Chemical Engineering Science 134 (2015) 129–137

Contents lists available at ScienceDirect

Chemical Engineering Science

journal homepage: www.elsevier.com/locate/ces

Influence of fluid-mechanical parameters on volumetric mass transfer

coefficient in a spout–fluid bed with a draft tube
Milena Knežević, Dragan Povrenović n
University of Belgrade, Faculty of Technology and Metallurgy, Karnegijeva 4, 11000 Belgrade, Serbia

H I G H L I G H T S

Increase of water and air flow rate in the annulus increase leakage to the draft tube.

Reduction of particle circulation due to the presence of air in the annulus.

Increase of the particle diameter contributes to better fragmentation of bubbles.

Particle circulation improves oxygen mass transfer in spout–fluid bed with draft tube.

art ic l e i nf o a b s t r a c t

Article history: In order to apply a spout–fluid bed reactor with a draft tube for the nitrification process of wastewater
Received 4 January 2015 treatment, the influence of fluid-mechanical parameters on volumetric mass transfer coefficient was
Received in revised form analyzed. Experiments were carried out in a 1000 mm high 2D semi-column with a draft tube. The draft
30 March 2015
tube had a cross section of 50  50 mm2 and the annular region of the reactor had a cross section of
Accepted 17 April 2015
50  140 mm2. The work described in this paper reports the influence of several fluid-mechanical
Available online 29 April 2015
parameters including liquid and gas flow rates, particle diameters and particle circulation on volumetric
Keywords: mass transfer coefficient, kL a. The liquid flow ranged from 3.5–4.5 m3/h and gas flow ranged from 300 to
Three-phase system 800 L/h. The particles used were glass spheres with diameters of 3, 4, 5 and 6 mm. Tap water and air
Spout–fluid bed
were used as the liquid and gas phase, respectively. The experimental results have shown that particle
Draft tube
circulation through the draft tube contributes to kL a increase as well as the increase in the particle
Volumetric mass transfer coefficient
Particle circulation diameter, the gas and liquid flow rates.
& 2015 Elsevier Ltd. All rights reserved.

1. Introduction per unit volume compared to suspended growth reactors (Mendoza-

Espinosa and Stephenson, 1999).
The excess of nitrogen compounds in water-receptors is often the Numerous studies of different biofilm reactors for nitrogen removal
result of insufficient or inadequate wastewater treatment. The negative include fixed-film reactors (Hamoda et al., 1996; Villaverde et al.,
influence these compounds present on the environment reflects in 1997), fluid-bed reactors (Aslan and Dahab, 2008; Andalib et al., 2011;
eutrophication of natural water resources, toxicity to aquatic organ- Wang et al., 2012), airlift reactors (Heijnen et al., 1993; Nogueira et al.,
isms, and dissolved oxygen consumption (He et al. 2009; Zhang et al. 2002). However, in the available literature, no application of spout or
2011). Removing nitrogen compounds from wastewater is a necessary spout–fluid bed for the nitrification process was found.
measurement for prevention of environmental pollution. Biological The advantages of a spout system with a draft tube over a
removing based on nitrification–denitrification process is considered regular fluidized bed were also noticed in biomass pyrolysis
to be the most cost-effective treatment, so consequently it is the most process and were mostly reflected in better mass and heat transfer
widely used one. Biofilm reactors are often used for this kind of (Fernandez-Akarregi et al., 2013; Makibar et al., 2011; Zhang et al.,
treatment since they offer a higher concentration of microorganisms 2009). It was also noted that it was possible to use bigger and
coarser particles. Spout bed with a draft tube systems can also be
applied in coating processes (Ijichi et al., 2000; Publio and Oliveira,
n
Corresponding author. Tel.: þ 381 113303790. 2004), phenol degradation (Safont et al., 2012) and a denitification
E-mail address: [email protected] (D. Povrenović). process (Keshava et al., 2014).

http://dx.doi.org/10.1016/j.ces.2015.04.030
0009-2509/& 2015 Elsevier Ltd. All rights reserved.
130 M. Knežević, D. Povrenović / Chemical Engineering Science 134 (2015) 129–137

Two-phase spout–fluid bed systems come in two versions, as gas- suitability for usage in nitrification processes of wastewater
particle or liquid-particle systems. Intense particle–fluid contact is treatment.
typical for both versions, so numerous successful applications have
already been reported (Aguado et al., 2005; Plawsky et al., 2010).
The three-phase variation of the aforementioned system also 2. Experimental system
has two versions. The aerated liquid-particle system one has not
been thoroughly examined yet, but it is considered that it could be The experiments were conducted in a 1000 mm high 2D semi-
successfully applied in certain bioprocesses (Merchuk and Siegel, column made of Plexiglas with a cross section of 200  50 mm2.
1988) while the gas-particle system with dispersed liquid has The draft tube was 400 mm long and had a 50  50 mm2 cross
found its application in various drying processes (Altzibar et al., section. The distance between the inlet flow distributor and the
2008; Arsenijević et al., 2002; Berghel and Renström, 2014; draft tube was 40 mm. A schematic diagram of the experimental
Povrenović et al., 1992). Previous examinations of these systems system is shown in Fig. 1.
(Grbavčić et al., 1992; Littman et al., 2009; Povrenović, 1996) have Water was introduced to the system through a spouting inlet
shown that the spout–fluid bed reactor with a draft tube has nozzle (50  50 mm2) and through the annulus inlet nozzle over a
several advantages over the reactor with simple spout or spout– fixed bed distributor made out of glass particles having 6 mm in
fluid bed reactors. diameter. In order to prevent dead zone emersion, the bottom of
Due to its flexibility and intensive fluid–particle contact, the the annular part of the column was set at the angle of 301. Annular
three-phase spout–fluid bed reactor with a draft tube can be and spout flows were introduced by two separate pumps of
applied in aerobic processes of wastewater treatment. The limiting 350 W, with maximum work flow rate of 7.5 m3/h and the
factor in these processes is the oxygen mass transfer from the pressure of 1 bar, and for gas introducing a compressor was used.
gaseous to the liquid phase as a certain concentration of dissolved The gas was introduced on the bottom of the annulus over a tube
oxygen is a necessary precondition for the development of the distributor with 1 mm diameter orifices. The fluid flows were
desired nitrifying bacteria. Considering that oxygen is poorly measured by rotameters.
soluble in water, the absence of dissolved oxygen, as the only Liquid flow rate through the draft tube was set on 2.5 m3/h and
form available to the microorganisms, is one of the major causes of the flow rate through the annulus had values of 0.5, 1.0, 1.5, 2.0 m3/h.
process failure. An additional problem is the development of These flow rates were used with each particle diameter in order to
heterotrophic bacteria and the competition between them and examine the particle diameter influence on oxygen mass transfer.
autotrophic nitrifying bacteria for oxygen as an electron donor Liquid was recirculated through the column. Air flow rate varied
(Morgenroth and Wilderer, 2000; Okabe et al., 1995; Rittmann and from 300 to 800 L/h.
Manem, 1992; Rittmann et al., 2002), as well as for the space in The dependence of particle circulation through the draft tube
biofilms (Ohashi et al., 1995; Tijhuis et al., 1994; Van Benthum from fluid flow was examined by using a specially designed
et al., 1997). The efficiency of the oxygen mass transfer from the catcher that would not disturb the fluid or particle flow pattern
gas phase to the liquid phase is described through values of in any way. It had negligible resistance to free flow of water and
volumetric mass transfer coefficient kL a (Benyhaia et al., 1996). particles. The catcher was set on the surface of annulus by a long
Numerous investigations of various three-phase systems with handle. This handle enabled quick descending and withdrawal of
fluidized beds, biofilters, airlift reactors etc. have been performed, the catcher in a specific moment of time. The particle mass flow
examining the effects of gas and liquid velocity, type of gas rate was calculated from the mass of particles caught in the
distributor, characteristics of the solid phase in the system and catcher for a certain period of time.
liquid viscosity on volumetric mass transfer coefficient (Chen and Experiments were performed using tap water at room tem-
Leu, 2001; Freitas and Teixeira, 2001; Hamdad et al., 2007; perature. Dissolved oxygen concentration in the water was
Maldonado et al., 2008; Miura et al., 2012). reduced to concentration below 1 mg/L by sparging with nitrogen
Particles of different diameters and densities, such as glass, through the gas distributor. After the system was depleted from
polyethylene or silicon-dioxide ones were used in different experi- oxygen, the air was introduced to the system until the concentra-
ments as solid phase, as well as nickel powders and Ca-alginate, tion of the dissolved oxygen reached the 95% of the saturation
(Bukur et al. 1990; Chen and Leu, 2001; Freitas and Teixeira, 2001; concentration for the given temperature. Oximeter WTW Oxi 340i
Hamdad et al. 2007; Herskowitz and Merchuk, 1986; Lee et al., was used to measure dissolved oxygen concentration variation in
1993; Schumpe et al., 1989; Sivasubramanian, 2010; Vandu and the column with time. The oximeter probe was set on the top of
Krishna, 2004; Yang et al., 2001). the column. A dynamic oxygen desorption method, using air to N2
Tap water, ethanol, paraffin and tellus oil, CMC and tetradecan substitution, was used to measure the volumetric gas-liquid mass
are some of the substances used as the liquid phase (Freitas and transfer coefficient kL a (Letzel et al., 1999; Vandu and Krishna,
Teixeira, 2001; Hamdad et al. 2007; Vandu and Krishna, 2004). 2004) with assumption that liquid phase was perfectly mixed. This
Available literature provides a great deal of contradictory assumption was tested by injecting methyl violet in both the draft
results and conclusions regarding the effects of process para- tube region and the annular region and it was confirmed by the
meters on oxygen mass transfer in three-phase systems. Various visual mixing of these currents in the area where the oxi-probe
experiments have proved that volumetric mass transfer coefficient was set. This assumption was also confirmed by obtaining similar
depends on distribution and size of the bubbles in the system values of dissolved oxygen concentration at the top and at the
(Camarasa et al., 1999) and that bubble behavior, mostly affected bottom of the column.
by the dispersion system, has a great influence on both mass
dC

transfer and system's fluid-dynamics (Deckwer and Schumpe, ¼ kL a C n C t ð1Þ
dt
1993). It is also experimentaly established that the specific
n
influence of particular process parameters on bubble formation In Eq. (1), C denotes the oxygen saturation concentration and
and mass transfer relies on system's fluid-mechanics. C t is the dissolved oxygen concentration for a specific moment in
This paper describes experimental examination of fluid- time. After integration of Eq. (1), presuming that C 0 is the
mechanical parameters in the spout–fluid bed reactor with a draft dissolved oxygen concentration for t ¼ 0:
tube and their impact on oxygen mass transfer in the system. The


investigation was conducted in order to determine the system's ln C n C t ¼ ln C n  C 0  kL a U t ð2Þ
 M. Knežević, D. Povrenović / Chemical Engineering Science 134 (2015) 129–137 131

Fig. 1. Scheme of the experimental system.

By plotting the lnðC n  C t Þ as a function of time, a linear curve is 3. Results and discussion
obtained. The value of kL a can be calculated from the slope of
the curve. Throughout the experimental examinations of water-air-
Spherical glass particles of 3, 4, 5 and 6 mm in diameter, with a particle system reported in this paper, the draft tube distance
density of 2540 kg / m3, were used as a solid phase. The particle from the column bottom (the height of the entrance region)
bed mass was 3 and 4 kg. In order to compare the impact of the seemed to have no influence on spouting stability. According to
solid phase on the volumetric mass transfer coefficient,kL a values certain previous studies (Epstein and Grace, 2011; San José et al.,
were also determined in a two-phase system, with no particles 2007) of gas-particles conical spout bed systems, the ratio
present. between the entrance region and particle diameter had to be less
Pressure variations in the draft tube and the annulus were than 10 for a stable system operation. In the water-particles
measured by means of piezometers connected to pressure taps on system described in this paper the same ratio ranged from 6.66
the column. The variations were measured depending on the type to 13.33 and no system instability was noticed. The only system
of the particles and the flow rate value of water and air in both the instability was noticed in the case of smaller particle diameters
spout and the annular region by using piezometers set at 35 points combined with high inlet flow rates through annulus region due to
along the columns, regularly distributed by axial and radial the intense particle agitation.
direction. In defining the resulting quantity of the fluid leaking
out of the inlet spout flow into the annulus and vice versa, a 3.1. Influence of particle presence and size on volumetric mass
calibration diagram was used. This calibration diagram was transfer coefficient
acquired by measuring pressure drops throughout the particle
bed for specific flow rate values Q A ð0Þ while the entrance region of One of the main goals of the research described in this paper
the spout section and draft tube were mechanically blocked. After was to examine the influence of particle bed in a spout–fluid
that, the pressure drop was measured when a stable spout–fluid three-phase water-air-glass spheres system on oxygen mass
bed was established and based on these values and data from the transfer from the gaseous to the liquid phase. The results are
calibration diagram, the flow rates in the annular region Q A were expressed through volumetric oxygen mass transfer coefficient kL a
determined. This calibration diagram gives the relation between and show that the presence of particles significantly improves
pressure drop and flow rate and it differs for each air flow rate and mass transfer efficiency (Fig. 2), i.e. higher volumetric mass
particle type. transfer coefficient values are obtained in the three-phase system
The methyl violet was introduced for the visual display of fluid compared with the two-phase system without particles. Improve-
exchange between the spout and annular flows by pulse injecting ment of mass transfer in the presence of particles can be attributed
methyl violet into the spout and the annulus liquid flow line. to the constant renewal of mass transfer area between two fluids
In order to examine the effects of particle circulation in the (Lee et al., 2013), and it is considered that in this way the mass
spout–fluid bed with a draft tube on the oxygen mass transfer transfer resistance, expressed through mass transfer coefficient in
values were also determined when the particle circulation in the a liquid phase kL , is reduced. This conclusion is also consistent with
system was disabled with a mechanical barrier between the the experimental results obtained during other experimental
annulus and the draft tube. The inlet fluid flow rate values used studies. In one of these studies experiments were performed in a
were the same as in previous tests. fluidized bed reactor with glass spheres having 3, 5, 7 mm in
132 M. Knežević, D. Povrenović / Chemical Engineering Science 134 (2015) 129–137

The increase in liquid velocity enhances turbulence in the

three-phase system, increases the contact surface between the
phases as well as the surface renewal rate. All these effects
contribute to a more efficient oxygen mass transfer. The increase
of the liquid flow rate through the particle bed has the same effect
on kL a because it intensifies the change of velocity and direction of
the liquid flowing through the bed (Fig. 4). The positive effect that
velocity and liquid flow rate increase have on kL a is attributable to
better bubble fragmentation and consequent increase in specific
surface area (Kang et al., 2003). In contrast to these findings,
available literature offers results obtained in other studies which
show that increasing liquid velocity substantially reduces the
residence time of turbulent eddies which devalues the overall
impact of increasing liquid velocity to the increase in kL a value
(Lee et al., 2013). Results of some studies (Miura et al., 2012) even
suggest that kL a value decreases with increasing liquid velocity
whether a non-Newtonian or Newtonian fluid is in question. The
contradiction between these conclusions and the results obtained
in this study indicate that the kL a value is extremely affected by
Fig. 2. Influence of particle diameter on kL a values.
the fluid-mechanical and geometric characteristics of the system.

diameter using both Newtonian and non-Newtonian fluids (Miura

et al., 2012). In another an airlift reactor with both porous 3.3. Influence of the inlet anular flow rate and air flow rate on fluid
(alumina particles) and non-porous (ceramic) particles was used leakage over the entrance region
(Herskowitz and Merchuk, 1986). Similar findings can also be
found in a study where a slurry bubble column with iron oxide Fluid leakage over the entrance region from the annulus into
and silica particles was used (Bukur et al., 1990). the draft tube and vice versa appears in spout–fluid bed systems
The extent of particles' contribution to the improvement of mass with a draft tube (Povrenović 1996; Erbil, 2003). It is necessary to
transfer between the gas and the liquid reflects the combination of know the value of liquid flow rate which interacts with air in the
their features and the system´s characteristics. The results obtained annulus region if a spout–fluid bed with a draft tube and two inlet
during the tests described in this paper show that the value of kL a liquid streams (when the annulus stream is aerated and the one in
increases with increasing particle diameter in the range of 3–6 mm. spout is not) is to be applied in a process. The ratio between the
A similar conclusion was derived during other studies which inlet liquid flow rate on the bottom of the annulus and the liquid
included a fluidized bed of glass particles with diameters of 3, flow rate through the annulus is directly related to the intensity of
5 and 7 mm (Miura et al., 2012). leakage out of the annulus towards the draft tube and vice versa.
The particle diameter increase contributes to better fragmenta- In other words, it is related to the resultant value of leakage. The
tion of bubbles due to the increase of inertial force and its above-mentioned phenomenon would have a profound influence
momentum. By reducing the diameter of bubbles the specific in the implementation of these systems in biological treatment of
surface area between gas and liquid increases, and consequently wastewater since the feed wastewater stream would be intro-
the volumetric mass transfer coefficient increases as well (Han duced to the system through annulus and circulation of particles
et al., 1998; Kim and Kang, 1997, 2006). For smaller particles, with immobilized microorganisms would be provided by introdu-
increased pressure drop causes compression of air in the bed cing the treated wastewater through the draft tube. The resultant
bottom layer, which than causes bubble coalescence and weaker of the fluid leakage in the examined system, with and without air
air-particle contact in the bed (Fig. 3). Larger bubbles have a lower flow, is shown in Fig. 5. In a water-particle system, the increase of
specific surface area value, shorter retention time and a higher inlet anular flow rate causes intensified fluid leakage from the
rising velocity along the column compared to the smaller ones, annulus to the draft tube. In one of the experiments performed as
which reduces oxygen mass transfer. a part of this research methyl-violet was introduced to the system
by pulse injections. The results presented in Fig. 6 show that the
3.2. Influence of air and liquid flow rates on kL a leakage from the annulus into the draft tube increases with
annular flow rate increase.
With air flow rate increase and a constant liquid flow rate The presence of air in the annulus aditionally increases the
expressed as a sum of spout and annular flow rates in the spout– leakage from the annulus into the draft tube. With the increase of
fluid bed, kL a value increases, as shown in Fig. 4. This result is in air flow rate through the annulus, the leakage from the annulus to
compliance with a few other examined three-phase systems like the draft tube also incrases. An experiment performed in a
the fludized bed system with carbonate/bicarbonate buffer solu- cylindrical column with a flat bottom, where the fluid leakage in
tion, CO2 and glass beads (Chang et al., 1986), or the water-air- a spout–fluid bed with a draft tube was examined by means of
glass spheres (dp ¼0.4 mm) system in a circulating fluidized bed “tracer” method, has also shown that air affects leakage from the
reactor (Yang et al., 2001). Similar findings were also acquired in a annulus (Erbil, 2006).
fluidized bed system having nickel powder as the solid phase The results given in Fig. 5 show that smaller leakage is noted
(dp ¼0.194 mm) and carboxymethyl cellulose solution as the liquid within the system with bigger particles. With particle diameter
phase (Chen and Leu, 2001). decreasing, the bed porosity also decreases, and consequently,
Increasing the air flow rate at constant process parameters resistance to fluid flow increases causing intensified leakage from
increases the amount of retained gas, and with it the amount of anulus to the draft tube. This enabled a higher fraction of inlet
bubbles in the system. This way the specific surface area a annular fluid to be aerated. This further contributes to kL a increase
increases and therefore the kL a value rises as well (Chang et al., (beside the better bubble breakage which comes with bigger
1986; Lee et al., 2013) particles).
 M. Knežević, D. Povrenović / Chemical Engineering Science 134 (2015) 129–137 133

Fig. 3. Shapes of bubbles formed in the column during the introduction of air through the annulus (a) dp ¼ 3 mm; (b) dp ¼ 6 mm; mass of the particle bed is 3 kg in both
cases, while the particle diameter varies; Q g ¼ 300 L=h; Q s ð0Þ ¼ 2:5 m3 =h; Q A ð0Þ ¼ 1:5 m3 =h.

Fig. 4. Volumetric mass transfer coefficient value dependence of air flow rate and Fig. 5. Influence of gas flow rate on the leakage resultant for the system with a
total liquid flow rate through the column. different particles diameter, at same bed mass.

3.4. Influence of particle circulation and bed mass on kL a The presence of air in the annulus caused a reduction in the
flow rate of particles through the central tube despite contrary
The increase of the liquid flow rate through the draft tube, expectations based on the appearance of increased liquid leakage
happening as a consequence of leakage from the annulus, from the annulus to the draft tube in the presence of air (Fig. 7).
increased the intensity of particle circulation, and consequently Countercurrent movement of gas bubbles hampered the descend-
their downward velocity in the annulus. This way the intensity of ing of particles in the annulus toward the inlet region. The
fluid–particle contact is reinforced and mass transfer from the gas expansion of the particle bed in the annulus was also noted.
to the liquid phase is improved. Reduction of particle circulation due to the presence of air in the
The contribution of the increased particle circulation to kL a annulus has been shown in other studies (Erbil, 2006).
value increase has also been noted in various types of three-phase The increase of inlet fluid flow rate within the annulus
systems with fluidized bed and circulation (Lee et al., 2013; Son increases the particle circulation, while on the other hand, the
et al., 2007, 2008, 2009). air flow increase affects the actual bed weight decrease, reducing
134 M. Knežević, D. Povrenović / Chemical Engineering Science 134 (2015) 129–137

Fig. 6. Visual representation of annulus leakage with methyl violet pulse injection into the inlet annulus liquid flow, dp ¼ 6 mm; M ¼ 3 kg, Q s ð0Þ ¼ 2:5 m3 =h; Q g ¼ 0 L=h,
(a) Q A ð0Þ ¼ 0:5 m3 =h; (b) Q A ð0Þ ¼ 1:0 m3 =h, (c) Q A ð0Þ ¼ 1:5 m3 =h, (d) Q A ð0Þ ¼ 2:0 m3 =h.

Fig. 8. Influence of particle mass flow rate on a volumetric mass transfer coefficient
Fig. 7. Influence of air flow rate and liquid flow rate throught the draft tube on the
depending on air flow, with spouting inlet flow of 2:5 m3 =h, and annulus inlet flow
particle mass flow rate.
of 1:0 2:0 m3 =h.

in this way the quantity of particles transferring from the annulus been shown that a smaller amount of liquid leaks from the draft
into the draft tube over the inlet region. The cumulative effect of tube to the annulus when the bed mass is 4 kg compared to a 3 kg
these operating conditions is that kL a value increase is affected by bed (Fig. 9). With higher values of inlet flow rate through the
the gas flow increase through the annulus and by particles' annulus, the fluid leaks from the annulus to the draft tube and a
circulation increase through the draft tube, Fig. 8. larger amount of fluid leaks from the annular region to the draft
The amount of particles in the system also has a significant tube when the bed weight is 4 kg compared to the system with a
influence on the gas bubbles formation. The research results 3 kg bed. The impact of particles on the exchange of fluids through
concerning the three-phase systems found in available literature the inlet region is reflected in the increased aeration of the liquid
are contradictory. In a number of studies it was concluded that the in the system with 3 kg of particles compared to a system with
increase in particle quantity in the system increases the value kL a, 4 kg of particles. It can be deduced from the previous statement
most likely due to intense bubble fragmentation (Miura et al., that increasing the bed height in a spout–fluid bed system with a
2012). On the other hand, results in a couple of other studies draft tube used in this research above the minimum height
indicate that increasing the amount of particles causes a more required for a steady spout–fluid bed formation causes a slight
pronounced decrease in kL a values due to the coalescence of decline in the value of kL a, as can be seen in Fig. 10.
bubbles (Freitas and Teixeira, 2001). This behavior can be attributed to the fluid-mechanical condi-
A higher fluid flow resistance occurs when the particle bed tions at the bottom of the particle bed and at the spout-annulus
mass increases. When the inlet annulus liquid flow rate values are boundary below the central tube, rather than to the influence of
low, the leakage occurs in the draft-annulus direction and it has the bed height on the process of mass transfer.
 M. Knežević, D. Povrenović / Chemical Engineering Science 134 (2015) 129–137 135

Fig. 11. Influence of particle circulation on kL a values.

Fig. 9. Influence of the bed mass on the resultant leakage flow rate.

them, gave the following ranges of values: 0.005–0.0075 s  1 (Mena

et al., 2011), 0.008–0.07 s  1 (Lee et al., 2013), 0.007–0.068 s  1 (Chang
et al., 1986), 0.008–0.057 s  1 (Hamdad et al., 2007), 0.07–0.22 s  1
(Miura et al., 2012).
The magnitude of the kL a values obtained in these examina-
tions show that three-phase spout–fluid bed with a draft tube can
potentially be used in microbial wastewater treatment because
their unique configuration allows intense contact between the
phases and enables stable operation of aerobic processes.

4. Conclusions

The main objective of this experimental study was to examine

the volumetric mass transfer coefficient kL a dependence of fluid-
mechanical system parameters in spout–fluid bed with a draft tube.
Distribution of the inlet fluid flow between the annulus and the
spout is characteristic for spout–fluid bed reactors with a draft
tube. Fluid leakage occurs regardless of whether air is present in
Fig. 10. Influence of the bed mass on kL a values.
the annulus or not. The amount of fluid that will move from the
spout to the annulus and vice versa, depends on fluid flow rate
Despite the increased leakage, improved circulation which also values, the geometrical characteristics of the system and the
occurs with the increase of bed mass contributes to the rise in kL a characteristics of the solid phase.
values, hence the kL a value dependence of the bed mass is not Experimental results obtained in this study show that increase
significant. in particle diameter from 3 to 6 mm, for glass spheres used in this
In spout–fluid bed reactors with a draft tube particles move in a system, contributes to the increase of the volumetric mass transfer
circular way. Their movement goes from the bottom of the central coefficient value. Increase in gas and liquid flow rates, as well as
tube to its top and then continues through the annulus down to particle circulation, also contribute to the kL a value increase.
the inlet region, where again they are “sucked” into the draft tube. Mass transfer efficiency is considerably affected by the circular
The influence of particle motion on volumetric mass transfer movement of the particles. Higher kL a values were obtained in the
coefficient was tested by comparing two systems with the same system with a formed spout–fluid bed than in the system with a
particle size and the same inlet fluid flow rate. The first system fixed or a fluidized bed.
was a regular spout–fluid bed reactor with particle circulation, The results obtained in these examinations can be utilized in
while in the second system a barrier was placed under the central defining conditions for an efficient oxygen mass transfer in a
tube, so the particle circulation was disabled. The obtained results particular system and maintaining optimal aerobic conditions in a
are shown in Fig. 11, and the kL a values for the two systems nitrification reactor.
differed significantly. The higher value of the volumetric mass
transfer coefficient characterized the system with a well developed
spout–fluid bed, indicating that particle circulation improves Notation
oxygen mass transfer between the two phases.
A similar experiment, studying the volumetric mass transfer in a Q s ð0Þ Spout inlet flow rate, m3 =h
system with a spout–fluid bed and a draft tube, hasn´t been found in Q A ð0Þ Annulus liquid inlet flow rate, m3 =h
the available literature and therefore it is not possible to directly Qs Liquid flow rate in the draft tube, m3 =h
compare the results of this study with other studies. Examinations of QA Liquid flow rate in the annulus, m3 =h
different three-phase systems, and determining the kL a values in Q Total liquid flow rate through a column, m3 =h
136 M. Knežević, D. Povrenović / Chemical Engineering Science 134 (2015) 129–137

Qg Annulus air flow rate, L=h Kang, Y., Cho, Y.J., Lee, C.G., Song, P.S., Kim, S.D., 2003. Radial liquid dispersion and
Gp Particles mass flow rate, g=s bubble distribution in three-phase circulating fluidized beds. Can. J. Chem. Eng.
81, 1130–1138.
dp Particle diameter mm Joshi, Keshava, Rajan, Reshma, Srinikethan, G., Saidutta, M.B., 2014. Biological
kL a Volumetric mass transfer coefficient, s  1 denitrification with immobilized pseudomonas syringae on granular activated
M Particle bed mass in the column, kg carbon using three phase draft tube spouted bed reactor. Int. J. Curr. Eng.
Technol. 4 (5), 3304–3309.
Ct Dissolved oxygen concentration in the liquid, mg=L
Kim, S.D., Kang, Y., 1997. Heat and mass transfer in three-phase fluidized-bed
Cn Saturation dissolved oxygen concentration in the reactor: an overview. Chem. Eng. Sci. 52, 3639–3660.
liquid, mg=L Kim, S.D., Kang, Y., 2006. Hydrodynamics, heat and mass transfer in inverse and
C0 Dissolved oxygen concentration in the liquid for t ¼ 0, circulating three-phase fluidized-bed reactors for waste treatment. Stud. Surf.
Sci. Catal. 159, 103–108.
mg=L Lee, D.H., Kim, J.O., Kim, S.D., 1993. Mass transfer and phase holdup characteristics
in three-phase fluidized beds. Chem. Eng. Commun. 119, 179–196.
Lee, J.S., Jin, H.R., Lim, H., Lim, D.H., Kang, Y., Kim, S.D., Jun, K.W., 2013. Interfacial
area and liquid-side and overall mass transfer coefficients in a three-phase
circulating fluidized bed. Chem. Eng. Sci. 100, 203–211.
Acknowledgment Letzel, H.M., Schouten, J.C., Krishna, R., van den Bleek, C.M., 1999. Gas holdup and
mass transfer in bubble column reactors operated at elevated pressure. Chem.
Eng. Sci. 54, 2237–2246.
This work has been supported by the Ministry of Education, Littman, H., Paccione, J.D., Plawsky, J.L., 2009. New type of draft tube spout–fluid
Science and Technological Development of the Republic of Serbia bed. Part 1: hydraulic transport of 1.94 mm glass particles in water. Ind. End
under Grant no. 172022. Eng. Chem. Res. 48, 9286–9298.
Makibar, J., Fernandez-Akarregi, A.R., Alava, I., Cueva, F., Lopez, G., Olazar, M., 2011.
Investigations on heat transfer and hydrodynamics under pyrolysis conditions
References of a pilot-plant draft tube conical spouted bed reactor. Chem. Eng. Process.:
Process Intensif. 50 (8), 790–798.
Maldonado, J.G.G., Bastoul, D., Baig, S., Roustan, M., Hebrard, G., 2008. Effect of solid
Aguado, R., Prieto, R., San José, M.J., Alvarez, S., Olazar, M., Bilbao, J., 2005. characteristics on hydrodynamic and mass transfer in a fixed bed reactor
Defluidization modelling of pyrolysis of plastics in a conical spouted bed operating in co-current gas–liquid up flow. Chem. Eng. Process. 47, 1190–1200.
reactor. Chem. Eng. Process. 44, 231–235. Mena, P., Ferreira, A., Teixeira, J.A., Rocha, A., 2011. Effect of some solid properties on
Altzibar, H., Lopez, G., Alvarez, S., Jose, M., Barona, A., Olazar, M., 2008. A draft-tube gas–liquid mass transfer in a bubble column. Chem. Eng. Process. 50, 181–188.
conical spouted bed for drying fine particles. Dry. Technol. 26 (3), 308–314. Mendoza-Espinosa, L., Stephenson, T., 1999. A review of biological aerated filters
Andalib, M., Nakhla, G., Sen, D., Zhu, J., 2011. Evaluation of biological nutrient (BAFs) for wastewater treatment. Environ. Eng. Sci. 16 (3), 201–216.
removal from wastewater by Twin Circulating Fluidized Bed Bioreactor Merchuk, J.C., Siegel, M.H., 1988. Airlift reactors in chemical and biological
(TCFBBR) using a predictive fluidization model and AQUIFAS APP. Bioresour. technology. J. Chem. Technol. Biotechnol. 41, 105–120.
Technol. 102, 2400–2410. Miura, H., Katoh, T., Kawase, Y.i, 2012. Gas–liquid mass transfer in co-current three-
Arsenijević, Z., Grbavčić, Ž., Garić-Grulović, R., 2002. Drying of solutions and phase fluidized beds with non-Newtonian fluids: theoretical models based on
suspensions in the modified spouted bed with draft tube. Therm. Sci. 2, 47–70. the energy dissipation rate. Chem. Eng. J. 185–186, 337–346.
Aslan, S., Dahab, M., 2008. Nitrification and denitrification of ammonium rich Morgenroth, E., Wilderer, P.A., 2000. Influence of detachment mechanisms on
wastewater using fluidized-bed biofilm reactors. J. Hazard. Mater. 156, 56–63. competition in biofilms. Water Resour. 34 (2), 417–426.
Benyhaia, Jones, L., Plantaz, D., 1996. Mass transfer studies in pneumatic reactors. Nogueira, R., Melo, L.F., Purkhold, U., Wuertz, S., Wagner, M., 2002. Nitrifying and
Chem. Eng. Technol. 19, 425–431. heterotrophic population dynamics in biofilm reactors: effects of hydraulic
Berghel, J., Renström, R., 2014. An experimental study on the influence of using a retention time and the presence of organic carbon. Water Res. 36, 469–481.
draft tube in a continuous spouted bed dryer. Dry. Technol. 32 (5), 519–527. Ohashi, A., de Silva, D.G.V., Mobarry, B., Manem, J., Stahl, D., Rittmann, B.E., 1995.
Bukur, D.B., Patel, S.A., Daly, J.G., 1990. Gas holdup and solids dispersion in a three Influence of the substrate C/N ratio on the structure of multi-species biofilms
phase slurry bubble column. AIChE J. 36, 1731–1735.
consisting of nitrifiers and heterotrophs. Water Sci. Technol. 32 (8), 75–84.
Camarasa, E., Vial, C., Poncin, S., Wild, G., Midoux, N., Bouillard, J., 1999. Influence of
Okabe, S., Hirata, K., Watanabe, Y., 1995. Dynamic changes in spatial microbial
coalescence behaviour of the liquid and of gas sparging on hydrodynamics and
distribution in mixed- population biofilms: experimental results and model
bubble characteristics in a bubble column. Chem. Eng. Process. 38, 329–344.
simulation. Water Sci. Technol. 32 (8), 67–74.
Chang, S.K., Kang, Y., Kim, S.D., 1986. Mass transfer in two and three-phase fluidized
Plawsky, J.L., Littman, H., Paccione, J.D., 2010. Design, simulation, and performance
Beds. J. Chem. Eng. Jpn. 18, 524–530.
of a draft tube spout fluid bed coating system for aerogel particles. Powder
Chen, C.M., Leu, L.P., 2001. A highly elevated mass transfer rate process for three-
Technol. 199, 131–138.
phase, liquid-continuous fluidized beds. Chem. Eng. J. 81, 223–230.
Povrenović, D.S., Grbavčić, Ž.B., Hadžismajlović, Dž.E., Vuković, D.V., Littman, H.,
Deckwer, W.D., Schumpe, A., 1993. Improved tools for bubble column reactors
1992. Fluid mechanical and termal characteristics of spout–fluid drier with a
design and scale-up. Chem. Eng. Sci. 48, 889–911.
draft tube. Dry. Technol. 117, 343–351.
Epstein, Grace, 2011. Spouted and Spout Fluid Beds. Cambridge University Press.
Povrenović, D.S., 1996. Fluidmechanical characteristics and stability of a large
Erbil, A.C., 2003. Annulus leakage and distribution of the fluid flow in a liquid
diameter spout–fluid bed with a draft tube. J. Serb. Chem. Soc. 61 (45),
spout–fluid bed with a draft tube. Chem. Eng. Sci. 58, 4739–4745.
355–365.
Erbil, A.C, 2006. Effect of the annulus aeration on annulus leakage and particle
Publio, M.C.P., Oliveira, W.P., 2004. Effect of the equipment configuration and
cicculation in a three-phase spout–fluid bed with a draft tube. Powder Tehnol.
162, 38–49. operating conditions on process performance and on physical characteristics of
Fernandez-Akarregi, A.R., Makibar, J., Lopez, G., Amutio, M., Olazar, M., 2013. Design the product during coating in spouted bed. Can. J. Chem. Eng. 82 (1), 122–133.
and operation of a conical spouted bed reactor pilot plant (25 kg/h) for biomass Rittmann, B.E., Manem, J.A., 1992. Development and experimental evaluation of a
fast pyrolysis. Fuel Process. Technol. 112, 48–56. steady-state, multispecies biofilm model. Biotechnol. Bioeng. 39 (9), 914–922.
Freitas, C., Teixeira, J.A., 2001. Oxygen mass transfer in a high solids loading three- Rittmann, B.E., Stilwell, D., Ohashi, A., 2002. The transient-state, multiple-species
phase internal-loop airlift reactor. Chem. Eng. J. 84, 57–61. biofilm model for biofiltration processes. Water Resour. 36 (9), 2342–2356.
Grbavčić, Ž., Vuković, D., Jovanović, S., Garić, R., Hadžismajlović, Dž., Littman, H., Safont, B., Vitas, A.I., Penas, F.J., 2012. Isolation and characterization of phenol
Morgan III, M.H., 1992. Fluid flow pattern and solids circulation rate in a liquid degrading bacteria immobilized onto cyclodextrin-hydrogel particles within a
phase spout–fluid bed with draft tube. Can. J. Chem. Eng. 70, 895–904. draft tube spouted bed reactor. Biochem. Eng. J. 64, 69–75.
Hamdad, I., Hashemi, S., Rossi, D., Macchi, A., 2007. Oxygen transfer and hydro- San José, M.J., Alvarez, S., Ortiz de Salazar, A., Olazar, M., Bilbao, J., 2007. Operating
dynamics in three-phase inverse fluidized beds. Chem. Eng. Sci. 62, 7399–7405. conditions of conical spouted beds with a draft tube. Effect of the diameter of
Hamoda, M.F., Zeidan, M.O., Al-Haddad, A.A., 1996. Biological nitrification kinetics the draft tube and of the height of entrainment zone. Ind. Eng. Chem. Res. 46,
in a fixed-film reactor. Bioresour. Technol. 58, 41–48. 2877–2884.
Han, S., Zhou, J., Jin, Y., Loh, K.C., Wang, Z., 1998. Liquid dispersion in gas–liquid– Schumpe, A., Deckwer, W.-D., Nigam, K.D.P., 1989. Gas–liquid mass transfer in
solid circulating fluidized beds. Chem. Eng. J. 70, 9–14. threephase fluidized beds with viscous pseudoplastic liquids. Can. J. Chem. Eng.
He, S., Xue, G., Wang, B., 2009. Factors affecting simultaneous nitrification and 67, 873–877.
denitrification (SND) and its kinetics model in membrane bioreactor. J. Hazard. Sivasubramanian, V., 2010. Gas–liquid mass transfer in three-phase inverse flui-
Mater. 168, 704–710. dized bed reactor with Newtonian and non-Newtonian fluids. Asia-Pacific
Heijnen, J.J., van Loosdrecht, M.C.M., Mulder, R., Weltevrede, R., Mulder, A., 1993. J. Chem. Eng. 5, 361–368.
Development and scale-up of an aerobic biofilm air-lift suspension reactor. Son, S.M., Kang, S.H., Kang, T.G., Song, P.S., Kim, U.Y., Kang, Y., Kang, H.K., 2007. Gas
Water Sci. Technol. 27, 253–261. holdup and gas–liquid mass transfer in three-phase circulating fluidized bed
Herskowitz, M., Merchuk, J.C., 1986. A loop three-phase fluidized bed reactor. Can. bioreactors. J. Ind. Eng. Chem. 13, 14–20.
J. Chem. Eng. 64, 57–61. Son, S.M., Kang, S.H., Kim, U.Y., Shin, I.S., Kang, Y., Kim, S.D., 2008. Oxygen transfer
Ijichi, K., Uemura, Y., Hatate, Y., 2000. Coating of particles with finer particles using in a three-phase circulating fluidized bed with viscous liquid medium. J. Chem.
a draft-tube spouted-bed. J. Chem. Eng. Jpn. 33 (3), 526–528. Eng. Jpn. 41, 655–661.
 M. Knežević, D. Povrenović / Chemical Engineering Science 134 (2015) 129–137 137

Son, S.M., Kim, U.Y., Shin, I.S., Kim, S.D., Jung, H., 2009. Enhancement of gas holdup Wang, B., Wang, W., Han, H., Hu, H., Zhuang, H., 2012. Nitrogen removal and
and oxygen transfer in three-phase circulating fluidized-bed bioreactors with simultaneous nitrification and denitrification in a step-feed process. J. Environ.
swirling flow. J. Chem. Eng. Jpn. 42, 142–148. Sci. 24 (2), 303–308.
Tijhuis, L., Rekswinkel, H.G., van Loosdrecht, M.C.M., Heijnen, J.J., 1994. Dynamics of Yang, W., Wang, J., Wang, T., Jin, Y., 2001. Experimental study on gas–liquid
population and biofilm structure in the biofilmairlift suspension reactor for interfacial area and mass transfer coefficient in three-phase circulating flui-
carbon and nitrogen removal. Water Sci. Technol. 29, 377–384. dized beds. Chem. Eng. J. 84, 485–490.
Van Benthum, W.A.J., Van Loosdrecht, M.C.M, Heijnen, J.J., 1997. Control of Zhang, H., Xiao, R., Pan, Q., Song, Q., Huang, H., 2009. Hydrodynamics of a novel
heterotrophic layer formation on nitrifying biofilms in a biofilm airlift suspen- biomass autothermal fast pyrolysis reactor: flow pattern and pressure drop.
sion reactor. Biotechnol. Bioeng. 53 (4), 397–405. Chem. Eng. Technol. 32 (1), 27–37.
Vandu, C.O., Krishna, R., 2004. Volumetric mass transfer coefficients in slurry Zhang, J.B., Wu, P.X., Hao, B., Yu, Z.N., 2011. Heterotrophic nitrification and aerobic
bubble columns operating in the churn-turbulent flow regime. Chem. Eng.
denitrification by the bacterium Pseudomonas stutzeri YZN-001. Bioresour.
Process. 43, 987–995.
Technol. 102 (21), 9866–9869.
Villaverde, S., Garcia-Encina, P.A., Fdz-Polanco, F., 1997. Influence of pH over
nitrifying biofilm activity in submerged biofilters. Water Res. S0043–1354
(96), 00376 4.