Packed-Fluidized Bed Reactors – Batch Experiments and Fundamental Modeling with Random Metal Packings

Licentiatavhandling, 2022

The aim of this work is to investigate the effect of random packings on heat and mass transfer phenomena, when applied in bubbling fluidized beds. For this purpose, first, the heat transfer coefficient to a horizontal tube submerged in a fluidized bed containing various types of random packings was investigated. Then, the conversion of different gaseous fuels during chemical-looping combustion (CLC) was studied in packed fluidized reactors with selected random packings. The experimental set-ups consisted of cylindrical laboratory-scale bubbling fluidized-bed reactors with an inner diameter of 78 mm and a height of 1.27 m.

For the first set of experiments, a horizontal tube (do=6 mm), through which there was a flow of water, was submerged in the bed. Air was used as fluidizing gas. Silica sand in the size range of 212-300 µm was used as bed material. Heat transfer coefficient, pressure drop and vertical segregation of solids were evaluated experimentally for bed temperatures ranging from 400 °C to 900 °C and superficial gas velocities from 0.04 m/s to 0.411 m/s. The bed height was 13 cm at rest for experiments, with and without packings. Five different types of packings were evaluated for the heat transfer experiments: i) RMSR (25 mm stainless steel thread saddle ring), ii) Hiflow (25 mm stainless steel pall ring), iii) RR6 (6 mm ceramic Raschig ring), iv) RR10 (10 mm ceramic Raschig ring) and v) ASB (12.7 mm aluminum silicate balls). For the second set of experiments, three of the packings (ASB, RMSR and Hiflow) were selected. CLC experiments were conducted using three fuels: CH4, CO and syngas (50/50% H2/CO), at temperatures between 840-940°C.

The results show that the nature of the packings has significant impact on the behavior of a packed-fluidized bed. Packings with low void factor such as RR6, RR10 and ASB had lower heat transfer coefficient, higher pressure drop and more significant vertical segregation, compared to a bubbling bed without packings. Packings with high void factor were quite different. The RMSR packing showed an improvement in heat transfer coefficient (up to 1243 W/m2K) at higher gas velocities, as compared to bubbling bed with no packings (up to 1124 W/m2K). Also, beds with RMSR and Hiflow packings had lower pressure drop, lower vertical segregation and higher fuel conversion in CLC compared to a bubbling bed with no packings.

It is concluded that packings with high void factor such as RMSR and Hiflow can be used in bubbling fluidized bed with small impact on pressure drop and solids segregation, and potentially positive effect on both heat and mass transfer. Packings with low void factor may be of interests for other applications, which remains to be explored.