• Energy Research

Abstract Several investigations have shown that the differences between deposits obtained in oxy-firing and air-firing of coal mainly are due to differences in the flame temperature. Consequently, deposit rate predictions not taking the in-flight history into account are unlikely to be successful. In this paper, a model for predicting the deposit formation propensity of pulverized coal in oxy-fuel and air combustion due to the inertial impaction mechanism is developed and tested. The model builds on the use of viscosity as an indicator of the sticking probability. The composition and amount of the amorphous slag phase in the coal ash are calculated assuming thermodynamic equilibrium. Further, it is assumed that the maximum temperature the ash particle has experienced will control the composition and amount of the amorphous slag phase. As the ash particle impacts the probability to stick is estimated using the viscosity of this melt composition, but with the temperature of particle temperature at the moment of impaction. In the equilibrium calculation no material exchange with the gas phase is assumed. This assumption is based on X-ray diffraction (XRD) investigations of coal ash samples produced in a lab-scale burner simulating oxy-fuel and air combustion. The XRD showed that there was no significant impact on the mineralogy of the coal ash caused by the gas atmosphere. The probability of an ash particle to stick as a function of maximum experienced temperature and impact temperature was evaluated for three coals. For one of the coals a CFD study on particle deposit is done for a 300 kWth test facility.

Abstract Several investigations have shown that the differences between deposits obtained in oxy-firing and air-firing of coal mainly are due to differences in the flame temperature. Consequently, deposit rate predictions not taking the in-flight history into account are unlikely to be successful. In this paper, a model for predicting the deposit formation propensity of pulverized coal in oxy-fuel and air combustion due to the inertial impaction mechanism is developed and tested. The model builds on the use of viscosity as an indicator of the sticking probability. The composition and amount of the amorphous slag phase in the coal ash are calculated assuming thermodynamic equilibrium. Further, it is assumed that the maximum temperature the ash particle has experienced will control the composition and amount of the amorphous slag phase. As the ash particle impacts the probability to stick is estimated using the viscosity of this melt composition, but with the temperature of particle temperature at the moment of impaction. In the equilibrium calculation no material exchange with the gas phase is assumed. This assumption is based on X-ray diffraction (XRD) investigations of coal ash samples produced in a lab-scale burner simulating oxy-fuel and air combustion. The XRD showed that there was no significant impact on the mineralogy of the coal ash caused by the gas atmosphere. The probability of an ash particle to stick as a function of maximum experienced temperature and impact temperature was evaluated for three coals. For one of the coals a CFD study on particle deposit is done for a 300 kWth test facility.

Abstract In torrefaction, the mass yield depends on the biomass type, size of the biomass, torrefaction temperature, and residence time. Mass yield curves vs. residence time are usually modeled based on biomass type at different torrefaction temperature. This work is the second part of a study on the effect of alkali metals on torrefaction in which the K content affects the degradation of biomass during torrefaction. In this part of the study, the mass loss of spruce wood with different content of K was modeled using a two-step reaction model based on four kinetic rate constants. The results show that it is possible to model the mass loss of spruce wood doped with different levels of K using the same activation energies but different pre-exponential factors for the rate constants. Three of the pre-exponential factors increased linearly with increasing K content, while one of the pre-exponential factors decreased with increasing K content. Therefore, a new torrefaction model was formulated using the hemicellulose and cellulose content and K content. The new torrefaction model was validated against the mass loss during the torrefaction of aspen, miscanthus, straw and bark. There was good agreement between the model and the experimental data for the other biomasses, except bark. For bark, the mass loss of acetone extractable material was also needed to be taken into account. The new model can describe the kinetics of mass loss during torrefaction of different types of biomass. This is important for considering fuel flexibility in torrefaction plants.

Abstract In torrefaction, the mass yield depends on the biomass type, size of the biomass, torrefaction temperature, and residence time. Mass yield curves vs. residence time are usually modeled based on biomass type at different torrefaction temperature. This work is the second part of a study on the effect of alkali metals on torrefaction in which the K content affects the degradation of biomass during torrefaction. In this part of the study, the mass loss of spruce wood with different content of K was modeled using a two-step reaction model based on four kinetic rate constants. The results show that it is possible to model the mass loss of spruce wood doped with different levels of K using the same activation energies but different pre-exponential factors for the rate constants. Three of the pre-exponential factors increased linearly with increasing K content, while one of the pre-exponential factors decreased with increasing K content. Therefore, a new torrefaction model was formulated using the hemicellulose and cellulose content and K content. The new torrefaction model was validated against the mass loss during the torrefaction of aspen, miscanthus, straw and bark. There was good agreement between the model and the experimental data for the other biomasses, except bark. For bark, the mass loss of acetone extractable material was also needed to be taken into account. The new model can describe the kinetics of mass loss during torrefaction of different types of biomass. This is important for considering fuel flexibility in torrefaction plants.

Abstract Torrefaction is a promising heat-treatment method being developed for biomass to increase the use of biomass in its thermochemical conversion processes. This type of pre-treatment can improve the properties of biomass for thermal conversion by improving grindability, heating value, reducing the hydrophilic nature, and increasing its resistance to biodegradation. In this work, we studied the impact of organically bound K, Na, Ca and Mn on mass loss of biomass during torrefaction. These elements were of interest because they have been shown to be catalytically active in solid fuels during pyrolysis and/or gasification. In this work, we studied spruce and pine as coniferous woods, aspen as a deciduous wood and miscanthus as an herbaceous biomass. The biomasses were first acid washed to remove the ash-forming elements and then organic sites were doped with K, Na, Ca or Mn. The doping was performed in a nitrate solution of each metal. The resultant fuels were then torrefied at fixed temperatures between 240 and 280 °C in a thermogravimetric analyzer. The results show that K and Na bound to organic sites can significantly increase the mass loss during torrefaction at temperatures between 240 and 280 °C for a fixed time and temperature. It is also seen that Mn bound to organic sites increases the mass loss while Ca addition does not influence the mass loss rate during torrefaction. This increase in mass loss during torrefaction with alkali addition is unlike what has been found in the case of pyrolysis where alkali addition resulted in a reduced mass loss. These results are important for the future operation of torrefaction plants which will likely be designed to handle various biomasses with significantly different contents of K. The results imply that shorter retention times are possible for high K-containing biomasses.

Abstract Torrefaction is a promising heat-treatment method being developed for biomass to increase the use of biomass in its thermochemical conversion processes. This type of pre-treatment can improve the properties of biomass for thermal conversion by improving grindability, heating value, reducing the hydrophilic nature, and increasing its resistance to biodegradation. In this work, we studied the impact of organically bound K, Na, Ca and Mn on mass loss of biomass during torrefaction. These elements were of interest because they have been shown to be catalytically active in solid fuels during pyrolysis and/or gasification. In this work, we studied spruce and pine as coniferous woods, aspen as a deciduous wood and miscanthus as an herbaceous biomass. The biomasses were first acid washed to remove the ash-forming elements and then organic sites were doped with K, Na, Ca or Mn. The doping was performed in a nitrate solution of each metal. The resultant fuels were then torrefied at fixed temperatures between 240 and 280 °C in a thermogravimetric analyzer. The results show that K and Na bound to organic sites can significantly increase the mass loss during torrefaction at temperatures between 240 and 280 °C for a fixed time and temperature. It is also seen that Mn bound to organic sites increases the mass loss while Ca addition does not influence the mass loss rate during torrefaction. This increase in mass loss during torrefaction with alkali addition is unlike what has been found in the case of pyrolysis where alkali addition resulted in a reduced mass loss. These results are important for the future operation of torrefaction plants which will likely be designed to handle various biomasses with significantly different contents of K. The results imply that shorter retention times are possible for high K-containing biomasses.