Rennebaum, Hannah SophieHannah SophieRennebaumBrummerloh, Daniel LukasDaniel LukasBrummerlohBenders, StefanStefanBendersPenn, AlexanderAlexanderPenn2025-12-122025-12-122025-11-20Chemical Engineering Science 322: 122977 (2026)https://hdl.handle.net/11420/60190Real-time magnetic resonance imaging (MRI) was used to compare the hydrodynamics of three-dimensional (3D) and pseudo-two-dimensional (2D) gas-solid fluidized beds of various wall separation distances by measuring the particle distribution and velocity. Five different superficial gas velocities from one to two times the minimum fluidization velocity and four pseudo-2D bed thicknesses between 15 and 50 mm were investigated. A neural network model was used to segment gas bubbles, revealing that thinner beds homogenize the radial bubble distribution and reduce the average equivalent bubble diameter compared to 3D beds. Furthermore, pseudo-2D beds showed lower bubble rise and average particle velocities. Based on the findings, an existing correlation for predicting bubble rise velocity is extended by incorporating an additional term that accounts for the influence of bed thickness. The minimum fluidization velocity and the expansion ratio increases as the bed thickness decreases. Yet, the shape of the bubbles in pseudo-2D fluidized bed remains unaffected when compared to 3D fluidized beds. Pseudo-2D fluidized beds with a smaller thickness, typically employed in optical experiments, exhibited the greatest deviation in hydrodynamics compared to 3D fluidized beds.en0009-25092025Elsevier BVhttps://creativecommons.org/licenses/by/4.0/Bubbling fluidized bedsPseudo-two-dimensional fluidized bedsMagnetic resonance imagingWall effectsBubble dynamicsParticle velocityTechnology::660: Chemistry; Chemical EngineeringTechnology::621: Applied Physics::621.3: Electrical Engineering, Electronic EngineeringJournal Articlehttps://doi.org/10.15480/882.1630810.1016/j.ces.2025.12297710.15480/882.16308Journal Article