CFD-DEM Modeling of Bubble Movement in a Rectangular Fluidized Bed

In this research, CFD-DEM was implemented to study a single bubble formation, elevation and eruption in a rectangular fluidized bed. In order to validate the CFD-DEM model, experiments were conducted in a 1(H) × 0.2(W) × 0.02(D) m bed. Glass beads particles (d_p = 1300 µm, density = 2500 kg/m³) were used to fill the bed at the static bed height of 0.22 m. Pressure probe was placed on the bed wall at the elevation of 0.025 m from the distributer to record the pressure fluctuations with the frequency of 400 Hz. Furthermore, a high-speed camera (240 fps) was installed in front of the bed (with the distance of 30 cm from the bed wall) to capture the bubble different states in the bed. While the bed was kept at the minimum fluidization (U_mf = 0.129), a pulse air flow (U_injection =0.85 m/s) was injected through embedded inlet (inlet diameter = 0.02m) for 0.23 s to form the single bubble. CFD-DEM model was conducted for the same bed and particle conditions. 463000 glass bead particles filled the rectangular bed. In order to validate the model, images from high-speed camera at different bubble states were compared with the images gained from CFD-DEM. For a more quantitative comparison, bubbles diameter was calculated from the images at different states, and plotted through time. The diagram revealed that CFD-DEM predicted the bubble diameter and rise time accurately. Estimated pressure fluctuations were showed that CFD-DEM was able to predict experimental pressure fluctuations.

It was observed that pressure fluctuations from the simulations had a maximum peak and a minimum. According to the images and pressure fluctuations time steps, when the bubble was introduced to the bed this maximum peak was observed. It was observed that after the start of the injection, the gas velocity started to increase near the injection tube. After 0.085 s the whole bed had a uniform velocity, which was in the order of magnitude of the injection velocity, and after this time the gas velocity decreased. Particles velocities distribution were also studied at the beginning of the injection step. Although the same pattern as the gas velocity was found for the particles, their maximum velocity reached smaller number. The particles hold-up was also drawn through time at the location where pressure signals were gained. The solid hold-up plot illustrated an increasing until 0.085 s, and after this time hold-up decreased. Regarding this information, it was concluded that after the beginning of the injection, the upward travelling compression wave increased the gas velocity through the whole bed. Then due to the friction force between gas and particles, particles were also accelerated and moved through the void spaces in the bed. By bubble growing on the distributer, the bed was expanded and hold-up decreased. Therefore, pressure decreased during bubble formation on the distributer. After the bubble detachment, injection was stopped and a minimum was observed on the pressure-time figure. Due to the stop of injection, compression wave stopped in the bed, and the mentioned minimum was observed. During bubble elevation in the bed, particles at the bottom part of the bubble formed the bubble wake. It was observed that particles move from the regions near the bottom of the bubble inside the bubble wake. Therefore, the hold up at the bottom of the bubble was less than the hold-up of the fluidization. When bubble passed the pressure probe and the distance between the bubble wake and the probe increased, particles moved back to the fluidization state. This phenomenon explained the pressure increase trend from the minimum to the steady part of the figure. Eruption of the bubble took place when it reached the surface of the bed. When the bubble erupted, a small increase was observed in the pressure. Eruption leaded to a relative weak compression wave (comparing to the compression wave caused by the bubble injection) which travelled downward through the bed. The increase at the end of the pressure signal was due to this downward compression wave.

The gas velocity profile inside the bubble was also studied during the formation, elevation, and eruption of the bubble using CFD-DEM. Results revealed that at the formation phase, gas velocity inside the bubble was uniform upward. During bubble rising in the bed, two vortices (one clockwise and one anticlockwise) were observed inside the bubble. When the bubble moved to the upper parts of the bed, these vortexes tend to separate. At this moment, particles start to fall into the bubble through the border of the vortexes. At the eruption point, the inside gas of the bubble traveled through the breakage area, and exited from the bubble from the breakage point. Particles velocity around the bubble at different phases were also illustrated using CFD-DEM results. It was shown that by starting the injection, the velocity vectors of all the particles were upward. Particles at the sides of the bubble in its growing phase, tend to travel downward and make the bubble wake. In the elevation phase same pattern of particles orientation as formation phase was observed. In this phase, particles at the bottom of the bubble had larger velocity than particles at the top of the bubble. Furthermore, there are few particles inside the bubble which their velocity vector is in accordance with the gas vortex inside the bubble.