This thesis focuses on the design and characterisation of miniature bioreactors and evaluates their potential as a scale-down device for microbial cultivation processes. Miniature bioreactors, such as the one detailed in this work, have been developed by many research groups and companies, and seek to increase throughput at the early stages of bioprocess development. Power input was measured in two prototype stirred-tank miniature bioreactors (10 ml and 25 ml) as a function of impeller speed and the vessels were characterised alongside a 7 L bioreactor. The results show that both miniature bioreactors used in this study were able to be characterised using the same methods developed for larger vessels and that the key engineering parameters of volumetric oxygen transfer coefficient and mixing time compared favourably with those of a conventionally-sized bioreactor when expressed as a function of specific power input. An Escherichia coli plasmid DNA cultivation was successfully scaled down to the 10 ml miniature bioreactor from a 7 L bioreactor on the basis of equal specific power input, and demonstrated equivalent performance under oxygen-rich and oxygen- limited conditions. An intermittently-fed process to produce a Fab' antibody fragment using E. coli and a batch cultivation of the filamentous bacterium Saccharopolyspora erythraea producing erythromycin were also evaluated in the 25 ml miniature bioreactor and three other small scale cell cultivation devices (i.e. microtitre plate, miniature bubble column reactor and shake flasks). Their relative performances in terms of growth and product formation were related to that of the 7 L bioreactor. The results obtained demonstrated the ability of the 25 ml miniature stirred tank bioreactor to perform both of these technically-demanding, industrially-relevant bioprocesses to a comparable degree as the 7 L vessel that was not achievable using the other miniature devices tested. The results shown in this thesis highlight the potential of miniature bioreactors to be used to deliver a fully-integrated, high-throughput solution for cell cultivation process development.

The availability of large numbers of engineered organs would offer significant benefits to the clinical management of surgery. Tissue engineering offers the potential of providing tissues that can mimic the morphology, function and physiologic environment of native ones. Cells could grow in vitro within a biodegradable polymer to construct tissue for implantation. However no generic bioreactor design currently exists. There is now a need to establish a robust process for the production of engineered tissues using autologous cells. A key challenge will be the prediction of the supply of nutrients and removal of metabolites. Models of transport phenomena were developed in order to predict the fluid flow and mass transfer requirements of a prototype bioreactor for the formation of engineered tissues. These models were solved to generate windows of operation which relate key operating parameters with the feasibility of tissue preparation. Examples highlight how the windows of operation can be used to visualize rapidly the region of operating conditions that satisfy the design constraints. The impact of the cell concentration, tube geometry, alginate diffusivity, substrate and metabolite concentration levels, feed and recycle rate on the design of the bioreactor is illustrated. The result of this analysis determines the best configuration of the bioreactor which can meet the cellular transport requirements as well as being reliable in performance whist seeking to reduce the amount of valuable nutrients to be used. Micro scale experiments were designed in order to evaluate from measurements, effective diffusivities of substrates and metabolites in alginate matrices as well as substrate consumption and metabolite production rates in matrices with immobilized growing cells. The oxygen diffusivity and oxygen uptake rate of alginate immobilized neonatal fibroblasts were evaluated using integrated oxygen sensor spots. Additionally, alginate cylindrical constructs with immobilized neonatal fibroblasts were prepared in transwells in order to evaluate the effective diffusivities of glucose and lactate as well as the glucose consumption and lactate production rate. The advantage of such micro scale experiments was that greater data sets could be generated with the small number of cells available but in a way which predicts the larger scale. The database which was created was used to construct the windows of operation to give quantitative solutions of how engineered tissues may be prepared and to visualize process operability in a more explicit way.

Expression of recombinant protein in bacteria such as Escherichia coli can result in the formation of inclusion bodies. Active protein is derived from inclusion bodies by protein refolding. Refolding yields are often poor, and are a bottleneck in such recombinant protein processes. Work has addressed refolding process issues, but several areas remain poorly understood. This thesis aims to understand the importance of process parameters upon refolding yields. Previous studies highlighted that mixing can affect refolding, but have failed to establish why. Use of a two impeller system with operation of small paddle impeller (Re= 2000) in the proximity of the injection point, revealed the importance of energy dissipation experienced by injected denatured protein upon lysozyme refolding yields. A factorial experiment studying the effect of factors on lysozyme refolding yield in fed- batch revealed effects and interactions between physical and chemical process parameters. GdHCl concentration (1.2M) and redox ratio (2:1 red:ox) had the greatest effect. A graphical (windows of operation) approach revealed high GdHCl concentration (1.2 M) and redox ratios at or above unity gave greatest yields in the minimum time. Industry typically refolds from impure inclusion bodies. A series of studies using trypsinogen IB detailed the effect of process contaminants and the efficacy of steps to remove them, upon refolding. Analysis of the effect of centrifugation conditions on IB purity demonstrated the compromise between the levels of IB purity and recovery achieved, together with the removal of key contaminants and the refolding yield obtained, with an optimum of Q/I= 23.08 x 10'g m/s. A second study looked at IB washing and revealed that maximising the area for washing is key to IB purity but not to refolding yield. Use of Triton X was key to maximising refolding yields.

No description available.

This thesis describes research carried out by the author between 1988 and 1991 at the Department of Chemical Engineering, University of Edinburgh, under the supervision of Dr Donald Glass and Dr Bruce Ward. The aim of the research was to design, build, operate and control a dual-hollow fibre bioreactor. The principle behind the design is that of the blood supply system in animals. The nutrients are supplied in one set of fibres to the growth region, similar to the arteries in the blood system, and another separate set of fibres takes waste products away from the growth region, in a manner analogous to the venous system. The design, construction and operation of the bioreactor is described. The development of novel building techniques are explained, covering new ground in fibre bioreactor construction. The monitoring equipment required is described with a number of successful experimental runs demonstrating the data collection capabilities of the apparatus. During the research, areas of work not initially envisaged were explored, with the aim to provide a basis for future control strategies. This included the development of a fibre testing rig, so that different fibres and various medium preparations could be tested outside a reactor system. This was done due to the lack of basic information available on fibre performance. This leads into work on the modelling the bioreactor by means of a numerical solution run on a computer. The model provides new areas of simulation, the fouling of fibres and the changing nutrient concentrations supplied to the bioreactor. The work is now at a stage where experimental work and modelling work should be brought closer together to help understand problems experienced in both areas.

The basis of this work was to understand how the performance of Microbial Fuel Cells (MFCs) can be understood and improved by analysing the behaviour of the microbial communities in the anodic chamber. It was hypothesized that specific types of species generally become more abundant in MFCs over time leading to enhanced power production. An acclimatised microbial consortium obtained from a tubular MFC was used as the inoculum for the MFC described in this study: It was found to lead to a different bacterial composition, but similar power density, to those observed in an MFC inoculated with the unacclimatised community (anaerobic sludge). Using anaerobic sludge as inoculum in four replicate MFCs, both the anodic biofilm and the suspended communities evolved differently. The spatial and temporal dynamics of microbial communities were studied in the tubular MFCs. Although the removal of organic compounds was spatially different, the dynamics of the dominant bacteria showed spatial similarity, probably attributed to the versatile metabolic capabilities of species. No specific species were found the relative abundance of which would have clearly enhanced and correlated with the power production. Using similar substrate feeds and inocula, the communities consisted of metabolically different species in the two reactor types studied. Functional redundancy was observed in the anodic communities of both reactor designs. These findings suggest that the exoelectrogenic ability could be present among a range of bacteria wider than generally thought. 2 The results of this study suggest that the development of the microbial communities in MFCs with a given inoculum and substrate are determined by the reactor design and the operational conditions. Secondly, the adaptation of bacterial communities to produce electricity may not require specific changes in community composition but instead be based on the ability of bacteria to adapt generating electricity and enhance their exoelectrogenic capacity over time.

No description available.

No description available.

No description available.

No description available.