Options
Characterization of bubble dynamics in vibrated gas-solid fluidized beds with magnetic resonance imaging and numerical modelling
Publikationstyp
Conference Presentation
Date Issued
2026
Sprache
English
Author(s)
Citation
10th World Congress on Particle Technology, WCPT10 2026
Contribution to Conference
Publisher Link
Bubble properties in gas-solid fluidized beds play a crucial role in determining process efficiency, as they directly influence gas-solid contact. A thorough understanding of these properties is therefore essential for evaluating overall system performance. Due to the opaque nature of three-dimensional (3D) systems, characterization of these properties has predominantly relied on intrusive probes, which distort the flow, or on simulations, which require validation with 3D data. Recently, magnetic resonance imaging (MRI) has emerged as a powerful, non-invasive tomographic technique capable of resolving bubble properties and particle velocities within 3D fluidized beds with high spatio-temporal resolution [1]. In these recent work, clinical MRI systems have been employed, consequently, the horizontal orientation of the magnet bore limits the maximum sample height that can be investigated. In this work, a vertical MRI system is employed, allowing investigation of fluidized beds up to 2000 mm in height and 300 mm in outer-diameter. The fluidized beds can be moved inside the magnet bore, thereby different positions along the bed can be measured. Furthermore, this unique imaging system enables the investigation of vertically vibrated beds which are used in many processes due to enhanced fluidization behavior [2]. The investigation of vertically vibrated beds is mechanically challenging with a clinical MRI system because the radiofrequency (RF) shielding is typically integrated into the room in which the MRI system is located, preventing the use of vibration sources, such as electrodynamic shakers or motors, which generates RF signals. The compact
RF shield design of this MRI system allows for the placement of an electrodynamic shaker, an RF-noise emitting peripheral, along the axis of the fluidized bed in close proximity to the magnet. To transmit the vibrations from the electrodynamic shaker to the fluidized bed located 4 m above the floor level, a set of glass fiber reinforced epoxy tubes was used. A custom-designed bearing system ensures that the fluidized bed is radially centered inside the magnet bore, while still allowing for axial movement meanwhile isolating the magnet from vibrations.
MRI studies of fluidized beds often use agricultural seeds such as poppy and mustard seeds (Geldart D) due to their high oil content. To study a wider range of processes, the range of particles that can be investigated has been expanded in this work. For Geldart groups A, C and B, porous particles soaked in liquids and dried on their surface were used as the MRI-active solid phase.
For the different particle systems, the impact of vibration frequency and amplitude on bubble properties was investigated. It was found that for Geldart group B and D particles, increased vibration amplitude enlarged the bubbles, while frequency had no significant effect on the bubble properties. For Geldart group C, vibration caused expansion of the bed and initiated bubble formation (Figure 1c). Additionally, dynamically structured bubbling, where the normally chaotic motion of bubbles can be transformed into more organized and predictable patterns [3], was studied in 3D vibrated gas-solid fluidized beds.
MRI data were further compared with numerical simulations combining Computational Fluid Dynamics and the Discrete Element Method (CFD-DEM) to validate frequently used models in the simulations. CFD-DEM simulations showed good agreement with the MRI data for the conditions studied. We anticipate that this novel technique and expanded particle range will allow direct experimental observations of various processes, enabling the testing and further development of numerical models as well as empirical correlations.
References
[1] Penn, A. et al., Industrial & Engineering Chemistry Research, (2018)
[2] Guo, Q. et al., Current Opinion in Chemical Engineering, (2023)
[3] Guo, Q. et al., PNAS, (2021)
RF shield design of this MRI system allows for the placement of an electrodynamic shaker, an RF-noise emitting peripheral, along the axis of the fluidized bed in close proximity to the magnet. To transmit the vibrations from the electrodynamic shaker to the fluidized bed located 4 m above the floor level, a set of glass fiber reinforced epoxy tubes was used. A custom-designed bearing system ensures that the fluidized bed is radially centered inside the magnet bore, while still allowing for axial movement meanwhile isolating the magnet from vibrations.
MRI studies of fluidized beds often use agricultural seeds such as poppy and mustard seeds (Geldart D) due to their high oil content. To study a wider range of processes, the range of particles that can be investigated has been expanded in this work. For Geldart groups A, C and B, porous particles soaked in liquids and dried on their surface were used as the MRI-active solid phase.
For the different particle systems, the impact of vibration frequency and amplitude on bubble properties was investigated. It was found that for Geldart group B and D particles, increased vibration amplitude enlarged the bubbles, while frequency had no significant effect on the bubble properties. For Geldart group C, vibration caused expansion of the bed and initiated bubble formation (Figure 1c). Additionally, dynamically structured bubbling, where the normally chaotic motion of bubbles can be transformed into more organized and predictable patterns [3], was studied in 3D vibrated gas-solid fluidized beds.
MRI data were further compared with numerical simulations combining Computational Fluid Dynamics and the Discrete Element Method (CFD-DEM) to validate frequently used models in the simulations. CFD-DEM simulations showed good agreement with the MRI data for the conditions studied. We anticipate that this novel technique and expanded particle range will allow direct experimental observations of various processes, enabling the testing and further development of numerical models as well as empirical correlations.
References
[1] Penn, A. et al., Industrial & Engineering Chemistry Research, (2018)
[2] Guo, Q. et al., Current Opinion in Chemical Engineering, (2023)
[3] Guo, Q. et al., PNAS, (2021)
Subjects
Magnetic resonance imaging
Vibrated gas-solid fluidized beds
MRI-active solid particles
Numerical modelling
CFD-DEM
DDC Class
600: Technology