Particle Mixing in Bubbling Fluidized Bed Reactors with Immersed Heat Exchangers and Continuous Particle Exchange

Abstract

Tracer experiments were conducted in a lab-scale cold flow model of 0.4 m by 0.2 m in cross section to investigate the solid mixing behavior of bubbling fluidized beds under continuous particle exchange with and without immersed tube bundle heat exchangers. The superficial gas velocity, the solid circulation rate, and the heat exchanger geometry were varied thereby. An established inductance measurement setup detecting pulse-injected ferromagnetic tracer particles was used to obtain the residence time distribution of the continuously exchanged solids. A profound mathematical routine was applied to calculate the particles’ mean residence time and characteristic values describing solid mixing phenomena. In this study, it was shown that the bubbling fluidized bed mixing characteristics are similar to mixed flow but superimposed by dispersed plug flow to a greater or lesser extent depending on the geometrical configuration of the immersed tube bundle heat exchangers. Furthermore, the results show that solid mixing enhances with increasing superficial gas velocity, whereby short-circuiting and sluggish turnover of solids were observed at fluidization numbers below 5. It was shown that these phenomena are increasingly promoted as the tube spacing decreases, and in general, quantifying them fundamentally improves the understanding of the fluidized bed reactor design process.

Wall-to-Bed Heat Transfer in Bubbling Fluidized Bed Reactors with an Immersed Heat Exchanger and Continuous Particle Exchange

“…Interestingly, the scattering range of the box plot at 206 kg·h –1 is slightly higher compared to the other box plots at low gas velocity. One reason for this may be the poor homogeneous mixing of the solids that was previously observed . The experiments with the high solids circulation rate of ṁ s = 323 kg·h –1 show that the median heat transfer coefficient reaches the observed maximum of 362 W·m –2 ·K –1 .…”

“…The particle size distribution analysis was done with a Mastersizer 2000 in accordance with the validation guidelines for laser diffraction systems in ISO 13320, USP ⟨429⟩, and EP 2.9.31. For the sake of completeness, it must be mentioned that in addition to the glass beads 0.2 kg of tracer particles, i.e., a maximum mass concentration of <1% by weight, were used in the experiments and were separated over time because the residence time distribution of the particles was measured in parallel to the heat transfer measurements (compare Eder et al ). The tracers were 1.4742 ferritic stainless-steel particles with a Sauter mean diameter d sv = 72 μm .…”

“…Therefore, various methods such as measurements, , correlative approaches, − or numerical simulations − and often several examinations with different approaches are necessary. As for the measurement method, a lab-scale cold flow model was utilized for the investigation of the wall-to-bed heat transfer , and subsequently used for the quantification of the solids residence time distribution and mixing characteristics. , A heat transfer measurement test device was developed to measure cumulated gas- and particle-convective heat transfer coefficients. , The superficial gas velocity, the tube bundle geometries, and the particle diameter were varied during these experiments. The obtained results from the experiments were compared with the results derived from mathematical models by Molerus et al, Lechner et al, Natusch et al, and Petrie et al This extensive experimental campaign revealed the general behavior that heat transfer coefficients decrease with decreasing tube spacing, which is also predicted by the models introducing a so-called tube bundle reduction factor, as proposed by Lechner et al or Natusch et al In view of maximizing the flexibility of the cold flow model and to also investigate cross-flow bubbling beds, the narrower side walls of the cold flow model that enclosed the fluidized bed were redesigned with openings, allowing the solids to flow from and to the fluidized bed to simulate a realistic stage of a multistage system with an effective countercurrent movement of gas and particles, that is, a cross-flow bubbling fluidized bed with continuous solids exchange.…”

“…The obtained results from the experiments were compared with the results derived from mathematical models by Molerus et al, Lechner et al, Natusch et al, and Petrie et al This extensive experimental campaign revealed the general behavior that heat transfer coefficients decrease with decreasing tube spacing, which is also predicted by the models introducing a so-called tube bundle reduction factor, as proposed by Lechner et al or Natusch et al In view of maximizing the flexibility of the cold flow model and to also investigate cross-flow bubbling beds, the narrower side walls of the cold flow model that enclosed the fluidized bed were redesigned with openings, allowing the solids to flow from and to the fluidized bed to simulate a realistic stage of a multistage system with an effective countercurrent movement of gas and particles, that is, a cross-flow bubbling fluidized bed with continuous solids exchange. In addition to the modified cold flow model, an inductance measurement device detecting pulse-injected ferromagnetic tracer particles was developed, built, and used to obtain the residence time distribution of the continuously exchanged solids. , Thereby, the superficial gas velocity, the solids circulation rate, and the tube bundle geometry were varied. It was shown that the particle mixing behavior inside the bubbling fluidized bed was mainly characterized by mixed flow but increasingly superimposed by the characteristics of dispersed plug flow when the tube spacing and/or the fluidization number decreased.…”

“…It was shown that the particle mixing behavior inside the bubbling fluidized bed was mainly characterized by mixed flow but increasingly superimposed by the characteristics of dispersed plug flow when the tube spacing and/or the fluidization number decreased. In addition, an increase of the solids circulation rate (lateral cross-flow) caused the proportion of the dispersed plug flow to increase as well . The knowledge obtained from the tracer experiments and the fact that the continuous temperature swing adsorption process is enabled via the solids circulation rate and that previous wall-to-bed heat transfer measurements were carried out without solids circulation rates lead to the conclusion that further heat transfer measurements with solids circulation rates are necessary to also investigate the impact of the solids circulation rate on the heat transfer coefficient.…”