Engineering Science and Technology, an International Journal 19 (2016) 189196

Contents lists available at ScienceDirect

Engineering Science and Technology,

an International Journal Press: Karabuk University, Press Unit
ISSN (Printed) : 1302-0056
ISSN (Online) : 2215-0986
ISSN (E-Mail) : 1308-2043

j o u r n a l h o m e p a g e : h t t p : / / w w w. e l s e v i e r. c o m / l o c a t e / j e s t c h
H O S T E D BY

Available online at www.sciencedirect.com

ScienceDirect

Full Length Article

Agitation of yield stress uids in different vessel shapes

Houari Ameur *
Institut des Sciences et Technologies, Centre Universitaire Salhi Ahmed CUN-SA, BP 66, Naama 45000, Algeria

A R T I C L E I N F O A B S T R A C T

Article history: The Agitation of yields stress uids with a six-curved-blade impeller (Scaba 6SRGT) is numerically in-
Received 18 February 2015 vestigated in this paper. The xanthan gum solution in water which is used as a working uid is modeled
Received in revised form by the HerschelBulkley model. The main purpose of this paper is to investigate the effect of vessel design
20 June 2015
on the ow patterns, cavern size and energy consumption. Three different vessel shapes have been per-
Accepted 20 June 2015
Available online 20 August 2015
formed: a at bottomed cylindrical vessel, a dished bottomed cylindrical vessel and a closed spherical
vessel. The comparison between the results obtained for the three vessel congurations has shown that
the spherical shapes provide uniform ows in the whole vessel volume and require less energy con-
Keywords:
Mechanical agitation sumption. Effects of the agitation rate and the impeller clearance from the tank bottom for the spherical
CFD vessel are also investigated. Some predicted results are compared with other literature data and a sat-
Yield stress uid isfactory agreement is found.
Flow pattern Copyright 2015, The Authors. Production and hosting by Elsevier B.V. on behalf of Karabuk
Power consumption University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).

1. Introduction Armenante et al. [6] determined, by experiments and CFD simu-

Agitation is of a great importance and it is used in many indus-
trial processes. In the various applications, stirred tanks are required
to fulll several needs like suspension of solid particles, disper-
sion of gases into liquids, heat and mass transfer, etc. Although many
experimental as well as numerical studies on liquid ows in cylin-
drical vessels have been published, very little attention has been
paid to the study of ow elds and energy consumption in fully
closed vessels.
Nagata [1] proposes the use of hemispherical bottomed cylin-
drical reactors, envisaging probably the improvement of the reactor
eciency, due to the cancellation of the prole discontinuity at the
wall-bottom junction. Other researchers [24] studied the power
consumption in closed vessels for Newtonian uids; they noticed
a discrepancy between their results (closed vessels) and the same
type of results for open vessels.
For a Newtonian uid, Medek and Fort [5] studied experimen-
tally the distribution of pressure along the lid of fully lled closed
cylindrical vessels at mixing by high-speed impellers (a Rushton
turbine or a pitched bladed turbine). They found that the power input
is not greater than that of an open cylindrical reactor, and the
pumping number of the same impeller is increased by approxi-
mately 10% probably due to the uniformization of the ow pattern
at the top end of the reactor.
* Tel.: +213770343722, fax: +21349797640.
E-mail address: houari_ameur@yahoo.fr
Peer review under responsibility of Karabuk University.
lations, the velocity proles and the turbulent kinetic energy
distribution for the ow generated by a pitched-blade turbine in
an unbaed, at-bottom, cylindrical tank provided with a lid, and
completely lled with water.
By numerical simulations, Ciofalo et al. [7] studied the Newto-
nian turbulent ow in closed and free surface unbaed tanks stirred
by at-bladed impellers. For a low viscous Newtonian uid and tur-
bulent ow regime, Taca and Paunescu [8] studied experimentally
the power input in a spherical closed vessel stirred by a Rushton
turbine or six pitched blade impeller. Taca and Paunescu [9] re-
ported that the optimum shape a vessel used for the suspension
of solid particles should have is the spherical one. These authors
reported also that some anomalies have also been noticed for the
fully lled lidded cylindrical reactors, as compared to the open re-
actors: the decrease of the power number for turbulent regime
(Re > 75,000), with the increase of Reynolds number and the im-
peller diameter.
By CFD simulations, Yapici and Basturk [10] studied the conju-
gate heat transfer and homogeneously mixing two immiscible
different uids in a stirred and heated hemispherical vessel. For a
Newtonian uid and turbulent ow regime, Ammar et al. [11] studied
by numerical simulations the effect of vessel design on the ow
pattern generated with a pitched blade turbine.
Agitation of yield stress uids results in the formation of a cavern
(well mixed region) around the impeller [12,13] and isolated regions
far away. Some works have been published using curved blade im-
pellers to evaluate the cavern size as a function of the power drawn
by yield stress uids including those by Galindo and Nienow [14,15]
for Lightnin A315 and Scaba 6SRGT impellers; Amanullah et al. [16]

http://dx.doi.org/10.1016/j.jestch.2015.06.007
2215-0986/Copyright 2015, The Authors. Production and hosting by Elsevier B.V. on behalf of Karabuk University. This is an open access article under the CC BY-NC-
ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
190 H. Ameur/Engineering Science and Technology, an International Journal 19 (2016) 189196

Fig. 1. Agitation system.

for axial ow SCABA 3SHPI impeller; Serrano-Carreon and Galindo
[17] for four different impellers (Rushton turbine, Chemineer He-
3, CD-6 and Scaba 6SRGT) in individual and dual arrangements;
Pakzad et al. [13,18,19] and Ameur et al. [20] for SCABA 6SRGT im-
pellers. Pakzad et al. [21] were interested with the agitation of
viscoplastic uids by Scaba-anchor coaxial mixers. For stirring shear
thinning uids, Ameur and Bouzit [22] studied the effect of curva-
ture blade on the power consumption. They found that the curve
bladed impeller requires less power consumption compared with
the at bladed impeller.
Our search in the literature shows that a little space has been
reserved to the agitation of viscoplastic uids by curved-bladed im-
pellers within closed vessels. Therefore, the main purpose of this
paper is to investigate the ow elds and the energy required for
the agitation of viscoplastic uids by a Scaba 6SRGT impeller. We
focus on the effects of vessel design on the ow elds, cavern size
and energy consumption. Three geometric congurations are re-
alized to perform the test: a at bottomed cylindrical vessel, a dished
bottomed cylindrical vessel and a spherical vessel. Effects of the ag-
itation rate and the impeller clearance within a closed vessel are
also investigated.
2. Mixing system
Effects of the vessel design are investigated in this paper by re-
alizing three types of vessels: a at bottomed cylindrical vessel, a
dished bottomed cylindrical vessel and a spherical vessel (Fig. 1).
Each vessel is equipped by a Scaba 6SRGT impeller (Fig. 2) which
consists of six curved blades xed on a disc with 8 mm of thickness.
The disc is attached on a cylindrical central shaft of diameter ds/
D = 0.05. The vessel height (H) is equal to its diameter (D),
D = 400 mm. All other parameters are listed in Table 1.
Effects of the impeller clearance from the tank bottom are also
studied. Three different congurations are realized for this purpose
and which are: c/D = 0.2, 0.35 and 0.5, respectively.
3. Mathematical background
The xanthan gum solution in water used in this study has a yield
stress behavior modeled by the HerschelBulkley model [18].
Thus, its apparent viscosity () is given by:
y
= + K  n1 (1)

where y is the yield stress,  is the shear rate, and K and n are the
consistency index and the ow behavior index, respectively.
According to the measurements conducted by Galindo and
Nienow [15], the rheological properties of the xanthan gum solu-
tion used were summarized in Table 2.
The HerschelBulkley model used causes a numerical problem
during the CFD simulations because the non-Newtonian viscosity
becomes unbounded at small shear rate. This behavior causes in-
stability during computation [23]. Thus, the modied Herschel
Bulkley model was employed to avoid the numerical instability. It
was assumed that the xanthan gum solution acts as a very viscous
uid with viscosity o at y and the uid behavior is described
by a power law model at > y [23]:

= 0 at y
n

n

y + K  n y (2)
0
= at > y


Table 1
Vessel parameters.

D H d ds dt bt

[mm] 400 400 60 20 8 6

Table 2
Rheological properties of the xanthan gum solution used in this work.

Xanthan gum concentration % K [Pa sn] n [] y [Pa]

(in mass content)

3.5 33.1 0.18 20.6

Fig. 2. Geometrical parameters of the impeller.
 H. Ameur/Engineering Science and Technology, an International Journal 19 (2016) 189196 191

The MetznerOtto correlation [24] was employed to calculate

the modied Reynolds number for the HerschelBulkley uids: V*z Num [present work]
Exp [13]
Nd Nd
2
Nd2 2
N d ks
2 2
N d ks 2 2
(3)
0.08
Re y = = = = =
 avg ks N y + K (k s N )
n
0.06
where  avg is the average shear rate and ks is a MetznerOtto
0.04
constant.
The power consumption is a macroscopic result obtained by in-
0.02
tegration on the impeller surface of the local power transmitted by
the impeller to the uid: 0.00
P= Qdv (4) -0.02
vessel volume

-0.04
The power number is calculated according to this equation:

P -0.06
NP = (5)
N 3d 5
-0.08
0.0 0.2 0.4 0.6 0.8 1.0
4. Numerical method Z*
Fig. 3. Axial velocity for Rey = 80.9, n = 0.12.
In this section, only a brief description of the computational tech-
niques used in this work is presented. Further details can be found
in a previous paper [25].
was ejected horizontally from the impeller and it was directed in
Once the discretization of the ow domain (tetrahedral mesh)
one of 2 ways. Either the uid owed up the wall toward the top
has been performed using ICEM-CFD, the CFD solver CFX 13.0, Ansys
surface and then back down the impeller shaft, forming a recircu-
Inc., is used to obtain values for the velocity components and pres-
lation zone above the impeller, or the material was directed down
sure at each of the node points. The solver uses a full nite volume
to the dish center and was pulled up through the center of the dish
formulation to solve the fully-coupled mass and momentum con-
back to the blade, forming a second recirculation zone below the
servation equations. The ow elds are solved in a rotating frame
impeller. In a comparison between the three cases, the recircula-
of reference in which the ow is time-independent.
tion loop center is very close to the blade tip for the spherical vessel,
In this study, the Reynolds number is varying from 1 to 3000.
and this is due to the vessel design.
Even if the ow is not fully into the turbulent regime, we should
The pressure distribution inside the three vessel shapes is also
simulate the ow as being turbulent. Fortunately, the turbulent prop-
presented (Fig. 6). As remarked, the pressure is strong in the area
erties vanish when in the laminar ow, so the correct way of
swept by the impeller and highest value is reached at the blade tip
simulating in the transitional regime is using a turbulence model.
for the three vessels. In a comparison between the three cases, the
In our study, we have used the SST model with specied intermit-
spherical shape yields the stronger pressure on the blade tip.
tency of turbulence.
For a at bottomed cylindrical vessel (Fig. 7a), we remark that
In each simulation, the solution is considered converged when
the liquid tends to move mainly along circular trajectories, result-
the normalized residuals for the three velocity components and the
ing in weak radial ows directed toward the tank walls and small
pressure all fall below 106. Calculation of the velocity and pres-
relative velocities between the impeller and the uid. This results
sure elds takes between 90 and 300 iterations to converge and each
in a poor axial mixing and in the formation of a vortex on the free
run takes less than ve hours of CPU time. The computations were
run in Core i7 CPU 2.20 GHz with 8.0 GB of RAM.

5. Results and discussion Np

Exp [26]
10
At the beginning, it is necessary to check the validity of the CFD Num [Present work]
code and the numerical method performed. For this objective, we
referred to the experimental work presented by Pakzad et al. [13].
With the same geometry (i.e. a baed cylindrical vessel with a at
bottom), variations of the axial velocity along the vessel height are
predicted and presented on Fig. 3. Another validation is made with
the experimental data of Woziwodzki and Sowinski [26]. These
authors used other impellers but the same type of uid. For a
Rushton turbine and a Xanthan gum solution with 1% of concen-
tration, values of the power number versus Reynolds number are
presented on Fig. 4. As observed on these gures (Figs. 3 and 4),
the comparison between our numerical results and the other ex-
perimental data shows a satisfactory agreement.

1 10 100
5.1. Effect of vessel shape Rey
Figure 5 presents the ow patterns in a vertical plane contain- Fig. 4. Power consumption for a Rushton turbine and 1% of Xanthan gum
ing the blade. It can be observed that for any vessel shape, the uid concentration.
192 H. Ameur/Engineering Science and Technology, an International Journal 19 (2016) 189196

Fig. 5. Flow patterns for Rey = 2500, c/D = 0.5, (a) at bottomed cylindrical vessel, (b) dished bottomed cylindrical vessel, (c) spherical vessel.

surface of the liquid, whose depth depends on the impeller rota-
tional speed. This is due to the absence of baes.
At the same impeller rotational speed, if we change the vessel
shape by using a dished bottom, it can reduce the size of the vortex
which is formed on the free surface of the liquid (Fig. 7c).
The agitation of viscoplastic uids results in the formation of a
well stirred region (cavern) near the impeller and stagnant zones
elsewhere. The stagnant zones yield poor heat and mass transfer
rates, high temperature gradients, etc. [27]. Therefore, the cavern
size should be increased as well as much as possible.
Figure 8 presents the cavern size generated by the Scaba for dif-
ferent vessel shapes. As observed, the cavern is limited in the area
swept by the turbine. Another remark is that the difference is slight
between the three cases studied for low Reynolds number. The

Fig. 6. Pressure distribution for Rey = 2500, c/D = 0.5, (a) at bottomed cylindrical vessel, (b) dished bottomed cylindrical vessel, (c) spherical vessel.

Fig. 7. Flow patterns for Rey = 1000, c/D = 0.2, (a) at bottomed cylindrical vessel, (b) dished bottomed cylindrical vessel, (c) spherical vessel.
 H. Ameur/Engineering Science and Technology, an International Journal 19 (2016) 189196 193

Fig. 8. Cavern size for c/D = 0.5, (a) at bottomed cylindrical vessel, (b) dished bottomed cylindrical vessel, (c) spherical vessel.

increasing Reynolds number yields a wider cavern for cylindrical
vessels than the spherical vessel (Fig. 8). But and if the impeller is
placed very close to the vessel base (Fig. 7), toroidal vortices will
appear near the free surface of liquid within cylindrical vessels. These
vortices may cause huge vibrations and consequently a damage of
the rotating impeller. This issue can be eliminated by using a spher-
ical closed design.
At the vessel mid-height (Z* = 0.5), the radial velocity compo-
nent is presented along the vessel radius for the three types of vessel
(Fig. 9a). We remark that the at bottomed cylindrical vessel gen-
erates a powerful radial jet of ow when compared with the two
other cases.
The axial velocity component (V*z) is also presented for more de-
scription of the hydrodynamics induced (Fig. 9b). The V*z proles
are plotted along the vessel height (Z*) for a radial position near
the blade tip (R* = 0.6). The minus sign of velocity indicates the ex-
istence of a recirculation zone. At the lower part of the vessel, the
axial velocity magnitude is intense for the case 1. However, at the
upper part of the vessel, this velocity component becomes more
intense within the spherical vessel.
The uniformization of the ow pattern at the vessels upper part
(for a closed spherical vessel) results in the recovery of centrifugal
forces and the increase of impeller ow rate.
As reported by Taca and Paunescu [8], the closed vessels are
highly useful in several mixing processes, especially if steep pres-
sures are needed and when the operation of agitation without the
presence of dispersed gases (air) are needed. A closed vessel is also
favorable as the whole vessel can be used since it may be lled all
the way to the top. Moreover, the vessel bases are not required for
preventing the vortex formation.
5.2. Effect of impeller clearance
The impeller location has an important effect on the perfor-
mance of a mixing system. In this section, we investigate the effect
of impeller clearance from the tank bottom. Three geometric

Vr* Flat bottomed cylindrical vessel Z* Flat bottomed cylindrical vessel

Dished bottomed cylindrical vessel 1.2 Dished bottomed cylindrical vessel
0.30 Spherical vessel
Spherical vessel
1.0
0.25
0.8
0.20
0.6
0.15
0.4
0.10
0.2
0.05
0.0
0.00
-0.04 -0.03 -0.02 -0.01 0.00 0.01 0.02 0.03 0.04
Vz*
0.2 0.4 0.6 0.8 1.0 R*
(a) (b)
Fig. 9. Velocity proles for Rey = 1000, = 0, c/D = 0.2, (a) Radial component at Z* = 0.5 (b) Axial component at R* = 0.6.
194 H. Ameur/Engineering Science and Technology, an International Journal 19 (2016) 189196

Fig. 10. Flow patterns for Rey = 70.

congurations have been tested for a spherical vessel and which
are: c/D = 0.5, 0.35 and 0.2. As observed on Fig. 10a, the uid ow
is blocked near the vessel base when the impeller clearance is very
small, resulting in higher power consumption (Fig. 11). On the other
hand, the rest of the vessel is poorly stirred (Fig. 10b).
Knowledge of the power consumption is very important for the
choice of the system installed. The power consumption depends on
all parameters characterizing the impeller geometry, the vessel ge-
ometry and the impeller rotational speed.
Figure 11 presents the numerical values of the power consump-
tion required for different vessel shapes and different clearances from
the vessel base. The numerical investigations showed the superior
performances of spherical reactors, as compared to the classical at
bottomed and dished bottomed cylindrical reactor. The improved
Flat bottomed cylindrical vessel
Np Dished bottomed cylindrical vessel
Spherical vessel
4.8
4.7
eciency of the spherical reactor is due to its shape, having non
contour discontinuities at the wall-bottom and wall-lid junctions,
as would be the case for a cylindrical vessel. That demonstrates the
existence of the so-called sphere effect [9].
5.3. Effect of Reynolds number
The ow patterns generated by the Scaba for different agita-
tion rates are presented on Fig. 12. With a low Reynolds number
(Case 1: Rey = 50), we remark that the radial ow impinging from
the impeller blade is weak, and centers of the recirculation loops
which are formed above and below the impeller are located in the
area swept by the blade. With increasing Reynolds number (Case
2: Rey = 200), these recirculation loops are displaced horizontally
toward the tank sidewalls. The center of each loop is located at the
blade tip in this case. With a greater Reynolds number (Case 3:
Rey = 3000), these loops are detached from the blade tip. This is due
to the higher radial jet which can increase the size of these structures.
For more details on this phenomenon, we presented on Fig. 13
the variation of axial velocity along the vessel radius for two loca-
tions: Z* = 0.26 (below the impeller) and Z* = 0.44 (above the
impeller). The minus sign of the velocity indicates the existence of
a recirculation loop. As observed, the increase of Reynolds number
yields a larger recirculation zone and a wider cavern (Fig. 13b).

4.6 6. Conclusion

The ow elds and the energy required for the agitation of

4.5 viscoplastic uids by a Scaba 6SRGT impeller have been numeri-
cally investigated. Three different vessels were used to study the
design effect on the mixing system characteristics, namely: a at
4.4
bottomed cylindrical vessel, a dished bottomed cylindrical vessel
and a spherical vessel.
4.3 For any vessel shape, the ow impinging from the Scaba is radial,
0.20 0.25 0.30 0.35 0.40 0.45 0.50 c/D resulting in the formation of two recirculation loops. The compar-
ison between the results obtained for the three vessel congurations
Fig. 11. Power consumption for Rey = 4. has shown that the spherical shape provides uniform ows in the
 H. Ameur/Engineering Science and Technology, an International Journal 19 (2016) 189196 195

Fig. 12. Flow patterns for a spherical tank, c/D = 0.35.

whole vessel volume and it requires less energy consumption, rec- Nomenclature
ommending this type of vessel for a more frequent usage in the eld
of the processing industries equipment.
Effects of the impeller clearance from the vessel bottom are also bt blade thickness, m
studied for the case of a spherical vessel. It was found that the middle c impeller off-bottomed clearance, m
of the vessel is the most appropriate position. d blade diameter, m
The agitation rate has also an important effect. With a curved- ds shaft diameter, m
bladed impeller and low Reynolds number, the cavern size is dt disc thickness, m
very limited. However, a sucient impeller rotational can promote h blade height, m
the axial circulation of uid, enlarge the cavern size and enhance n ow behavior index, dimensionless
mixing. D tank diameter, m

*
Vz Vz* Rey = 40
Rey = 600
0.05 Rey = 40 Rey = 1200
Rey = 600 0.02 Rey = 2500
0.04 Rey = 1200
Rey = 2500
0.03 0.01

0.02
0.00
0.01
-0.01
0.00

-0.01 -0.02
-0.02
-0.03
-0.03
-0.04
-0.04
0.0 0.2 0.4 0.6 0.8 1.0 R* 0.0 0.2 0.4 0.6 0.8 1.0 R*
(a) (b)
Fig. 13. Radial proles of the axial velocity component for a spherical vessel, c/D = 0.35, (a) below the impeller (Z* = 0.26), (b) above the impeller (Z* = 0.44).
196 H. Ameur/Engineering Science and Technology, an International Journal 19 (2016) 189196

H liquid level, m [9] C.D. Taca, M. Paunescu, Suspension of solid particles in spherical stirred vessels,
Chem. Eng. Sci. 55 (2000) 29892993.
K consistency index, Pa sn
[10] H. Yapici, G. Basturk, CFD modeling of conjugate heat transfer and
Ks MetznerOttos constant, dimensionless homogeneously mixing two different uids in a stirred and heated
N impeller rotational speed, 1/s hemispherical vessel, Comp. Chem. Eng. 28 (2004) 22332244.
P power, W [11] M. Ammar, Z. Driss, W. Chtourou, M.S. Abid, Effect of the tank design on the
ow pattern generated with a pitched blade turbine, Int. J. Mech. App 2 (2012)
Np power number, dimensionless 1219.
Qv viscous dissipation function, 1/s2 [12] D. Anne-Archard, M. Marouche, H.C. Boisson, Hydrodynamics and Metzner-Otto
R, Z radial and axial coordinates, respectively, m correlation in stirred vessels for yield stress uids, Chem. Eng. J. 125 (2006)
1524.
R* dimensionless radial coordinate, R* = 2R/D [13] L. Pakzad, F. Ein-Mozaffari, P. Chan, Using computational uid dynamics
Z* dimensionless axial coordinate, Z* = Z/D modeling to study the mixing of pseudoplastic uids with a Scaba 6SRGT
Rey Reynolds number for a yield stress uid, dimensionless impeller, Chem. Eng. Proc. 47 (2008) 22182227.
[14] E. Galindo, A.W. Nienow, Mixing of highly viscous simulated xanthan
Vz, V, Vr axial, tangential and radial velocities, respectively, m/s fermentation broths with the Lightnin A-315 impeller, Biotech. Progress 8 (1992)
V* dimensionless velocity, V* = V/ND 233239.
[15] E. Galindo, A.W. Nienow, Performance of the Scaba 6SRGT agitator in
the mixing of simulated xanthin gum broths, Chem. Eng. Tech. 16 (1993)
Greek letters 102108.
 avg average shear rate, 1/s [16] A. Amanullah, S.A. Hjorth, A.W. Nienow, Cavern sizes generated in highly shear
shear stress, Pa thinning viscous uids by Scaba 3SHP1 impellers, Food Bioprod. Proc. 75 (1997)
232238.
y suspension yield stress, Pa
[17] L. Serrano-Carreon, E. Galindo, Studies on cavern development in mixing a yield
uid density, kg/m3 stress uid in a pilot-scale proto-fermentor, Rc. Prog. Gnie Proc. 11 (1997)
apparent viscosity, Pa s 161168.
0 yielding viscosity, Pa s [18] L. Pakzad, F. Ein-Mozaffari, P. Chan, Using electrical resistance tomography and
computational uid dynamics modeling to study the formation of cavern in
angular coordinate, degree the mixing of pseudoplastic uids possessing yield stress, Chem. Eng. Sci. 63
angular velocity, rad/s (2008) 25082522.
[19] L. Pakzad, F. Ein-Mozaffari, S.R. Upreti, A. Lohi, Characterization of the mixing
References of non-Newtonian uids with a Scaba 6SRGT impeller through ERT and CFD,
Can. J. Chem. Eng. 91 (2013) 90100.
[20] H. Ameur, M. Bouzit, M. Helmaoui, Numerical study of uid ow and power
[1] S. Nagata, Mixing. Principles and Applications, Wiley, New York, 1975. consumption in a stirred vessel with a Scaba 6SRGT impeller, Chem. Proc. Eng.
[2] A.W. Nienow, D. Miles, The effect of impeller/tank congurations on uid 32 (2011) 351366.
particle mass transfer, Chem. Eng. J. 15 (1978) 1324. [21] L. Pakzad, F. Ein-Mozaffari, S.R. Upreti, A. Lohi, Agitation of Herschel-Bulkley
[3] C. Xanthopoulos, M. Stamatoudis, Effect of impeller and vessel size on impeller uids with the Scaba-anchor coaxial mixers, Chem. Eng. Res. Des. 91 (2013)
power number in closed vessels for Reynolds number above 75000, Chem. Ing. 761777.
Tech. 60 (1988) 560562. [22] H. Ameur, M. Bouzit, Numerical investigation of ow induced by a disc turbine
[4] N. Papastefanos, M. Stamatoudis, Effect of vessel and impeller geometry on in unbaed stirred tank, Acta Sci. Tech 35 (2013) 469476.
impeller power number in closed vessels for Reynolds number between 40 and [23] S. Saeed, F. Ein-Mozaffari, S.R. Upreti, Using computational uid dynamics
65000, Chem. Eng. Comm. 80 (1989) 6979. modeling and ultrasonic doppler velocimetry to study pulp suspension mixing,
[5] J. Medek, I. Fort, Liquid circulation in mechanically agitated closed closed vessel, Indus. Eng. Chem. Res. 46 (2007) 21722179.
Proc. 8th Europ. Conf. Mixing, Cambridge, UK 136 (1994) 473480. [24] A.B. Metzner, R.E. Otto, Agitation of non-Newtonian uids, AIChE J. 3 (1957)
[6] P.M. Armenante, C. Luo, C.C. Chou, I. Fort, J. Medek, Velocity proles in a closed, 310.
unbaed vessel: comparison between experimental LDV data and numerical [25] H. Ameur, M. Bouzit, M. Helmaoui, Hydrodynamic study involving a
CFD predictions, Chem. Eng. Sci. 52 (1997) 34833492. Maxblend impeller with yield stress uids, J. Mech. Sci. Tech. 26 (2012)
[7] M. Ciofalo, A. Brucato, F. Grisa, N. Torraca, Turbulent ow in closed and 15231530.
free-surface unbaed tanks stirred by radial impellers, Chem. Eng. Sci. 51 (1996) [26] S. Woziwodzki, J. Sowinski, Power requirements for yield stress uids in a vessel
35573573. with forward reverse rotating impeller, Proc. Eng. 42 (2012) 14371444.
[8] C.D. Taca, M. Paunescu, Power input in closed stirred vessels, Chem. Eng. Sci. [27] T.P. Elson, The growth of caverns formed around rotation impellers during the
56 (2001) 44454450. mixing of a yield stress uid, Chem. Eng. Comm. 96 (1990) 303391.