EasyChair Preprint

№ 6696

CPFD Simulations on a Chlorination Fluidized

Bed Reactor for Aluminum Production: an
Optimization Study

Zahir Barahmand, Chameera Jayarathna and

Chandana Ratnayake

EasyChair preprints are intended for rapid

dissemination of research results and are
integrated with the rest of EasyChair.

September 26, 2021

 CPFD Simulations on a Chlorination Fluidized Bed Reactor for
Aluminum Production: An Optimization Study
Zahir Barahmand1 Chameera Jayarathna2 Chandana Ratnayake1,2
1
Department of Process, Energy and Environmental Technology, University of South-Eastern Norway
zbarahmand@[Link]
2
SINTEF Tel-Tek, SINTEF Industry, Porsgrunn, Norway

Abstract is dissolved in an electrolyte predominantly composed

of liquid cryolite (Na3AlFe6). The electrolyte is altered
Early CPFD simulation studies on designing a fluidized with calcium fluoride, aluminum fluoride, and other
bed reactor for alumina chlorination showed that the additives (Peterson & Miller, 2007).
model suffers from high particle outflow and dense In a typical alumina reduction cell, multiple prebaked
phase bed channeling. The present study is aimed to carbon anodes are immersed in the electrolyte, and as an
optimize the previous alumina chlorination fluidized intermediate product, oxide ions from alumina
bed reactor model through modified geometry, dissolution are discharged electrolytically onto the
parameter modifications, and improved meshing. To anodes. On the other hand, the oxide intermediate reacts
optimize the performance of the reactor, complex further with the carbon anodes, eventually consuming
geometry with an extended top section was combined them by producing gaseous carbon dioxide (CO2).
with a regular cylindrical reactor. Besides, the gas inlet Inside the electrolyte is a molten aluminum reservoir
pattern was changed from an ideal uniform distribution enclosed in a preformed composite lining and thermally
to a non-uniform one. Besides, the reactor’s inlet sealed by refractory and insulation components inside a
diameter is reduced, and the value for the particle steel shield (Thonstad, 2001). Hence, aluminum is
sphericity and voidage has been updated based on molded by reducing aluminum-containing anions at the
experimental observations. The results show that the electrolyte-metal interface. Although the term cathode
new reactor with an extended cross-sectional area on top is often used to refer to the whole tank of liquid metal
has a significantly lower particle outflow even with the and electrolyte, the actual acting cathode is the metal
higher inlet superficial gas velocity. The paper discusses pad or aluminum pool's top surface. The following
the optimization steps and relevant changes in reactor reaction is the overall reaction of dissolved alumina with
performances in detail. carbon to form the products (Barahmand, Jayarathna, et
Keywords: Optimization, fluidized bed reactor (FBR), al., 2021d).
alumina chlorination, Barracuda, CPFD simulation
½ 𝐴𝑙2 𝑂3 (𝑑𝑖𝑠𝑠𝑜𝑙𝑣𝑒𝑑) + ¾ 𝐶 (𝑠) (1)
1 Introduction → 𝐴𝑙 (𝑙) + ¾ 𝐶𝑂2 (𝑔)
Aluminum is now the world's second most used metal This process's total energy conversion is
(Donaldson & Raahauge, 2013). Since aluminum has a approximately 0.16–0.2 MJ/ton (Barahmand, 2021a).
unique combination of appealing properties and Using titanium diboride cathodes will also substantially
functionalities, it allows significant energy savings in reduce energy consumption (up to 20%). The
many applications, such as vehicles and buildings. manufacturing of a permanent anode is a more
Although this energy-saving leads to lower CO2 challenging task, and while large-scale experiments are
emissions, the production process of aluminum still in progress in Japan, Europe, and the United States, no
dramatically impacts the environment. success has been reported. More immediately,
In 1889, the melted cryolite-alumina electrolysis sophisticated controls can increase H-H cell efficiency
process known as Hall-Héroult (H-H) was started, and by up to 5%. To anticipate anode effects and optimize
the commercial production of metallic aluminum its positioning, the pattern of individual cell voltage
started, and this process has been used almost variations can be monitored and analyzed. Higher
exclusively in the aluminum production industry energy costs also induce cells to run at lower current
(Prasad, 2000). This process has been independently densities, resulting in higher efficiency (Survey of
developed and patented, after two young men, oceans Potential Processes for the Manufacture of Aluminum,
apart, around 140 years ago in the United States and 1979).
France. This discovery in 1886 by C. Hall and P. Héroult The search for feasible alternative processes for
has provided the world with the gleaming light metal at aluminum production has been accelerated by rising
reasonable prices. In this method, solid alumina (Al2O3) prices and a lack of large blocks of electrical energy. For
instance, the electrolysis of aluminum chloride, sulfide, In continuation of the studies on a New Sustainable
nitride; carbothermic reduction of ore or alumina; and Aluminum Production (NSAP) process (Barahmand,
the disproportioning reactions of aluminum sulfide or Aghaabbasi, et al., 2021; Barahmand, Jayarathna, et al.,
the mono-chloride route can be considered possible 2021d, 2021a, 2021c, 2021b), this optimization study
alternatives (Barahmand, 2021a). The aluminum aims to design an industrial alumina chlorination
industry is undergoing significant changes. It can no fluidized bed reactor (isothermally) to achieve the
longer be selective when it comes to developing sites minimum particle outflow and to improve the
based on the cost of electricity. Due to the limited hydrodynamics inside the reactor by minimizing the
electricity available for expansion, the next generation channeling effect in the dense phase bed.
of aluminum smelters must optimize their energy
efficiency. This shift has also impacted the importance 2 Overall Design Criteria
given to alternative process technology (Grjotheim &
Welch, 2016). All of the non-electrolysis processes The present study aims to design an industrial fluidized
necessitate extremely high temperatures that can only be bed reactor for pure γ-alumina chlorination in the
reached in an electric furnace. In each case, electrical presence of equimolar carbon monoxide and chlorine
energy consumption (17.6-22 kWh/kg) is estimated to gas mixture under the isothermal condition at 700℃.
be higher than that of a Hall-Héroult cell (13.2-17.6 The reactor should be designed for handling 0.6 kg/s of
kWh/kg) (Survey of Potential Processes for the alumina feed. There are no specified limitations for the
Manufacture of Aluminium, 1979). reactor dimensions or the geometry. However, it is
Except for the Alcoa chlorination process, there is no recommended to minimize internals (specifically for
technology for producing aluminum metal that can cooling and solid circulation). Moreover, because of
compete with the Hall-Héroult process in electrical some technical considerations, such as the possibility of
energy consumption. There are some other significant having a considerable percentage of α-alumina in the
advantages, although, that make the chlorination feed, at the first step, the rector has been designed with
process attractive. This process does not necessitate the no circulation. In the following sections, the main
use of pure aluminum oxide as a raw material design factors and considerations are discussed.
exclusively. Consequently, the Bayer process could be Circulation System: An early study (Barahmand,
skipped, eliminating the issue of disposing of vast Aghaabbasi, et al., 2021) ended up having a turbulent
amounts of red sludge (Barahmand, Jayarathna, et al., regime using an internal circulation system, but the
2021a). Carbochlorination can result in relatively high reactor will be designed for a bubbling regime with
CO2 concentrations in the process gas, making CO2 better hydrodynamics. On the other hand, considering
capture and storage easier to implement (Barahmand, the highly corrosive environment inside the reactor and
2021a). Carbon's mechanical properties, which are the existence of α-alumina impurity in the system, which
merely a chemical reactant in the chlorination of is not favorable, it avoids circulation internally or
aluminum chloride, are not required. Consequently, externally.
biocarbon may be used instead of coke from petroleum Gas-Solid Separator: The project's primary goal is to
refineries, which the Hall-Héroult process requires, minimize the solid escape from the system. Therefore,
necessitating high mechanical strength and density an external high-efficiency cyclone with an efficiency
anodes (Øye, 2019). of 99% will be designed to handle maximum solid
The industry has turned its attention to the two-step carryover. The cyclone is designed for half of the inlet
process of converting alumina to aluminum chloride and solid flow rate to the reactor (0.01-0.3 kg/s). The other
then further reducing the aluminum chloride to design parameters, such as pressure, fluid properties,
aluminum metal after failing to find a cost-effective and average particle size, will be calculated from
procedure for direct carbothermic reduction of alumina. simulations.
In the patent literature (Rao & Soleiman, 1986), two Regime and Bed Type: As mentioned earlier, the
chlorination processes are mentioned. The first is a reactor should be designed for the bubbling regime. A
fluidized bed that converts aluminum to aluminum free bubbling bed with no internal baffles is
chloride (AlCl3) at a temperature of 590°C. Hydrogen recommended to use. To have a smaller bubble size and
chloride, aluminum hydroxy chloride, aluminum lower rise velocity, the superficial velocity is chosen in
oxychloride, and sodium chloride are also delivered a range close to the minimum bubbling velocity.
significantly. The sodium comes from the alumina, Bed Aspect Ratio (H/D): The bed height (H) to the
which contains sodium as an impurity from the Bayer bed diameter (D) ratio is known as the bed aspect ratio
process (Barahmand, 2021a). The stoichiometry of (H/D), which is one of the most crucial factors for
chlorination of gaseous reactants is as follows, reactor design calculations (Shaul et al., 2012). The
superficial gas velocity by matching the required
𝐴𝑙2 𝑂3 (𝑠) + 3𝐶𝑙2 + 3𝐶𝑂 (2) fluidization regime is used to determine the bed
→ 2𝐴𝑙𝐶𝑙3 + 3𝐶𝑂2
diameter. The bed is generally called tall or deep if the
aspect ratio is more significant than unity. On the other
side, a shallow bed has an aspect ratio of one or less than
one. The precise aspect ratio that marks the transition
between a deep and shallow fluidized bed has yet to be
determined (Sathiyamoorthy & Horio, 2003). To have
better hydrodynamics in the bed, it has been avoided
using a shallow bed. The minimum aspect ratio has been
about unity as a safe value (Kunii & Levenspiel, 1991).
The given bed aspect ratio selected by the authors in an
early study (Barahmand, Jayarathna, et al., 2021a) has
been used in the current study.
Reactor Diameter: Based on the carbon monoxide
and chlorine mixture's stoichiometry and physical
properties, the needed volumetric flow rate of the fluid
at the inlet to handle 0.6 kg/s of solid can be calculated
easily. On the other hand, the fluid’s superficial velocity
range is chosen before being very close to minimum
bubbling velocity (Regime and Bed Type), which can be Figure 1. (a) Sieve plate sparger, (b) multiple ring sparger,
calculated too (Barahmand, Aghaabbasi, et al., 2021). (c) spider, and (d) pipe sparger (Kulkarni et al., 2009).
Reactor height: The height of a fluidized bed reactor Alumina: Many types of alumina have different
can be divided into dense and lean phases. Solids lose properties (Barahmand, Jayarathna, et al., 2021d). In the
density as they rise in height. The lean process's height present study, pure γ-alumina has been taken into
(or freeboard) can be separated into two zones, with the account, and the effect of impurities is neglected. Each
lower section known as the Transport Disengaging solid particle has many characteristic properties which
Height (TDH). If there is no secondary reaction in the affect the fluidized bed system. Parameters properties
freeboard region, the reactor outlet can be located on top such as particle size distribution, sphericity, voidage,
of TDH or above. Otherwise, it can reduce the freeboard and density should be defined as accurately as possible
and add a more efficient cyclone to the system. Both (Barahmand, Jayarathna, et al., 2021b).
dense-phase and TDH can be calculated for a bubbling Reactor Geometry: A simple cylindrical reactor with
regime (Kunii & Levenspiel, 1991). a uniform circular cross-section has been used in early
Operating Pressure: These parameters highly affect studies (Barahmand, Aghaabbasi, et al., 2021;
hydrodynamics by influencing fluid’s physical Barahmand, Jayarathna, et al., 2021a). The reactor's
properties (Barahmand, Aghaabbasi, et al., 2021). There optimum dimensions should be chosen by changing the
is no specific pressure in the system because the bed aspect ratio, superficial velocity, and reactor height.
pressure at the bottom and top of the reactor is not An exit geometry should be selected between smooth
identical, and there is a pressure drop to overcome the and abrupt (Harris et al., 2003; Mabrouk et al., 2008).
bed height of the reactor. The upstream pressure is the Any change in the reactor geometry can be applied to
bottleneck to define a pressure for the system. It is achieve lower particle escape and desirable
enough to calculate the needed pressure at the reactor's hydrodynamics. For example, cylindrical FB reactors
bottom by choosing a pressure for the fluidized bed inherited several weak points. Yang et al. have
outlet. Depending on how the flow boundary at the inlet investigated the effect of the expanded cross-section at
is defined, this pressure can be calculated automatically the top of the reactor and related hydrodynamics.
by the simulation software (Barracuda®) or defined Reaction kinetics: The present study simulates the
manually. single overall reaction (equation 2) in Barracuda®. The
Gas Distributor: The distribution mechanism in the reaction kinetics are based on (Barahmand, Jayarathna,
gas inlet of the fluidized bed reactor significantly affects et al., 2021d). Although the alumina chlorination at
hydrodynamics. For example, it can contribute to 700℃ is very fast and the reactor height can be reduced,
channeling in the bed or change the bubble size or the hydrodynamics inside the reactor plays a crucial
regime in the bed (Kunii & Levenspiel, 1991). At the role. The H/D highly affects the overall reactor height,
first step, a uniform and flat distribution throughout the and the reported value is minimum equal to unity (Kunii
whole inlet area will be used. In the next step, a grid & Levenspiel, 1991) for Geldart A particles. On the
plate or sparger should be considered a uniform other hand, reaction kinetics is highly correlated with
distributing system to be more realistic. Figure 1 shows temperature, particle size distribution, and porosity
different types of sparger arrangements in a fluidized (Barahmand et al., 2021d).
bed.
 Heat Transfer: Although the primary goal of the blending both the Wen-Yu and Ergun functions as the
present study is to touch the project objectives under following (Gidaspow, 2012).
isothermal conditions at 700℃, the thermal study of the For 𝜃𝑝 < 0.75𝜃𝑐𝑝 , the drag model can be calculated
chlorination process is one of the most critical design by the Wen-Yu model (𝐷1 ).
considerations (Barahmand et al., 2021c). The literature
confirms that the best and optimum temperature for the 24 −2.65
𝜃 𝑅𝑒 < 0.5
alumina chlorination process is 700℃, but taking the 𝑅𝑒 𝑓
industrial design limitations into account, 600℃ will be 𝐷1 = 24 −2.65 (1
𝜃 + 0.15𝑅𝑒 0.687 ) 0.5 ≤ 𝑅𝑒 ≤ 1000 (3)
the optimum temperature (Gokcen, 1983). 𝑅𝑒 𝑓
Solid Feeder: The current fluidized bed is a { 0.44𝜃𝑓 −2.65 𝑅𝑒 > 1000
continuous reactor that alumina is fed to the reactor with
a feeding rate of 0.6 kg/s. In general, the powder can be For 𝜃𝑝 > 0.85𝜃𝑐𝑝 , the drag model can be calculated by
transported mechanically or pneumatically (or air- the Ergun drag function (𝐷2 ) as below.
assisted). The screw feeder, as an example of a
mechanical conveyor or pneumatic conveying system, 𝜌𝑓 |𝑢𝑓 − 𝑢𝑝 |
180𝜃𝑝
can be used in the design. In this project, it is considered 𝐷2 = 0.5 ( + 2) (4)
that the powder is injected pneumatically using CO2. 𝜃𝑓𝑅𝑒 𝑟𝑝 𝜌 𝑝
Another critical point is the location and direction of the For 0.75𝜃𝑐𝑝 ≤ 𝜃𝑝 ≤ 0.85𝜃𝑐 , the following equation can
injection. In some cases, particles' downward movement derive the drag model:
positively affects the reaction (For example, alumina
chlorination in fluidized bed (Gokcen, 1983)). However, 𝜃𝑝 − 0.75𝜃𝑐𝑝
taking the other considerations, such as possible particle 𝐷 = (𝐷2 − 𝐷1 ) ( ) (5)
outflow, into account, placing the particle injection at 0.85𝜃𝑐𝑝 − 0.75𝜃𝑐𝑝
the reactor's bottom side-wall is beneficial. Where 𝜃𝑝 and 𝜃𝑓 are the particle and fluid volume
Construction Material: Although the current study
fraction respectively, 𝜃𝑐𝑝 is the particle volume fraction
does not directly deal with the materials, reactor design
is affected by general considerations. The typical at the close pack1, Re is the Reynolds number, 𝑢𝑓 and
fluidized bed uses a carbon steel shell lined with a 𝑢𝑝 are fluid and particle velocity, 𝜌𝑝 is particle density
particular alumina refractory (Barahmand, Aghaabbasi, and 𝑟𝑝 is the average particle radius.
et al., 2021).
Erosion: There are three primary sources for erosion 3 3D Multiphase Particle-in-Cell
in a fluidized bed, temperature, chemicals, and solid
particles (Barahmand, 2021b). All the internal surfaces
Approach
that contact a corrosive or very high-temperature fluid Barracuda®'s technology is based on 3D Multiphase
are in danger of erosion. On the other hand, in higher Particle-in-Cell (3D-MP-PIC), a patented
velocities, solid particles can cause erosion, and usually, computational technique for CFD simulation of gas-
it is associated with transitional and directional changes particle flows that includes close fluid-particle coupling
in the system. For example, most erosive wear may as well as careful consideration of thermal physics and
occur in the internal cyclone wall or near the elbow of reaction chemistry (Ahmadpour Samani et al., 2020).
bent pipes. In the present study, a particular alumina For dense particle flows, a three-dimensional,
refractory is considered a reactor lining to protect multiphase particle-in-cell approach is presented. The
against very high temperatures and chemical corrosion. computational technique uses a continuum model to
Although alumina particles are highly abrasive solve the governing equations of the fluid phase and a
(Haugland et al., 2019), this effect may be minimal Lagrangian model to solve the governing equations of
because of the low superficial velocity in the system. the particle phase (Snider, 2001). Through mapping
The erosion has not been considered in this study. particle properties to an Eulerian grid and then mapping
Drag Model: The force acting on a particle by the back-calculated stress tensors to particle positions, the
fluid flow around it is determined by the particle's drag difficulties associated with estimating inter-particle
model. The Barracuda® provides a range of predefined interactions with dense particle flows with volume
drag models that the WenYu-Ergun blended drag model fractions above 5% have been removed. A robust sub-
could be more suitable for the current study. Since the grid particle normal stress model for isolated particles
Wen and Yu correlation is appropriate for more dilute that eliminates the need for an implicit measurement of
systems and the Ergun relationship is appropriate at normal particle stress on the grid has been presented.
higher packing fractions, proposed a drag function The properties of interpolation operators that provide
compact support, conservatism, and a quick solution for

1
The close pack volume fraction specifies the maximum
volume fraction of particles when they are packed randomly.
a broad particle population are defined. The solution 4 CPFD Simulations
scheme allows for particle forms, sizes, and mass
distributions with no numerical diffusion from the The standard range for the bed aspect ratio (H/D) is not
Lagrangian particle equations. The fluid momentum and thoroughly investigated and discussed in the literature.
pressure equations are indirectly solved, resulting in a The best aspect ratio used in the present study is based
stable solution. on the authors' previous work (Barahmand et al., 2021a),
equal to 2. The reaction kinetics and alumina properties
There are two approaches to this, the Continuum and
are explained elsewhere (Barahmand et al., 2021d). The
the Particulate Phase. The continuity equation for a fluid
base model was developed with simple cylindrical
with no interphase mass transfer is (Verma & Padding,
geometry and a smooth exit in the first step. The smooth
2020).
exit does not affect the hydrodynamics of the reactor top
𝜕𝜃𝑓 (6) (Mabrouk et al., 2008).
+ ∇. (𝜃𝑓 𝑢𝑓 ) = 0
𝜕𝑡
The momentum equation for the fluid will be as
4.1 Base Model
equation (7). The chlorine concentration (as a factor of conversion
rate) and particle distribution (as a factor of reactor
𝜕(𝜃𝑓 𝑢𝑓 ) hydrodynamics) have been studied in the model. In the
+ ∇. (𝜃𝑓 𝑢𝑓 𝑢𝑓 )
𝜕𝑡 (7) first step, the gas reactants are distributed
1 1 homogeneously from the bottom of the reactor (ideal
= − ∇𝑃 − F + g𝜃𝑓
𝜌𝑓 𝜌𝑓 distribution). Figure 3 illustrates the particle mass flux
where, g is the acceleration gravity, F is “the rate of in different heights in the reactor. As seen in the figure,
momentum exchange per volume between the fluid and the red ring emphasizes particle escape through the
particle phases,” 𝑃 is fluid pressure, and 𝜌𝑓 is the fluid reactor wall.
density. In the Particulate Phase, the particle probability Studying the particle outflow shows that for uniform
inflow, the average particle outflow at the pseudo-
distribution function ∅ (𝑋, 𝑢𝑝 , 𝜌𝑝 , 𝛺𝑝 , 𝑡) is used to
steady-state is about 0.38 kg/s, almost 63% of the
define the dynamics of the particle process, where X is
particle inflow (Figure 2). On the other hand, the particle
the particle position, 𝑢𝑝 is the particle velocity, 𝜌𝑝 is the
escape through the wall harms reaction conversion. 'Due
particle density, and 𝛺𝑝 is the particle volume (Snider, to high resistance within the particle bed, the fluid temps
2001). For the time being, it is thought that each to escape close to the reactor wall, increasing fluid
particle's mass remains stable over time (i.e., no mass velocity in the near-wall region. As a result, the
transfer between particles or to the fluid), although reactants have less time to react.
particles may vary in size and density. The time
evolution is obtained by solving a Liouville equation
(Williams, 1985) for the particle distribution function.
𝜕∅ (8)
+ ∇. (∅ 𝑢𝑝 ) + ∇𝑢𝑝 . (∅𝐴) = 0
𝜕𝑡

where ∇𝑢 is the divergence operator concerning

velocity. Using the definition from (Andrews &
O’Rourke, 1996), the discrete particle acceleration, A,
can be defined as,
1 1 (9)
𝐴 = 𝐷𝑝 (𝑢𝑓 − 𝑢𝑝 ) − ( ∇𝑃 + τ) + g
𝜌𝑝 𝜃𝑝 𝜌𝑝
where, the terms describe acceleration due to
aerodynamic drag, pressure gradient, interparticle stress
gradient, and gravity, respectively, the Gidaspow drag
model (Gidaspow, 2012), which is a combination of the
Wen and Yu (Wen & Yu, 1966) and the Ergun (Ergun, Figure 2. Particle outflow (kg/s) with uniform distribution
1952) drag models (equations 3-5), can be used in
equation (9). The present study optimizes the alumina A non-uniform distribution pattern (Figure 4) with
chlorination in a fluidized bed reaction having a higher velocity in the middle and gradually decreasing
bubbling regime. The superficial velocity of the fluid is toward the inner walls (Figure 5) has been applied to the
close to the minimum bubbling velocity. system to solve this problem. The results have shown a
significant change in the reactor hydrodynamics and
particle escape.
 Figure 5. Inlet gas (CO+Cl2) mass flow rate of each cell-
color scale refers to Figure 4.

Figure 3. The particle mass flux (kg/sm2) in different

heights with a uniform distribution. c) Particle mass flux at
the specific time, and d) Average particle mass flux in the
last 300 seconds.

Figure 6. The particle mass flux (kg/s.m2) in different

heights with the non-uniform distribution. a) Particle mass
flux at the specific time, and b) Average particle mass flux
in the last 300 seconds.
Figure 4. Defined non-uniform ring distribution in Although the hydrodynamics has experienced a
Barracuda® considerable change, the chlorine concentration through
Changing the fluid distribution pattern shows a the reactor is almost constant due to the very high
considerable change in the bed’s hydrodynamics. As reaction rate. Table 1 gives the chlorine concentration
seen in Figure 6, the ring area is turned to a crescent through the reactor from bottom to top in the specified
shape. Compared with uniform distribution, the particle heights in Figure 3 and Figure 6. Although the results
outflow is 0.59 kg/s which is two times more and almost confirm that the conversion rate becomes complete at
identical to the particle feed rate. the bottom of the reactor, reducing the reactor height is
not suggested to keep the best hydrodynamics
(Barahmand, 2021a).
Table 1. Average Cl2 concentration (mg/l) in different
heights of the reactor.

Level Uniform Non-Uniform

0 289.27 65.23
1 0.35 0.59
2 0.05 0.04
3 0.03 0.02
4 0.02 0.02
5 0.02 0.04
6 0.03 0.03
7 0.02 0.03

In these simulations, the particle escape from the top of

the reactor is the biggest challenge.

4.2 Optimized model

As discussed in section 2, Yang et al. have used an Figure 8. Particle distribution through the reactor with the
expanded section to reduce the slugging and to solve this non-uniform flow.
problem effectively. Therefore, it may positively affect
reducing particle outflow (Yang & Keairns, 1980). As seen in Figure 8, the expanded bed is located in the
Figure 7 illustrates a schematic view of the new bottom cylindrical section of the reactor. As a result, a
geometry with the expanded cross-sectional at the top. negligible effect on particle escape through the reactor
In the current simulation, a total number of 65000 cells wall has been expected. In comparison with cylindrical
have been used. models, Figure 9 and Figure 10 confirm this
phenomenon.

Figure 7. The meshed geometry with the expanded cross-

sectional area on top
Similarly, for uniform and non-uniform distributions,
the bed hydrodynamics, particle outflow, and Cl2
concentration have been studied. As expected, the Figure 9. The particle mass flux (kg/s.m2) in different
geometry shows no effect on the reaction and chlorine heights with a uniform distribution. a) Particle mass flux at
consumption.
the specific time, and b) Average particle mass flux in the simple cylindrical geometry, non-uniform gas
last 300 seconds. distribution contributes to reducing high gas escape
Nevertheless, in both cases, particle outflow has been close to the wall, enhancing the reaction. Combining
dropped significantly. Table 2 gives comparative data geometrical modification and change in gas injection,
for all cases. the reactor has now shown quite promising
performances. It is crucial to validate the CFD
Table 2. Particle outflow (kg/s) in different cases simulation data with a lab-scale experimental unit as
Uniform gas Non-Uniform future work. Moreover, even though current simulations
Distribution gas Distribution are done based on mess specifications from CPFD
Cylindrical 0.38 0.59 software (Barracuda®), the mesh convergence can be
New Design 0.15 0.0004 helpful to find the right mesh size and further
improvement of the model.
Although the optimized geometry has shown a
remarkable performance in reducing the particle escape
from the top of the reactor, this design may suffer from
the possibility of particle deposition and caking
phenomenon in the conical top sections. Applying a
non-uniform inlet flow pattern has reduced the
channeling effect, and has resulted less scape through
the sidewall has been observed. Figure 11 shows a
channel created inside the bed.

Figure 11. Channeling effect in a cylindrical reactor with

uniform distribution.

References
N. Ahmadpour Samani, C. Jayarathna, and L. A. Tokheim.
CPFD simulation of enhanced cement raw meal fluidization
through mixing with coarse, inert particles. In proceedings -
61st SIMS Conference on Simulation and Modelling SIMS
Figure 10. The particle mass flux (kg/s.m2) in different
2020, Finland, 2020. doi:10.3384/ecp20176399
heights with the non-uniform distribution. a) Particle mass
flux at the specific time, and b) Average particle mass flux M. J. Andrews and P. J. O’Rourke. The multiphase particle-
in the last 300 seconds. in-cell (MP-PIC) method for dense particulate flows.
International Journal of Multiphase Flow, 22(2), 379–402.
1996. doi:10.1016/0301-9322(95)00072-0
5 Conclusion
Z. Barahmand. Design of an Industrial Chlorination Reactor
The modified geometry leads to minimizing the particle using CPFD Simulations, Master’s Thesis. University of
outflow significantly and helps the reactor’s South-Eastern Norway, 2021a.
hydrodynamics. On the other hand, in contrast with
Z. Barahmand, O. Aghaabbasi, E. K. L. Rustad, J. L. Salcido, R. Mabrouk, J. Chaouki, and C. Guy. Exit effect on the
C. Jayarathna, and C. Ratnayake. Designing of a medium- hydrodynamics of the internal circulating fluidized bed
scale circulating fluidized bed reactor for chlorination of riser. Powder Technology - POWDER TECHNOL, 182,
processed aluminum oxide. In proceedings - 1st SIMS 406–414, 2008. doi:10.1016/[Link].2007.07.008
EUROSIM Conference on Modelling and Simulation,
B. Øye. Could the chloride process replace the Hall-Héroult
Finland, 2021.
process in aluminum production?. 2019, March 28.
Z. Barahmand, C. Jayarathna, and C. Ratnayake. CPFD
W. S. Peterson and R. E. Miller. Hall-Héroult Centennial:
modeling of the hydrodynamics and reaction kinetics of
First Century of Aluminum Process Technology, 2007.
alumina chlorination in an industrial fluidized bed reactor.
In proceedings - 1st SIMS EUROSIM Conference on S. Prasad. Studies on the Hall-Héroult aluminum electro
Modelling and Simulation, Finland, 2021a. winning process. Journal of the Brazilian Chemical Society,
11, 245–251, 2000. doi:10.1590/S0103-
Z. Barahmand, C. Jayarathna, and C. Ratnayake. Sensitivity
50532000000300008
and uncertainty analysis in a fluidized bed reactor modeling.
In proceedings - 1st SIMS EUROSIM Conference on Y. K. Rao and M. K. Soleiman. Alumina chlorination. United
Modelling and Simulation, Finland, 2021b. States Patent No. US4565674A, 1986.
[Link]
Z. Barahmand, C. Jayarathna, and C. Ratnayake. Study of the
thermal performance of an industrial alumina chlorination D. Sathiyamoorthy and M. Horio. On the influence of aspect
reactor using CPFD simulation. In proceedings - 1st SIMS ratio and distributor in gas fluidized beds. Chemical
EUROSIM Conference on Modelling and Simulation, Engineering Journal, 93(2), 151–161, 2003.
Finland, 2021c. doi:10.1016/S1385-8947(02)00257-7
Z. Barahmand, C. Jayarathna, and C. Ratnayake. The effect of S. Shaul, E. Rabinovich, and H. Kalman. Generalized flow
alumina impurities on chlorination in a fluidized bed regime diagram of fluidized beds based on the height to bed
reactor: A CPFD study. In proceedings - 1st SIMS diameter ratio. Powder Technology, 228, 264–271, 2012.
EUROSIM Conference on Modelling and Simulation, doi:10.1016/[Link].2012.05.029
Finland, 2021d.
D. M. Snider. An Incompressible Three-Dimensional
D. Donaldson and B. Raahauge. Essential Readings in Light Multiphase Particle-in-Cell Model for Dense Particle Flows.
Metals, Alumina and Bauxite. John Wiley & Sons, 2013. Journal of Computational Physics, 170(2), 523–549, 2001.
doi:10.1006/jcph.2001.6747
S. Ergun. Fluid flow through packed columns. Fluid Flow
Through Packed Columns, 48, 89–94. Scopus, 1952. Survey of potential processes for the manufacture of
aluminium (ANL/OEPM-79-4). Little (Arthur D.), Inc.,
D. Gidaspow. Multiphase Flow and Fluidization: Continuum
Cambridge, MA (USA), 1979. doi:10.2172/5669730
and Kinetic Theory Descriptions, 2012.
J. Thonstad, J. Aluminium electrolysis: Fundamentals of the
N. A. Gokcen. Rates of chlorination of aluminous resource,
Hall-Héroult process, 2001.
pages 28, U.S. Department of the Interior, Bureau of Mines,
1983. V. Verma and J. T. Padding. A novel approach to MP-PIC:
Continuum particle model for dense particle flows in
K. Grjotheim and B. Welch. Impact of Alternative Processes
fluidized beds. Chemical Engineering Science: X, 6,
for Aluminium Production on Energy Requirements. In G.
100053, 2020. doi:10.1016/[Link].2019.100053
Bearne, M. Dupuis, & G. Tarcy (Eds.), Essential Readings
in Light Metals: Volume 2 Aluminum Reduction C. Wen and Y. Yu. Mechanics of fluidization. The Chemical
Technology, pages 1049–1055. Springer International Engineering Progress Symposium Series, 62, 100–111,
Publishing, 2016. doi:10.1007/978-3-319-48156-2_154 1966.
A. T. Harris, J. F. Davidson, and R. B. Thorpe. Influence of F. A. William. Combustion Theory. The Benjamin/Cummings
exit geometry in circulating fluidized-bed risers. AIChE Publishing Company, Inc, 1985.
Journal, 49(1), 52–64, 2003. doi:10.1002/aic.690490107
W. Yang and D. Keairns. The effect of an expanded section on
I. B. Haugland, O. Kjos, A. Røyset, P. E. Vullum, T. A. slugging, 1980. doi:10.1002/AIC.690260124
Aarhaug, and M. Halstensen. Alumina Scale Composition
and Growth Rate in Distribution Pipes. In C. Chesonis (Ed.),
Light Metals 2019 (pp. 697–706). Springer International
Publishing, 2019. doi:10.1007/978-3-030-05864-7_86
A. V. Kulkarni, S. V. Badgandi, and J. B. Joshi. Design of ring
and spider type spargers for bubble column reactor:
Experimental measurements and CFD simulation of flow
and weeping. Chemical Engineering Research and Design,
87(12), 1612–1630, 2009. doi:10.1016/[Link].2009.06.003
D. Kunii and O. Levenspiel. Fluidization Engineering.
Butterworth-Heinemann, 1991.