The objective of this work is to improve the understanding of continuous bioreactor performance. Firstly, the consequences of immobilised cell loading, or crowding, of the biomass hold-up: in the free cell liquid phase, in the Calcium Alginate gel entrapment matrix, and by partition within a dialysis membrane tube on the functioning of the yeast <I>Saccharomyces cerevisiae</I> NCYC 1026 have been investigated. Secondly, an evaluation has been made of the operational performance of experimental anaerobic, aseptic continuous bioreaction systems containing <I>Saccharomyces cerevisiae</I> NCYC 1026 as liquid phase free cells and as both a naturally and man-made immobilised cell solid phase matrix, in terms of mass transfer limitations of the packed, or crowded cells and the contribution of each biomass hold-up to the design performance models of the bioreactor. The kinetic state of the free cell biomass hold-up was measured independently of the continuous bioreaction system, such that, these data, along with on-line substrate, biomass and dilution rate data, were available for analysis of the continuous experimental bioreaction systems, using mathematical models developed in this work. Mathematical modelling of continuous free cell bioreaction systems, with the additional complication of a solid phase immobilised biomass hold-up, revealed a dual root solution to the limiting substrate mass balance, a consequence of the free cell small floc hold-up term, neither of which could be discounted. Attempts to remove the free cell liquid phase biomass hold-up from the experimental bioreaction system proved to be unsuccessful. The immobilised cell hold-up activity was identified as a key modelling parameter, which could not be effectively estimated, nor measured, by current measurement techniques.

No description available.

No description available.

No description available.

No description available.

This dissertation is concerned with physical processes occurring in chocolate at temperatures near 20°C during the "cold extrusion" forming process. The microstructure of chocolate at these temperatures consists of solid particles dispersed in a continuous fat matrix, which itself consists of liquid regions coexisting with fat crystals. The approximate volume fractions of the solid particles and the continuous matrix were calculated from chocolate recipes and the total volume fraction of the solid particles is close to 60%. The chocolate microstructure has some similarities with other highly concentrated suspensions and granular pastes, and certain phenomena observed for these materials, such as "wall slip", were recognised to be important in understanding the behaviour of chocolate during cold extrusion. However, the coexistence of solid and liquid phases that form the continuous matrix introduced other phenomena to the deformation, which were investigated experimentally and theoretically. The extrusion behaviour of a milk chocolate and a cocoa butter was characterised at temperatures of 20-24°C using a "MultiPass Rheometer". This allowed sequences of extrusions to be performed, thus the effect of processing history could be explored. The extrusion pressures of both the chocolate and the cocoa butter strongly depended on the processing history: each extrusion pass softened the material, first strongly, then by smaller amounts in subsequent extrusions. If a delay time was introduced between the extrusions, the flow pressures were larger than during a continuous extrusion sequence and increased with the length of the delay time, indicating a re-hardening effect. For delay times of more than 60 minutes, the flow pressure of milk chocolate was identical to the original flow pressure, while for cocoa butter, this was the case after only 10 minutes. The previously reported "post-extrusion flexibility" of cold extruded chocolate was therefore shown to have direct consequences for the extrusion processing itself. An in-situ X-ray investigation, using a combination of the MultiPass Rheometer with an X-ray diffraction system, was then carried out for the two materials. Changes of the triglyceride crystal structure and crystal content in these materials were monitored during processing. Clear evidence for a reduction of the crystal fat content during extrusion processing and for re-crystallisation thereafter was obtained, for both cocoa butter and chocolate. The results were consistent with previous observations of changes in the liquid phase content of cocoa butter and chocolate after extrusion processing. Some information about the kinetics of this re-crystallisation process was also obtained. These kinetics were similar to those of chocolate extrudate re-hardening and to the recovery of the extrusion pressure seen in the multi-pass extrusion experiments. The insight gained through the experimental program was used to develop an extended model for the deformation of chocolate during cold extrusion. Macroscopic energy balances and finite element analyses were used in this development. Firstly, the effects of wall friction on the extrusion pressure were studied. The effects of extrusion processing on the temperature and on the amounts of the solid and liquid phases of the fat matrix were then considered. Finally, it was estimated how these changes affect subsequent processing and the extrusion pressure itself. The new model was used to discuss the experimental results from this work and from previous experimental investigations. It provided a qualitative explanation of previously observed localised concentrations of high liquid fat content and suggests that, for typical extrusion geometries, the softening does not alter the extrusion pressure in a single extrusion strongly compared to a material with a constant yield stress.

Much of the evidence supporting a mitochondrial uncoupling function for the close homologues of UCP1 (e.g. UCP2 and UCP3) comes from experiments in which the levels of protein expression have been genetically manipulated. Evidence is presented, from yeast genetically modified to express UCP1 and UCP3, that these experimental manipulations can lead to an uncoupling of oxidative phosphorylation that does not necessarily represent a native activity of the protein. UCP1, whose native uncoupling activity can be fully inhibited by GDP, was used to demonstrate that artefactual uncoupling occurs at higher levels of protein expression. Similarly uncoupling was observed when UCP3 was expressed at high levels. UCP3 was expressed in yeast mitochondria at levels similar to, 2-fold and 7-fold greater than those found in mouse skeletal muscle (135 ± 45 ng/mg mitochondrial protein). No increase in uncoupling was seen than can be observed in fully inhibited UCP1. This uncoupling was not stimulated by the addition of palmitate nor was it inhibited by GDP. Therefore it is concluded that all uncoupling reported to be due to the expression of UCP3 is artefactual and does not represent a native activity of the protein.

The objective of this thesis was to gain an understanding of the cold extrusion of chocolate, both physically and mathematically. This was achieved by quantifying the effect of pertinent process variables upon the pressure difference across a die as a chocolate flowed through it. Changes in the chocolate's cross-sectional area were found to have a large effect upon the extrusion pressure, whereas increases in its flowrate did not. It was also observed that the pressure on cessation of extrusion remained at a finite value. In addition, the boundary condition during cold extrusion was directly visualised and it was concluded that slip was occurring at the wall. This slippage was quantified by a low Coulomb friction coefficient. Therefore, it was concluded that during cold extrusion chocolate was deforming as a near-perfect plastic material. This characteristic allowed the successful use of a relatively simple theory to model the extrusion of chocolate through axisymmetric dies. The extension of this modelling by finite element techniques was investigated with the use of a commercial code (<I>ABAQUS/Standard</I>). The pressure difference across a die was found to be very sensitive to the temperature of the chocolate. The relationship between the pressure and the reciprocal of the temperature was fitted with a straight line. This relationship was extended to other chocolate compositions by scaling the chocolate's temperature by its nominal melting point. Across most of the flowrate range studied the flow pressure difference was either independent of, or increased slightly with, increases in the flowrate. However, at low flowrates and temperatures, the flow pressure was seen to decrease on increasing the flowrate. For dies with long capillaries, extruding in this region resulted in a flow instability. This instability showed a stick-spurt extrudate motion which was accompanied by large pressure oscillations, from which the chocolate's compressibility was calculated to be approximately the same as for some common polymers. The instability was successfully modelled by the combination of two simultaneous differential equations and a non-monotonic pressure-flowrate curve.

Extrusion is a low cost, versatile process widely used in industry for the production of a broad range of products of various shapes, including foodstuffs, pharmaceutical pellets and catalysts. The primary tool used to study the extrusion process in this work is Nuclear Magnetic Resonance (NMR). An extruder system made from PEEK and aluminium, suitable for use in strong magnetic fields, has been designed and commissioned. It has been used to acquire fully-quantitative spatially resolved real time NMR velocity measurements of two soft-solid materials representative of those commonly extruded in industry; a mixture of 5 cSt polydimethyl siloxane (PDMS) added to a commercial soap (Dove^TM), and a stiff biscuit dough. Flow patterns were found to be material dependent, with the soap and PDMS mix exhibiting a much flatter velocity profile than that of the biscuit dough. These spatially resolved velocity profiles are reliable and extensive, providing data suitable for evaluating theoretical models. Comparison of the experimentally determined velocity profiles with velocity profiles predicted from geometric and CFD models highlighted the limitations of these existing models, none of which proved to be entirely satisfactory. Self-diffusion measurements were used to investigate extrusion induced changes in the distribution of PDMS within soap and PDMS mixtures prepared by (i) warm-mixing and (ii) cold-mixing. Before extrusion the PDMS exists in domains of ~ 6mm diameter. Extrusion of the warm-mixed sample leads to a ~17% reduction in domain size, whereas extrusion of the cold-mixed sample causes elongation of the domains but not reduction in domain size. NMR has also been used to make <i>in situ</i> measurements of the extent of rejoining of two streams formed during ram extrusion through a die with a slit obstruction, and of the filling of a cylindrical mould following ram extrusion.

A model fat system to study the phenomenon of undesired deposition in food fat distribution lines was developed and characterised after consideration of the major triglyceride fractions that were found to be present in core analyses and in related systems exhibiting unwanted crystal formation in palm oil based products and fat blends. The model fat solution utilised was a blend of tripalmitin, the highest melting point triglyceride present in palm oil, and an inert paraffin solvent. A batch crystalliser was commissioned and techniques were developed and validated for sampling the crystallising solution and isolating crystal samples without the introduction of artefacts. Protocols and methods for analysis of crystal samples <i>e.g.</i> particle size distribution were developed and verified. Results from the batch crystallisation experiments indicated that there were two main modes of crystal formation which determined the type and nature of the crystals formed. A deposition flow cell apparatus was commissioned in which the mean flow velocity and coolant temperatures could be varied. A heat flux sensor was incorporated into the flow cell to allow in-situ monitoring of the extent of disposition. The apparatus was designed so that the deposit formed could be monitored visually during deposition and could be readily removed for closer examination of its microstructure. The data from the heat flux sensor in the flow cell was used to generate real-time deposit thickness profiles and fouling rates. Results from the deposition experiments indicated that there are two main modes of deposition. The first occurs at low saturation levels and at a small temperature driving force, where the deposit grows slowly and steadily to reach an asymptote with no evidence of removal, therefore indicating that the deposit is relatively strong in nature. The second type of deposition observed involved rapid growth with frequent removal (sawtooth behaviour) indicating a weaker deposit. This latter type of behaviour was seen predominantly at high concentrations and with large temperature differentials.