aroma en smaak komponente vry te stel uit die doppe, asook om sapverheldering en filtrasie te verbeter. Die pektolitiese ensieme werk op ‘n sinergistiese wyse om pektien in wyn af te breek. Protopektinase produseer wateroplosbare en hoogs gepolimeriseerde pektien uit protopektien, slegs uit niegemetileerde galakturoonsuur eenhede. Pektien metielesterase verwyder metielester groepe van die poligalakturoonsuurketting. Die glikosidiese bindings tussen galakturoonsuur eenhede word deur poligalakturonase afgebreek. Pektien- en pektaat-liase het ‘n β-eliminasie aanslag op die ketting en as gevolg daarvan word dubbelbindings tussen C4 en C5 in die terminale residue gevorm. Vanuit bogenoemde is dit dus duidelik dat ensieme ‘n kardinale rol speel in die wynbereidingsproses. Ongelukkig is daar ‘n verskeidenhied van faktore wat die werking van ensieme in die wynbereidingsproses kan beïnvloed. Een moontlike faktor is die teenwoordigheid van ‘n suur-protease, van fungisidiese en/of gis oorsprong, in die wynmedium, omdat dit ander ensieme as substraat kan benut en degradeer. Suiwer ensiem preparate is gebruik om die ensiem interaksie tussen ‘n gis suur-protease en ‘n verslag aktiwiteit (β-glukosidase) in vitro te ondersoek. ‘n Gebotteleerde wyn en ‘n buffer is gebruik om die in vitro kondisies na te boots. Relatiewe ensiem aktiwiteit is ontleed oor ‘n aantal dae. Beide die ensieme het aktiwiteit getoon in die media, maar gis protease het geen statisties beduidende invloed gehad op die aktiwiteit van die verslag ensiem nie. Daaropvolgend is wyn berei van Sauvignon blanc druiwe, met verskillende ensiempreparaat toevoegings. Die ensiemaktiwiteit is deurlopend tydens fermentasie gemeet. Na afloop van stabilisasie is chemiese, sowel as sensoriese ontledings op die wyn gedoen. Die resultate het bevestig dat gis protease, onder hierdie kondisies, geen beduidende invloed op die verslag aktiwiteit gehad het nie. Die protease se onvermoë om die verslag aktiwiteit beduidend te beinvloed blyk onwaarskynlik aangesien die suurprotease aktief is by lae pH vlakke en dit as die enigste protease voorgestel is wat die fermentasie proses kan oorleef. Dit blyk asof ‘n wyn-verwante faktor, moontlik etanol, hiervoor verantwoordelik kan wees. Dus hou protease geen gevaar in vir die gebruik van kommersiële ensieme in wynbereiding nie. Navorsing kan in die toekoms fokus op meer gedetailleerde ensiem interaksie studies tussen protease en ander ensiem aktiwiteite, in gespesifiseerde kondisies. Die degradasie kapasiteit kan moontlik aangewend word om ongewenste ensiem aktiwiteite, wat byvoorbeeld oksidasie en verbruining veroorsaak, te verminder. Die karakterisering van die interaksies tussen protease en β-glukosidase kan dus die sleutel wees tot die produksie van wyne met verhoogde aroma potensiaal, asook die eliminasie van ongewenste ensiematiese aktiwiteite.

Thesis (Msc (Viticulture and Oenology. Institute for Wine Biotechnology))--University of Stellenbosch, 2009. / Monoterpenes are important aroma compounds in certain grape varieties such as Muscat, Gewürztraminer and Riesling and are present as either odourless, glycosidically bound complexes or as free aromatic monoterpenes. These complexes occur as monoglucosides or, when present as diglycosides, most commonly as 6-O-α-L-arabinofuranosyl-β-D-glucopyranosides of mainly linalool, geraniol, nerol and citronellol. The release of monoterpenes from non-volatile glycosidically bound precursors occurs either by acid hydrolysis or enzymatic hydrolysis. High temperature acid hydrolysis causes a rearrangement of the monoterpene aglycones and a decrease in the aroma and changes in the aromatic characteristics of monoterpenes and is therefore not suitable. Enzymatic hydrolysis does not modify the monoterpene aglycones and can be an efficent method to release potentially volatile monoterpenes. α-L-arabinofuranosidase and β-glucosidase are important enzymes responsible for the liberation of monoterpene alcohols from their glycosides. Glycosidases from Vitis vinifera and Saccharomyces cerevisiae are severely inhibited by winemaking conditions and this leads to unutilized aroma potential, while commercial preparations of aroma liberating enzymes are crude extracts that often have unwanted and unpredictable side effects. It is therefore of interest to investigate alternative measures to release glycosidically bound monoterpenes to increase the floral aroma of wine without side activities that impact negatively on wine. Heterologous α-L-arabinofuranosidases and β-glucosidases have previously been expressed in S. cerevisiae and these studies have evaluated and found increased glycosidic activities against both natural and synthetic substrates. In this study, we expressed the Aspergillus awamori α-L-arabinofuranosidase (AwAbfB) in combination with either the β-glucosidases Bgl2 from Saccharomycopsis fibuligera or the BglA from Aspergillus kawachii in the industrial yeast strain S. cerevisiae VIN13 to facilitate the sequential enzymatic hydrolysis of monoterpene diglycosides. Enzyme assays and GC-FID (Gas Chromatography with a Flame Ionization Detector) results show a significant increase in the amount of free monoterpene concentrations under winemaking conditions in the strain co-expressing the AwAbfB and the Bgl2. The increases in free monoterpene levels obtained were similar to those obtained with a commercial enzyme preparation, LAFAZYM AROM. Sensorial evaluation confirmed the improvement in the wine aroma profile, particularly the floral character. This yeast strain permits a single culture fermentation which improves the sensorial quality and complexity of wine. Further investigations on the factors influencing the stability and reactivity of monoterpenes during alcoholic fermentation are needed.

Thesis (PhD (Food Science))--University of Stellenbosch, 2007. / Thermal pasteurisation failures in the dairy industry have often been found to cause end-products of poor quality and short shelf-life. Therefore, alternative methods to eliminate microbial contaminants in raw milk are being studied. Ultrasonication is one such non-thermal technology that could offer the dairy industry an alternative to traditional pasteurisation. The main objective of this dissertation was to evaluate the use of high-power lowfrequency ultrasound (20 kHz, 750 W, 124 μm) applied in batch mode to eliminate a selection of spoilage and potentially pathogenic microbes, commonly associated with milk. These included Gram-positive and negative microbes, comprising of rods and cocci, an endospore-former, and a yeast (Escherichia coli, Bacillus cereus, Chryseobacterium meningosepticum, Lactobacillus acidophilus, Lactococcus lactis, Listeria monocytogenes, Micrococcus luteus, Pseudomonas fluorescens and Saccharomyces cerevisiae). Three strains of E. coli (1 x 106 cfu.ml-1) tested, viz. ATCC 11775, a wild strain from raw milk, and an O157:H7 strain from milk were sensitive to ultrasonication. Complete elimination of viable cells occurred within 10 min. Viable counts of P. fluorescens were reduced by 100% within 6 min of ultrasonication and L. monocytogenes was reduced by 99.0% within 10 min. Lactococcus lactis was reduced by 97.0% and M. luteus, B. cereus and C. meningosepticum by 88.0%, 87.0% and 85.0% respectively. Lactobacillus acidophilus showed the most resistance to ultrasound with only 78.0% of viable cells being eliminated. Under similar conditions, S. cerevisiae was reduced by 99.7%. Microbial cell morphology, size and Gram status did not necessarily influence the efficacy of ultrasonication. Sterile saline solution and UHT milk were used as the suspension media, and the reputed protective effect of milk fat was not observed under the parameters used in this study. A higher wave amplitude (100%; 124 μm) was found to be more efficient in eliminating microbes than a lower wave amplitude (50%; 62 μm). Pulsed-ultrasonication did not enhance the efficiency of ultrasonication indicating that standing waves were absent. Limited success was achieved by ultrasonication itself, and the long batch treatment time (10 min or more) was found to be unrealistic for industrial implementation. Hence the simultaneous application of ultrasound and heat (thermoultrasonication) was examined. Thermo-ultrasonication proved to be more effective than either an ultrasonic or heat treatment with all viable M. luteus cells being eliminated within 4 min (100% amplitude at 72°C). Similarly, to eliminate E. coli and Lb. acidophilus from milk, only 2 min and 4 min thermo-ultrasonication was required, respectively. Bacillus cereus endospores remained resistant and after a 10 min thermo-ultrasonic treatment only 78.04% were eliminated. During this investigation both extensive surface (SEM) and internal (TEM) cell damage caused by ultrasonication were observed in E. coli, Lb. acidophilus and S. cerevisiae. Hence ultrasonication physically/mechanically damages these microbial cells causing cell death/injury. Microbial proteins and DNA released from cells into the environment after an ultrasonic treatment was measured and an increase in released microbial proteins and DNA was found to be indicative of a decrease in the number of viable cells, providing that the initial cell concentration was high enough. It was, however, not possible to correlate the concentration of released microbial proteins and DNA with the exact number of viable cells eliminated, rendering it an ineffective quality indicator for the industry. Ultrasonication had no statistically significant influence on the protein, fat and lactose content of both raw and pasteurised milk. The somatic cell count of raw and pasteurised milk was found to decrease after ultrasonication. Unlike with heating, activity of alkaline phosphatase and lactoperoxidase were not reduced by ultrasonication. Hence neither enzyme can be used to indicate a successful ultrasonic treatment of milk. This study has demonstrated that ultrasonication offers a viable alternative to pasteurisation as it is effective in eliminating microbes, and does not alter native milk components. However, to attain a more effective killing, thermo-ultrasonication is recommended for the treatment of milk to be used for the production of different dairy products.