Agglomeration mechanisms of fluidized beds during biomass combustion

The world experienced the warmest year recorded in 2024, also breaching the 1.5 °C warming limit. It has now become even more crucial to switch to more sustainable and CO2-neutral energy production to limit and hopefully reverse climate change someday. Therefore, the need for cheap but challenging biomass fuels, earlier classified as waste for fuidized bed combustion is increasing. Traditionally, fuels such as wood chips and bark, with low alkali and phosphorus content, have been used which induced only minor agglomeration problems. However, due to the limited availability of conventional wood-based fuels, interest in agricultural residue biomass is steadily increasing due to high availability and lower costs. As agricultural residue biomass usually has high alkali-, phosphorus-, and in some cases silica content, the risk of agglomeration is high due to the formation of reactive or low temperature melting alkali compounds.

Therefore, this study focused on identifying the most common agglomeration-causing compounds present during combustion of agricultural residue biomass and testing them in an electrically heated laboratory-scale fluidized bed reactor.

The critical components for bed agglomeration were fed as synthetic ash compounds into the reactor at 10-minute intervals until defluidization, or until a pre-determined maximum feed limit had been reached. After completing the experiments, samples of the bed, including agglomerates and bed material, were cast into epoxy, ground and polished to reveal the crosssections of agglomerates and bed particles. These cross-sections were analyzed using Scanning Electron Microscopy/Energy-Dispersive X-ray spectroscopy (SEM/EDX). In some cases, wet chemical analyses were also conducted on the bed materials to determine the remaining unreacted synthetic compounds.
In addition, thermal analyses (DSC/TGA) and X-Ray Diffraction analyses (XRD) were done on selected synthetic ash compounds and mixtures, with and without quartz bed particles, to elucidate the agglomeration mechanisms and involved compounds further.

Based on the results, the ash compounds could be categorized into three agglomeration mechanisms:
Type I agglomeration: The compound melted and glued the bed material together
Type II agglomeration: The compound reacted with the quartz bed particles to reaction products that glued the bed particles together.
Type III Agglomeration: A combination of the Type I and II mechanisms, where the molten compound glued the bed particles together with subsequent reaction with bed particles forming molten agglomeration causing alkali silicate.

The agglomeration-reducing effect of calcium was studied with synthetic ash compounds and fuels. In the experiments, CaCO3 was mixed with two different K-phosphates that had shown different agglomeration mechanisms. Each mixture was fed into the fluidized bed reactor using quartz sand as bed material. The results suggested a significant decrease in agglomeration for the phosphate compound that caused agglomeration via the Type I mechanism. In contrast, the phosphate compound that had caused agglomeration via the Type II mechanism showed an increase in agglomeration. In both cases, highmelting K-Ca phosphates formed, yet in the case with the Type II mechanism phosphate, an increase in the amount of potassium available for further melt forming reactions was observed. Moreover, the effect of Ca3(PO4)2 on the Type II mechanism phosphate with and without SiO2 was assessed from DSC/TGA experiments. The results indicated no melt formation or reactions with SiO2.

Moreover, defluidization experiments with four different agricultural residue biomass fuels were done with the fluidized bed reactor at various temperatures. The results were compared to those from the experiments with synthetic ash compounds.