Combining spatially distributed predictions from neural networks

In this report we discuss the problem of combining spatially-distributed predictions from neural networks. An example of this problem is the prediction of a wind vector-field from remote-sensing data by combining bottom-up predictions (wind vector predictions on a pixel-by-pixel basis) with prior knowledge about wind-field configurations. This task can be achieved using the scaled-likelihood method, which has been used by Morgan and Bourlard (1995) and Smyth (1994), in the context of Hidden Markov modelling

Non-zero mean Gaussian process prior wind field models

This report outlines the derivation and application of a non-zero mean, polynomial-exponential covariance function based Gaussian process which forms the prior wind field model used in 'autonomous' disambiguation. It is principally used since the non-zero mean permits the computation of realistic local wind vector prior probabilities which are required when applying the scaled-likelihood trick, as the marginals of the full wind field prior. As the full prior is multi-variate normal, these marginals are very simple to compute.

Gas distribution in shallow large diameter packed beds

It is important to maintain a uniform distribution of gas and liquid in large diameter packed columns to maintain mass transfer efficiency on scaling up. This work presents measurements and methods of evaluating maldistributed gas flow in packed columns. Little or no previous work has been done in this field. A gas maldistribution number, F, was defined, based on point to point velocity variations in the gas emerging from the top of packed beds. f has a minimum value for a uniformly distributed flow and much larger values for maldistributed flows. A method of testing the quality of vapour distributors is proposed, based on "the variation of f with packed height. A good gas distributor requires a short packed depth to give a good gas distribution. Measurements of gas maldistribution have shown that the principle of dynamic similarity is satisfied if two geometrically similar beds are operated at the same Reynold's number. The validity of f as a good measure of gas maldistribution, and the principle of dynamic similarity are tested statistically by Multi-Factor Analysis of the variance, and visually by the response "surfaces technique. Pressure distribution has been measured in a model of a large diameter packed bed, and shown to be associated with the velocity of the gas in a tangential feed pipe. Two simplified theoretical models are proposed to describe the flow of gases through packed beds and to support the principle of dynamic similarity. These models explain why the packed bed itself causes the flow of gas to become more uniformly distributed. A 1.2m. diameter scaled-down model was constructed geometrically similar to a 7.3m. diameter vacuum crude distillation column. The previously known internal cylinder gas distributor was tested. Three new distributors suitable for use in a large diameter column were developed and tested, these are: Internal Cylinder with Slots and Cross Baffles, Internal Cylinder with Guides in the Annulus, Internal Cylinder with Internal Cross Baffles - It has been shown that this is an excellent distributor.

The production, harvest and adsorptive recovery of an infectious herpes simplex virus vaccine

At present there is not a reliable vaccine against herpes virus. Viral protein vaccines as yet have proved unsuccessful to meet the challenge of raising an appropriate immune response. Cantab Pharmaceuticals has produced a virus vaccine that can undergo one round of replication in the recipient in order to produce a more specific immune reaction. This virus is called Disabled Infectious Single Cycle Herpes Simplex Virus (DISC HSV) which has been derived by deleting the essential gH gene from a type 2 herpes virus. This vaccine has been proven to be effective in animal studies. Existing methods for the purification of viruses rely on laboratory techniques and for vaccine production would be on a far too small a scale. There is therefore a need for new virus purification methods to be developed in order to meet these large scale needs. An integrated process for the manufacture of a purified recombinant DISC HSV is described. The process involves culture of complementing Vero (CR2) cells, virus infection and manufacture, virus harvesting and subsequent downstream processing. The identification of suitable growth parameters for the complementing cell line and optimal limes for both infection and harvest are addressed. Various traditional harvest methods were investigated and found not to be suitable for a scaled up process. A method of harvesting, that exploits the elution of cell associated viruses by the competitive binding of exogenous heparin to virus envelope gC proteins, is described and is shown to yield significantly less contaminated process streams than sonication or osmotic approaches that involve cell rupture (with> 10-fold less complementing cell protein). High concentrations of salt (>0.8M NaCl) exhibit the same effect, although the high osmotic strength ruptures cells and increase the contamination of the process stream. This same heparin-gC protein affinity interaction is also shown to provide an efficient adsorptive purification procedure for herpes viruses which avoids the need to pre-treat the harvest material, apart from clarification, prior to chromatography. Subsequent column eluates provide product fractions with a 100-fold increase in virus titre and low levels of complementing cell protein and DNA (0.05 pg protein/pfu and 1.2 x 104 pg DNA/pfu respectively).

Purification of plasmid DNA for use in human gene therapy

Two key issues defined the focus of this research in manufacturing plasmid DNA for use In human gene therapy. First, the processing of E.coli bacterial cells to effect the separation of therapeutic plasmid DNA from cellular debris and adventitious material. Second, the affinity purification of the plasmid DNA in a Simple one-stage process. The need arises when considering the concerns that have been recently voiced by the FDA concerning the scalability and reproducibility of the current manufacturing processes in meeting the quality criteria of purity, potency, efficacy, and safety for a recombinant drug substance for use in humans. To develop a preliminary purification procedure, an EFD cross-flow micro-filtration module was assessed for its ability to effect the 20-fold concentration, 6-time diafiltration, and final clarification of the plasmid DNA from the subsequent cell lysate that is derived from a 1 liter E.coli bacterial cell culture. Historically, the employment of cross-flow filtration modules within procedures for harvesting cells from bacterial cultures have failed to reach the required standards dictated by existing continuous centrifuge technologies, frequently resulting in the rapid blinding of the membrane with bacterial cells that substantially reduces the permeate flux. By challenging the EFD module, containing six helical wound tubular membranes promoting centrifugal instabilities known as Dean vortices, with distilled water between the Dean number's of 187Dn and 818Dn,and the transmembrane pressures (TMP) of 0 to 5 psi. The data demonstrated that the fluid dynamics significantly influenced the permeation rate, displaying a maximum at 227Dn (312 Imh) and minimum at 818Dn (130 Imh) for a transmembrane pressure of 1 psi. Numerical studies indicated that the initial increase and subsequent decrease resulted from a competition between the centrifugal and viscous forces that create the Dean vortices. At Dean numbers between 187Dn and 227Dn , the forces combine constructively to increase the apparent strength and influence of the Dean vortices. However, as the Dean number in increases above 227 On the centrifugal force dominates the viscous forces, compressing the Dean vortices into the membrane walls and reducing their influence on the radial transmembrane pressure i.e. the permeate flux reduced. When investigating the action of the Dean vortices in controlling tile fouling rate of E.coli bacterial cells, it was demonstrated that the optimum cross-flow rate at which to effect the concentration of a bacterial cell culture was 579Dn and 3 psi TMP, processing in excess of 400 Imh for 20 minutes (i.e., concentrating a 1L culture to 50 ml in 10 minutes at an average of 450 Imh). The data demonstrated that there was a conflict between the Dean number at which the shear rate could control the cell fouling, and the Dean number at which tile optimum flux enhancement was found. Hence, the internal geometry of the EFD module was shown to sub-optimal for this application. At 579Dn and 3 psi TMP, the 6-fold diafiltration was shown to occupy 3.6 minutes of process time, processing at an average flux of 400 Imh. Again, at 579Dn and 3 psi TMP the clarification of the plasmid from tile resulting freeze-thaw cell lysate was achieved at 120 Iml1, passing 83% (2,5 mg) of the plasmid DNA (6,3 ng μ-1 10.8 mg of genomic DNA (∼23,00 Obp, 36 ng μ-1 ), and 7.2 mg of cellular proteins (5-100 kDa, 21.4 ngμ-1 ) into the post-EFD process stream. Hence the EFD module was shown to be effective, achieving the desired objectives in approximately 25 minutes. On the basis of its ability to intercalate into low molecular weight dsDNA present in dilute cell lysates, and be electrophoresed through agarose, the fluorophore PicoGreen was selected for the development of a suitable dsDNA assay. It was assesseel for its accuracy, and reliability, In determining the concentration and identity of DNA present in samples that were eleclrophoresed through agarose gels. The signal emitted by intercalated PicoGreen was shown to be constant and linear, and that the mobility of the PicaGreen-DNA complex was not affected by the intercalation. Concerning the secondary purification procedure, various anion-exchange membranes were assessed for their ability to capture plasmid DNA from the post-EFD process stream. For a commercially available Sartorius Sartobind Q15 membrane, the reduction in the equilibriumbinding capacity for  ctDNA in buffer of increasing ionic demonstrated that DNA was being.adsorbed by electrostatic  interactions only. However, the problems associated with fluid distribution across the membrane demonstrated that the membrane housing was the predominant cause of the .erratic breakthrough curves. Consequently, this would need to be rectified before such a membrane could be integrated into the current system, or indeed be scaled beyond laboratory scale. However, when challenged with the process material, the data showed that considerable quantities of protein (1150 μg) were adsorbed preferentially to the plasmid DNA (44 μg). This was also shown for derived Pall Gelman UltraBind US450 membranes that had been functionalised by varying molecular weight poly-L~lysine and polyethyleneimine ligands. Hence the anion-exchange membranes were shown to be ineffective in capturing plasmid DNA from the process stream. Finally, work was performed to integrate a sequence-specific DNA·binding protein into a single-stage DNA chromatography, isolating plasmid DNA from E.coli cells whilst minimising the contamination from genomic DNA and cellular protein. Preliminary work demonstrated that the fusion protein was capable of isolating pUC19 DNA into which the recognition sequence for the fusion-protein had been inserted (pTS DNA) when in the presence of the conditioned process material. Althougth the pTS recognition sequence differs from native pUC19 sequences by only 2 bp, the fusion protein was shown to act as a highly selective affinity ligand for pTS DNA alone. Subsequently, the scale of the process was scaled 25-fold and positioned directly following the EFD system. In conclusion, the integration of the EFD micro-filtration system and zinc-finger affinity purification technique resulted in the capture of approximately 1 mg of plasmid DNA was purified from 1L of E.coli  culture in a simple two stage process, resulting in the complete removal of genomic DNA and 96.7% of cellular protein in less than 1 hour of process time.

Intumescent formulations incorporating melamine phosphate

Once familiar with the fire test rig constructed by M Kay, and modified to allow incorporation of both video and computer facilities, Melamine Phosphate production was scaled up from small to large laboratory scale, and then commercial scale production was considered. Samples produced at each stage were compared analytically, visually and in fire testing. The separation and drying stages on a commercial scale lay unresolved practically, due to lack of test facilities. Different cure regimes for the Araldite MY753 and Versamid system were investigated along with weathering tests and cured samples. Surface priming is suggested for large scale application, though on a small scale a clean unprimed surface was thought sufficient. Some samples heat, aired, cracked at the edges but remained bonded on fire testing. An intumescent sample containing Melamine Phosphate, Araldite and Versamid could not be applied to a vertical surface successfully, the viscosity had to be increased to allow application and curing, various additives were tested, two successful ones being fumed silica and a solvent, isopropanol. The low percentages fumed silica used was incorporated into the sample and the viscosity and fire test results compared with a `standard sample'. An expanding graphite incorporated into a standard sample made mixing and application increasingly difficult, due to the lubricating affect of graphite, but the char produced was a good quality, stable char. A suitable formulation could now be mixed, applied and cured, and assuming no adverse interaction between the additives would protect the sample in the event of a fire.

The atomisation of detergent slurries

A study was made to determine the conditions under which the optimum droplet size distribution (ie., narrowest size range with a minimum of fines and over-sized agglomerates), is generated in sprays from centrifugal pressure nozzles. A range of non-Newtonian detergent slurries were tested but the results are of wider application and parallel work was undertaken with water, ionic solutions and chalk slurries. Six centrifugal pressure nozzles were used and the drop-size distributions correlated as a function of fluid properties, pressure, fiowrate, feed temperature, and nozzle characteristics. Measurements were made using a Malvern Particle Size Anayser slung across a specially-designed transparent tower section of approximately 1.2m diameter in order to reduce obscuration caused by the spray and improve existing droplet sizing techniques. The results obtained were based upon the Rosin-Rammler distribution model and the Size Analyser provided a direct print-out of size distribution and the parameters characterising it. A Spraying System nozzle, AAASSTC8-8, produced the optimum spray distribution with the detergent slurry at a temperature of 60°C whilst operating at 1200 psi. With other fluids the Delevan 2.2SJ nozzle produced the optimum spray distribution operating at 1200 psi but with the Spraying Systems nozzles there was no clear-cut optimum set of conditions, ie. the nozzle and pressure varied depending upon the fluid being sprayed. The mechanisms of liquid sheet break-up and droplet dispersion were investigated in specially-constructed, scaled-up, transparent nozzles. A mathematical model of centrifugal pressure nozzle atomisation was developed based upon fundamental operating parameters rather than resorting to empirical correlations. This enabled theoretical predictions to be made over a wide range of operating conditions and nozzle types. The model predictions for volumetric fiowrate, liquid sheet length and air core diameter showed good agreement with the experimentally determined results. However, the model predicted smaller droplet sizes than were produced experimentally due to inaccuracies identified in the initial assumptions.

The casing layer in the culture of Agaricus bisporus, the cultivated mushroom

The casing layer is an essential component of the system employed in the culture of Agaricus bisporus. The literature appropriate to the casing layer is fully reviewed, including aspects relating to fructification and morphogenesis in A.bisporus, together with an appraisal of the various media employed, their properties and functions, and the commercial significance of the casing layer. Equipment is described for use in experiments in mushroom culture, based on a scaled-down version of normal growing technique, allowing the analysis of both weights and number of fruitbodies forming, which was useful in assessing the effects of different casing treatments. The basic steps in the production of fruitbodies in A.bisporus.are described, including a photographic study of the colonisation of casing and fructification. Various alterations to the physical structure of peat/chalk casing mixtures were found to have an effect on fructification; those causing an opening-out of the casing structure tended to give better yields, especially in the early stages of production. It was shown that, in order to obtain greater yield through casing amendment, fructification must be stimulated, giving increased numbers of fruitbodies, disproportionate to their total weight and consequently of lower mean weight. A synthetic casing medium based on the light glass-like mineral, perlite, was developed. The best formula obtained was -.1 part perlite: 1 part montmorillonite clay (by weight): 3 parts 0.01% glucose solution. Perlite/montmorillonite casing could be improved by adding compost colonised by mycelium of A.bisporus, or adding a peat-chalk casing extract. Perlite was also found to be suitable for admixture with the standard casing medium and a mixture of equal parts by volume performed as well as the peat/chalk casing normally used.

Automated assessment of visual fields and their inter-relation to evoked potentials in visual disorders

The Octopus Automated Perimeter was validated in a comparative study and found to offer many advantages in the assessment of the visual field. The visual evoked potential was investigated in an extensive study using a variety of stimulus parameters to simulate hemianopia and central visual field defects. The scalp topography was recorded topographically and a technique to compute the source derivation of the scalp potential was developed. This enabled clarification of the expected scalp distribution to half field stimulation using different electrode montages. The visual evoked potential following full field stimulation was found to be asymmetrical around the midline with a bias over the left occiput particularly when the foveal polar projections of the occipital cortex were preferentially stimulated. The half field response reflected the distribution asymmetry. Masking of the central 3° resulted in a response which was approximately symmetrical around the midline but there was no evidence of the PNP-complex. A method for visual field quantification was developed based on the neural representation of visual space (Drasdo and Peaston 1982) in an attempt to relate visual field depravation with the resultant visual evoked potentials. There was no form of simple, diffuse summation between the scalp potential and the cortical generators. It was, however, possible to quantify the degree of scalp potential attenuation for M-scaled full field stimuli. The results obtained from patients exhibiting pre-chiasmal lesions suggested that the PNP-complex is not scotomatous in nature but confirmed that it is most likely to be related to specific diseases (Harding and Crews 1982). There was a strong correlation between the percentage information loss of the visual field and the diagnostic value of the visual evoked potential in patients exhibiting chiasmal lesions.

Topographic studies of scalp potentials evoked by pattern presentation

The waveform and scalp distribution of the visual evoked potentials elicited by stimuli in the foveal and parafoveal regions have been investigated in a group of normal humans using a 16-channel `brain mapping' system. The waveform and topography of the responses to pattern onset and pattern reversal stimulation were investigated, using 4 x 4o full field and 4 x 2o lateral and altitudinal half-field stimuli. The responses were composed of several successive peaks which are in some respects consistent with those demonstrated by other workers using larger field sizes. The differences in the behaviour of these components with respect to the position of the stimulus in the visual field were suggestive of origins in different areas of the visual cortex and/or different visual mechanism. Of particular interest were the major early positive components `P90' and `P95' of the responses to pattern onset and pattern reversal stimulation respectively. More detailed exploration of the behaviour of these major early positive components was carried out using `M-scaled' stimuli selected to activate one square centimetre patches of striate cortex and associated extrastriate re-projections, positioned at different points in the foveal and parafoveal area of the visual field. The inter- and intra-subject variability in amplitude and localisation of the signals elicited by these targets was considered to be a reflection of the individual variations in relationship of visual field projections with the pattern of gyri and fissures on the proximal surface of the occipital lobe. The behaviour of component P90 of the onset response is consistent with a lateral origin in extrastriate visual cortex; that of P95 of the pattern reversal response is consistent in some respects with a striate cortical origin, but in others with a partial origin in extrastriate cortex.

The investigation of the sensitivity gradient of the visual field, as a function of stimulus dynamic range, in the normal and abnormal eye

The study utilized the advanced technology provided by automated perimeters to investigate the hypothesis that patients with retinitis pigmentosa behave atypically over the dynamic range and to concurrently determine the influence of extraneous factors on the format of the normal perimetric sensitivity profile. The perimetric processing of some patients with retinitis pigmentosa was considered to be abnormal in either the temporal and/or the spatial domain. The standard size III stimulus saturated the central regions and was thus ineffective in detecting early depressions in sensitivity in these areas. When stimulus size was scaled in inverse proportion to the square root of ganglion cell receptive field density (M-scaled), isosensitive profiles did not result, although cortical representation was theoretically equivalent across the visual field. It was conjectured that this was due to variations in the ganglion cell characteristics with increasing peripheral angle, most notably spatial summation. It was concluded that the development of perimetric routines incorporating stimulus sizes adjusted in proportion to the coverage factor of retinal ganglion cells would enhance the diagnostic capacity of perimetry. Good general and local correspondence was found between perimetric sensitivity and the available retinal cell counts. Intraocular light scatter arising both from simulations and media opacities depressed perimetric sensitivity. Attenuation was greater centrally for the smaller LED stimuli, whereas the reverse was true for the larger projected stimuli. Prior perimetric experience and pupil size also demonstrated eccentricity-dependent effect on sensitivity. Practice improved perimetric sensitivity for projected stimuli at eccentricities greater than or equal to 30o; particularly in the superior region. Increase in pupil size for LED stimuli enhanced sensitivity at eccentricities greater than 10o. Conversely, microfluctuation in the accommodative response during perimetric examination and the correction of peripheral refractive error had no significant influence on perimetric sensitivity.

Tactile sensing with fibre Bragg gratings and neural networks

Fibre Bragg grating sensors are usually expensive to interrogate, and part of this thesis describes a low cost interrogation system for a group of such devices which can be indefinitely scaled up for larger numbers of sensors without requiring an increasingly broadband light source. It incorporates inherent temperature correction and also uses fewer photodiodes than the number or sensors it interrogates, using neural networks to interpret the photodiode data. A novel sensing arrangement using an FBG grating encapsulated in a silicone polymer is presented. This sensor is capable of distinguishing between different surface profiles with ridges 0.5 to 1mm deep and 2mm pitch and either triangular, semicircular or square in profile. Early experiments using neural networks to distinguish between these profiles are also presented. The potential applications for tactile sensing systems incorporating fibre Bragg gratings and neural networks are explored.

The study of ultra-low energy deposition of hydrogenated amorphous carbon thin films

This thesis is dedicated to the production and analysis of thin hydrogenated amorphous carbon films. A cascaded arc plasma source was used to produce a high density plasma of hydrocarbon radicals that deposited on a substrate at ultra low energies. The work was intended to create a better understanding of the mechanisms responsible for the film formation, by an extensive analysis on the properties of the films in correlation with the conditions used in the plasma cell. Two different precursors were used: methane and acetylene. They revealed a very different picture for the mechanism of film formation and properties. Methane was less successful, and the films formed were soft, with poor adhesion to the substrate and decomposing with time. Acetylene was the better option, and the films formed in this case were harder, with better adhesion to the substrate and stable over time. The plasma parameters could be varied to change the character of films, from polymer-like to diamond-like carbon. Films deposited from methane were grown at low deposition rates, which increased with the increase in process pressure and source power and decreased with the increase in substrate temperature and in hydrogen fraction in the carrier gas. The films had similar hydrogen content, sp3 fractions, average roughness (Ra) and low hardness. Above a deposition temperature of 350°C graphitization occurred - an increase in the sp2 fraction. A deposition mechanism was proposed, based upon the reaction product of the dissociative recombination of CH4+. There were small differences between the chemistries in the plasma at low and high precursor flow rates and low and high substrate temperatures; all experimental conditions led to formation of films that were either polymer-like, soft amorphous hydrogenated carbon or graphitic-like in structure. Films deposited from acetylene were grown at much higher deposition rates on different substrates (silicon, glass and plastics). The film quality increased noticeably with the increase of relative acetylene to argon flow rate, up to a certain value, where saturation occurred. With the increase in substrate temperature and the lowering of the acetylene injection ring position further improvements in film quality were achieved. The deposition process was scaled up to large area (5 x 5 cm) substrates in the later stages of the project. A deposition mechanism was proposed, based upon the reaction products of the dissociative recombination of C2H2 +. There were large differences between the chemistry in the plasma at low and medium/high precursor flow rates. This corresponded to large differences in film properties from low to medium flow rates, when films changed their character from polymer-like to diamond-like, whereas the differences between films deposited at medium and high precursor flow rates were small. Modelling of the film growth on silicon substrates was initiated and it explained the formation of sp2 and sp3 bonds at these very low energies. However, further improvements to the model are needed.

Numerical modeling of holmium-doped fluoride fiber lasers

We combine all the known experimental demonstrations and spectroscopic parameters into a numerical model of the Ho3+ -doped fluoride glass fiber laser system. Core-pumped and cladding-pumped arrangements were simulated for all the population-bottlenecking mitigation schemes that have been tested, and good agreement between the model and the previously reported experimental results was achieved in most but not in all cases. In a similar way to Er3+ -doped fluoride glass fiber lasers, we found that the best match with measurements required scaled-down rate parameters for the energy transfer processes that operate in moderate to highly concentrated systems. The model isolated the dominant processes affecting the performance of each of the bottlenecking mitigation schemes and pump arrangements. It was established that pump excited-state absorption is the main factor affecting the performance of the core-pumped demonstrations of the laser, while energy transfer between rare earth ions is the main factor controlling the performance in cladding-pumped systems.

Low severity Fischer-Tropsch synthesis for the production of synthetic hydrocarbon fuels

Currently, the main source for the production of liquid transportation fuels is petroleum, the continued use of which faces many challenges including depleting oil reserves, significant oil price rises, and environmental concerns over global warming which is widely believed to be due to fossil fuel derived CO2 emissions and other greenhouse gases. In this respect, lignocellulosic or plant biomass is a particularly interesting resource as it is the only renewable source of organic carbon that can be converted into liquid transportation fuels. The gasification of biomass produces syngas which can then be converted into synthetic liquid hydrocarbon fuels by means of the Fischer-Tropsch (FT) synthesis. This process has been widely considered as an attractive option for producing clean liquid hydrocarbon fuels from biomass that have been identified as promising alternatives to conventional fossil fuels like diesel and kerosene. The resulting product composition in FT synthesis is influenced by the type of catalyst and the reaction conditions that are used in the process. One of the issues facing this conversion process is the development of a technology that can be scaled down to match the scattered nature of biomass resources, including lower operating pressures, without compromising liquid composition. The primary aims of this work were to experimentally explore FT synthesis at low pressures for the purpose of process down-scaling and cost reduction, and to investigate the potential for obtaining an intermediate FT synthetic crude liquid product that can be integrated into existing refineries under the range of process conditions employed. Two different fixed-bed micro-reactors were used for FT synthesis; a 2cm3 reactor at the University of Rio de Janeiro (UFRJ) and a 20cm3 reactor at Aston University. The experimental work firstly involved the selection of a suitable catalyst from three that were available. Secondly, a parameter study was carried out on the 20cm3 reactor using the selected catalyst to investigate the influence of reactor temperature, reactor pressure, space velocity, the H2/CO molar ratio in the feed syngas and catalyst loading on the reaction performance measured as CO conversion, catalyst stability, product distribution, product yields and liquid hydrocarbon product composition. From this parameter study a set of preferred operating conditions was identified for low pressure FT synthesis. The three catalysts were characterized using BET, XRD, TPR and SEM. The catalyst selected was an unpromoted Co/Al2O3 catalyst. FT synthesis runs on the 20cm3 reactor at Aston were conducted for 48 hours. Permanent gases and light hydrocarbons (C1-C5) were analysed in an online GC-TCD/FID at hourly intervals. The liquid hydrocarbons collected were analyzed offline using GC-MS for determination of fuel composition. The parameter study showed that CO conversion and liquid hydrocarbon yields increase with increasing reactor pressure up to around 8 bar, above which the effect of pressure is small. The parameters that had the most significant influence on CO conversion, product selectivity and liquid hydrocarbon yields were reactor temperature and catalyst loading. The preferred reaction conditions identified for this research were: T = 230ºC, P = 10 bar, H2/CO = 2.0, WHSV = 2.2 h-1, and catalyst loading = 2.0g. Operation in the low range of pressures studied resulted in low CO conversions and liquid hydrocarbon yields, indicating that low pressure BTL-FT operation may not be industrially viable as the trade off in lower CO conversions and once-through liquid hydrocarbon product yields has to be carefully weighed against the potential cost savings resulting from process operation at lower pressures.