Numerical modelling of biomass thermochemical conversion and potassium release

The use of biomass as a renewable source of energy has been increasing over the past few decades, and biomass is regarded as a promising replacement for fossil fuels. Thermochemical conversion of biomass is a common approach for biomass utilization with a relatively short conversion time. Despite the many advantages that biomass offers over fossil fuels, the thermochemical conversion of biomass has some drawbacks including harmful emissions such as NO and negative impacts of alkali metals such as potassium, which need to be further studied. Numerical modelling is a powerful tool that compliments the experimental measurements for studies on biomass conversion and its challenges.

In this thesis, numerical models are developed with a focus on different scales of biomass conversion, ranging from single particles to reactor-scale simulations. A detailed mesh-based particle model is developed to study the intra-particle phenomena, as well as the reactions surrounding the particle. The model is used to study the effects of particle shrinkage, anisotropic heat transfer, and the stiff problem of particle-surrounding gas coupling. Moreover, a new anisotropic shrinkage model is proposed that correlates the axial and radial shrinkage of particles to the decomposition of main wood components, i.e., cellulose, hemicellulose, and lignin. Finally, the particle model is used to study the stages of hydrogen release during pyrolysis and the effects of wood inhomogeneities on particle conversion.

Potassium release from biomass can cause numerous problems in the reactors, namely, corrosion, slagging, fouling, and aerosol formation. A new detailed potassium release model is developed in the current study, which for the first time, takes into account the effects of variables such as fuel type and ash composition, conversion temperature, and surrounding atmosphere. The model is validated against different experimental measurements from the literature in terms of the prediction of different types of potassium in residual solid, the total potassium, chloride, and sulfur in solid, and also the gas-phase potassium-containing species downstream of reacting particles. Later, the developed potassium model is used in a CFD simulation of an entrained flow gasifier to study the gasification process and the potassium release in the reactor.

Numerical studies are also carried out to investigate the performance and NO emissions from fixed-bed biomass reactors. A novel method is proposed to handle the complicated tar species and their decomposition in the freeboard of a lab-scale fixed bed reactor. The model is used to study NO emissions from the same reactor and the performance of selective non-catalytic reduction (SNCR) for NO reduction. In a later study, the bed is also included in the simulations with a fast-solving 2D bed model. The model is based on the detailed simulation of thermally-thick particle conversion at different conditions. The simulation results are tabulated and then used to extract the source terms for particles in the fuel bed at different times and locations.