(77b) Bubble Self-Arrangement in Annular Gas-Solid Fluidized Beds

Bubble Self-Arrangement in Annular Gas-Solid
Fluidized Beds

Kaiqiao Wu¹, Victor
Francia², Marc-Olivier Coppens¹

^{12.0pt;line-height:107%;font-family:" arial>1} EPSRC
“Frontier Engineering” Centre for Nature Inspired Engineering & Department
of Chemical Engineering, University College London, London.

^{12.0pt;line-height:107%;font-family:" arial>2}Institute
of Mechanical, Process and Energy Engineering, Heriot-Watt University, Edinburgh

*Email
of the corresponding author: m.coppens@ucl.ac.uk

line-height:107%;text-autospace:none"> line-height:107%;font-family:" arial>The dynamics of the bubbling gas
flow are critical to the performance of gas-solid fluidized
beds font-family:" arial>, but they are still hard to control, even in small,
simplified systems. Gas-solid contact time and transport processes rely heavily
on the bubble characteristics, such as size and velocity. Without the ability to
control them properly, operation and scale-up remain largely empirical
exercises. Some disadvantages associated with chaotic dynamics may be overcome by
imposing more “structure” to the flow. Conventional practices entail the use of
internals via baffles and heat exchanger tubes, which are typically used on an
ad hoc basis and suffer from additional costs in energy, pressure drop and
particle attrition. Instead, we have followed a more fundamentally grounded approach
by using the self-organizing nature of granular systems to create ordered
structures, as shown in Fig.1(a). In a line-height:107%;font-family:" arial>pseudo-2D system, an oscillatory perturbation of the gas
flow rearranges bubbles into triangular lattices, as shown in Fig.1(b) [1]. The flow structure is fully
scalable and leads to much more controllable properties, such as line-height:107%;font-family:" arial>bubble size
distribution, residence time, and spatial distribution. These unique features could
allow us to bypass some of the challenges in conventional units, like flow
maldistribution and non-uniform contact, as well as decouple conflicting design
objectives, such as good solid mixing and gas-solid contact.

line-height:107%;text-autospace:none"> line-height:107%;font-family:" arial> 

line-height:107%;text-autospace:none"> line-height:107%;font-family:" arial>Scaling up such structured
flows to an industrially relevant set-up will allow engineers to design a new
class of fluidized bed operation. Recently, based on the knowledge obtained
from our fundamental investigations [2-4], we have been able to recreate a similarly
structured flow in a cylindrical geometry, extending it from a flat-2D to an annular-3D configuration, Fig. 1(c). Nevertheless, the periodic
walls and strongly curved boundaries may result in instabilities affecting the
pattern formation process line-height:107%;font-family:SimSun">. In
this contribution, results from experiments are shown to investigate the
reproducibility of the 3D periodically structured pattern under different oscillatory
flows. These results also sketch
a multi-parametric operating window, which identifies a range of frequencies
and amplitudes of the applied flow to produce structured flows. Comparison of flow
properties in the flat-2D and the annular-3D system allows us to quantify the impact
of curved and periodic boundaries on the flow behavior.

line-height:107%;text-autospace:none"> line-height:107%;font-family:" arial> 

11.0pt;font-family:" arial font-style:normal>Figure 1 – Natural granular patterns (a) and formation of the structured bubble flow in
a pulsed quasi-2D bed (b), and in an annular-3D bed (c).

text-autospace:none">[1]
M.-O. Coppens, J.R. van Ommen, Structuring chaotic fluidized beds, Chem. Eng.
J., 96 (2003) 117-124

text-autospace:none">[2]
K. Wu, L. de Martín, M.-O. Coppens, Pattern formation in pulsed gas-solid fluidized
beds - the role of granular solid mechanics. Chem. Eng. J., 329 (2017) 4-14.

text-autospace:none">[3]
K. Wu, L. de Martín, L. Mazzei, M.-O.
Coppens. Pattern formation in fluidized beds as a tool for model
validation: a two-fluid model based study. Powder Technol., 295 (2016) 35-42.

text-autospace:none">[4]
L. de Martín, C. Ottevanger, J.R. van Ommen, M.-O. Coppens. Phys. Rev. Fluids,
3 (2018) p.034303.