Research Doctorate - Doctor of Philosophy (PhD) / An increased demand worldwide for the reduction in pollutants emitted from coal-fired power stations has meant that advanced coal utilisation technologies are being sought as alternatives to pulverised fuel (pf) fired plants. The leading systems use coal gasification to produce a fuel gas which is cleaned and used in a combined-cycle gas turbine system. This produces electricity at high efficiencies and with significant reductions in the emissions of CO2, N- and S- gases and particulates. These systems offer the emission levels approaching those of natural gas combined-cycle plants, with the low fuel cost of coal. Modern coal gasification technologies operate with high temperatures and at pressures many times that of pf boilers: the reliability and efficiency of gasification-based systems are strongly influenced by the performance of the coal used under these conditions. The high-intensity nature of these processes means that generating experimental coal performance data for coal assessment and reactor design is time consuming, expensive or even impossible due to the lack of suitable facilities. Using process models based on fundamental gasification phenomena, coal performance can be predicted over a range of conditions. This is beneficial for both the development of new gasification technologies and in the assessment of Australian coals for use in the evolving international market. The slowest stage of the coal gasification process, i.e. the conversion of the char, has been identified as an important parameter for the design and implementation of such models. In particular, the intrinsic reactivity—characterised by data measured under conditions where chemical processes alone control the reaction rates—is extremely important, as intrinsic data can be readily combined with char physical properties to predict the high-temperature reaction rates of coal chars. The lack of intrinsic data generated at pressures relevant to modern gasification systems has meant that kinetic input into process models has been somewhat unreliable. In particular, there are no high-pressure reactivity data—intrinsic or otherwise—available for Australian black coals. To address this need, work in this thesis has used a pressurised thermogravimetric analyser to measure the effects of pressure (up to 30 atm) on the intrinsic reactivities to O2, CO2 and H2O of several Australian black coal chars, at emperatures between 350 and 900°C. These chars were made under a range of pressures and heating rates, and were in the size range 100 μm to 1.0 mm. In particular, the experiments were performed under conditions where chemical processes alone controlled the reaction rates, and where inhibition of the respective reactions by the products was negligible. It was found (using chars made in bulk at atmospheric pressure with slow heating rates) that whilst the reaction rate increased with reactant pressure in all gases, at pressures above approximately 15–20 atm the rates of reaction with CO2 and H2O ceased to increase with pressure. There was no such observation for the char–O2 reaction up to 16 atm. Activation energies of the reactions were unaffected by pressure. Samples made at high pressures and with high heating rates were found to be orders of magnitude more reactive than the chars made at atmospheric pressure under slow heating rates. These differences were found to be largely due to an increased microporous surface area, such that the intrinsic reactivities (calculated using the CO2 adsorption surface area) were similar. The effects of variations in pyrolysis pressure and parent coal petrography (and consequently char morphology) were also largely accounted for by char surface area, such that intrinsic rates were not greatly affected by these variables. These intrinsic reaction rate data were examined to produce a modified version of the nth order rate equation. This incorporated a pressure order that decreased as the reactant pressure increased, based on the physical process of available surface saturation. This model was compared with measured data and it was shown that the predictive capability of the nth order rate equation over a range of pressures was improved. There is scope for further refinement of this model by investigating the effects of reactant pressure on the development of the surface area of the char during conversion, since it was shown in this work that pressure has a strong effect on the such development. Moreover, this effect of pressure was not consistent between reactant and coal char type. This kinetic model was combined with measured char properties such as surface area, pore size, etc. to crudely predict the high temperature reactivity of a sample. This demonstrated the usefulness of reliable intrinsic data in the development of high temperature gasification models, and highlighted the need for experimental data obtained under process conditions of high temperature and pressure that can be used to validate such models.

This investigation provides a comprehensive experimental dataset and kinetic model for biomass gasification, over a wide temperature range (1150-1350 °Ï¹) in CO2, H2O and the combination of these two reactant gases over the mole fraction ranges of 0 to 0.5 for H2O and 0 to 0.9 for CO2. The data come from a unique experimental facility that tracks continuous mass loss rates for poplar wood, corn stover and switchgrass over the size range of 6-12.5 mm. In addition, the data include char size, shape, surface and internal temperature and discrete measurements of porosity, total surface area, pore size distribution and composition. This investigation also includes several first-ever observations regarding char gasification that probably extend to char reactivity of all types and that are quantified in the model. These include: the effect of ash accumulation on the char surface slowing the apparent reaction rate, changes in particle size, porosity and density as functions of burnout, and reaction kinetics that account for all of these changes. Nonlinear least-squares regression produces optimized power-law model parameters that describe gasification with respect to both CO2 and H2O separately and in combination. A single set of parameters reasonably describes rates for all three chars. Model simulations agree with measured data at all stages of char conversion. This investigation details how ash affects biomass char reactivity, specifically the late-stage burnout. The ash contents ratios in the raw fuels in these experiments are as high as 40:1, providing a clear indication of the ash effect on the char reactivity. The experimental results definitively indicate a decrease in char reaction rate with increasing initial fuel ash content and with increasing char burnout -- most pronounced at high burnout. This investigation postulates that an increase in the fraction of the surface covered by refractory material associated with either higher initial ash contents or increased burnout decreases the surface area available for reaction and thus the observed reaction rate. A quantitative model that includes this effect predicts the observed data at any one condition within the data uncertainty and over a broad range of fuel types, particle sizes, temperatures, and reactant concentrations slightly less accurately than the experimental uncertainty. Surface area, porosity, diameter, and density predictions from standard models do not adequately describe the experimental trends. Total surface area increases slightly with conversion, with most of the increase in the largest pores or channels/vascules not measurable by standard surface area techniques but most of the surface area is in the small pores. Porosity also increases with char conversion except for abrupt changes associated with char and ash collapse at the end of char conversion. Char particle diameters decrease during these kinetically controlled reactions, in part because the reaction is endothermic and therefore proceeds more rapidly at the comparatively warmer char surface. SEM images qualitatively confirm the quantitative measurements and imply that the biomass microstructure does not appreciably change during conversion except for the large pore diameters. Extant char porosity, diameter, surface area, and related models do not predict these trends. This investigation suggests alternative models based on these measurements.

This investigation provides a comprehensive experimental dataset and kinetic model for biomass gasification, over a wide temperature range (1150-1350 °Ï¹) in CO2, H2O and the combination of these two reactant gases over the mole fraction ranges of 0 to 0.5 for H2O and 0 to 0.9 for CO2. The data come from a unique experimental facility that tracks continuous mass loss rates for poplar wood, corn stover and switchgrass over the size range of 6-12.5 mm. In addition, the data include char size, shape, surface and internal temperature and discrete measurements of porosity, total surface area, pore size distribution and composition. This investigation also includes several first-ever observations regarding char gasification that probably extend to char reactivity of all types and that are quantified in the model. These include: the effect of ash accumulation on the char surface slowing the apparent reaction rate, changes in particle size, porosity and density as functions of burnout, and reaction kinetics that account for all of these changes. Nonlinear least-squares regression produces optimized power-law model parameters that describe gasification with respect to both CO2 and H2O separately and in combination. A single set of parameters reasonably describes rates for all three chars. Model simulations agree with measured data at all stages of char conversion. This investigation details how ash affects biomass char reactivity, specifically the late-stage burnout. The ash contents ratios in the raw fuels in these experiments are as high as 40:1, providing a clear indication of the ash effect on the char reactivity. The experimental results definitively indicate a decrease in char reaction rate with increasing initial fuel ash content and with increasing char burnout -- most pronounced at high burnout. This investigation postulates that an increase in the fraction of the surface covered by refractory material associated with either higher initial ash contents or increased burnout decreases the surface area available for reaction and thus the observed reaction rate. A quantitative model that includes this effect predicts the observed data at any one condition within the data uncertainty and over a broad range of fuel types, particle sizes, temperatures, and reactant concentrations slightly less accurately than the experimental uncertainty. Surface area, porosity, diameter, and density predictions from standard models do not adequately describe the experimental trends. Total surface area increases slightly with conversion, with most of the increase in the largest pores or channels/vascules not measurable by standard surface area techniques but most of the surface area is in the small pores. Porosity also increases with char conversion except for abrupt changes associated with char and ash collapse at the end of char conversion. Char particle diameters decrease during these kinetically controlled reactions, in part because the reaction is endothermic and therefore proceeds more rapidly at the comparatively warmer char surface. SEM images qualitatively confirm the quantitative measurements and imply that the biomass microstructure does not appreciably change during conversion except for the large pore diameters. Extant char porosity, diameter, surface area, and related models do not predict these trends. This investigation suggests alternative models based on these measurements.

Pyrolysis of biomass consists of heating solid biomass inthe absence of air to produce solid, liquid and gaseous fuels.In the first part of this thesis rapid pyrolysis of wood(birch) and some agricultural residues (olive waste, sugarcanebagasse and wheat straw in untreated and in pelletized form) athigh temperature (800ºC–1000ºC) is studied ina free fall reactor at pilot scale. These conditions are ofinterest for gasification in fluidized beds. Of main interestare the gas and char yields and compositions as well as thereactivity of the produced char in gasification. A higher temperature and smaller particles increase theheating rate resulting in a decreased char yield. The crackingof the hydrocarbons with an increase of the hydrogen content inthe gaseous product is favoured by a higher temperature and byusing smaller particles. Wood gives more volatiles and lesschar than straw and olive waste. The higher ash content inagricultural residues favours the charring reactions. Charsfrom olive waste and straw are more reactive in gasificationthan chars from birch because of the higher ash content. Thecomposition of the biomass influences the product distribution.Birch and bagasse give more volatiles and less char thanquebracho, straw and olive waste. Longer residence time inrapid pyrolysis increase the time for contact between tar andchar which makes the char less reactive. The secondary charproduced from tar not only covers the primary char but alsoprobably encapsulates the ash and hinders the catalytic effectof the ash. High char reactivity is favoured by conditionswherethe volatiles are rapidly removed from the particle, i.e.high heating rate, high temperature and small particles. The second part of this thesis deals with slow pyrolysis inpresence of steam for preparation of active carbon. Theinfluence of the type of biomass, the type of reactor and thetreatment conditions, mainly temperature and activation time,on the properties and the yield of active carbons are studied.The precursors used in the experiments are birch (wood) anddifferent types of agricultural residues such as sugarcanebagasse, olive waste, miscanthus pellets and straw in untreatedand pelletized form. The results from the pyrolysis of biomass in presence ofsteam are compared with those obtained in inert atmosphere ofnitrogen. The steam contributes to the formation of solidresidues with high surface area and good adsorption capacity.The yield of liquid products increases significantly at theexpense of the gaseous and solid products. Large amount ofsteam result in liquid products consisting predominantly ofwater-soluble polar compounds. In comparison to the stationary fixed bed reactor, therotary reactor increases the production of energy-rich gases atthe expense of liquid products. The raw materials have strong effect on the yields and theproperties of the pyrolysis products. At equal time oftreatment an increase of the temperature results in a decreaseof the yield of solid residue and improvement of the adsorptioncapacity until the highest surface area is reached. Furtherincrease of the temperature decreases the yield of solidproduct without any improvement in the adsorption capacity. Therate of steam flow influences the product distribution. Theyield of liquid products increases while the gas yielddecreases when the steam flow is increased. <b>Keywords</b>: rapid pyrolysis, pyrolysis, wood, agriculturalresidues,biomass, char, tar, gas, char reactivity,gasification, steam, active carbon

<p>Pyrolysis of biomass consists of heating solid biomass inthe absence of air to produce solid, liquid and gaseous fuels.In the first part of this thesis rapid pyrolysis of wood(birch) and some agricultural residues (olive waste, sugarcanebagasse and wheat straw in untreated and in pelletized form) athigh temperature (800ºC–1000ºC) is studied ina free fall reactor at pilot scale. These conditions are ofinterest for gasification in fluidized beds. Of main interestare the gas and char yields and compositions as well as thereactivity of the produced char in gasification.</p><p>A higher temperature and smaller particles increase theheating rate resulting in a decreased char yield. The crackingof the hydrocarbons with an increase of the hydrogen content inthe gaseous product is favoured by a higher temperature and byusing smaller particles. Wood gives more volatiles and lesschar than straw and olive waste. The higher ash content inagricultural residues favours the charring reactions. Charsfrom olive waste and straw are more reactive in gasificationthan chars from birch because of the higher ash content. Thecomposition of the biomass influences the product distribution.Birch and bagasse give more volatiles and less char thanquebracho, straw and olive waste. Longer residence time inrapid pyrolysis increase the time for contact between tar andchar which makes the char less reactive. The secondary charproduced from tar not only covers the primary char but alsoprobably encapsulates the ash and hinders the catalytic effectof the ash. High char reactivity is favoured by conditionswherethe volatiles are rapidly removed from the particle, i.e.high heating rate, high temperature and small particles.</p><p>The second part of this thesis deals with slow pyrolysis inpresence of steam for preparation of active carbon. Theinfluence of the type of biomass, the type of reactor and thetreatment conditions, mainly temperature and activation time,on the properties and the yield of active carbons are studied.The precursors used in the experiments are birch (wood) anddifferent types of agricultural residues such as sugarcanebagasse, olive waste, miscanthus pellets and straw in untreatedand pelletized form.</p><p>The results from the pyrolysis of biomass in presence ofsteam are compared with those obtained in inert atmosphere ofnitrogen. The steam contributes to the formation of solidresidues with high surface area and good adsorption capacity.The yield of liquid products increases significantly at theexpense of the gaseous and solid products. Large amount ofsteam result in liquid products consisting predominantly ofwater-soluble polar compounds.</p><p>In comparison to the stationary fixed bed reactor, therotary reactor increases the production of energy-rich gases atthe expense of liquid products.</p><p>The raw materials have strong effect on the yields and theproperties of the pyrolysis products. At equal time oftreatment an increase of the temperature results in a decreaseof the yield of solid residue and improvement of the adsorptioncapacity until the highest surface area is reached. Furtherincrease of the temperature decreases the yield of solidproduct without any improvement in the adsorption capacity. Therate of steam flow influences the product distribution. Theyield of liquid products increases while the gas yielddecreases when the steam flow is increased.</p><p><b>Keywords</b>: rapid pyrolysis, pyrolysis, wood, agriculturalresidues,biomass, char, tar, gas, char reactivity,gasification, steam, active carbon</p>

It is recognised that the reactions with steam and CO2 are the rate limiting step during coal gasification, and a vast number of studies has been dedicated to the kinetics of these reactions. Most studies were carried out by using a single reactant (CO2 or H2O), either pure or diluted with an inert gas. Research using gas mixtures of CO2 and steam and their effects on gasification kinetics have been undertaken but are limited. The objective of this study is to determine the effects of CO2 on the steam gasification rate of a typical Highveld seam 4 coal. The South African medium ranked high volatile bituminous coal was charred at 950 °C. 2.0 g samples of ± 1 mm particles were analysed in a modified large particle thermo gravimetric analyser under various reactant gas concentrations. Experiments were conducted at atmospheric pressure (87.5 kPa) and temperatures from 775 to 900 °C, such that the conversion rate was controlled by chemical reaction. Reagent mixtures of steam-N2, steam-CO2 and CO2-N2 at concentrations of 25-75 mol%, 50-50 mol%, 75-25 mol% and 100 mol% were investigated. Arrhenius plots for steam and CO2 gasification produced activation energy values of 225 ± 23 kJ/mol and 243 ± 32 kJ/mol respectively. The calculated reaction orders with respect to reagent partial pressure were 0.44 ± 0.08 and 0.56 ± 0.07 for steam and CO2 respectively. Comparisons of the experimental data showed a higher reaction rate for the steam-CO2 mixtures compared to steam-N2 experiments. The semi empirical Wen model (m = 0.85) with an additive Langmuir-Hinshelwood styled rate equation predicted the mixed reagent gasification accurately. Reaction constants that were determined from the pure reactant experiments could directly be applied to predict the results for the experiments with mixtures of steam and CO2. The conclusion was made that under the investigated conditions steam and CO2 reacts simultaneously on different active sites on the char surface. / Thesis (MIng (Chemical Engineering))--North-West University, Potchefstroom Campus, 2013.

It is recognised that the reactions with steam and CO2 are the rate limiting step during coal gasification, and a vast number of studies has been dedicated to the kinetics of these reactions. Most studies were carried out by using a single reactant (CO2 or H2O), either pure or diluted with an inert gas. Research using gas mixtures of CO2 and steam and their effects on gasification kinetics have been undertaken but are limited. The objective of this study is to determine the effects of CO2 on the steam gasification rate of a typical Highveld seam 4 coal. The South African medium ranked high volatile bituminous coal was charred at 950 °C. 2.0 g samples of ± 1 mm particles were analysed in a modified large particle thermo gravimetric analyser under various reactant gas concentrations. Experiments were conducted at atmospheric pressure (87.5 kPa) and temperatures from 775 to 900 °C, such that the conversion rate was controlled by chemical reaction. Reagent mixtures of steam-N2, steam-CO2 and CO2-N2 at concentrations of 25-75 mol%, 50-50 mol%, 75-25 mol% and 100 mol% were investigated. Arrhenius plots for steam and CO2 gasification produced activation energy values of 225 ± 23 kJ/mol and 243 ± 32 kJ/mol respectively. The calculated reaction orders with respect to reagent partial pressure were 0.44 ± 0.08 and 0.56 ± 0.07 for steam and CO2 respectively. Comparisons of the experimental data showed a higher reaction rate for the steam-CO2 mixtures compared to steam-N2 experiments. The semi empirical Wen model (m = 0.85) with an additive Langmuir-Hinshelwood styled rate equation predicted the mixed reagent gasification accurately. Reaction constants that were determined from the pure reactant experiments could directly be applied to predict the results for the experiments with mixtures of steam and CO2. The conclusion was made that under the investigated conditions steam and CO2 reacts simultaneously on different active sites on the char surface. / Thesis (MIng (Chemical Engineering))--North-West University, Potchefstroom Campus, 2013.

Ce travail porte sur la modélisation de gazéification de la biomasse dans un réacteur solaire à lit fixe. Nous avons développé et validé différents modules pour la résolution des problèmes physiques associés à de tels gazéifieurs (volumes finis, équilibre thermodynamique, facteur de forme, radiosité, évolution du maillage, etc.). Le réacteur est ensuite modélisé et confronté à des mesures expérimentales pour sa validation. Ces résultats nous laissent supposer que, pour les conditions opératoires rencontrées dans ce réacteur, l’évolution de la vitesse de chauffage et du taux de centre sont des facteurs d’influence importants de la cinétique pour le bois de hêtre. Une étude du réacteur en supposant que la réaction atteint l’équilibre thermodynamique est ensuite conduite et montre que cette approche doit être employée avec précaution pour le domaine de température étudié. Une étude paramétrique concernant la mise en place d’un échangeur air-air entre le gaz de sortie et le gaz d’entrée est proposée. Elle nous montre que l’ajout d’un gaz inerte peut être intéressant d’un point de vue énergétique dans de telles conditions. Enfin, un absorbeur solaire adapté au gazéifieur est ensuite modélisé et validé afin de pouvoir évaluer la sensibilité du système complet. / This work focuses on the modeling of biomass gasification in a solar fixed bed reactor. We developed and validated different modules to compute physical problems associated with such gasifier (finite volume, thermodynamic equilibrium, view factor, radiosity, evolution of the mesh, etc.). The reactor is then modeled, and results are compared to experimental measurements for its validation. These results suggest that for the operating conditions encountered in this reactor, the evolution of the heating rate and the ash concentration may influence the kinetics of the beech wood gasification. A study of the reactor by assuming thermodynamic equilibrium is conducted and shows that this approach should be used with caution for the studied temperature range. Parametric study of an air-air exchanger between the output and input gas is proposed. It shows that the addition of an inert gas can be interesting from an energy point of view in such conditions. Finally, a solar absorber suited to the reactor is modeled and validated in order to assess the sensitivity of the complete system.