7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
Augmentation of Airlift Pump Performance in Step Geometry
A. Karimi, P. Hanafizadeh, S. Ghanbarzadeh and M. H. Saidi
Multiphase Flow Research Group, Centre of Excellence in Energy Conversion, School of Mechanical Engineering, Sharif
University of Technology, Tehran, Iran, P O. Box: 111559567
Email: saman@tsharif.edu
Keywords: step upriser pipe, airlift pump, two phase flow pattern, flow regime
Abstract
Airlift pumps are devices which are widely used in industrial applications. Parameters such as diameter of the pipe, tapering
angle of the upriser pipe, submergence ratio (which is defined by the ratio of immersed length to the total length of the
upriser), the gas flow rate, bubble diameter, and inlet gas pressure affect the performance of these pumps. In the previous
work (Hanafizadeh et al. 2009) the effect of tapering angle and bubble diameter were considered. In this research, the
performance of airlift pump with a vertical riser length of 914 mm and initial diameters of 6 and 8 mm and various height for
steps namely: 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8 and 0.9 m in submergence ratio of 0.6 is investigated numerically. This paper
reports the improvement in performance of step airlift pump in comparison with ordinary type. Also in the present study effect
of height of steps and secondary pipe diameter were considered. The result shows that in constant gas flow rate exist special
height and secondary diameter for step which the performance of the pump can be optimized. The numerical results were
compared with the experimental data of White (2001) showing a reasonable agreement. The results have indicated that step
airlift pump has higher efficiency than the pump with constant pipe diameter.
Introduction
Two phase lifting pumps are simple devices for rising
liquids and mixtures of liquids and solid particles. Its
operation is based on the use of buoyancy force to pump the
liquid and solid particles through a partially submerged
vertical upriser pipe in the fluid which should be pumped. A
gas phase is injected at the bottom of the pipe to produce an
upward flow in the riser pipe (Figure 1).
Separator tank Do~H  
L2
Slug
Compressor
f Air injector
'
Li
i 1
Main tank
DP
Figure 1: Schematic of Airlift pump. 1. Overhead
collecting tank, 2. Upriser tube, 3. Compressor, 4. Air
injector, 5. Water reservoir.
The gas liquid two phase mixture is lighter than the liquid
and rises to the free surface. Injected gas decreases the
hydrostatic weight of the flow column.
This type of pumping has low efficiency, but great
advantages in utilization of them over mechanical pumps
are lower initial and maintenance costs, easy installation,
small space requirements, simplistic design and construction,
ease of flow rate regulation. These advantages accompanied
by the absence of moving mechanical parts cause that two
phase lifting pump can be used for pumping of different
fluids which are corrosive, abrasive or slurries, explosive,
toxic, sandy or salty (Giot 1982). These pumps are used for,
pumping of viscous liquids like hydrocarbons in oil field
industry Kato et al. 1975, underground well drilling (Giot
1982), under sea mining (Gibson 1961 and Mero 1968,
bioreactors (Chisti 1992 and Trystam and Pigache 1992)
Besides, they are used to prevent icing on some high
altitude (Abed 1977).
Fundamentally the flow in air lift pump can be modeled by
two phase flow in vertical tube. Therefore, different two
phase flow patterns such as bubbly, slug, chum and annular
flow are used to describe the flow in these pumps.
White and Beardmore (1962) and Zukoski (1966) realized
that the effects of surface tension on the dynamics of
vertical slug flow are very important when the tube diameter
is decreased below 20 mm. Lately, Kouremenos and Staicos
(1985) preformed their investigations on small diameter air
lift pumps down to 12 mm diameters and low length upriser
in the range of 1 to 3 m, with submergence ratios between
0.55 and 0.7. More recently, a wide range of investigating
the application of two phase pumps in moving liquids at
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
nuclear fuel reprocessing plants has been realized, such as
de Cachard and Delhaye (1996). These studies have been
mostly concerned in the accuracy of the air lift pump rather
than the efficiency. Previous laboratory experiments (Geest
et al. 2001 and Guet et al. 2002) with water and air showed
that the increasing the size of the injected bubbles can
improve the efficiency. Most of the work that has been done
on air lift pumps involves the usage of an upriser pipe with
constant diameter (Darbandi et al. 2007, Hanafizadeh and
Saidi 2008 and Kassab et al. 2009). Therefore not much
information is available on effect of step geometry for
upriser pipes on airlift pump performance.
In this research, the performance of two phase pump is
investigated numerically for different submergence ratios
and different diameter. In order to verify, the numerical
results are compared with the experimental data (White
2001). The comparison shows the numerical results are in
good agreement with the experimental data. Also the
performance of two phase pump is shown for different
tapered angles of the upriser pipe and different bubble
diameter (102 to 1 nmm).
Nomenclature
constant
diameter
force
g gravity
H depth of water
K interphase momentum exchange coefficient
k turbulent kinetic energy
L
LPM
upriser tube length
liter per minute
secondary phase
Governing equations
The numerical simulations presented here are based on the
twofluid, EulerianEulerian model (Simonin 1990, Mudde
et al. 1997, Lehr et al. 2002, Dhotre and Joshi 2007, Dhotre
et al. 2007). The Eulerian modeling system is based on
ensembleaveraged mass and momentum transport
equations for each phase. In the present work, the liquid
phase behaves as the continuum and the gaseous phase
(bubbles) as the dispersed phase.
The continuity equation for phase q is
n
S(Pqaq) + V. (PqaqVq) = (mlpq
p=l
rhqp) + Sq (1)
where, o(q, Vq and pq are the void fraction, velocity and
density of phase q, respectively. rhpq characterizes the
mass transfer from the pth to qth phase, and rhqp
characterizes the mass transfer from phase q to phase p. It
is supposed that mass transfer between two phases is zero
(rilpq = rilqp = 0).
The momentum conservation for multiphase flow is
described by the volume averaged momentum equation as
follows
 (qpqVq) + V. (aqpq;qVq)
= oqVp + V. Tq + qpqg
+ aqpq(Fq + Flift,q + Fvm,q)
n
+ (Kpq(p q)+ ilpqVpq)
p=l
Ti mass flow rate
pressure
source term
time
velocity
here g is the acceleration due to gravity, Tq is the phase
stressstrain tensor, Fq is an external body force, lift,q is
a lift force, Fvm,q is a virtual mass force, p is the pressure
shared by all phases, and Kpq is the interphase momentum
exchange coefficient. Vpq is the interphase velocity (If
rilpq > 0 Vpq = Vp ; if rilqp > 0 Vqp = Vq ). It is
assumed that interphase conversion does not occur in this
case, so interphase velocity is set to be zero.
void fraction
Tq = a"q[q(VVq + VVq) + q (Aq
bulk viscosity
shear viscosity
density
stressstrain tensor
turbulent dissipation
variable
3 Iq) V. q
here itq and Aq are the shear and bulk viscosities of phase
q, respectively.
Lift forces mainly act on a particle due to velocity gradients
in the primary phase flow field. The lift force is computed
from the following equation.
Flift = 0.5pqp(Vq Vp) X (V X Vq)
Suhbcribe
initial
primary phase
The lift force will be more important for larger particles.
Therefore, the inclusion of lift forces is not appropriate for
closely packed particles or very small particles.
For multiphase flows, virtual mass occurs when a secondary
v
Greek
symbols
phase accelerates relative to the primary phase.
F =oq(dV dpV,
Fvm = 0.5aqpq ( dt dt
where, the term denotes the phase material time
dt
derivative of the form
dq (4) 0(4)
dt = at + (Yq V) (6)
The virtual mass effect is significant when the secondary
phase density is much smaller than the primary phase
density e.g., for a transient bubble column, however the
virtual mass effect in this study is negligible.
Turbulence model
Turbulence is taken into consideration for the continuous
phase. The dispersed gas phase is modelled as laminar flow,
but the influence of the dispersed phase on the turbulence of
the continuous phase is taken into account with Sato's
additional term (Sato and Sekoguchi 1975). The wellknown
singlephase turbulence models are usually used to model
turbulence of the liquid phase in EulerianEulerian
multiphase simulations. In the present case the standard
k model is used (Launder and Spalding 1972). The
governing equations for the turbulent kinetic energy k and
turbulent dissipation e are:
a
 (Xqpqkq) + V. (aqpqUqkq)
=V. tq Vk)
Lik
+ (OqGk,q OqpqEq)
N
+ Kpq(Cpqk Cqpkq)
p=1 (7)
N
Kpq (Up Uq). t'p Vat
Yp1p a p
p=l
N
+ Kpq (Up q). It'q VXq
p=1
a
(QqPqEq) + V. (aqpqiqEq)
= v.(a Utqq/)
Cq
+ C[lQLqGkq C21ZtqPqEq
kq
+ C3, Kq(Cpqkp Cqpkq) (8)
\p=l
N
 Kpq (Up
p=l
N
+ Kpq (Up
p=l
Up Up
Uq). LVq
Ct (T 4 /
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
the terms Cpq and Cqp canbe approximated as
Cpq = 2, Cqp = 2 1 + pq
pq (' +p [lpJ
where, r1pq is available in (Csanady 1963).
Geometry and grid mesh specification
In this research, the domain of solution is selected in a
manner that boundaries are conformed on cylindrical
coordinate system, also, the mesh lines divide the geometry
span to control volume forming the hexahedral sector.
Figure 2 shows the air lift pump geometry with applied
meshing. The airlift pump geometry consists of a vertical
pipe having a length of L (upriser pipe) within a liquid
reservoir. Air is injecting in lower part of the vertical pipe.
In this model, the reservoir has a cylindrical shape, the riser
tube with 914 mm long, 6 mm initial diameter, different
tapered angles. Air injector pipe diameter is 3.175 mm.
Figure 2: Air lift pump with step geometry showing the
applied mesh
To make certain grid independency of the results from nodes
number, four meshes (5000, 7000, 9000, and 11000 nodes)
have been tested and pressure at axial of upriser have been
compared. Figure 3 shows that the pressure at 9000 nodes is
least nodes number which the results are finally independent
of them. Table 1 shows the characteristics of grid mesh that
have been applied for numerical modeling.
7000
0 0.2 0.4 0.6 0.8 1
Position (m)
Figure 3: Validation of mesh independency
Table 1 Specification of grid mesh
Cells 8340
Faces 17246
Nodes 8905
minimum volume 5.09
(m3)5.09e8
(m3)
maximum volume 715e
(m3)7.15e5
(m3)
Numerical solution
In this study computational method of CFD packaged which
is used for discretizing the governing equation, is based on
the control volume frame work which is proposed by
Patankar (1980). A collocated grid is used to all variables
are stored at the centre of control volume. The governing
equations are solved using the SIMPLE algorithm. The
details of discretization are found in Fogt and Peric (1994).
The time dependent equations are solved to increase the
stability of the numerical solution. For any iteration the
system of two continuity and two momentum equations with
the transport equations of turbulent energy and dissipation
are solved. Turbulent variables and velocity near the wall of
the control volume are estimated from the wall laws.
Velocities of both phases are calculated from the respective
momentum equations. After determination of velocity,
pressure computed from the liquid continuity equation.
Volume fraction is calculated from the continuity equation
of the gas phase.
Results and discussion
The present study deals with various step height and inlet
and outlet diameter tube size of an airlift pump. The effect
of step geometry is compared with ordinary type of airlift
pumps. Also, various step positions and diameters are
considered in operation of airlift pumps.
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
For verification, which is depicted in Figure 4, the obtained
results at submergence ratio of 0.4 and pipe diameter of 10
mm are compared with experimental data (White 2001).
The comparison showed the predictions were in reasonable
agreement with experimental data and can predicted the
overall behaviour of it. The difference justification with
experimental results can be attributed to the bubble size
estimation, isotherm and incompressible flow assumption,
relinquishment of surface adhesion influences, and
approximations employed on numerical modelling.
0.008
0.007
0.006
0.005
0.004
0.003
0.002
0.001
(H/L=0.4, D =10mm)
S Experimental [17]
 Present work S
0 0.25 0.5 0.75
1 1.25 1.5
Air Flow Rate (LPM)
Figure 4 Comparison between experimental results
and present work
Comparison between step and ordinary airlift
pump
Figure 5 shows the comparison between ordinary and step
airlift pump. In this case the step is mounted at the height of
0.2 m with inlet and outlet diameters of 6 and 12 mm,
respectively. It is clearly shown that the SALP has more
outlet liquid mass flow rate than the OALP in the same inlet
air flow rate. So it is obvious in Fig. 5b that SALP will
have more efficient than OALP
Comparisons between present work and
experimental results
0005
S0 004
 1
S0 003
0 002
0001
0
ooo
0 0.2 0.4
Air Flow Rate (LPM)
50
40
S30
20
10
0.6 0.8
a) comparison of exit liquid mass flow rate in terms of
injected air mass flow rate for airlift with and without
0 0.2 0.4
Air Flow Rate (LPM)
b) comparison of efficiency in terms of injected air mass
flow rate for airlift with and without step
step
Fig. 5 comparison of airlift pump with and without step; for submergence ratio of 0.6, height of step 0.2 m, inlet diameter
 OALP
  SALP
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
6mm and outlet diameter 12mm
In Figure 6, the flow pattern map of Taitel et al. (1980) is
marked with the Solid line and the flow pattern map of
HewittRoberts (1969) is marked with the Round Dot line.
It is demonstrated that the SALP is located in the slug
region but the OALP incline to the chum region. It is
mentioned before by other researchers (Kassab et al. 2007)
the best flow regime for operating of airlift systems is slug
flow. So as it is expected the efficiency of SALP is higher
than OALP.
SBubb
10 Bubbly Slug Bubble Wispy
.. Annular
S1
slug
/Annular
.0.01
ChC
 SALP
  OALP
0.001
0.01 0.1 1 10 100
Air Superficial Velocity (m/s)
Figure 6 Comparison two flow pattern maps for SALP
and OALP
COMPARISON DIFFERENT STEP AIRLIFT PUMP
Figure 7 shows the variation of water flow rate with air
flow rate for SALP with various heights of steps. The
submergence ratio is 0.6 and the inlet and outlet diameter
0014
S0012
4
S001
S0008
m 0006
S0004
 0002
0
0 0.5 1
Air Flow Rate (LPM)
of step are 6 and 12 mm, respectively. Figure 8 shows the
variation of outlet water flow rate with injected air flow
rate for various steps heights in submergence ratio of 0.6
and inlet and outlet step diameters of 8 and 12 mm,
respectively. The variation of water mass flow rate with
step height is illustrated in Fig. 9 in constant submergence
ratio of 0.6. Two air flow rate namely 0.5 and 0.8 LPM
correspond to Fig. 9a, b, respectively. It is realized that for
a particular air flow rate always exist a specific height for
step which maximize the outlet water flow rate.
Comparison between Figs. 9a and b shows that the
optimum height of step is decreased when the injected air
flow rate increases. It can be described by regards to the
fact that increase in air flow rate advance the transition of
slug flow regime to chur flow. Increase in pipe area
section can postpone the flow pattern transition and hence
the slug flow can be abided in the upriser pipe. It can be
concluded that setting the step in the optimum height can
improve the efficiency of the pump.
Figure 10 shows the variation of water mass flow rate with
step height in submergence ratio of 0.6. In this case the
inlet and outlet diameters are 8 and 12 mm which are
illustrated in Figs. 10a and b, respectively. The
comparison between Figs. 9 and 10 shows that increase in
inlet diameter increases the optimum height of step.
Increase in pipe diameter in constant air flow rate
postpones the transition of slug flow regime to chum. It
means that this transition happens in higher height of
upriser so the optimum height of step must be increase
when the initial diameter of the pipe is increased.
0 009
0 008
_,0 007
* 0006
0 005
0 004
o 003
Z 0002
0 001
1.5 2
0 0.5 1 1.5
Air Flow Rate (LPM)
Figure 7 Variation of water mass flow rate with air mass flow rates for different SALP in submergence ratio of 0.6, inlet diameter of 6mm
and outlet diameter of 12mm
............. step=0.4(m)
  step=0.2(m)
  step0.5(m)
step0.3(m)
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
0 02
to 0016
4
0 012..
0 008 step
........*** .. step
c 0 004   ste
S step
0
0  i  i  i
0 0.5 1 1.5
Air Flow Rate (LPM)
002
S 0016
a
S0012
o
0008
 0 004
0
s  step 0.5(m)
S step=0.4(m)
............. step=0.3(m)
0 0.5 1 1.5 2
Air Flow Rate (LPM)
Figure 8 Variation of water mass flow rate with air mass flow rates for different SALP in submergence ratio of 0.6, inlet diameter of 8mm
and outlet diameter of 12mm
00045
"0 0044
" 00043
S00042
0 0041
a
0 004
00039
0 0.2 0.4 0.6 0.8 1
Step Height (m)
00062
S0006
4
00058
00056
S00054
a
S00052
0005
0 0.2 0.4 0.6 0.8 1
Step Height (m)
a b
Figure 9 Variation of water mass flow rate with step height in submergence ratio of 0.6, inlet diameter of 6mm and outlet diameter of
12mm; a: for injected air of 0.5 LPM, b: for injected air of 0.8 LPM
0006
0 0 0056
4
0 0052
0 0048
7a
.Z 0 0044
0004
0 0.2 0.4 0.6 0.8 1
Step High (m)
0 0092
n 0 0088
S00084
0 008
a
. 0 0076
00072
0 0.2 0.4 0.6 0.8 1
Step High (m)
a b
Figure 10 Variation of water mass flow rate with step height in submergence ratio of 0.6, inlet diameter of 8mm and outlet diameter of
12mm; a: for injected air of 0.5 LPM, b: for injected air of 0.8 LPM
Effect of Step Height on the Efficiency of the
Pump
Moreover, in this study the influence of step height on the
airlift pump performance has been investigated. The
Efficiency of air lift pump is defined by Niklin (1963) as
below :
Qppg(L H)
 QqPaLf()
QqPaLn( )
(Ta
where, Qp is liquid mass flow rate, Qq is air flow rate, p
is the liquid density, Po and Pa are injection and
atmospheric pressures, respectively.
Figures 11 and 12 present the results of the efficiency as a
function of air flow rate at various step heights in constant
submergence ratio. Figure 11 is depicted for inlet diameter
of 6 mm in submergence ratio of 0.6 but Fig. 12 is depicted
for 8 mm inlet diameter.
Moreover, these figures clearly reveal that the maximum
50
40
S30
20
10
0
1.5 2
1
Air Flow Rate (LPM)
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
efficiency occur at the specific step height. Besides, as it
shows from Figure 11 and Figure 12, the maximum
efficiency occurs in higher step height when the diameter
of the pipe is increased. So it seems that increase in pipe
diameter can postpone the transition of slug flow pattern to
chum flow and increase the efficiency of the pump.
60
50
S40
S30
S20
10
0
0.5 1 1.5
Air Flow Rate (LPM)
Figure 11 Variation of efficiency with air flow rates for different SALP in submergence ratio of 0.6, inlet diameter of 6 mm
and outlet diameter of 12 mm
0.5 1 1.5
Air Flow Rate (LPM)
1.5 2
1
Air Flow Rate (LPM)
Figure 12 Variation of efficiency with air flow rates for different SALP in submergence ratio of 0.6, inlet diameter of 8 mm
and outlet diameter of 12 mm
EFFECT OF OUTER DIAMETER IN SALP
The liquid flow rate versus air flow rate is illustrated in Fig.
13a. Also, the efficiency of SALP is depicted in Fig. 13b.
these figures clearly show that there is an optimum outer
diameter for SALP which the efficiency of the pump is
maximized in it. It is obvious that the very large outer
diameter can destroy the slugs and therefore reduce the
efficiency of the pump. Figure 13b reveals that very large
outer diameter not only improve the efficiency of airlift
pump but also deteriorate the operation of the pump.
The liquid mass flow rate versus pipe outer diameter is
shown in Fig. 14. It seems that there is a specific outer
diameter which can optimize the operation of the pump.
This cross section can stable the slug flow regime and
therefore the efficiency of the pump improve in this
diameter.
  step=0.2(m)
..  step=0.3(m)
._ ............. step=0.4(m)
S .. . .. step=0.5(m)
. ... "! P 05 
step0.2(m) ...
 step=0.6(m)
  step0.7(m)
 step 0.8(m)
........... step=0.9(m)
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
0 0.1 0.2 0.3 0.4 0.5 0.6
Air flow rate(kg/s)
70
60
50
S40
30
0 0.002 0.004 0.006
Air flow rate (LPM)
a b
Figure 13 comparison of effect of outer diameter in SALP with 6 mm inlet diameter; a: liquid flow rate versus air flow rate, b:
efficiency versus air flow rate
0.006
0.0059
0.0058
S0.0057
0.0056
6 11 16 21 26
Outer diameter (mm)
Figure 14 variation of liquid mass flow rate with outer diameter of the pipe
00089
00088
00087
a 0 0086
0 0085
0 0084
0 0083
0 0082
. 00081
0 008
00079
6mm
1 1.5 2 2.5 3
D2/D1
1 1.2 1.4 1.6
D2/D1
1.8 2 2.2
a b
Figure 15 variation of liquid mass flow rate with diameter ratio, a: inlet diameter of 6mm; b: inlet diameter of 8 mm
The variation of water mass flow rate versus of diameter
ratio, which is defined by ratio of outlet to inlet diameter, is
demonstrated in Figs. 15a and b for inlet diameters of 6
and 8 mm, respectively. These figures show that the
maximum liquid mass flow rates, moreover the diameter
ratio, depends on the other parameters such as inlet
diameter. Generally, the dimensionless parameter of
diameter ratio cannot be introduced as the only effective
parameter in performance of SALP.
In Fig. 17 the operation of two SALPs with outer diameters
of 8 and 10 mm is demonstrated on the flow pattern map of
Hewitt and Roberts (1969) and Teital et al. (1980). As it is
clear the SALP with outer diameter of 10 is located in the
slug flow region and consequently has a higher efficiency.
0.007
0.006
S0.005
0.004
0.003
g 0.002
0.001
0 *
D out= 8 mm
  D out= 10 mm
  D out= 21.94 mm
0.0064
0.0063
0.0062
" 0.0061
 0.006
0.0059
1 0.0058
 0.0057
0.0056
0.0055
10
5
01
0 01
0001
0.01
0.1 1
Jg
10 100
Figure 16 Comparison of flow pattern maps for two SALP
with outer diameters of 8 and 10 mm
Conclusion
In this paper, the performance of airlift pump with step
geometry is investigated numerically. It was seen that in
SALP there is an optimum height for step position which
the efficiency of the pump can be maximized in that
position. The outlet diameter of upriser pipe in step
geometry is the other important parameter which can affect
the performance of the pump. The best position of step is
where the transition of flow regime from slug flow to churn
flow occurs in the pipe. Maximum amount of liquid is lifted
if the pump operates in the slug regimes so the best
efficiency for this type of pumps is always occurred in this
flow regime. Increase in cross sectional area along the
upriser pipe due to step geometry can increases the
efficiency of the pump. The results showed that the outer
diameter of step also has an optimum value which can
retain the flow pattern in slug regime.
Acknowledgements
This research was funded by Iran Supplying Petrochemical
Industries Parts, Equipment and Chemical Design
Corporation (SPEC) as a joint research project with Sharif
University of Technology (project no. KPR8628077). The
contribution is greatly appreciated.
References
Abed, K., A. Theoretical study on the performance of airlift
pumps. The Inst. of Engineers, MC, Vol. 77: 202 (1977).
Chisti, Y Assure bioreactor sterility, Chem. Eng. Prog. Vol.
88(9): 80 (1992).
Csanady, G. T. Turbulent diffusion of heavy particles in the
atmosphere. J. Atmos. Science Vol. 20: 201208 (1963).
Darbandi M., Saidi M. H., Hanafizadeh P, A numerical
approach to simulate twophase flow in airlift pumps, Int.
Conf. on Comp. Meth. ICMM2007, Japan (2007).
De Cachard F., Delhaye J. M., A slugchur model for
smalldiameter airlift pumps. Int. J. Multiphase Flow, 22
(4): 627649 (1996).
Dhotre, M., T. and Joshi, J., B. Design of a gas distributor:
threedimensional CFD simulation of a coupled system
consisting of a gas chamber and a bubble column. Chem.
Eng. Journal Vol. 125: 149163 (2007).
Dhotre, M., T., Niceno, B., N., Smith, B., L. Large eddy
simulation of a bubble column using dynamic subgrid
scale model. Chem. Eng. J., doi:10.1016/j.cej.2007.04.016.
Fogt, H., and Peric, M. Numerical calculation of gas_ liquid
flow using a two fluid finite volume method. Num.
Method in Multiphase Flows, Vol. 185: 7380 (1994).
7th International Conference on Multiphase Flow
ICMF 2010, Tampa, FL USA, May 30June 4, 2010
Geest S. van., Ellepola J. H., Oliemans, R.VA. Comparison
of different air injection methods to improve gaslift
performance. Proc. the 10t Con. Multiphase, France (2001).
Gibson, A., H. Hydraulics and its applications. 5th Edition,
Constable, London (1961).
Giot, M. 1982. Three phase flow, Handbook of multiphase
systems, edited by G. Hetsroni, Hemisphere, McGraw Hill.
Guet, S., Ooms, G. and Oliemans, R. V A. Influence of
bubble size on the transition from lowre bubbly flow to
slug flow in a vertical pipe. Exp. Th. Fluid Sc., Vol. 26 (6):
635641 (2002).
Hanafizadeh, P. and Saidi, M. H., Integral solution for
gravity driven gasliquid two phase flow, Proc. of 11th Fluid
Dyn. Conf. FD2008 1211, Iran (2008).
Hanafizadeh P, Saidi M. H., Zamiri A., Karimi A., Effect of
Bubble Size and Angle of Tapered Upriser Pipe on the
Effectiveness of a Two Phase Lifting Pump,
FEDSM200978214, Proc. FEDSM2009, ASME (2009).
Hewitt, G.F., Roberts, D.N., Studies of twophase flow
patterns by simultaneous Xray and flash photography.
UKAEA Report AEREM2159 (1969).
Kassab, S. Z., Kandil, H. A., Warda, H. A., and Ahmed, W.
H., Airlift pumps characteristics under twophase flow
conditions. Int. J. Heat and Fluid Flow, 30: 8898 (2009).
Kato, H., Miyazawa, T, Timaya, S., and Iwasaki, T; A
study of an airlift pump for solid particles. Bull. JSME 18:
286294 (1975).
Kouremenos, D. A. and Staicos, J. Performance of a small
airlift pump. Int. J. Heat Fluid Flow 6: 217222 (1985).
Launder, B., E., Spalding, D., B. Mathematical models of
turbulence, GB: Academic Press, London (1972).
Lehr, F., Millies, M., and Mewes, D. Bubblesize
distributions and flow fields in bubble columns. A.I.Ch.E. J.
Vol. 48 (11): 24262443 (2002).
Mero, J. L. Seafloor minerals, A Chemical Engineering
Challenge. Chem. Eng. J. Vol. 43 (2): 73 (1968).
Mudde, R. F, Lee, D. J., Reese, J., and Fan, L. S. Role of
coherent structures on Reynolds stresses in a 2d bubble
column. A.I.Ch.E. Journal, Vol. 43: 913926 (1997).
Nicklin, D.J. The airlift pump theory and optimization,
Trans. Inst. Chem. Eng., J. 41: 2939 (1963).
Patankar, S. V Numerical heat transfer and fluid flow,
McGrawHill, New York (1980).
Sato Y, Sekoguchi, K., Liquid velocity distribution in
twophase bubbly flow, Int. J. Multiphase Flow Vol. 2:
7995 (1975).
Simonin O., Eulerian formulation for particle dispersion in
turbulent twophase flows, Proceedings of the 5th Workshop
on TwoPhase Flow Predictions, Germany (1990).
Taitel, Y, Barea, D., and Dukler, A.E. Modeling flow
pattern transitions for steady upward gasliquid flow in
vertical tubes. AIChE J. 26: 345354 (1980).
Trystam, G. and Pigache, S. Modeling and simulation of a
large scale air lift fermenter. Proc. Eur. Symp. Comp.Aid.
Proc. Eng.2: 51715176 (1992).
White, E. T., and Beardmore, R. H., The velocity of rise of
single cylindrical air bubbles through liquids contained in
vertical tubes. Chem. Eng. Sci. Vol. 17: 351361 (1962).
White, S., J. Bubble pump design and performance. MSc
Thesis Georgia Institute of Technology (2001).
Zukoski, E. E. Influence of viscosity, surface tension, and
inclination angle on motion of long bubbles in closed tubes.
J. Fluid Mech. 20: 821832 (1966).
