Conclusion
To enhance the understanding of the difference between the fast and
turbulent fluidization regimes, the cluster, mass flux, and segregation
datasets of Geldart Group B particles were comparatively analyzed. While
the fluidized beds were of different sizes, the measurement instruments
used were the same, specifically, fiber optic probes for clusters,
extraction probes for local mass flux and sieve for segregation. The
particle systems investigated in both regimes included narrow PSDs,
binary mixtures, and broad PSDs.
Regarding clusters, all three characteristics (namely, probability,
duration, and frequency) were statistically different between the two
regimes. Relative to the fast fluidized bed, the turbulent bed gave
higher cluster probability and frequency due to higher bed density, and
lower duration due to more chaotic hydrodynamics. For cluster
probability, the relative dominance of the variables was similar for
both fast and turbulent fluidized beds, with h/H being most
dominant and Ug least. The mechanisms underlying
cluster formation appeared to be different in the two regimes, because
(i) the relative influences on cluster duration and frequency were
different for the two fluidized beds; and (ii) although the monodisperse
(i.e., narrow PSD) systems gave distinctly larger clusters than the
non-monodisperse (i.e., binary mixtures and broad PSDs) ones in the fast
fluidized bed, the different dispersity of the particle systems gave
similar cluster durations in the turbulent bed.
Regarding local mass flux, the data from the fast and turbulent
fluidized beds were statistically different, with the latter giving a
lower median. Although the median values were of the same order of
magnitude, the prevalence of downflow was much greater for the turbulent
bed. While an earlier study has shown that particle-related properties
had little influence on the overall mass flux in the fast fluidized bed,
it was shown here that particle-related properties played a more
significant role in the turbulent bed.
Regarding segregation, although the segregation extents were expected to
be different in the two fluidization regime to the much greater
dissipation and collisional stress in fast fluidization, the segregation
datasets from the two fluidized beds were found to be statistically
similar. Radial position (r/R ) was the most influential variable
on segregation in the fast fluidized bed, but the least in the turbulent
one, because of the more pronounced core-annulus profiles in the former.
While h/H had a negligible influence in the fast fluidized bed,
it was more significant in the turbulent bed.
The comparative analysis of the two fluidization regimes here,
therefore, provides more insights into the differences in flow
phenomena. Further studies recommended include investigation of whether
and under what circumstances does the core-annulus profile exist for the
turbulent regime, and why radial position influences local mass flux but
not segregation in the turbulent bed.