Doctoral Degrees (Chemical Engineering)

Permanent URI for this collectionhttps://hdl.handle.net/10413/6657

Browse

Now showing 1 - 20 of 99

Dispersion in slowly moving fluids.
(1970) Te Riele, Wolter A. M.; Woodburn, Edward T.; King, R. P.
This work is concerned with the characterization of slowly moving fluids and was carried out on the flow of water through a narrow sedimentation tank. Dispersion in the type of flow structure involved is caused mainly by the presence of large eddies and, due to the fact that shear stresses are small, these eddies persist for a considerable period of time. Two flow models are presented : The first model assumes the X- Y- velocity component pair to form a discrete state Markov process in time and dispersion equations for the mean concentration at a point, the variance as well as concentration cross correlations are generated. In the second model the velocity fluctuation components are assumed to be independent, time-stationary Markov processes with normal probability density functions. The stochastic differential equation describing dispersion of tracer is formulated with and without the effect of molecular diffusion and solutions to both cases are presented. Comparison of the model with experimental data obtained from tracer and anemometer measurements show that the model is capable of describing mean dispersion in a relatively small region of the tank and that the tracer experiments were insensitive to molecular diffusion.
A study of the use of statistical turbulence parameters in correlating axial dispersion data in the central core of air flowing in a pipe.
(1970) Exall, Douglas Ian.; Woodburn, Edward T.
The longitudinal fluctuations at a point in the core of air flowing through a 15 cm. diameter pipe at a mean centerline velocity of 13.4 and 29.5 m/sec. were measured with a hot-wire anemometer. This signal, after analog to digital conversion, was stored in the form of digital samples on an ICT computer drum storage device. This method of data recording includes the effect of all eddy frequencies from DC upwards and the presence of large, slow eddies in the longitudinal direction became apparent in the subsequent autocorrelations. The longitudinal dispersion of a tracer material injected on the axis of the pipe was measured over short distances with pulses of approx. 20 msecs. duration of radioactive Krypton-85, detected at two downstream stations by small surface-barrier radiation detectors. By varying the separation of these two stations, an asymptotic mixing coefficient was established which was very much greater than the corresponding transverse mixing coefficient measured by other workers. The method proposed by Philip (4) for the prediction of the Lagrangian time autocorrelation from the Eulerian velocity measurements was examined with the correlation data of Baldwin and the data obtained in this investigation. The method applied to the unfiltered correlation data in the present measurements in a non-isotropic field to predict a longitudinal turbulent Peclet no. was found to predict a value in the region measured experimentally. When the present velocity data was filtered to remove the low-frequency components and give a turbulence intensity equal to that measured in a radial direction in previous dispersion measurements, the mixing coefficient predicted with Philip's method was found to agree very well with the transverse mixing coefficient reported in these investigations. A value is also suggested for the longitudinal Peclet number in the absence of the low frequency fluctuations.
A study of the cane sugar diffusion process.
(1972) Rein, Peter William.; Woodburn, Edward T.
A mathematical model has been developed to represent the extraction of sugar from sugar cane in the diffusion process. As a consequence of the structure of prepared cane, extraction is postulated to occur via two first order relations in parallel. The model was found to represent accurately the extraction behaviour observed in two different experimental configurations. Experiments on a laboratory scale were undertaken to elucidate the mechanism and basic rates of extraction. The results are consistent with the postulate that extraction takes place by a combination of washing and molecular diffusion. The diffusion process was simulated experimentally in a fixed-bed pilot plant diffuser. Model parameter values estimated from this data provide evidence of the effect of liquid hydrodynamics on the extraction process. These results have been interpreted in terms of liquid holdup, liquid velocities and the efficiency of liquid- solid contacting. It is shown how the model can be applied to full scale diffusers of the moving-bed type, which will enable prediction of performance and the choice of optimum operating conditions.
An investigation of the regeneration of carbonised catalyst pellets in a packed bed reactor.
(1973) Jager, Berend.; King, R. P.
No abstract available.
A study of the Fischer-Tropsch synthesis at elevated temperatures in a shock tube.
(1973) Kelly, Raymond James.; Woodburn, Edward T.
The shock tube was used to investigate the product spectrum of the initial stages of the Fischer-Tropsch synthesis carried out at elevated temperatures. Special attention was paid to the relationship between methane selectivity and temperature. The range of reaction environments studied are summarised below:- Reaction temperature 780 K - 1425 K. Reaction pressure 160 psia - 330 psia. Mean reaction time 628 u sec. - 727 u sec. Test gas composition - argon 81 - 87 mol. %. - hydrogen 6,5 - 9 mol. %. - carbon monoxide 6,5 - 9,5 mol.%. Catalyst type - fused iron, triply promoted. Catalyst loading - 0,12 - 0,14 mass catalyst / mass gas. The experiments were conducted in the incident shock region and quenching was achieved by the reflected rarefaction wave. Percentage conversion of hydrogen and carbon monoxide to useful products (hydrocarbons) varied between 0,1 and 2. Products detected in measurable quantities were methane, ethylene, ethane and propylene. The theory of shock tube wave propagations through heterogeneous medi a was studied in detail and unique theory developed for handling conditions of varying temperature and pressure. This enabled characterisation of the reaction environment so that multilinear regression could be used to find a correlation between H2 + CO consumption and system variables. Major information gleaned on the initial stages of the Fischer-Tropsch synthesis at elevated temperatures was; (i) contrary to observed trends under normal synthesis conditions, methane selectivity decreased and propylene selectivity increased with increasing temperature; (ii) the process appeared to be hydrogen adsorption, pate controlled; (iii ) molecular degradation processes played a negligible part in the format ion of final reaction products, and (iv) oxygen compounds, such as methanol, did not appear to be important intermediate products. It has been shown that the heterogeneous shock tube offers a possible means of obtaining initial reaction rate data for highly complex systems.
Changes in the chemical composition of sugar cane (Saccharum officinarum) during storage.
(1973) Bruijn, Jacob.; Sutton, Donald A.; Matic, M.
An outline is given of the South African sugar industry, with particular emphasis on the unit operations which make up the industrial process for manufacturing sugar from cane. Current knowledge of the chemistry of soluble polysaccharides is reviewed and the structures of several polysaccharides, including starch, dextran, and pullulan, are discussed. It has been found that changes take place in the chemical composition of the juice in sugar cane (Saccharum officinarum) during post-harvest storage. With increasing storage time, there is a proportional decrease in the starch content of the juice, and a considerably larger proportional increase in the soluble polysaccharide content. The increased polysaccharide content was found to be due to a single glucan which, contrary to most previous publications on this subject, is definitely not a dextran. Following structural analysis, it has been established that the polysaccharide formed in stored cane had not been described before and the name "sarkaran" , derived from the Sanskrit word "Sarkara", meaning "sugar" is proposed for it. The polysaccharide was isolated from cane juice by precipitation with ethanol after the starch in the juice had been removed by centrifugation. The polysaccharide was purified by repeated dissolution in water and reprecipitation with ethanol. Analysis by gel chromatography resulted in a single symmetrical peak, indicating that the isolated polysaccharide is homogeneous. This was confirmed by hydrolysing fractions representing a section of the ascending and a section of the descending part of the peak of the chromatogram, using the enzyme pullulanase. Chromatographic separation and quantitative analysis of the isolated oligosaccharides showed that the compositions of the two enzymes digests were identical. Acid hydrolysis of the polysaccharide resulted in a single hexose. This was identified as glucose by paper chromatography, comparing the Rf value with that of pure glucose. Confirmation was obtained by comparing the osazone with that of glucose, using microscopic examination and determination of the melting points. Paper electrophoresis showed the molecule to be uncharged. Several techniques, both absolute and non absolute, were used to determine the molecular weight of the polysaccharide. A method involving viscosity determination indicated a molecular weight of 34 000 while a figure of 50 000 was obtained by gel chromatography on a Sephadex column, comparing the peak elution volume of the polysaccharide with that of dextrans of a defined molecular weight. Both these techniques are non absolute and yield rough estimates of the molecular weight. Osmometric measurement, an absolute method, showed the number average molecular weight to be 51 500. An absolute value for the weight average molecular weight of 250 000 was obtained by light scattering techniques. Data from the light scattering experiments were also used to determine a value of 200 - 250 A for the radius of gyration RG of the polysaccharide. End group analysis after exhaustive methylation resulted in a value of 24 000 for the number average molecular weight Mn. This indicates either that some degradation of the polysaccharide molecule occurs , during the methylation procedures or that there is a certain degree of association between individual molecules. Periodate oxidation showed that 32 percent of the glucosidic linkages are in ( 1 + 6 ) position. The polysaccharide was exhaustively methylated by several Haworth methylations followed by a number of Kuhn methylations. The fully methylated product was methanolysed and the methyl glucopyranosides analysed by gas liquid chromatography. The results were compared with those obtained from fully methylated starch and dextran. From the absence of disubstituted methyl derivatives in the methanolysate it was concluded that the polysaccharide is an unbranched glucan. From the quantities of Methyl 2,3,4,6 tetramethyl-O-Dglucopyranoside, Methyl 2,3,6, trimethyl-O-D-glucopyranoside and Methyl 2,3,4? trimethyl-0-D-glucopyranoside, it was concluded that the only linkages in the glucan are ( 1 + 4 ) and ( 1 + 6 ) and that these are present in the ratio 68:32. Enzymic hydrolysis, using pullulanase, was followed by paper chromatographic separation. Quantitative determination of the oligo-saccharides present in the enzyme digest resulted mainly in two oligosaccharides, maltotriose and maltotetraose, in nearly equal proportions. For this reason it was postulated that the polysaccharide is a maltotriose-maltotetraose polymer, and that the individual units are linked in ( I + 6 ) position, a linkage for which pullulanase is specific in certain configurations. The sequence of the maltotriose and maltotetraose units in the polymer has not been investigated further, although this could be carried out by partial acid hydrolysis, followed by isolation and identification of the various oligosaccharides formed. An alternate method for the determination of the sequence of the monomers is discussed. It was subsequently shown that the linkages in the polysaccharide are in the a configuration. The polysaccharide is highly dextra rotary and the magnitude of the rotation is comparable to that of other polysaccharides linked in a position, . such as starch and dextran. Infrared spectroscopy was used to confirm the configuration. The spectrogram of the polysaccharide contained an absorption peak at 840 cm-1 , which is typical of the a-anomeric absorption occurring, for example, in the IR spectrum of starch. The spectrogram exhibited no absorption peak at 891 cm-1 , the wavelength typical of the B-anomeric absorption in the IR spectrum of cellulose. In addition, it was found that all polysaccharides containing a ( 1 + 4 ) linkages show an absorption peak at 700 cm 1. This absorption peak was absent in all IR spectra obtained from various dextrans. This phenomenon has not been reported previously and it is suggested that the presence of this absorption peak in the IR spectrum of a glucan can be used to support the evidence of the presence of a( 1 + 4 ) linkages. It was not possible to correlate the formation of the polysaccharide with the occurrence of a specific micro organism. It is suggested that the formation of the polysaccharide is the result of enzymic reactions in the sugar cane after harvesting. The investigation of the composition of juices from deteriorated cane has not been confined to polysaccharides. Ethanol has been isolated from the juice of some samples of stored cane which had been burnt before harvesting. The ethanol was isolated by fractional distillation and identified by measurement of the boiling point. It was confirmed, by the formation of the molybdate-xanthate complex, that the product isolated was an alcohol. The identification was further confirmed by oxidising the ethanol to acetic acid and proving the identity of the acids by paper chromatography. It has been shown that, with the exception of two acids, the carboxylic acid composition of cane juice remains unaltered during post-harvest storage of the cane. The two exceptions , succinic and aconitic acids, were identified from their melting points and by specific spot tests. Ion exchange was used to isolate the acids from the juice. The eluate from the ion exchange column was concentrated and the acids separated by liquid-liquid chromatography, using a silica gel column. The levels of both aconitic and succinic acids were found to increase during the early period of storage but decreased again slowly thereafter. The percentage change was greater in the case of succinic acid, although aconitic acid was the most abundant carboxylic acid in the juice. Lactic acid was absent from the cane juices analysed. This is surprising, as lactic acid is a common product of the metabolism of carbohydrates by micro organisms. It is suggested that the changes in acid composition during the storage of harvested cane are caused by deactivation of enzymes of the Krebs cycle. Post-harvest deterioration of sugar cane can have serious consequences which can affect the whole Sugar Industry. Not only is crystallisable sugar lost but the products of the deterioration have adverse effects on factory processing and laboratory analysis. The problem, which will become more acute with the introduction of mechanical cane harvesting, can only be resolved through the cooperative efforts of all the parties concerned.
The predictive modelling of the performance of a crude oil fractionator in terms of gas chromotographic characterization.
(1973) Van Zyl, Ockert Jacobus.; Judd, Michael Robin.
No abstract available.
Kinetic and mechanistic studies on the biological and chemical leaching of nickel from sulphide ores.
(1974) Corrans, Ian James.; Scholtz, M. T.; Kistner, A.
The aim of this investigation was to extend the knowledge of the bacterial leaching of copper and zinc sulphides into the area of nickel sulphide leaching. By far the major portion of both theoretical and practical expertise which is available in the field of bacterial leaching is based on the treatment of copper and zinc sulphides. As yet there is little information available on the bacterial leaching of nickel sulphides to meet the growing interest in this field both in South Africa and elsewhere. To a large degree, it was necessary to start from basic principles in this novel extension of bacterial leaching technology so that the work covers a fairly wide field in general rather than one particular aspect in detail. A strain of 'nickel adapted' Thiobacillus ferrooxidans was isolated from the tailings dam of a disused nickel mine. The growth characteristics of this strain were studied in some detail on sulphur using both batch and continuous techniques. This was done as it was considered that growth on sulphur would provide useful information which could be correlated with the mineral leaching results. The mineral pentlandite (NiFeS2) was chosen as the one with which to work because of its economic importance. This was prepared in a highly purified form from a concentrate of the Rhodesian Shangani deposit. Bacterial leaching tests in both batch and continuous operation were then carried out in order to define the effects of various physico-chemical parameters on the leaching of nickel from this mineral. As a preliminary to these tests, a detailed chemical kinetic study in the absence of bacteria of the leaching of nickel was carried out using similar physico-chemical conditions. The results of the bacterial and chemical leaching tests were then compared and used to postulate a mechanism and model for the process. It was found that the rate of leaching of nickel from pentlandite in acid ferric sulphate solutions was directly proportional to the concentration of ferric ions and speed of agitation of the stirrer and to the square root of the oxygen concentration. The form of the rate expression was interpreted in terms of a mixed diffusive and chemical rate controlling mechanism. Bacterial growth rates on flowers of sulphur were found to be controlled by the rate of dissolution of oxygen from the gas bubbles into the bulk solution. When this latter condition was made non-rate limiting, it was found that growth rates were still dependent on the rate of agitation, implying mass transport control by another mechanism. The batch bacterial leaching results showed a linear pattern of nickel leaching and bacterial growth, with a marked dependence on oxygen concentration and rate of agitation. A mechanism in accordance with the batch data was postulated, which proposed that the rate of bacterial leaching was proportional to the concentration of bacteria attached at the mineral surface and to the square of the oxygen concentration. The rates of bacterial leaching were computed by taking the difference between the overall measured leach rate and the chemical leach rate based on the chemical kinetic data. The leach rates in continuous bacterial leaching were higher than those predicted from the batch data. This effect was interpreted in terms of higher specific growth rates being achieved in continuous operation. An economic assessment was made of the process based on the optimum leach rates obtained in continuous leaching and found to show some promise.
A study of an on-line recursive filter applied to a milling circuit.
(1975) Barker, Ian James.; King, R. P.; Everson, R. C.; Dawson, M. F.
No abstract available.
Industrial application of flotation models.
(1975) Cramer, Larry Arthur.; Woodburn, Edward T.
No abstract available.
The prediction of multicomponent ion exchange equilibria with particular reference to the system involved in the recovery of uranium.
(1976) Smith, Robert Pitts.; Woodburn, Edward T.
The problem of predicting the general ion exchange equilibria pertaining to systems of industrial significance is generated by the multicomponent nature of such systems and the nonidealities which may be present in both the solution and exchanger phases. A general framework applicable to multicomponent systems incorporating nonideal effects in both phases is presented. For the solution phase a well established procedure for calculating activity coefficients is adopted. Deviations from ideal behaviour in the exchanger phase are modelled by the Wilson equation, which expresses the excess Gibbs free energy of mixing of the resinates as a function of composition. A Significant advantage is afforded by this equation in that theoretically a multicomponent system may be predicted from the binary interaction coefficients of this equation which are determined experimentally, thereby reducing the otherwise extensive experimental program. These ideas are applied to systems of increasing complexity from simple binary characterisation experiments to the prediction of a six component system related to that encountered in the recovery of uranium from sulphuric acid leach liquors. Experimentation for the systems involving the ions S04 2-, Cl- and NO-3 and a strong base anion exchanger have provided a severe test for the procedure proposed. The agreement between the predicted and experimental resin phase composition data for this ternary system is within ± 5%. The addition of complexing agents complicates the procedure in that it becomes physically impossible to decompose the system into the desirable experimental binary systems. In this case higher order systems are characterised. Introducing a mineral acid to the ternary system discussed previously generates the HSO-4 ion which necessitates the characterisation of other ternary systems before the quaternary system may be predicted. The agreement between the predicted and experimental resin phase composition for the quaternary system is shown to be within ± 10%. The work is easily extended to include the more complex systems generated by the complexation of metal ions with the various ligands present. .Provided the stoichiometry of the complex species in the exchanger phase is well defined the complexes present no difficulties in the characterisation procedures. Experimental studies on the acidic uranyl sulphate quaternary system provide the desired ion exchange equilibrium constants and the interaction coefficients. In order that the interaction coefficients for the ion pairs such as UO2 (SO4)2-2, Cl- and U02 (SO4)2-2, NO-3 may be estimated it is necessary to characterise two quinary systems. Nevertheless the characteristics of lower order systems are employed to reduce the number of unknown parameters. Finally it is possible to predict the resin phase composition of the six component system which results from chloride and nitrate species being included in the acidic uranyl sulphate system. The quantitative effects of all the components in the solution phase on the extent of uranium loading are predicted. Although the ferric ion is an important component in the industrial situation this ion has been excluded from this work because at this stage it is not possible to identify or measure the quantity of the various ferric complexes present in the resin phase for a particular solution condition.
A kinetic study of the dissolution of natural and synthetic sphalerite in aqueous sulphuric acid and in acidic ferric suplhate media.
(1977) Verbaan, Bernard.; Woodburn, Edward T.
Four sphalerites (synthetic, high grade natural, moderately impure flotation concentrate and highly impure flotation concentrate) were leached in acid sulphate media without and with ferric ions present under the following conditions :- Case (i) [Fe3+]o : [H2S04]o = 0,0 Case ( ii) [Fe3+]o : [H2S04]o = 1,8 Case (iii) [Fe3+]o : [H2S04]o = 0,1 Extensive data for leaching under these conditions are tabulated. Kinetic mechanisms based on Langmuir-Hinschelwood adsorption theories were proposed, and leaching models were developed for different assumed rate limiting steps. The initial rate and overall forms of the models were tested using experimental data.Leaching under case (i) conditions Non-oxidative dissolution took place with Zn2+ and H2S the predominant reaction products. The H2S partial pressure was monitored continuously and solution samples were taken for analysis at discrete time intervals. Vibratory (i.e. attrition) milling eliminated very large differences observed in the leaching characteristics of course size fractions of the natural sphalerites. The initial rate form of a model based on a dual site reaction mechanism and on either H+ adsorption or reaction product desorption rate control was found to fit the data for the synthetic and vibratory milled forms of sphalerite. The most impure vibratory milled sphalerite adsorbed Zn2+ and H2S very strongly, and this resulted inproduct desorption rate control. Vibratory milled forms of the high grade natural sphalerite and the moderately impure flotation concentrate, exhibited virtually identical initial rate dissolution kinetics, despite large differences in their chemical compositions. Leaching under case (ii) conditions Oxidative dissolution took place with Zn2+ and elemental sulphur the predominant reaction products. Scanning electron microscope photographs of leached and unleached particles showed the sulphur present on the particle surface. These photographs, and optical microscope photographs of etched polished sections, showed that dissolution took place in a complex way. A model based on ferric ion adsorption as the rate limiting step was proposed and confirmed experimentally. The model demonstrated a proportional dependency of the rate on the area and ferricion concentration, and an inverse dependency on the hydrogen ion concentration. For a -90,0 + 63,0 um size fraction, the three natural sphalerites exhibited virtually identical dissolution rates per unit area. The effect of ball milling or vibratory milling the sphalerites fine, was to increase the rate per unit area for the most impure natural sphalerite but decrease the rate per unit area for the high grade natural sphalerite.It was shown that for course size fractions of sphalerite, the most impure sphalerite which leached slowest under case (i) conditions (i.e. adsorbed H+ poorly) leached fastest under case (ii) conditions (i.e. adsorbed Fe3+ strongly). The reverse was true for the high grade natural sphalerite. Except in the case of synthetic sphalerite leaching under case (i) conditions, no correlation was shown to exist between the way the B.E.T. measured area changed, and the way the calculated active area changed during leaching. Leaching under case (iii) conditions Oxidative and non-oxidative dissolution, as well as H2S oxidation by Fe3+ occured simultaneously. The extents to which oxidative or non-oxidative dissolution occured could be explained in terms of the hydrogen ion and ferric ion adsorption characteristics of the sphalerites.The ferric ion oxidation of H2S was studied in the absence and presence of solids, and the presence of sphalerite or activated charcoal catalysed this reaction. No advantage was gained by leaching in the presence of activated charcoal with or without Fe3+ present, unless conditions were such that H2S was formed as a product of reaction.
An investigation of a supersonic chemical reactor.
(1977) Flemmer, Rory Loiveig Christian.
The development of a supersonic chemical reactor is examined. The central concept of such a reactor is that gases can be expanded to supersonic speed which will result in a decrease in their static temperature to the extent that no reaction will occur when the gases are mixed together. After mixing has occurred the mixture can then be passed through a standing shock to raise this temperature very rapidly, thus controlling the product spectrum. The main areas of interest for such a reactor, namely the establishment of a shock and the mixing of two gases at supersonic speed are examined.
Formulation and application of a dynamic model for atmospheric point sources.
(1977) Mulholland, Michael.; Scholtz, M. T.; Woodburn, Edward T.
No abstract available.
A study of processes occurring in flotation froths.
(1979) Moys, Michael Hugh.; Dawson, M. F.
While the significant effect that the froth phase has on the performance of a flotation operation has recently been widely recognised, little work has been published which promotes an understanding of the physical processes occurring in the froth phase. A more intimate understanding of these processes and their relative importance and interactions would lead to a more rational design of froth chambers and froth removal methods, with resultant improvements in flotation plant performance. In pursuit of this understanding the following investigations were performed: (1) In a specially designed cell the variation of mineral grade with height above the froth-slurry interface - as affected by gas rate, frother concentration, the presence of froth baffles (to minimise mixing) and final product removal rate - was measured. A mathematical model was formulated to assist in the interpretation of this data. (2) The residence time distribution of a 2-phase froth (air and water without solid particles) was measured as a function of froth height, gas rate and frother concentration. Small polystyrene balls were used as a tracer. The results were interpreted using two theoretical models: (a) a streamline model which involved the solution of the 2 - dimensional Laplace equation for frictionless flow of froth in the froth chamber; (b) a semi-phenomenological model which relates the froth residence time distribution to cell dimensions, gas rate and froth stability. On the basis of insights gained in these investigations, a number of objectives which should be met by froth removal methods were formulated: (1) froth removal efficiency should be maximised, i.e. there should be no stagnant zones in the froth; (2) froth stability should be optimised; (3) the minimum residence time of froth elements in the froth phase should be maximised. One novel method of froth removal was designed, and this and a number of other methods of froth removal were compared experimentally. It was found that substantial improvement in flotation performance could be obtained by (1) sprinkling the froth with water, which removed entrained particles and improved froth stability; (2) inserting a baffle in the froth phase near the concentrate weir which increased the minimum residence time of a froth element in the froth phase, thereby improving the drainage of entrained particles from the froth; and (3) placing a froth scraper near the back of the cell, thus ensuring that no stagnant froth zones developed.
An investigation into the Fischer-Tropsch synthesis over supported ruthenium in a stirred gas-solid reactor.
(1979) Kirk, Alan Robert McKenzie.; Everson, R. C.
A stirred gas-solid reactor has been used to study the Fischer-Tropsch synthesis over a supported ruthenium catalyst. Detailed catalyst characterisations were conducted on both fresh and used catalyst samples and included such techniques as physisorption and chemisorption, electron microprobe examination, carbon determination and mercury porosimetry. The product obtained from the synthesis consisted largely of n-alkanes and the product selectivity of individual carbon number fractions is demonstrated to have certain advantages over the corresponding selectivities (adapted from the literature) obtained over iron, cobalt and molybdenum catalyst. The rate data obtained is fitted to two empirical models: a simple power law model and a previously-described mechanistic model. Energies of activation for methanation and carbon monoxide removal (at approximately 34,3 and 19 kcal/mol respectively) are compared with regressed values obtained by other workers. Rate data for the synthesis of higher hydrocarbons are fitted to a simple power law model and energies of activation for individual carbon number fractions up to C8 are presented.
Design and operation of a multistage pressurized fluidized bed combuster.
(1981) Eleftheriades, Christos Mimi.
A three-stage Pressurized Fluidized Bed Combustor (PFBC) of principal dimensions, O,4Sm internal diameter by4m high was designed and fabricated to burn South African coals, with particular reference to coals unsuitable for burning in conventional boilers. The combustor which is the first of its kind and probably one of very few operational PFBCs in the world, was made of three jacketed sections positioned vertically one above the other and bolted together at the flanges. Distributor plates were located at the flanges which gave the combustor a multistage capability. A three, two, or one deep Fluidized Bed (FB) configurations were possible by removing the interstage distributors. Interstage solids circulation was made possible by the use of downcomers transporting solids downwards between the FBs. The solids were returned to the top FB using a pneumatic conveyor. The design of the PFBC was a sequence to a series of experimental and theoretical investigations which were carried out in order to provide us with the necessary PFBC design parameters. These investigations dealt with the following areas of research: (a) the development of a new type of cyclonic tuyere capable of transmitting through it high quantities of solids with the fluidizing gas, without choking, (b) the transfer and control of the downward flow of solids through downcomer pipes, (c) the control of the circulation of solids in a Circulatory system using a non-mechanical solids flow control valve, (d) the development of a new type of start up burner which could operate immersed under the solids, and (e) the combustion of coal in a small FB under batch conditions and the study of reaction kinetics of South African coals. On the basis of the results of the investigation in these research areas and the findings of research of individuals and of .organizations working in the field of fluidization technology the PFBC was designed, built, and successfully commissioned. A series of 12 runs, with each run lasting between 2 and 8 days, totalling more than 1500 hours, were carried out on the PFBC. Char and coal with ash content between 30 and 70 per cent were burnt in the combustor using various combinations of feeding ports and number of FBs. System pressures ranged between atmospheric and 6 bar(abs). For some of the runs the reactor was operated in a counter-current mode with solids and combustibles descending against the upflowing fluidizing air in order to study the effect that counter-current flow had on the efficiency of combustion. The combustion trials showed that the two-FB combustor, operated preferably without solids circulation, with the bottom FB acting as the main combustion cell and the top FB as a smuts burn-out cell, proved to be the most practical and most suitable combustor for burning South African high ash coals and fines or, in general, any low-grade carbonaceous materials of any size. With this configuration combustion efficiencies of up to 99 per cent, based on the combustibles in the feed and the ash, were achieved. The department computer (COC1700) was successfully linked with the PFBC for real time data logging and data processing. A mathematical model which was based on our research findings and the work of T.P. Chen and S.C. Saxena, C. Fryer and O.E. Potter, and D. Levenspiel was successfully developed and applied to the twoFB PFBC. The model describes the devolatilization and combustion of coal particles in the FB in accordance with a shrinking core type model and uses a population balance over all particles for the overall mass balance. The results from this model, which was put onto the computer, compared favourably with the experimental results and the model can be confidently used to predict the behaviour of the PFBC. It can also be easily adapted for use on any other single or multifluidized bed reactors provided that the assumptions made for the derivation of this mathematical model still hold. A mathematical model based on the work of H.C. Hottel and A.F. Sarofim, and L. Wender and G.T. Copper was also developed. This model describes the transfer of heat from the FB to the cooling coils using a stepwise heat and mass balance along the length of the cooling coil. Although this mathematical model was developed specifically for the cooling coils of our combustor it is strongly believed that it can also form the basis of a general purpose model.
Gas residence time testing and model fitting : a study of gas-solids contacting in fluidised beds.
(1984) Dry, Rodney James.
This work is concerned with the effect of vessel geometry on the hydrodynamics of fluidisation of a bed of milled iron oxide. The effect of going from a cold model representative of a typical pilot plant reactor to one simulating a semi-commercial unit is quantified, and various reactor internal configurations on the latter are evaluated. The experimental approach is one based on residence time testing and model fitting with parameter optimisation. A model screening aimed at identifying the most reasonable modelling approach is included, and altogether seven models in two categories are formulated and solved in the dynamic mode. Three of these models are considered novel at present, along with the dynamic solutions to two of the others. The residence time technique involves methane as an inert tracer in air, and continuous analysis of gas withdrawn from the bed via sample probes by a pair of flame ionisation detectors. The process stimulus is governed by a pseudo-random binary sequence, and correlation analysis is employed for noise reduction. A Fourier transform routine, developed from first principles, converts a pair of correlation functions to a process frequency response, and model predictions are compared with the experimental data in this form. Two parameters per model are fitted, and the residual error at the optimum parameter combination provides a means of identifying the best-fitting model. The optimised parameters of this model are regarded as estimates of those of the actual process. Five models compete in the first screening category. Four of these have appeared in the literature in one form or another, and the fifth is novel in that it accounts for axial mixing in the bubble phase by employing multiple plug flow units. This model, referred to as the multiple bubble-track or MBT model, is shown to fit the experimental data better than any of the other models in both bubbling and slugging systems. This suggests that employing multiple plug flow units in parallel for the bubble phase is mechanistically more correct than employing a single plug flow unit. The second screening category is related to the situation in which gas is sparged into an already fluidised bed at some height above the main distributor. The two models in this category are both considered novel, and describe opposite extremes of possible behaviour in one particular sense: one assumes rapid coalescence between grid and sparger bubbles, and the other none at all. The laterally segregated bubble phase or LSBP model emerges as the better process description.The formulation of this model suggests that physically, bubbles from the sparger tend to retain their identity as they pass through the bed. Crossflow ratios estimated on the basis of the best-fitting model in each category point to the existence of a very strong scale-up effect. From the shape of the crossflow profiles it appears that most of the interphase mass transfer occurs in the bottom meter or so of the bed, and it is suggested that grid design is the most significant controlling factor. The presence or otherwise of vertical coils in the bed is shown to have no significant effect on crossflow, and mass transfer between sparger bubbles and the dense phase is shown to be similar to that between grid bubbles and the dense phase. Finally, it is demonstrated that the axial crossflow profile in the bubbling bed is consistent with the concept of an axially invariant mass transfer coefficient based on bubble to dense phase interfacial area.
Zonal separation and solids circulation in a draft tube fluidized bed applied to coal gasification.
(1984) Rudolph, V.; Judd, Michael Robin.
In this thesis a fluidized bed containing a draft tube has been studied with the aim of developing the apparatus for coal gasification. The process has the capability of producing synthesis quality gas using air for combustion, and of being able to accomodate poor quality coal feeds containing heavy fines loads. These advantages arise from two special features of a draft tube fluidized bed. In the first place, the bed may be operated as two separate and independent reaction zones, one contained within the draft tube and the other in the annulus region surrounding it. As a result, the gasification reactions may be carried out in one compartment and the combustion reactions in the other, allowing the useful gasification products to be taken off separately and undiluted with the combustion flue gases. Secondly, the fluidized material in the bed may be induced to circulate up the draft tube and down the annulus. These circulating solids provide the heat carrier from the combustion to the gasification zones within the bed. Furthermore, circulation of the bed in this way leads to a much longer residence time of fine particles within the bed and results in a high fine coal utilization efficiency. In order to achieve these benefits in practice, it is necessary to separate the gases supplied to and emitted from the draft tube from those of the annulus, but at the same time allowing free movement of solids between these regions. The thesis deals with how this may be accomplished in three parts: Firstly, the principles underlying division of a fluidized bed with a draft tube into discrete reaction zones are formulated, and strategies for achieving zonal separation, based on these arguments, are experimentally tested. As a result a reactor configuration and operating conditions suitable for coal gasification have been empirically identified. Secondly, a model describing the bulk circulation of solid material in the bed is presented, for the draft tube operating in the slugging mode. This model allows the average solids residence time and the particle velocities in the annulus and draft tube to be predicted, provided that slug velocities and spacings are known. The necessary correlations between hydrodynamic behaviour and the system properties are available in the literature for round nosed and wall slugs, but not for square nosed slugs, which appear to be characteristic in the apparatus used here. The third part consequently examines the square nosed slugging regime, and a theory to describe this behaviour, based on interparticle stress analysis, is presented. This regime is identified as having significant advantage over other bubbling modes because of the high dense phase gas flow rates which are sustained, and the resulting improved gas-solid contacting. The three models together mathematically describe the operation of the draft tube fluidized bed, allowing gas partition between the annulus and the draft tube regions as well as solids circulation to be predicted, for different bed configurations and operating conditions. The predictions compare well with experimental results. The last part of the thesis deals with the application of the system to coal gasification on a one ton coal per day pilot plant. A high quality gas, containing up to 80% CO + H2, (balance CO2), has been produced by steam gasification in the draft tube, using air for the combustion reaction in the annulus. The H2/CO ratio can be varied from about 1 to 3, by changing the operating temperature of the reactor.
A real time fluorescent particle counter for atmospheric dispersion studies.
(1985) Davey, William Lewis Errol.; Scholtz, M. T.; Woodburn, Edward T.; Raal, Johan David.
No abstract available.

Browse

Browsing Doctoral Degrees (Chemical Engineering) by Issue Date