Numerical simulation of the heat transfer process in a fluidized bed with an inmersed tube

Abstract

This work aims to perform a numerical simulation of the heat transfer process of an immersed spherical surface in a bubbling fluidized bed. The experimental conditions of Di Natale et al. [1] were numerically reproduced: a hot sphere of 28 mm diameter immersed in a fluidized bed of 600 mm height. The sphere was in the middle of the bed, at a height of 300 mm, and with constant surface temperature of 373 K. The numerical simulation was approximated to a 2-D geometry, with a thickness of the bed of only 15 mm, in which the hot sphere is replaced by a horizontal cylinder resembling a tube. The bed was filled with spherical glass particles with a mean particle diameter of 0.5 mm, which were fluidized with atmospheric air at 293 K and with an air velocity of 0.3 m/s. The numerical simulations were carried out with the software CPFD-Barracuda, which is based on multiphase particle in cell (MP-PIC) method. This methodology is specific to simulate granular flows and solve the motion of groups of particles called “clouds”. In this way, the computational cost is notably reduced in comparison with a full Lagrangian simulation, in which the motion of all individual particles is solved. The numerical results permited a detailed analysis of the local heat transfer coefficient between the hot tube and the fluidized bed. Different heat transfer rates were observed around the tube. According to the results, the heat transfer coefficient is high at the bottom and at both sides of the tube (with values close to 200 W/(m2·K)), where the bubbles motion continuously replaces the heated particles in contact with the tube surface with new cold particles, creating a high heat transfer rate in those regions. In contrast, on the top of the surface, the particles are not fluidized. This was clearly observed in the simulation results: on the top of the tube the resulting particle velocity was close to zero and the particle volume fraction was close to the one at minimum fluidization conditions (0.6). This means that particles on top of the tube are at rest and they are not replaced by new cold particles by the action of the bubbles. Consequently, the heat transfer in this region remains low all the time (with values around 20 W/(m2·K)).