@phdthesis{Tanui2020, author = {Tanui, Josephat}, title = {Detailed study of wood combustion in a fixed bed reactor under oxy-fuel condition}, doi = {10.15771/DISS_2020_1}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:526-opus4-15320}, school = {Technische Hochschule Wildau}, pages = {272}, year = {2020}, abstract = {Biomass fuel is the major source of energy for a significant proportion of the world population. Conventionally, biomass is burned in combustors that use air as oxidizer. The need for efficiency and reduced emissions has led to development of clean combustion technologies such as oxy-fuel combustion. In oxy-fuel combustion technique, fuel is burnt in O2/CO2 environment instead of O2/N2 environment. In spite of biomass being used under oxy-fuel conditions using fixed bed furnaces, there is no detailed study of biomass combustion in fixed bed reactors under oxy-fuel condition. The studies only cover combustion under air-fuel condition. There is no report in the literature on important combustion parameters, viz., flame propagation speed, temperature profiles, ignition time, and species evolution in the fixed bed combustion under oxy-fuel conditions; the studies have been only in air. This work extends the research on biomass conversion to combustion under oxy-fuel conditions. To address these knowledge gaps, this PhD thesis focused on establishing and isolating the role of various CO2 effects on biomass combustion properties using a laboratory fixed bed reactor through experiments and numerical modeling. The fixed bed reactor used in this study was operated in counter-current flame propagation mode. Oxidizer was a mixture of pure O2 and N2 or CO2. Measurement were made for air-fuel condition, oxy-fuel condition of different O2/CO2 mixtures and various initial bed porosity. In this work, all numerical simulations were carried out using a commercial CFD software CD-Adapco (STAR CCM+ version 11.04). Euler-Lagrange (CFD-DEM) approach was used to model biomass conversion in a packed bed. Solid particles and their conversion were modeled using DEM while fluid phase was modeled as a continuous phase using CFD. Biomass conversion was modeled by four sub-models, namely, drying, pyrolysis, gas-phase reactions and char oxidation models. These models were adopted and improved from previous research on wood combustion. To account for oxy-fuel combustion environment, reaction involving CO2 which has been identified as the dominant reaction pathway as well as its associated chain-branching reactions were incorporated in homogeneous reaction sub-model. In order to separate effects of CO2 on wood combustion, four different mixtures of O2 and Ar/CO2/N2 were designed and simulated. Temperature in O2/CO2 combustion environment was adjusted to be equal to that of O2/N2 environment by adding an appropriate amount of Ar while O2 amount remained the same. The results showed that flame front propagation speed in oxy-fuel atmosphere reduced to 78\% of that of the air-fuel condition with similar O2 concentration. The CFD-DEM model agreed very well with experimental values for mass loss, propagation speed and flame front positions. However, peak temperatures were poorly predicted at lower oxygen concentrations. The accuracy of temperature prediction improved at higher oxygen concentrations. It was noted that oxy-fuel peak temperature is lower than air-fuel temperature by about 200 K. For oxy-fuel combustion, peak temperature value and propagation speed increases as O2 concentration is increased. The results showed that at any given oxygen concentration, ignition time in oxy-fuel environment is almost twice that of corresponding air-fuel condition. It was established that the packing density affects combustion process by changing the burning conditions and limits. Furthermore, the study revealed that there is an optimum packing density, χ = 0.71, beyond which the efficiency falls due to the onset of quenching in the spaces. It was concluded that dilution effect is the most influential parameter on the burning rate of wood combustion in an oxy-fuel system. In a future study, the model prediction could be improved by including ash inhibition sub-model and using a full chemical kinetic mechanism for hydro-carbon combustion. It was also recommended that a detail analysis of heat transfer in the bed could be investigated in a future study. The findings from this work are important and may be useful when designing burners which operate under oxy-fuel combustion conditions.}, language = {en} }