Development of solid phase viscosity models in turbulent fluidised beds

fingersfieldMechanics

Feb 22, 2014 (3 years and 1 month ago)

104 views

fingersfield_415ce72b
-
5210
-
4fef
-
b2fb
-
f7ba8b17b261.doc

Development of solid phase viscosity models in
turbulent fluidised beds


Luca Mazzei
1
, Paola Lettieri
1
, Tim Elson
1
, Derek Colman
2

1

Department of Chemical Engineering, University College London,

Torrington Place, WC1E 7JE, London, UK, p.lettieri@ucl.ac.uk

2

BP Chemicals Ltd, Sunbury on Thames, UK



ABSTRACT



Fluidization, the operation consisting in passing a fluid upwards through a bed of
particles in such a way as to confer the latter a fluid
-
like behaviour, has proved over the years
a winning and effect
ive technology thus exiting more and more the interest of the industrial
world.

Employed in a wide range of industrial processes such as catalytic cracking,
combustion and biomass gasification, this technology is also deemed to be the most
promising for op
timizing the available potential of renewable energy.

Albeit used extensively in commercial operations, nonetheless fluidization still poses
a major challenge to engineers when tackling the design of new industrial plants. The
performance of such systems,
for their very nature, is highly dependent on their
hydrodynamic behaviour which in turn is affected by the reactor geometry and size. Critical
scale
-
up problems are thus related to how accurately the performance changes with plant
size are accounted for i
n order to ensure that a full
-
size commercial unit will attain satisfactory
performance.

In this regard, CFD has proved a valuable research means; the main goal is
succeeding in simulating and investigating directly the behaviour of full
-
size unit so as to

avoid having to draw heavily on uncertain results obtained from pilot plants.

To this purpose, it is nevertheless critical that accurate models be developed, along
with appropriate constitutive equations. Whilst for some hydrodynamic regimes this has
alre
ady been partially accomplished with fulfilling results, for turbulent flow regimes the
research still lags considerably behind.

Turbulent fluidization is commonly utilized in industrial fluidized bed reactors due to
vigorous gas
-
solid contacting, favourab
le bed
-
to
-
surface heat transfer, high solids hold
-
ups
(typically 25
-
35% by volume) and limited axial mixing of the gas. Despite its practical
importance, turbulent fluidization has hitherto received much less attention than other flow
regimes, probably on
account of the challenges of experimental and theoretical work that this
particular regime poses.

When observed on a macro
-
scale, through X
-
ray imaging for example, a turbulent
bed appears to be uniform with the gas evenly distributed. While uniformity at
the macro
-
scale
is convenient for describing the bulk behaviour of the bed, at a meso
-
scale (say, up to 100
times the particle diameter) there is a significant local movement of the particles as they
interact with the fluidizing medium. Particles rapidly f
orm into and break out of clusters thus
giving rise to a “local turbulence” that on a macro
-
scale results in enhanced mass, momentum
and energy diffusion within the solid phase.

The major challenge is therefore contriving a model tailored for turbulent flo
w and
able to thoroughly account for these added contributions due to the very nature of this regime.
This challenge is undertaken in this work.

With this ultimate purpose in mind, the work will start by assessing experimentally the
overall behaviour of a
fluidized turbulent bed in terms of its overall flow properties and then
compare the results with those obtained by means of CFD modelling. In order to carry out the
experimental analysis, a novel technique will be employed based on the use of a
mechanical
ly stirred fluidized bed rheometer, recently developed at UCL.



Keywords: Fluidization; Turbulent flow regimes; CFD modelling; Fluidized bed rheometers.