Numerical study on ignition, combustion, and pollutant formation processes using solid pulverized fuels

In this work, the ignition, combustion, and pollutant formation of solid pulverized fuels are studied under different operating conditions to provide guidelines for predictive reduced-order model developments using detailed high-fidelity simulations. In the employed validated numerical framework, detailed models for both the particle and the gas phase are utilized within the fully coupled Eulerian-Lagrangian framework. Using these detailed models, first, the ignition, combustion, and flame structure of coal and biomass particles in a laminar flow reactor configuration are studied, showcasing similar combustion characteristics between coal and biomass due to the cancellation of competing effects. The effects of particle/flow interaction on ignition and combustion are also studied, considering different particle shapes. The comprehensive investigation showed that higher slip velocities lead to faster ignition. Also, non-sphericity promotes faster ignition due to its effects on particle motion and heat transfer. Furthermore, the particle/particle interaction effects on the ignition and flame topology in the particle group configuration are investigated, indicating later ignition time and flame opening behavior for higher particle number densities. To study the effects of particle/chemistry/turbulence interactions on ignition, the particle cloud clustering phenomenon is investigated in homogeneous isotropic turbulence. The analyses show that ignition is more likely to happen outside particle clusters, where more suitable conditions for ignition exist. Finally, pollutant formation during solid fuel combustion is analyzed with a focus on NOx formation in laminar biomass flames. The highest NOx formation is found to occur near particle groups around stoichiometric mixture fractions. Important pathways for NOx formation under different operating conditions are studied, revealing the dominance of Fuel-NOx pathways. All DNS datasets are also used to assess reduced-order flamelet-based models, and the assumptions typically used for predicting the ignition, combustion, and pollutant formation of solid fuels. Also, the effects of simplifying assumptions in the solid kinetic model and the gas-phase chemistry are studied to examine the deficits in the reduced-order modeling assumptions under different operating conditions.