Simulations of mono-sized solid particles in the reflux classifier under continuous process conditions
 
More details
Hide details
1
University of Engineering and Technology Peshawar Pakistan
CORRESPONDING AUTHOR
Naveedul Hasan Syed   

University of Engineering and Technology Peshawar Pakistan, Department of Chemical Engineering, UET Peshawar, 25000 Peshawar, Pakistan
Publication date: 2019-04-23
 
Physicochem. Probl. Miner. Process. 2019;55(3):631–642
 
KEYWORDS
TOPICS
ABSTRACT
In this study, a fluidized bed separator incorporating inclined channels, the Reflux Classifier (RC), was modeled to describe the transport behavior of mono-sized solid particles using a 2D computational segregation-dispersion model. The model is a volume flux-based model comprising the dispersion and segregation fluxes. Simulations were performed to examine variations in the solid volume fraction of particle species of size 0.163 mm and density 2450 kg/m3 by altering variables such as fluidization velocity, underflow rate and water flux in the feed. The system achieved a maximum solid volume fraction of 0.50 (v/v) near the base at a fluidization velocity 0.00020 m/s, that reduced to 0.20 at the fluidization velocity 0.0060 m/s. Overall, the results showed a decrease in the average solid volume fraction from 0.37 to 0.21 for the corresponding fluidization velocities. Simulation results also successfully demonstrated the capacity of the RC in retaining the solid particles at a superficial fluidization velocity 0.020 m/s, significantly higher than the terminal settling velocity, 0.015 m/s, of the solid particles, due to the presence of an inclined channel. Similarly, with increasing the underflow rate, the average solid volume fraction decreased from 0.29 to 0.055 due to the discharge of a larger quantity of solid particles from the base. Furthermore, a higher concentration of solid particles was observed in the inclined section at lower water flux in the feed stream. Additionally, flux balance calculations were carried out at different points within the RC to ensure the accuracy of the model predictions.
eISSN:2084-4735
ISSN:1643-1049