(142q) Analysis of the Particle Viscosity and Drag Coefficient in Dilute Gas-Solid Flows in Ducts

In the chemical industry several
processes occur in the presence of particulate flows, such as for instance, the
upward flow of catalytic particles in riser reactors in the petroleum industry,
the pneumatic transport of solid particles in horizontal and vertical ducts,
among others. The study of particulate flow in ducts does not seem to be as
easy to study as expected. There are a lot of variables that may change the
characteristics of the gas-solid flow in ducts, such as particle diameter
distribution; particle-wall, particle-particle and particle-gas interactions; operational
conditions, design and so on. One way to get important information about the
flow conditions can be achieved by the acquisition of experimental data, which
can be also applied to configure different mathematical models and within
describe the entire system.

In order to analyze the behavior of
gas-solid flow, the experimental nonintrusive technique of PIV was applied.
This technique makes possible the acquisition of information about the
microscopic velocity field in the bidimensional plane.

The physical experiments were conducted
in the vertical and horizontal sections of a test facility. The operational
conditions in the inlet were 140 m³/h of air and a 40 g/m³
mass load ratio, which are typical conditions for dilute flows. Solid-phase
catalyst particles with a mean particle diameter of 56.7µm, similar to those in
petroleum FCC systems, were used. Experimental radial profiles for the axial
velocity data were compared with the respective numerical results obtained for
the CFD code FLUENT 12. Turbulence in the gas phase was modeled with a
k-epsilon model and the influence of the particle viscosity in the dilute phase
was evaluated. Furthermore, the analysis of the drag force comparing different
drag models, as Gidaspow, Morsi and Alexander, Syamlal and O'Brien, and a
procedure known as EMMS theory (Energy Minimization
Multi-Scale) was also investigated.

The numerical results obtained for
different positions in the vertical (L/D = 1.8, 5.8 and 10.3) and the
horizontal (L/D = 2.0, 5.5 and 13.0) duct showed a good agreement with the experimental
data obtained with PIV, although no difference can be observed among the particle
mean velocity profiles changing the viscosity for the solid phase. In
consonance with these results, the different drag models presented the same
behavior, showing that the mean particle velocity is not highly influenced by
the drag model in dilute flows. Furthermore, a positive result was observed
using the EMMS theory and presented in Figure 1 for the vertical section at L/D
= 1.8. The result showed that the EMMS theory avoids numerical problems for the
mean particle velocity profiles due to the small values of volume fraction for
the solid phase.

Figure 1 ? Mean
particle velocities using different drag models.