Flameless combustion has shown a great potential for nitric oxides (NOx) abatement, reaction zone and temperature field homogenization. It is characterized by a strong internal recirculation of hot flue gases and a strong dilution of the fresh reactants combustion air and pulverized coal, amongst others. On the one hand, the homogenized reaction zone and temperature field are the guarantors for the NOx abatement, jointly with the O2-deficient atmosphere. On the other hand, the homogenized temperature field allows for an increased flue gas temperature level, since lower safety factors against temperature peaks can be applied during furnace design, as this is currently possible because of the higher temperature fluctuations of conventional flame burners. Thus, an increased burnout due to better fuel conversion at higher temperatures can be expected.
This work investigates the flameless combustion of a pulverized high-volatile bituminous coal in both experiment and simulation at the pilot scale. It aims to show how different burner designs and different coal-carrier-gases influence the flameless combustion of pulverized coal. Therefore, two different burner prototypes are investigated. Both ensure sufficiently high internal flue-gas recirculation necessary for establishing flameless combustion. Detailed experimental data are available for the flow field by laser Doppler velocimetry (LDV), for the reaction zone, its topology and its reaction intensity, by OH* chemiluminescence imaging, and for the main reactive species and the prevailing temperatures by suction-probe sampling. In addition, the total and the radiative heat fluxes are analyzed. Two different flameless operating conditions have been tested for each prototype burner. It is shown that the choice of the coal-carrier-gas, CO2 or air, can strongly influence the burnout and NOx formation for the central coal-carrier-gas annulus. As the pulverized coal directly emerges into the hot and oxygen-deficient flue-gas recirculation zone, similar results are achieved, regardless of the applied coal-carrier-gas type.
A state-of-the-art low-NOx flame burner has been investigated in air-staged and unstaged operation modes for reference. It performs well regarding the low NOx concentration which is similar or lower to that of the flameless burners, if it is operated in an air-staged mode. This leads to the conclusion that a lower O2 concentration in the reaction zone of the flameless burner prototypes needs to be striven for to achieve lower NOx concentrations, either by air-staging or by emphasizing the internal flue gas recirculation.
Flameless combustion is secondly investigated by means of computational fluid dynamics (CFD). Therefore, IFK's in-house program code AIOLOS is deployed which describes the reactive fluid flow based on the Reynolds-averaged Navier-Stokes equations. The CFD program code is validated for flameless pulverized coal combustion using the experimental data. The main focus is drawn on the mixing of the entering reactants and the recirculated flue gas, and on the chemical reactions taking place in this highly diluted atmosphere. Thus, the turbulence model, the turbulence-chemistry interaction model and the chemical reaction model including pollutants, such as NOx, are investigated. Three k-ε turbulence models are compared against the velocity data obtained by LDV, and the RNG k-ε model gives the best agreement. Different model constants of the eddy-dissipation concept (EDC) model are tested to reproduce the turbulence-chemistry interaction, and one best-fitting set of constants is identified. The global combustion reaction model is assessed with regards to the application of a detailed pyrolysis preprocessing. The modeling of NOx formation and reduction is realized with a global reaction model respecting the fuel-NO and thermal NO paths. As for the combustion reaction model, the influence of detailed pyrolysis pre-processing is researched. The initial nitrogen split into char-N, tar-N and light-gas-N can strongly alter the result of the NOx post-processing. The char-N release path is found to be of key interest.
The experimental results highlight the requirement for establishing a highly diluted atmosphere at a temperature level well above the auto-ignition temperature of the fuel, in order to ensure good NOx reduction and fuel conversion. The computational results emphasize the importance of a detailed analysis of the pyrolysis products. This holds for both the combustion and the NOx reaction models.