Desalination 324 (2013) 3747

Contents lists available at ScienceDirect

Desalination
journal homepage: www.elsevier.com/locate/desal

CFD modeling and simulation of concentration polarization in

microltration of oilwater emulsions; Application of an Eulerian
multiphase model
Mehrdad Zare, Farzin Zokaee Ashtiani , Amir Fouladitajar
Department of Chemical Engineering, Amirkabir University of Technology, No. 424, Hafez Ave., Tehran, Iran

H I G H L I G H T S

CP phenomenon was modeled using CFD technique in microltration of oily water.

An Eulerian multiphase model was used to solve oil concentration for the rst time.
The model showed high capability to predict CP behavior in different conditions.
The model needs modications to predict permeate ux accurately.

a r t i c l e i n f o a b s t r a c t

Article history: In this study, a 2D CFD modeling and simulation of concentration polarization (CP) phenomenon in microltration
Received 15 October 2012 of oilwater emulsion in a narrow rectangular membrane module was presented. As the novelty of this research,
Received in revised form 27 May 2013 an Eulerian based multiphase method was investigated for the rst time to capture the concentration of emulsied
Accepted 28 May 2013
oil droplets in different extents of membrane module. It was shown that this new model is highly compatible
Available online 25 June 2013
with concentration polarization theory and is highly capable for determining CP prole and mass boundary
Keywords:
layer behavior in different operating conditions, as well. In this research it was proved that under developed
Microltration feed cross-ow shows better performance in terms of membrane surface cleaning. Also it was shown that a
Oilwater Emulsion small portion of oil droplets accumulates at the roof of membrane module feed channel due to gravity. The simu-
Concentration polarization lation results proved that for under developed feed cross-ows, lower feed Reynolds number induces a lower CP
CFD modeling prole peak despite increasing the oil droplet concentration on membrane surface and at the feed channel roof.
Multiphase model 2013 Elsevier B.V. All rights reserved.

1. Introduction
Oily waste-water is one of the most important pollutants emitted
into the environment. These oil contaminated waters are generated
daily worldwide in many diverse industries such as petroleum ren-
ery, petrochemical, metal and steel, textile and machine industries
[1,2]. For the oily wastewater treatment, the conventional technolo-
gies are not efcient enough to treat the stable oilwater emulsions,
with oil droplets smaller than 20 m, especially when the oil concen-
tration is pretty low and the droplets are nely dispersed. To address
this problem membrane processes have been increasingly investigat-
ed for treating oilwater emulsions. The rst attempts to treat oily
water by membrane separation processes were made in the beginning
of 70's decade [3].
In recent decades pressure driven membrane separation processes
have been playing an important role in industrial unit operations due
Corresponding author. Tel.: +98 21 6454 3124; fax: +98 2166405847.
E-mail address: zokaee@aut.ac.ir (F. Zokaee Ashtiani).
to their moderate operational conditions and cost efciency [4]. How-
ever these processes are suffered by fouling phenomena causing per-
meate ux decline which is undesirable and in some cases may be
irreversible [3]. A solute concentration which is higher in regions close
to the membrane surface in comparison to the bulk free stream, leads
to the concentration polarization, CP, phenomenon which arises from
the nature of microltration and ultraltration processes [5]. This phe-
nomenon affects the permeate ux negatively by introducing a positive
concentration prole toward the membrane surface causing back-
diffusion of solute rejectants from membrane surface to the bulk [5].
Therefore, numerous studies have been accomplished by re-
searchers experimentally, mathematically and numerically to achieve
insight knowledge about membrane fouling, especially concentration
polarization.
Many of these studies employed CFD simulation techniques due to
the accuracy of the results which are yielded through the numerical
solution of conservation of mass, momentum, and energy [6].
A large number of these studies have been performed over the
last two decades and are devoted to investigate the effects of different

0011-9164/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.desal.2013.05.022
38 M. Zare et al. / Desalination 324 (2013) 3747
types of bafes, spacers and other types of turbulent promoters
and their orientation and spatial conguration on ow hydrodynam-
ics and concentration boundary layer disruption. These effects have
been considered in terms of parameters such as wall sheer stress,
mass transfer coefcients, friction factor, power number, etc. [617].
Also some researches are about the CFD calculations for new mem-
brane module designs or special turbulence promoting modules
such as gas sparged ones to demonstrate their effects on ow hydro-
dynamics, mass boundary layer, and permeation ux [1827]. In this
category, some authors neglected the effects of low permeate ux
by considering the membrane as an impermeable wall and fully fo-
cused on studying the effects of the promoted turbulence on ow
hydrodynamics [7,12,14,15,17,20,21,2327]. Consequently the effect
of membrane transport mechanism on the ow hydrodynamics was
neglected.
Liu et al. [14] conducted a CFD simulation to evaluate the effects
of bafes on turbulent ow in a tubular membrane module. They
showed that using bafes signicantly increases the wall shear stress
and improves the separation performance. Furthermore, it was proved
that central bafes are more effective than wall bafes in producing
wall shear stress and result in a higher permeation ux.
Karode and Kumar [17] visualized the ow in a spacer lled rectan-
gular channel. Despite the previous reports, it was revealed that the
bulk of uid ows along the spacer laments and does not change direc-
tion at each spacer node. Also, it was proved that spacers with unequal
lament diameters exert lower pressure drop to the uid ow.
On the other hand, some researchers made the assumption of a
xed solute concentration, higher than that of the feed bulk, as a
boundary condition on the membrane wall (impermeable wall) to
consider the mass boundary layer formed due to concentration polar-
ization phenomenon. Then, they evaluated the effects of the pro-
moted turbulence on the mass boundary layer disruption [6,9,11,16].
These assumptions might lead to a more detailed but unrealistic eval-
uation of concentration polarization phenomenon.
Santos et al. [6] performed a CFD simulation to investigate the ow
pattern for 12 different ow-aligned using cyclic boundary condition.
They analyzed the results in terms of shear rate, mass transfer and a
modied form of friction factor. It was proved that the modied fric-
tion factor can be used to determine the best spacer conguration to
achieve higher mass transfer.
Shakaib et al. [9] carried out a CFD simulation to investigate the
ow and mass transfer in spacer lled channel of a spiral wound
membrane module. They proved that axial and transverse spacer
laments exert higher shear stress at the top membrane wall but
the mass transfer rate remained almost the same.
In contrast, some studies have been conducted to model the
empty membrane channel with laminar ow condition [13,2838].
In these studies the attempt was to consider the impact of membrane
transport mechanism on ow hydrodynamics. The calculations of con-
centration polarization were conducted through the solution of conti-
nuity, NavierStokes and solute continuity equations by incorporating
Film Theory which is based on true rejection and permeation ux.
In these studies the impact of membrane transport mechanism on
ow hydrodynamics was calculated by considering constant rejection
and constant or variable permeation ux as a boundary condition for
membrane in CFD calculations [30,32,33,38].
Ahmad et al. [32] performed a CFD modeling of concentration
polarization in a narrow rectangular membrane channel by assuming
xed permeate ux for the membrane. They investigated the effects
of different operating conditions on CP and showed that there was a
good agreement between simulation results and literature data.
But the assumption of xed/variable permeation-rejection might
lead to introducing errors to the calculation of concentration polari-
zation. Hence some authors tried to improve the accuracy of these
calculations by introducing the Darcy's permeation law to the CFD
codes to predict the membrane permeate ux locally, based on local
pressure distribution prole on the membrane surface in the compu-
tational domain [29,31,3437].
Pak et al. [29] developed a 2D CFD code to simulate concentration
polarization for a porous membrane using Darcy's permeation law as
a boundary condition for the membrane. They showed that with this
assumption the CFD model may have a better agreement with exper-
iment results.
Following this, some researchers employed Darcy's permeation
law, solute continuity equation and lm theory in conjunction with
turbulence promoters to calculate a more accurate solution of turbu-
lent dominated membrane modules [8,10,13]. Using Darcy's perme-
ation law, Wardeh and Morvan [13] carried out a CFD simulation
of a spacer-obstructed rectangular membrane module. They proved
that zigzagging type spacers are more efcient in concentration
polarization disruption.
Since the rst attempts for treating oily wastewater by membrane
in early 1970's, several studies have been done to achieve insight
knowledge about different membrane separation of oilwater emul-
sion systems and concentration polarization. Most of these studies
are limited to experiments and mathematical modeling where valu-
able results have been achieved; however numerical modeling of
such systems has received less attention [3,3947]. Besides, the avail-
ability of CFD modeling and simulations of concentration polari-
zation in membrane separation of oilwater emulsion was found to
be limited.
Addressing this problem, in the current study, a 2D nite volume
model was developed for CFD modeling of laminar ow in cross ow
microltration of oilwater emulsion in an empty rectangular channel.
The membrane actual permeation ux and its effect on ow hydrody-
namics were considered based on the Darcy's permeation law through
the calculation of local pressure distribution in the computational
domain. The oil was intended to be dispersed in water in the form of
spherical micelles. As the novelty of this study, multi-phase ow equa-
tions based on Eulerian multiphase model were investigated in CFD
calculations to yield more accurate and realistic results.
2. Theory
2.1. Module geometry and problem description
In the current study, the focus was on CFD modeling and simulation
of concentration polarization in cross-ow microltration of oilwater
emulsion in an empty rectangular membrane module. The feed channel
was a rectangular slit with dimensions of 0.3 5 10 cm as height,
width and length of the rectangle, respectively. The porous membrane
was as the lower 5 10 cm wall of the slit. The permeate channel
was at the same shape and dimensions, as well. The feed was a mixture
comprising water as continuous phase, containing emulsied oil as
dispersed phase in the form of micelles of diameters of about 1 m.
The oil density and viscosity were intended to be 850 kg/m3 and
0.00332 kg/ms, respectively. The membrane module was fed by differ-
ent oil concentrations of 200, 1000, 5000 and 20,000 ppm. Also differ-
ent operating conditions in terms of trans-membrane pressure (TMP)
and cross-ow velocity were applied in the problem. The process was
conducted in different TMPs of 0.5, 1, and 2 bar, and cross-ow veloci-
ties corresponding to the Reynolds numbers of 500, 1300 and 2200.
2.2. Geometry reconstruction and grid generation
According to geometry dimensions noted in previous section, the
ratio of height to width of the feed channel of membrane module
was far less than unity (H/W 0.06). Hence the effects of side walls
on ow hydrodynamics were neglected, and in order to reduce com-
putational costs, the model was intended to be two-dimensional for
CFD calculations. Therefore, a 0.3 10 cm rectangle (longitudinal
cross-section of feed channel) was considered as the computational
 M. Zare et al. / Desalination 324 (2013) 3747 39

Fig. 1. A 3D (a) and 2D (b) schematic view of the computational domain.

domain. Also, despite the fact that the study of ow hydrodynamics in condition due to the absence of turbulence promoters. The problem was
the permeate side was out of our scope in this research, a rectangle solved through the computational uid dynamic technique based
with dimension of 0.1 10 cm as permeate channel was considered on the multi-phase ow theory. Hence the governing equations of the
under the feed channel. This consideration was made only due to problem were divided into two main categories of continuous and
membrane transport phenomena calculations. The computational dispersed phases. All governing equations have been presented in
domain has been schematically shown in Fig. 1. Cartesian coordinate, as well.
The 2D computational domain was meshed with quadrilateral ele-
ments in order to minimize numerical diffusion errors in calculations. 2.4. Continuous phase
Finally after mesh dependency tests (Fig. 2), the short and long edges
of the feed side of the domain were divided into 300 and 200 intervals, As described above, the mixture contained water as continuous
respectively. To ensure that the grid is ne enough near the membrane phase and oil as dispersed phase in low concentrations. Consequently,
wall to capture the concentration polarization, a bell-shaped prole was the role of water as continuous phase was dominant in ow hydrody-
applied to the short edges of the feed channel, forcing height of the rst namics. For continuous phase, the ow hydrodynamics can be
rows of the cells adjacent to the membrane wall to be 0.176 m. Conse- achieved through the numerical solution of Continuity (Eq. (1)) and
quently the grid generated for feed side channel contained 60,000 cells NavierStokes (Eqs. (2), (3)) in their discretized form as governing
with minimum cell surface area of 8.8 1011 m2. equations.Continuity equation:
For the permeate side 0.1 10 cm channel, a grid of 20 200
cells was generated introducing 4000 cells to the computational do-
u v
main. This grid was ne enough just to capture pressure distribution 0 1
X Y
prole in permeate side of the membrane. A brief description about
problem parameters and operating conditions is presented in Table 1.
Cell height and vertical position of cell center points are shown in NavierStokes in X-direction:
Fig. 3 for each cell.



u u P u u
u v 2
2.3. Governing equations X Y X X X Y Y

In this study, the mixture was assumed to be incompressible and

isothermal. The calculations were performed for steady state in laminar
Table 1
Parameters considered for simulations in this study.

Property Value Unit

Feed channel dimensions 0.3 5 10 cm

Permeate channel dimensions 0.1 5 10 cm
Conc. (mg/L)

Computational domain 0.3 10 cm

Material properties
Water density 998.2 kg/m3
Fit 60,000 Water viscosity 0.001003 kg/ms
Fit 50,000 Oil density 830 kg/m3
Fit 40,000 Oil viscosity 0.003320 kg/ms
Fit 20,000
Fit 10,000 Operating conditiona
TMP 2, 1, 0.5 bar
0 0.02 0.04 0.06 0.08 0.1 0.12 Feed channel Reynolds no. 500, 1300, 2200
X-Position (m) Feed oil concentration 200, 1000, 5000, 20,000 ppm
a
Based on full factorial method, the simulations were conducted for 36 different
Fig. 2. Mesh dependency test (concentration vs. position for different grids). combinations of operating conditions.
40 M. Zare et al. / Desalination 324 (2013) 3747

where n is the number of phases, v m is the mass-averaged velocity,
m is the mixture density, and k is the volume fraction of phase k.
For a mixture containing continuous phase (q) and dispersed
phases (p), volume fraction () equation is:

  
 

p p : p p v m : p p v dr;p 7
t

X
n

v dr;p v pq k v qk 8
k1

k k
ck 9
m
 
p p m
v pq v p v q a 10
f drag p

2
p dp
p 11
18 q

 0:687
f drag 1 0:15Re Re1000 12
0:0183Re Re1000

 
v
a g v m : v m m 13
t

where v dr;p is drift velocity, ck is mass fraction for the phase k, v pq is
relative velocity, p is particle relaxation time, dp is mean particle

diameter, fdrag is drag coefcient and a is the acceleration.
According to the parameters introduced above, the momentum
equation can be expressed as:

    h  T
i
m v m : m v m v m p : m v m v m 14
Fig. 3. (a) Height of each cell. (b) Vertical position of each cell center point. The cells
t !
X n

are numbered in Y-direction from membrane surface toward the channel top wall. m g F : k k vdr;k v dr;k
k1

NavierStokes in Y-direction:
X
n


 m k k 15
v v P v v k1
u v 3
X Y Y X X Y Y
Where F is a body force, m is the viscosity of the mixture, and again

where is density, is viscosity and, u and are velocity components v dr;k is the drift velocity for dispersed phase k. The last term in
in X and Y directions, respectively. right-hand side of Eq. (14) keeps account for the impact of dispersed
phase on continuous phase hydrodynamics. Once more, it should
2.4.1. Dispersed phase be noticed that all time-dependent terms in equations above were
The calculations for emulsied oil micelles, in order to determine neglected in CFD calculations due to steady state nature of the
their accumulation and concentration in different regions of mem- problem.
brane module, were based on Eulerian multi-phase ow theory. From the governing equations, it can be concluded that the impact
Hence the formulation of governing equations introduced in this sec- of any change in momentum on dispersed phase volume fraction can
tion is based on phasic volume fraction, and concepts such as relative be taken into account through the solution of Eqs. (7) and (14). Sub-
velocity and drag force are incorporated in them. sequently it is expected that membrane transport mechanism
Continuity equation for a mixture is as follows:For a mixture (permeation-rejection) effects on oil micelles deposition on mem-
continuity equation is: brane surface to be captured via this formulation. Since oil micelles
are large spheres (dp 1m) the diffusion of rejected oil micelles
  from membrane surface to the bulk was assumed to be zero.
m : m v m 0 4
t
2.5. Boundary conditions
X
n

k k v k

The computational domain is represented in Eqs. (16)(20) as

k1
vm 5 boundary conditions. It was assumed that the feed was premixed and
m
homogenous, and both the continuous and dispersed phases enter the
X
n domain at the same velocity. Therefore, an identical plug velocity prole
m k k 6 was considered for both phases at inlet of the feed channel. Besides,
k1 a uniform volume fraction of dispersed phase corresponding to the
 M. Zare et al. / Desalination 324 (2013) 3747 41

homogenous mixture of each oil concentration was set as boundary

condition for the dispersed phase at the inlet.
For feed channel inlet:

u u0 ; v 0 for each phase
at X 0 and 0Y0:003 :
ak ak0 for dispersed phase
16

At the feed channel outlet, the pressure was set as boundary

condition for both phases.
For feed channel outlet:

at X 0:1 and 0Y0:003P P feed outlet : 17

All non-membrane walls were considered as impermeable walls.

Also no-slip boundary condition was used for all impermeable walls
and membrane interface.
For membrane, the Darcy's permeation law was used as boundary
condition to count for the transport mechanism imposed by perme-
ation phenomena. The permeate ux of each phase passing through
the porous membrane was calculated based on this law by consider-
ing the viscosity of each phase in calculations.
For impermeable walls:

at Y 0:003; 0X0:1 and Y 0:001; 0X0:1

u0 18
and X 0; 0:001Y0 :
v0

For porous membrane:

(
u0
at Y 0; 0X0:1 p 19
v Jv
lR Fig. 4. The solution method based on segregated algorithm.

where R is proportionality constant representing membrane resis-

tance which was obtained experimentally, and l is the porous mem-
brane thickness (120 m).
For permeate channel outlet, the atmospheric discharge pressure
was used as boundary condition
at X 0:1; 0:001Y0 P P disch arge : 20
2.6. Solution strategy
CFD simulations were performed through the numerical solution
of governing equations in their discretized form, based on boundary
conditions, using a computer code developed for this research.
In this study the worldwide known PISO (Pressure-Implicit with
Splitting of Operators) algorithm [48] was employed, which is a
member of the widely known SIMPLE (Semi-Implicit Method for
Pressure-Linked Equations) algorithm family. Also in order to achieve
higher level of accuracy, PRESTO (PREssure Staggering Option) and
QUICK (Quadratic Upstream Interpolation for Convective Kinetics)
discretization schemes were used for pressure and momentum equa-
tions, respectively.
Prior to multi-phase ow calculations, a converged solution for
single-phase ow (pure water) was obtained. Then the multi-phase
calculations were initiated based on this single-phase solution. At
this stage, for volume fraction calculations, QUICK discretization
scheme was used.
The numerical calculations were performed in segregated manner
to reduce computational costs. Fig. 4 shows the algorithm followed
by the solver to achieve convergence. In this method, in each loop
of calculations, the X and Y-velocity components were calculated
separately by solving momentum equation (Eq. (14)) in each direc-
tion. Then the resulting values were corrected through the solution
of continuity (Eq. (4)) as correction equation. After achieving a
corrected momentum eld in the domain, the volume fraction for
the dispersed phase was calculated in the domain using Eq. (7).
Then the obtained solution was examined based on convergence
criterions. The convergence criterions for the momentum and volume
fraction residuals were about 1016 and 1015 respectively. Also the
net error of inlet, outlet and membrane permeate uxes was about
1020 kg/s.
In this study a total number of 1.8 107 iterations were made for
a grid comprising of 64,000 quadrilateral elements with two Intel
Xenon E5620 2.4GHz CPU and 20 GB of RAM.
2.7. Experimental
A 0.45 m pore size at sheet Polyvinylidene uoride (PVDF)
membrane (Millipore, USA), with 70% porosity and hydrophilicity
behavior was used for all of the experiments. A at sheet module
was specially designed and fabricated for this study. Cross-ow mem-
brane operation was used in the microltration experiment using a
laboratory scale setup which has been explained in detail in previous
work [49].
The oil in water emulsions was prepared by adding gasoil in denite
concentrations, containing Tween 80 as the surfactant, to distillated
water. The emulsion was generated by mixing for 30 min at a speed
of 12,000 rpm. This oil water emulsion was considered stable and uni-
form, as measured by a laser light-scattering technique.
3. Results and discussions
As mentioned previously, a single-phase CFD simulation for pure
water owing through membrane channel was conducted prior to
42 M. Zare et al. / Desalination 324 (2013) 3747

Table 2
CFD simulation results for uxes compared with experimental results.

TMP (Barg) Permeate ux (L/m2h) Error %

CFD simulation Experiment

0.5 91.715 108 15.08

1 175.515 201.6 12.94
2 333.330 360 7.41

multi-phase ow simulation. Hence this section is divided into two

main parts of single-phase and multi-phase ow.

3.1. Single-phase ow (pure water)

3.1.1. Permeate ux
Simulation results for permeation ux are shown in Table 2. A
comparison with experimental results shows that there is a good
agreement between CFD simulation and experiments and the maxi-
mum and minimum errors are %15.08 and %7.41, respectively. By
taking a shallow look to the simulation results, it can be found that
the permeate ux increases by increasing the TMP. This can be simply
explained by direct proportionality of permeate ux with TMP in
Darcy's permeation law (Eq. (19)).

3.1.2. Velocity prole

According to the inlet boundary condition, the feed enters the
membrane feed channel with a plug velocity prole. Then the ow
tends to develop a parabolic velocity prole by proceeding along
the feed channel. In Figs. 57, the Y-component of ow velocity across
the feed channel was examined as a factor of ow development for
different distances from feed channel inlet (1 to 9 cm). It can be
seen that, at the beginning of the feed channel the Y-velocity prole
has relatively sharp peaks near both membrane and channel roof,
meaning that the ow is under developed. Then the ow got more
developed gradually and the peaks were damped by moving toward
the feed outlet due to the ow development.
It can be obviously found that ow velocity development and
Y-velocity prole are affected by feed Reynolds number.
Fig. 8 shows the effect of different feed Reynolds numbers on
Y-velocity prole at each TMP for typical distance of 8 cm from
channel inlet (0.8 of length of feed channel). It can be seen that at
Reynolds number of 2200 the ow is still under developed even at
the end of the feed channel. By moving from Reynolds number of
2200 to 500 the ow tends to get more developed, however it is not
fully developed. This can be concluded from presence of a peak in
the curve of Reynolds number of 500 around 0.5 mm above the mem-
brane surface for all TMPs.
At membrane surface (Y-position equal to zero) Y-velocity had a
negative value. This was due to the effects of membrane permeation
phenomena on ow eld prole. Therefore, this negative velocity
can be considered as membrane permeation velocity. Fig. 9 shows
the effect of different TMPs on Y-velocity prole at each feed Reynolds
number for typical distance of 8 cm from feed channel inlet (0.8 of
length of feed channel). By increasing TMP, the Y-velocity magnitude
increased at membrane wall and the ow prole was pushed down
toward the membrane. This was due to the direct proportionality
of permeate ux with TMP, discussed in Section 3.1.1.
3.2. Multi-phase ow (oilwater emulsion)
3.2.1. Permeate ux
Fig. 10 shows the effect of TMP on membrane permeation ux.
Accordingly, the higher the TMP is, the higher the permeate ux is.
By comparing the slopes of two sections of the graph, it can be
found that the rate of permeate ux increase is higher at lower
TMPs. This can be explained so that at higher TMPs, the permeation
Fig. 5. Variation of Y-velocity prole across the membrane feed channel for TMP of
2 bar and feed Re. of 2200. (a) From 1 cm to 5 cm. (b) From 5 cm to 9 cm.
ux and consequently, the rejection rate are higher, leading to a
higher rejectant concentration on membrane wall, and this high con-
centration exerts a greater resistance to permeation phenomena.
The results showed similar behavior for different combination of
feed Reynolds numbers and feed concentrations. Thus their graphs
were omitted due to similarity.
Fig. 11 shows the effect of feed oil micelles concentration on per-
meation ux. Obviously, it can be seen that an increase in concen-
tration leads to a permeate ux decline. This can be explained so
that by increasing concentration, the amount of rejected oil deposited
on feed side of the membrane grows and exerts a higher resistance
to permeation phenomena. Also a comparison between Fig. 11(a)
and (b) implies that the inuence of feed concentration on permeate
ux is more intense in higher TMPs. This is due to the fact that a
higher TMP causes a higher permeation ux and consequently a
higher rejection, increasing the concentration of rejectant more on
membrane surface and therefore exerting a higher resistance to per-
meation phenomena.
The results showed similar behavior for different combination of
TMPs and feed Reynolds numbers. Thus their graphs were omitted
due to similarity.
The effect of feed Reynolds number on membrane permeation ux
is shown in Fig. 12. Accordingly, an increase in feed Reynolds number
tends to increase permeate ux. Applying higher feed Reynolds num-
bers exerts higher shear force to the membrane surface and conse-
quently sweeps the deposited rejected oil on feed side membrane
wall more effectively. By comparing the slope of each section of
Fig. 12(a) with corresponding section of Fig. 12(b), it can be found
 M. Zare et al. / Desalination 324 (2013) 3747
Fig. 6. Variation of Y-velocity prole across the membrane feed channel for TMP of 2 bar
and feed Re. of 1300. (a) From 1 cm to 5 cm. (b) from 5 cm to 9 cm.
that the effect of feed Reynolds number on permeate ux is stronger
for lower TMPs. This is for the reason that at lower TMPs the rejectant
concentration on membrane wall is less; hence the sweeping feature
of cross ow velocity is more effective.
The results showed similar behavior for different combinations of
TMPs and feed concentrations. Thus their graphs were omitted due to
similarity.
3.2.2. Concentration polarization prole
Fig. 13 shows oil concentration prole for a height across the feed
channel, ranging from half of the channel height to membrane surface
(from 1.5 to 0 mm), at typical distance of 8 cm from feed channel
inlet (0.8 of feed channel length). From half of the feed channel to
about 0.5 mm above the membrane surface, the concentration has a
at prole equal to feed bulk concentration. Then at about 0.5 mm,
to 0.3 mm above the membrane, the concentration increases, making
a maximum point. The growth of concentration from that of the feed
bulk to the maximum point by moving toward the membrane surface
can be considered as concentration polarization prole. Afterward,
the concentration decreases by moving toward the membrane
wall (near-membrane-wall region). This can be explained according
to ow development condition and Y-velocity prole discussed in
Section 3.1.2. Accordingly the under developed ow prole intro-
duces a stream from membrane surface toward the bulk feed due to
the presence of positive Y-velocity prole, sweeping the rejected oil
droplets from membrane surface and reducing the concentration at
membrane wall.
43
Fig. 7. Variation of Y-velocity prole across the membrane feed channel for TMP of
2 bar and feed Re. of 500. (a) From 1 cm to 5 cm. (b) From 5 cm to 9 cm.
A comparison of concentration prole for different TMPs is shown
in Fig. 13. It can be seen that the concentration prole peak remains
constant by increasing the TMP from 0.5 to 2 bar. This can be
explained by the assumption of zero diffusion coefcient due to
large micelle diameters; meaning that even a slight amount of TMP
leads to formation of concentration polarization prole at its extreme
magnitude, due to large micelle diameter. For near-membrane-wall
region with positive slope of concentration prole, the concentration
increases by an increase in TMP. This is due to the increase of perme-
ation and rejection rates at higher TMPs, inducing a higher rejectant
concentration at feed side membrane wall.
The results showed similar behavior for different combination of
feed Reynolds numbers and feed concentrations. Thus their graphs
were omitted due to similarity.
The effect of feed bulk concentration on CP prole was modeled
successfully and is shown in Fig. 14. It is evident that an increase
in concentration leads to a rise in peak of CP prole. This happens
as a consequence of rejectant concentration growth due to rejection
increment. Then the rise in CP prole exerts a higher resistance to
permeation phenomena, lowering permeation ux and consequently
rejection rate, causing the reduction of rejectant concentration in
near-membrane-wall region. These results are highly compatible
with the effects of feed concentration on permeate ux, discussed in
Section 3.2.1.
The results showed similar behavior for different combination of
TMPs and feed Reynolds numbers. Thus their graphs were omitted
due to similarity.
44 M. Zare et al. / Desalination 324 (2013) 3747
Fig. 8. Variation of Y-velocity prole by feed Re. for different TMPs at 0.8 of feed channel
length. (a) TMP = 2 bar. (b) TMP = 1 bar. (c) TMP = 0.5 bar.
The effect of different feed Reynolds numbers on CP prole was
modeled and the results are shown in Fig. 15(a) and (b), for lower
half (membrane side) and upper half of the feed channel, respectively.
It's shown in Fig. 15(a) that applying lower feed Reynolds numbers
leads to development of a lower peak in CP prole but increases
the rejectant concentration, as a resistance for permeation, in near-
membrane-wall region. The reverse effect of feed Reynolds number
on the peak of CP prole can be explained according to Fig. 15(b). In
Fig 15(b), it can be seen that a large portion of oil droplets (in compar-
ison to the peak of CP prole) is deposited near the feed channel roof
due to the effect of gravity and density difference. Also this deposition
Fig. 9. Variation of Y-velocity prole by TMP for different feed Re. at 0.8 of feed channel
length. (a) Re. 2200. (b) Re. 1300. (c) Re. 500.
is more intense for lower feed Reynolds numbers. This is due to the
fact that, in ows with lower Reynolds numbers, the oil micelles have
further residence time, and consequently have more time to proceed
their migration from feed bulk toward the feed channel roof. It can be
concluded that the deposition of oil micelles near the roof of feed chan-
nel reduces the concentration in membrane side, and the more the oil
micelles deposited near feed channel roof, the lower the peak of CP
prole.
Applying a lower feed Reynolds number reduces the sweeping
effect of ow, and thus leads to a rise in rejectant concentration in
the near-membrane wall region as a resistance to the permeation
 M. Zare et al. / Desalination 324 (2013) 3747 45

Fig. 10. The effect of TMP on permeate ux for feed Re. of 2200 and concentration of
20,000 ppm.

phenomenon. Therefore, the permeate ux reduced by applying low

feed Reynolds numbers. This explanation is highly compatible with
the results of feed Reynolds number on permeate ux, discussed in
Section 3.2.1.
The results showed similar behavior for different combination of
TMPs and feed concentrations. Thus their graphs were omitted due
to similarity.

Fig. 12. The effect of feed Re. on permeate ux. (a) 2 bar and 20,000 ppm. (b) 0.5 bar
and 20,000 ppm.

4. Conclusions

In this research, an Eulerian based multi-phase model was investi-

gated for the rst time to perform CFD simulation of concentration
polarization phenomenon in micro-ltration of oilwater emulsions,
in a at rectangular membrane module with laminar ow. A compar-
ison between experimental data and simulation results of permeation

Fig. 13. The effect of TMP on oil concentration prole (or concentration polarization
Fig. 11. The effect of feed concentration on permeate ux. (a) 2 bar and Re. 2200. effect). For Re. of 2200 and oil concentration of 20,000 ppm. Values presented by the
(b) 0.5 bar and Re. 2200. plots should be added up with corresponding bulk feed concentration.
46 M. Zare et al. / Desalination 324 (2013) 3747
Fig. 14. The effect of feed bulk concentration on CP prole for TMP of 2 bar and Re. of
2200. For better comparison, (a) is divided to (b) and (c). Values presented by the plots
should be added up with corresponding bulk feed concentrations.
ux showed that despite great capability of this model for pure water
permeate ux prediction, the model cannot predict the permeate ux
very accurately in the case of multiphase oilwater emulsion. This
can occur as the result of assumptions which were made to simplify
the model such as neglecting the size distribution of oil droplets,
neglecting the inter particle interactions of oil droplets and their coa-
lescence based on population balance concept, neglecting the interac-
tion between oil droplets and the membrane surface, etc.
Nevertheless by comparing simulation results with concentration
polarization theory, it was shown that this innovative method is
Fig. 15. The effect for feed Reynolds number on CP prole for TMP of 1 bar and concen-
tration of 20,000 ppm. (a) For lower half of feed channel. (b) For upper half of feed
channel. Values presented by the plots should be added up with corresponding bulk
feed concentration.
highly capable of determining concentration polarization prole and
to predict the behavior of mass boundary layer in different operating
conditions, though it is expected that the variations in the behavior of
CP prole to be under estimated due to the simplifying assumptions
mentioned above. According to this, it was shown that by increasing
trans-membrane pressure and feed initial concentration, the CP will
be more intensive, as expected. Furthermore, it was proved that for
underdeveloped ows, presence of a positive Y-velocity component
near the membrane surface sweeps the rejected oil droplets accumu-
lating on membrane surface, and consequently reduces the concen-
tration in this region. This may be benecial in terms of membrane
fouling. The less the ow is developed, the lower the positive
Y-velocity near the membrane surface is, and consequently the higher
the membrane surface concentration is, due to less sweeping effect.
The simulation results also showed that a small portion of oil
droplets accumulates at the roof of the feed channel due to gravity ef-
fect and the higher the feed Reynolds number the smaller the portion
is, due to cross-ow sweeping effect. Applying a higher feed Reynolds
number, despite reducing the oil droplets concentration at feed chan-
nel roof and in near-membrane region (the region with positive slope
in CP prole, approximately in between membrane surface and about
0.3 mm above it), induces a higher peak in CP prole which acts as a
resistance to separation process. According to the simulation results
the sweeping effects of cross-ow velocity are more affecting on per-
meation performance than the higher CP peak induced at higher feed
Reynolds numbers.
 M. Zare et al. / Desalination 324 (2013) 3747
Nomenclature
CP Concentration Polarization
TMP Trans-Membrane Pressure (bar)
u X-Component of Velocity (m/s)
v Y-Component of velocity (m/s)
Density (kg/m3)
P Pressure (Pa)
Viscosity (kg/ms)
vm Mass-Averaged Velocity (m/s)
m Mixture Density (kg/m3)
ak Volume Fraction of Phase, K
v dr;p Drift Velocity (m/s)
ck Mass Fraction of Phase, K
v pq Relative Velocity (m/s)
p Particle Relaxation Time (s)
fdrag Drag Coefcient
Acceleration (m/s2)
F Body Force (N)
m Viscosity of the Mixture (kg/ms)
dp Mean Particle Diameter (m)
R Membrane Resistance (Pas/kgm)
l Membrane Thickness (m)
References
[1] H. Ohya, J.J. Kim, A. Chinen, M. Aihara, S.I. Semenova, Y. Negishi, O. Mori, M.
Yasuda, Effects of pore size on separation mechanisms of microltration of oily
water, using porous glass tubular membrane, J. Membr. Sci. 145 (1998) 114.
[2] A.B. Kotuniewicz, R.W. Field, Process factors during removal of oil-in-water
emulsions with cross-ow microltration, Desalination 105 (1996) 7989.
[3] A. Salahi, M. Abbasi, T. Mohammadi, Permeate ux decline during UF of oily
wastewater: experimental and modeling, Desalination 251 (2010) 153160.
[4] R. Ghidossi, D. Veyret, P. Moulin, Computational uid dynamics applied to mem-
branes: state of the art and opportunities, Chem. Eng. Process. 45 (2006) 437454.
[5] K. Madireddi, R.B. Babcock, B. Levine, J.H. Kim, M.K. Stenstrom, An unsteady-state
model to predict concentration polarization in commercial spiral wound mem-
branes, J. Membr. Sci. 157 (1999) 1334.
[6] J.L.C. Santos, V. Geraldes, S. Velizarov, J.G. Crespo, Investigation of ow patterns
and mass transfer in membrane module channels lled with ow-aligned spacers
using computational uid dynamics (CFD), J. Membr. Sci. 305 (2007) 103117.
[7] S. Ahmed, M.T. Seraji, J. Jahedi, M.A. Hashib, Application of CFD for simulation of a
bafed tubular membrane, Chem. Eng. Res. Des. 90 (2012) 600608.
[8] Y.-L. Li, K.-L. Tung, Y.-S. Chen, K.-J. Hwang, CFD analysis of the initial stages of
particle deposition in spiral-wound membrane modules, Desalination 287
(2012) 200208.
[9] M. Shakaib, S.M.F. Hasani, M. Mahmood, CFD modeling for ow and mass transfer
in spacer-obstructed membrane feed channels, J. Membr. Sci. 326 (2009)
270284.
[10] M. Rahimi, S.S. Madaeni, K. Abbasi, CFD modeling of permeate ux in cross-ow
microltration membrane, J. Membr. Sci. 255 (2005) 2331.
[11] M.A. Monfared, N. Kasiri, A. Salahi, T. Mohammadi, CFD simulation of bafes ar-
rangement for gelatin-water ultraltration in rectangular channel, Desalination
284 (2012) 288296.
[12] S. Ahmed, M. Taif Seraji, J. Jahedi, M.A. Hashib, CFD simulation of turbulence pro-
moters in a tubular membrane channel, Desalination 276 (2011) 191198.
[13] S. Wardeh, H.P. Morvan, CFD simulations of ow and concentration polarization
in spacer-lled channels for application to water desalination, Chem. Eng. Res.
Des. 86 (2008) 11071116.
[14] Y. Liu, G. He, X. Liu, G. Xiao, B. Li, CFD simulations of turbulent ow in bafe-lled
membrane tubes, Sep. Purif. Technol. 67 (2009) 1420.
[15] J. Schwinge, D.E. Wiley, D.F. Fletcher, A CFD study of unsteady ow in narrow
spacer-lled channels for spiral-wound membrane modules, Desalination 146
(2002) 195201.
[16] A. Saeed, R. Vuthaluru, Y. Yang, H.B. Vuthaluru, Effect of feed spacer arrangement
on ow dynamics through spacer lled membranes, Desalination 285 (2012)
163169.
[17] S.K. Karode, A. Kumar, Flow visualization through spacer lled channels by com-
putational uid dynamics I.: pressure drop and shear rate calculations for at
sheet geometry, J. Membr. Sci. 193 (2001) 6984.
[18] A. Drews, H. Prieske, E.-L. Meyer, G. Senger, M. Kraume, Advantageous and detri-
mental effects of air sparging in membrane ltration: bubble movement, exerted
shear and particle classication, Desalination 250 (2010) 10831086.
[19] T. Taha, Z.F. Cui, CFD modelling of gas-sparged ultraltration in tubular mem-
branes, J. Membr. Sci. 210 (2002) 1327.
47
[20] T. Taha, Z.F. Cui, CFD modelling of slug ow in vertical tubes, Chem. Eng. Sci. 61
(2006) 676687.
[21] S. Buetehorn, D. Volmering, K. Vossenkaul, T. Wintgens, M. Wessling, T. Melin,
CFD simulation of single- and multi-phase ows through submerged membrane
units with irregular ber arrangement, J. Membr. Sci. 384 (2011) 184197.
[22] L. Martinelli, C. Guigui, A. Line, Characterisation of hydrodynamics induced by
air injection related to membrane fouling behaviour, Desalination 250 (2010)
587591.
[23] S. Smith, T. Taha, Z. Cui, Enhancing hollow bre ultraltration using slug-ow a
hydrodynamic study, Desalination 146 (2002) 6974.
[24] N. Ratkovich, C.C.V. Chan, P.R. Berube, I. Nopens, Experimental study and CFD
modelling of a two-phase slug ow for an airlift tubular membrane, Chem. Eng.
Sci. 64 (2009) 35763584.
[25] T. Taha, W.L. Cheong, R.W. Field, Z.F. Cui, Gas-sparged ultraltration using hori-
zontal and inclined tubular membranesa CFD study, J. Membr. Sci. 279 (2006)
487494.
[26] A. Sarkar, S. Moulik, D. Sarkar, A. Roy, C. Bhattacharjee, Performance characteriza-
tion and CFD analysis of a novel shear enhanced membrane module in ultraltra-
tion of Bovine Serum Albumin (BSA), Desalination 292 (2012) 5363.
[27] M. Bubolz, M. Wille, G. Langer, U. Werner, The use of dean vortices for crossow
microltration: basic principles and further investigation, Sep. Purif. Technol. 26
(2002) 8189.
[28] M. Rahimi, S.S. Madaeni, M. Abolhasani, A.A. Alsaira, CFD and experimental
studies of fouling of a microltration membrane, Chem. Eng. Process. 48 (2009)
14051413.
[29] A. Pak, T. Mohammadi, S.M. Hosseinalipour, V. Allahdini, CFD modeling of porous
membranes, Desalination 222 (2008) 482488.
[30] D.E. Wiley, D.F. Fletcher, Computational uid dynamics modelling of ow and
permeation for pressure-driven membrane processes, Desalination 145 (2002)
183186.
[31] M.F. Gruber, C.J. Johnson, C.Y. Tang, M.H. Jensen, L. Yde, C. Hlix-Nielsen, Compu-
tational uid dynamics simulations of ow and concentration polarization in
forward osmosis membrane systems, J. Membr. Sci. 379 (2011) 488495.
[32] A.L. Ahmad, K.K. Lau, M.Z.A. Bakar, S.R.A. Shukor, Integrated CFD simulation of
concentration polarization in narrow membrane channel, Comput. Chem. Eng.
29 (2005) 20872095.
[33] S.K. Karode, Laminar ow in channels with porous walls, revisited, J. Membr. Sci.
191 (2001) 237241.
[34] E. Lyster, Y. Cohen, Numerical study of concentration polarization in a rectangular
reverse osmosis membrane channel: permeate ux variation and hydrodynamic
end effects, J. Membr. Sci. 303 (2007) 140153.
[35] R. Ghidossi, J.V. Daurelle, D. Veyret, P. Moulin, Simplied CFD approach of a
hollow ber ultraltration system, Chem. Eng. J. 123 (2006) 117125.
[36] P. Schausberger, N. Norazman, H. Li, V. Chen, A. Friedl, Simulation of protein ultra-
ltration using CFD: comparison of concentration polarisation and fouling effects
with ltration and protein adsorption experiments, J. Membr. Sci. 337 (2009)
18.
[37] K.-J. Lee, R.-M. Wu, Simulation of resistance of cross-ow microltration and
force analysis on membrane surface, Desalination 233 (2008) 239246.
[38] D.E. Wiley, D.F. Fletcher, Techniques for computational uid dynamics modelling
of ow in membrane channels, J. Membr. Sci. 211 (2003) 127137.
[39] M. Ebrahimi, K.S. Ashaghi, L. Engel, D. Willershausen, P. Mund, P. Bolduan, P.
Czermak, Characterization and application of different ceramic membranes for
the oil-eld produced water treatment, Desalination 245 (2009) 533540.
[40] T.C. Arnot, R.W. Field, A.B. Koltuniewicz, Cross-ow and dead-end microltration
of oilywater emulsions: part II. Mechanisms and modelling of ux decline,
J. Membr. Sci. 169 (2000) 115.
[41] T. Mohammadi, A. Kohpeyma, M. Sadrzadeh, Mathematical modeling of ux
decline in ultraltration, Desalination 184 (2005) 367375.
[42] H. Peng, A.Y. Tremblay, Membrane regeneration and ltration modeling in
treating oily wastewaters, J. Membr. Sci. 324 (2008) 5966.
[43] N. Ghaffour, Modeling of fouling phenomena in cross-ow ultraltration of sus-
pensions containing suspended solids and oil droplets, Desalination 167 (2004)
281291.
[44] T. Mohammadi, M. Kazemimoghadam, M. Saadabadi, Modeling of membrane
fouling and ux decline in reverse osmosis during separation of oil in water
emulsions, Desalination 157 (2003) 369375.
[45] X. Hu, E. Bekassy-Molnar, A. Koris, Study of modelling transmembrane pressure
and gel resistance in ultraltration of oily emulsion, Desalination 163 (2004)
355360.
[46] D.A. Masciola, R.C. Viadero Jr., B.E. Reed, Tubular ultraltration ux prediction for
oil-in-water emulsions: analysis of series resistances, J. Membr. Sci. 184 (2001)
197208.
[47] A.L. Ahmad, M.F. Chong, S. Bhatia, Ultraltration modeling of multiple solutes sys-
tem for continuous cross-ow process, Chem. Eng. Sci. 61 (2006) 50575069.
[48] R.I. Issa, Solution of the implicitly discretised uid ow equations by operator-
splitting, J. Comput. Phys. 62 (1986) 4065.
[49] H. Rezaei, F.Z. Ashtiani, A. Fouladitajar, Effects of operating parameters on fouling
mechanism and membrane ux in cross-ow microltration of whey, Desalination
274 (2011) 262271.