Show simple item record

dc.contributor.authorAgu, Cornelius Emeka
dc.description.abstractThe need to cut down the high dependency on the fossil fuels requires sustainable alternative energy resources. Aside that the stock of fossil fuels is depreciable, the energy source also accounts for the major contribution of greenhouse gas effects. Biomass in the categories of woody, herbaceous, marine and manure biomasses, are among the renewable energy sources, which can be grown everywhere in a sustainable manner. Biomass currently contributes to more than 10% global energy consumption in the different forms of application: direct combustion and conversion into a gaseous form for chemical synthesis. Gasification is a means of converting biomass into a higher energy gas containing mainly CO, CH4, H2 and CO2 for sustainable utilization. Among different technologies applied in biomass gasification, fluidized bed has wide industrial advantages in that a variety of feedstock can be gasified in addition to that the process can be easily controlled. To explore the numerous benefits of fluidized bed technology, an in-depth understanding of the fluid-particle interactions in the reactor at different operating conditions is necessary. This thesis investigates the behaviour of different powders in fluidized beds. The effects of particle size, bed height and gas velocity on the bubbling bed behaviour and on the transition from bubbling to slugging regime are outlined. The mixing and segregation patterns of biomass particles in binary mixtures with inert particles are also investigated. In addition, the study covers the measurement of residence time of biomass during conversion in an air-blown bubbling bed reactor and the yield of char particles during the devolatilization phase. The gasification of biomass in a laboratory scale reactor using different bed particle sizes, air-fuel ratios, steam-biomass ratios and biomass loading rates are also characterized. These studies are performed using two different experimental setups and a one-dimensional (1D) model developed for bubbling fluidized bed reactors. The two experimental setups have close internal diameter of 10 cm and effective total height in the range 1 – 1.4 m. The first setup is equipped with two electrical capacitance tomography (ECT) sensors, which measure the distribution of solids fraction at different bed positions for a given gas flowrate. At the ambient conditions, the ECT setup is used to characterize the behaviour of different particles at different gas velocities. The information from the two plane ECT sensors are also used to develop methods for determining different bubble properties, and the gas velocity and bed voidage at the onset of slugging regime. The second setup is used for hot flow processes and it is equipped with five different thermocouples and pressure transducers for monitoring the reactor performance along the vertical axis. The bubbling bed reactor model is developed to capture the average flow properties and product gas species at any position in the reactor. The model is unsteady and developed based on the conservation of mass, momentum and energy within and across the reactor. The basic assumption underlying the model is that the mean circulation velocity of the bed material is zero, which reduces the computational complexities in using the model. The results show that the ratio of superficial gas velocity to the minimum fluidization velocity at the onset of slugging regime increases with decreasing particle size. As the particle size increases, changes in the bed height has a negligible effect on the transition velocity. The bubble growth rate with increasing gas velocity increases with the particle size, resulting in the earlier occurrence of slugs in the beds of large particles. The bubble frequency increases with increasing gas velocity only when the bubble diameter is below a threshold value. The maximum bubble frequency over the range of operating gas velocities also indicates the transition from bubbling to slugging regime. For a mixture of biomass and bed material, the bubble diameter decreases with increasing amount of biomass, leading to a delay in the slug flow. The minimum fluidization velocity increases with increasing biomass load but for a high density biomass (~1000 kg/m3), the gas velocity slightly decreases due to a reduction in the bed voidage when the biomass load is below 20 vol.%. The sinking of biomass at a given gas velocity also increases with the biomass density while the spreads of biomass towards the walls decreases with increasing biomass density. The minimum gas velocity required to achieve a mixing over the bed increases with increasing biomass load, decreasing bed diameter and slightly unaffected by changes in the bed height. During conversion, the segregation pattern of the char particles is similar to the parent biomass in the cold condition. When introduced in a bubbling bed, the initial distribution of biomass particles is greatly influenced by the combined effects of the particle bulk density and the rising bubbles. As biomass devolatilizes, the particles rise upwards. The time for complete devolatilization increases with the amount of biomass charged and with decreasing air flowrate. Moreover, the amount of char released at the completion of devolatilization and the char residence time before complete conversion decrease with increasing air flowrate and decreasing amount of biomass loaded in the bed. The gasification of wood pellets with air shows that at the same air-fuel ratio, the particle size of the bed material has insignificant effect on the gas composition. With an increase in the air-fuel ratio, H2 yield increases and the yields of CO and CH4 decrease. Increasing the biomass flowrate from 2.7 to 3.6 kg/h increases the yields of CO and CH4 and decreases that of H2 at the same air flowrate. Similar behaviour with different particle sizes are also observed in the gasification of the same biomass with steam using the proposed 1D model. The model results also show that both H2/CO and CO2/CO ratios attain minimal values at certain bed temperatures. The method used in obtaining the bed expansion and bed voidage influences the model results. With an increase in the bed expansion within a certain range, the yields of CO and CH4 increase due to increasing char conversion. The increase in the biomass flowrate at a constant steam to biomass ratio increases the char accumulation. The biomass density also has a great influence on the particle distribution, and thus on the product quality. The higher the biomass density, the better the conversion efficiency. Different correlations are also proposed for prediction of bubble properties (bubble diameter, bubble flux, bubble velocity and bubble frequency), bed expansion, bed voidage of a binary mixture and the minimum gas velocity required to achieve particle mixing over a segregated layer of biomass. The proposed models for the bubble diameter and volumetric bubble flux averaged over the bed height account for the effect of particle and fluid properties on these variables. Applying the particle dependent-bubble diameter on the bed expansion model gives a good prediction for a given bed. The bubble velocity model gives better predictions for Geldart B and D particles than those in the existing literature. Using the proposed model for the bed voidage, accurate predictions can be achieved for different binary systems. The application of the voidage model to the Ergun equation shows that the minimum fluidization velocity of binary mixtures can be predicted with error of 15% for two inert materials and 7% for a mixture of biomass and an inert material. New correlations based on the air flowrate, biomass flowrate, mass of the bed material and the minimum fluidization velocity of the bed particles at the operating temperature are also proposed for the biomass residence time, the amount of char accumulated during the conversion and the total heat loss at the completion of devolatilization process. The results of this thesis can be useful for optimization of design and operational control of biomass gasification reactors. The proposed 1D model can also be incorporated into a circulating fluidized bed reactor to obtain the dynamic behaviour of the so-called dual fluidized bed reactors. As the model can accept all the possible inputs to a gasifier, it can be used to determine the optimum operating point for efficient conversion of biomass in a given bubbling fluidized bed reactor.nb_NO
dc.publisherUniversity of South-Eastern Norwaynb_NO
dc.relation.ispartofseriesDoctoral dissertations at the University of South-Eastern Norway;35
dc.relation.haspartArticle 1: Agu, C.E., Tokheim, L.-A., Eikeland, M. & Moldestad, B.M.E.: Determination of onset of bubbling and slugging in a fluidized bed using a dual-plane electrical capacitance tomography system. Chemical Engineering Journal 328, (2017), 997-1015.
dc.relation.haspartArticle 2: Agu, C.E., Eikeland, M., Tokheim, L.-A. & Moldestad, B.M.E.: Simulation of Bubbling Fluidized Bed Using a One- Dimensional Model Based on the Euler-Euler Method. Proceedings of the 9th EUROSIM Congress on Modelling and Simulation Oulu, Finland, September 12-16, 2016. ISBN 978-1-5090-4119-0.
dc.relation.haspartArticle 3: Agu, C.E., Tokheim, L.-A. & Halvorsen, B.: Measurement of bubble properties in a fluidized bed using electrical capacitance tomography. Proceedings of the 12th International Conference on Fluidized Bed Technology (CFB-12), Krakow, Poland, May 23–26, 2017nb_NO
dc.relation.haspartArticle 4: Agu, C.E., Ugwu, A., Pfeifer, C., Eikeland, M., Tokheim, L.-A. & Moldestad, B.M.E.: Investigation of bubbling behavior in deep fluidized beds at different gas velocities using electrical capacitance tomography. Industrial & Engineering Chemistry Research 58(5), (2019), 2084-2098.
dc.relation.haspartArticle 5: Agu, C.E., Pfeifer, C., Eikeland, M., Tokheim, L.-A. & Moldestad, B.M.E.: Models for Predicting Average Bubble Diameter and Volumetric Bubble Flux in Deep Fluidized Beds. Industrial & Engineering Chemistry Research 57(7), (2018), 2658-2669.
dc.relation.haspartArticle 6: Agu, C.E., Tokheim, L.-A., Eikeland, M. & Moldestad, B.M.E.: Improved Models for Predicting Bubble Velocity, Bubble Frequency and Bed Expansion in a Bubbling Fluidized Bed. Chemical engineering research & design 141, (2019), 361-371.
dc.relation.haspartArticle 7: Agu, C.E., Tokheim, L.-A., Pfeifer, C. & Moldestad, B.M.E.: Behaviour of biomass particles in a bubbling fluidized bed: A comparison between wood pellets and wood chips. Chemical Engineering Journal 363, (2019), 84-98.
dc.relation.haspartArticle 8: Agu, C.E., Pfeifer, C. & Moldestad, B.M.E.: Prediction of void fraction and minimum fluidization velocity of a binary mixture of particles: Bed material and fuel particles. Powder Technology 349, (2019), 99-107.
dc.relation.haspartArticle 9: Agu, C.E., Pfeifer, C., Eikeland, M., Tokheim, L.-A. & Moldestad, B.M.E.: Measurement and characterization of biomass mean residence time in an air-blown bubbling fluidized bed gasification reactor. Fuel 253, (2019), 1414-1423.
dc.relation.haspartArticle 10: Agu, C.E., Pfeifer, C., Eikeland, M., Tokheim, L.-A. & Moldestad, B.M.E.: Detailed One-Dimensional Model for Steam-Biomass Gasification in a Bubbling Fluidized Bed. Submitted version. Published in Energy & Fuels 33, (2019), 7385-7397.
dc.subjectfluidized bednb_NO
dc.subjectbinary mixturenb_NO
dc.subjectbubbling bednb_NO
dc.subjectslugging bednb_NO
dc.subjectbubble propertiesnb_NO
dc.subjectone-dimensional modelnb_NO
dc.titleBubbling Fluidized Bed Behaviour for Biomass Gasificationnb_NO
dc.typeDoctoral thesisnb_NO
dc.rights.holder© 2019 Cornelius E. Agunb_NO
dc.subject.nsiVDP::Teknologi: 500::Kjemisk teknologi: 560::Kjemiteknikk: 563nb_NO
dc.rights.license© The Author, except otherwise stated

Files in this item


This item appears in the following Collection(s)

Show simple item record
Except where otherwise noted, this item's license is described as