Ceramic Membranes: Theory

and Engineering Applications

H. L. F. Magalhães, Antonio Gilson Barbosa de Lima,

S. R. de Farias Neto, A. F. de Almeida, T. H. F. de Andrade
and V. A. A. Brandão

Abstract Porous membranes are equipment used to separate different phases,

restricting, totally or partially, the transport of one or more species present in a fluid
solution. Separation processes can be classified in microfiltration, ultrafiltration,
nanofiltration and reverse osmosis Filtration using porous membranes has presented
promising results in many industrial sectors, especially in water treatment. This
chapter provides theoretical and experimental information about ceramic and
polymer membranes, with particular reference to separation process. Herein, several
topics related to this theme, such as, theory, experiments, macroscopic mathemat-
ical modeling, and technological applications are presented and well discussed.
CFD simulations of the water/oil separation process using a tubular ceramic
membrane have been performed. The study clarified the importance of the CFD
technique to elucidate the fluid flow phenomena in porous membrane as used in
liquid filtration processes.

H. L. F. Magalhães
S. R. de Farias Neto
A. F. de Almeida

Department of Chemical Engineering, Federal University of Campina Grande,
Av. Aprígio Veloso, 882, Bodocongó, Campina Grande, PB 58429-900, Brazil
e-mail: [email protected]
S. R. de Farias Neto
e-mail: [email protected]
A. F. de Almeida
e-mail: arthur.fi[email protected]
A. G. Barbosa de Lima (&)
V. A. A. Brandão
Department of Mechanical Engineering, Federal University of Campina Grande,
Av. Aprígio Veloso, 882, Bodocongó, Campina Grande, PB 58429-900, Brazil
e-mail: [email protected]
V. A. A. Brandão
e-mail: [email protected]
T. H. F. de Andrade
Department of Petroleum Engineering, Federal University of Campina Grande,
Av. Aprígio Veloso, 882, Bodocongó, Campina Grande, PB 58429-900, Brazil
e-mail: [email protected]

© Springer International Publishing AG 2018 111

J. M. P. Q. Delgado and A. G. Barbosa de Lima (eds.), Transport Phenomena
in Multiphase Systems, Advanced Structured Materials 93,
https://doi.org/10.1007/978-3-319-91062-8_4
112 H. L. F. Magalhães et al.

Keywords Membrane
Separation process
Filtration
CFD simulation

1 Foundations in Porous Membrane

Around the 1970s, in conjunction with the classical processes of separation, a new
class of process, using synthetic membranes with selective barrier, emerged. Such
membranes arise in an attempt to imitate existing natural membranes, with respect
to the characteristics of selectivity and permeability [1].
Membranes are semipermeable and selective barriers, whose function is to
separate phases, restricting, totally or partially, the transport of one or more species
present in a solution.
The membranes are mainly developed from two distinct classes of materials: the
organic, mostly polymers; and inorganic, such as metals and ceramics.
Separation processes using membranes are commonly driven by two forces:
concentration gradient between two phases of a semipermeable membrane and
pressure-driven in the membrane, which occur in microfiltration, ultrafiltration,
nanofiltration and reverse osmosis processes [2].
Filtration, especially those using membranes as a filter mechanism, has presented
good results due to the several characteristics that lead them to have the best cost/
benefit ratio, simplicity of operation, low energy cost, long lifetime and uniformity
of permeate quality throughout the process.
Park et al. [3] report that the efficiency of separation and economic viability in
membrane processes depends on the cost, operating energy, permeate flux and
lifetime of the membrane, which is directly related to the fouling, caused by the
polarization effect by concentration, formation of a gel layer on its surface,
adsorption and blocked pores.
In the literature, several studies have been reported on the development and
application of membranes, for example, the experimental works of Ma et al. [4],
Jiang et al. [5], Cheng et al. [6], Matos et al. [7] and Zhu et al. [8]. In addition, we
can cite the following theoretical works that use computational fluid dynamics as a
tool in the study of membrane separation processes, Darcovich et al. [9] and
Geraldes et al. [10] using flat plate membranes; Serra et al. [11] and Serra and
Wiesner [12] using circular membranes, and Bellhouse et al. [13], De Souza [14]
and Magalhães [15], using tubular membranes.

1.1 Membrane Classification

Membranes can be classified into several categories assuming various configura-

tions and modules depending on the characteristics of the substances to be used in
the process. For each membrane category, there is a pre-established average pore
diameter (Fig. 1). In microfiltration, the pore diameter range from 0.1 to 10 lm; for
Ceramic Membranes: Theory and Engineering Applications 113

Particles
Microfiltration
10 μm

Ultrafiltration Macromolecules
0.1 μm
Sugars
Nanofiltration
0.002 μm

Water

Fig. 1 Types of filtration according to the degree of particle retention

ultrafiltration, the pore diameter range from 0.001 to 0.1 lm; for nanofiltration
membranes, the pore diameter range from 0.0005 to 0.002 lm; the choice depends
on the morphology and size of the particles in the solution [16]. Following this
topic will be detailed.

1.1.1 Nanofiltration Membranes

Nanofiltration consists in a membrane separation process indicated for molecules

separation of a wide range of molecular weights, being used in several areas, from
the fine chemical industry (separation of organic molecules) to high valued mole-
cule recovery (drugs, enzymes and biocatalysts) [17]. Considered as a relatively
new process, nanofiltration has characteristics that can be highlighted, such as its
capacity of anion fractionation of low molecular weight with different sizes and
valences through the ionic exclusion, and the fractionation of different organic
components in aqueous solution [18].

1.1.2 Ultrafiltration Membranes

Ultrafiltration is a process indicated for solute particle size 10 times larger than the
particle size of the solvent molecule and is intended for the concentration or
fractionation of macromolecules [2]. In the water supply treatment, in which the
organic contaminants is the main concern, the ultrafiltration process is considered
adequate, presenting lower energy consumption, greater efficiency for the removal
of organic pollutants and natural organic matter as a function of the molecular
weight [19].
114 H. L. F. Magalhães et al.

Table 1 Main characteristics of the separation process utilizing microfiltration membranes

Advantages Disadvantages
Low cost; Does not perform well for ionic substances and gases that
Low energy consumption; permeate the system;
Possible integration to other Are subjected to chemical attacks;
treatment process; Effluent must present low contaminants concentration at
Capable of metal selective the inlet of the membrane
removal

1.1.3 Microfiltration Membranes

Microfiltration may be defined as a separation process that employs a permeable

microporous membrane for phase separation as a barrier. It is a process applied to
several industrial fields, which primarily uses ceramics in its fabrication because of
its chemical stability, high durability and resistance to high temperature and pres-
sure [20]. However, there are some advantages and disadvantages (Table 1) of the
microfiltration process which must be taken into account according to the particle/
process desired.
Application of microfiltration membranes in the most various processes and
industrial fields is noticed in the literature, where the choice of membrane type and
the material used in its preparation will depend on the characteristics of the sub-
stances to be treated [21–25].

1.1.4 Polymeric Membranes

Polymer membranes are characterized by their versatility in obtaining different

morphologies in conjunction with low production cost, featuring numerous appli-
cations based on the characteristics of the material used for its manufacture.
However, there are few commercial polymers that can be applied in the manu-
facture of membranes, thus, in several cases, modification or combination of
polymers to obtain the adequate material to produce a membrane is performed [26].
In this sense, Leite et al. [27] report a growing use of second generation
membranes produced from synthetic polymers (such as polyetherimide) presenting
high chemical, thermal and chlorinated compounds resistance. In contrast, these
membranes present low resistance to mechanical compaction.
Perles [28] highlights the use of polymer membranes in combustion cells, which
are mechanisms responsible for converting Gibbs free energy variation of a redox
reaction into electric energy, using the polymer membrane electrolytes as a tool.
However, for polymer membranes to be well employed, the relationship between
the molecular and morphological structure of the polymer must be well known for
the well reactional development of the membrane.
Ceramic Membranes: Theory and Engineering Applications 115

In general, polymeric membranes, even showing some disadvantages in relation

to the ceramic ones, hold numerous applications, such as, in dehydration process of
ethanol, in fuel cells, in water treatment and in the food industry.

1.1.5 Ceramic Membranes

Ceramic membranes (Fig. 2) can be defined as a type of filter or a ceramic barrier

separating two phases delimiting, totally or partially, the transport of one or several
chemical species in the solution, where the separation capacity of the membranes
will depend on porosity and selectivity [29].
Ceramic microfiltration membranes usually have pore sizes ranging from 0.2 to
0.8 lm and can be fabricated by any synthesizing particle method or sol-gel pro-
cess, widely spread in the literature, presenting, as advantages, flexibility in the
synthesis, uniformity in pore size and optimal reproducibility. The less diffused
particle synthesis methods begin by coating a thin layer of particles to subsequently
perform a high temperature treatment (1000–1600 °C) to partially synthesize the
particles and create a porous separation layer [30].
Ceramic membranes present advantages over the polymeric ones, regarding their
chemical inertia, biological stability and resistance to high temperatures and pres-
sures. Also demonstrating advantages over the traditional methods of separation
such as, distillation and centrifugation, showing low energy consumption, long
lifetime, low space occupation and easy cleaning [31].
Zhu et al. [32] stress some advantages of ceramic membranes, when studying
low cost composite membranes of microfiltration. The authors report that ceramic

Fig. 2 Tubular ceramic membrane

116 H. L. F. Magalhães et al.

membranes excel in the treatment of emulsions with particle diameters less than
20 lm, compared to other treatment techniques such as gravity sedimentation,
adsorption and flotation. They also have several benefits compared to polymer
membranes due to their mechanical performance and easy regeneration.
The pore shape, size and distribution are parameters that influence the perme-
ability, in which the fragility is the main disadvantage of the ceramic membrane,
which can be circumvented with a support material [33]. Further, Chen et al. [34]
studying tubular ceramic membranes for nanofiltration, report that although they are
particularly important for processes involving molecular separation under adverse
conditions, they are usually made by the sol-gel process, which often presents
problems of low efficiency and unsatisfactory control of the membrane properties.
In their work, Chen et al. [34], present a strategy for confection of ceramic mem-
brane, based on the atomic layer deposition and calcination, thus reducing this
problem.
In synthesis, the ceramic membranes have high applicability, being commonly
used in several separation processes, with good results.

1.2 Behind the Permeation Process

Separation processes using membranes as a filter involve various physical phe-

nomena as permeation is initiated. Understanding some basic processes is
fundamental.

1.2.1 Permeation and Diffusion

Hwang and Kammermeyer [35] report that permeation can be defined as the phe-
nomenon that occurs when a particular species or component passes through
another substance, occasionally occurring, but not necessarily by the diffusion
mechanism. Since the term diffusion is specifically used to denote molecular dif-
fusion, the term permeation emerges as a phenomenological definition involving
various transport mechanisms.
Through the definition of the phenomenological permeability, it is possible to
express schematically the profile of concentration through a membrane. Figure 3
graphically shows this behavior. In this figure, we can see the existence of the initial
concentration (concentration of the contaminated fluid) and final concentration
(concentration do fluid after treatment with membrane), respectively, C1 and C2, as
well as the initial and final concentrations on the membrane surface, C1 and C2,
with formation of a boundary layer near the membrane.
When the permeation mechanism is the diffusion, the permeate flux, F, can be
written:
Ceramic Membranes: Theory and Engineering Applications 117

Fig. 3 Concentration profile

Γ1 Membrane

Boundary layer
through a membrane

C1

Boundary layer
Γ2
C2

C1  C2
F ¼ DAB S ð1Þ
e

where S is the area and e is the membrane thickness.

Exist too, possible to study the resistance of the boundary layer separately by
analyzing the membrane thickness on the Darcy permeability, K, defined by
(Eq. 2):
e
Rm ¼ ð2Þ
K

where K is the Darcy permeability.

1.2.2 Polarization by Concentration

The effect of polarization by concentration is pertinent to any membrane separation

process, being a phenomenon considered reversible. This effect occurs whenever a
solution permeates a selective membrane, which provokes an increase in the solute
concentration at the membrane interface, causing an increase in the solution
osmotic pressure close to the membrane and reducing the driving force of sepa-
ration [18].
According to Damak et al. [36], in a cross-flow filtration process with a mem-
brane, the particles contained in the feed stream are convectively transported to the
membrane surface, causing them to accumulate, until the equilibrium between the
convective and diffusive flows is reached, forming a polarized layer by concen-
tration. This becomes the major problem during filtration, because it causes the
permeate flux to decline along the time.
118 H. L. F. Magalhães et al.

Figure 4 illustrate schematically the appearance of the polarization layer by

concentration. In this figure, we can see a feed flow at a given distance, d, from the
membrane surface, with a concentration, Ca. Is considered that, near the membrane,
occurs the formation of a boundary layer that reaches its maximum value of con-
centration, Cm, being J
C the convective flux of solute in the membrane direction.
If the solute is not completely retained by the membrane, there will be a diffusion
flux return DAB ðdC=dxÞ, where the steady-state conditions will only be reached
when the convective transport is equal to the sum of the permeate flux and diffusive
transport [37].
The balance equation under permanent conditions can be written as follows:

dC
J
C  DAB ¼ J
Cp ð3Þ
dx

where Cp is the concentration after permeation and J
Cp is the flux of solute

through the membrane.
The following boundary conditions, is considered:

x ¼ 0 ) C ¼ Cm ð4Þ

x ¼ d ) C ¼ Ca ð5Þ

Thus, integrating Eq. 3, we obtain:


Cm  Cp Je
¼ exp ð6Þ
Ca  Cp DAB

where J is the volumetric flux and is the membrane thickness.

Boundary
layer
Feed flow
Cm
J .C p
J .C

Ca
dC
DAB
dx Cp

x 0

Fig. 4 Principle of polarization by concentration

Ceramic Membranes: Theory and Engineering Applications 119

This phenomenon of formation of the polarized layer depends mainly on the

hydrodynamics of the system, the nature and size of the solute molecules and the
membrane porosity and permeability; the latter limits the diffusive solute transport,
which can cause irreversible negative effects such as: fouling by deposition, pre-
cipitation and bioinfustrations [37].
Adams and Barbano [38] studying microfiltration ceramic membrane, report that
polarization by concentration can be attenuated by increasing the turbulence level in
the flow.

1.2.3 Incrustation

Membrane operations have been widely used today; however, its large-scale and
continuous use in filtration systems has suffered some limitations. The main
problem is the membrane clogging, which provokes reduction of the filtrate flow
and process efficiency as well as the increase in cost due to the energy consumption
and materials for the cleaning process and backwashing.
There are three types of incrustation: deposition, precipitation and biofouling.
When suspended solids deposition occurs, such as colloids, organic, corrosion
products, iron hydroxide, algae and fine particulate matter, blockage of the feed
channel of the membrane modules occur, often being difficult to remove, resulting
in loss of performance.
Precipitation is a type of incrustation that occurs by precipitation of the soluble
compounds present in the feed when its solubility limit is reached. Due to the
concentration polarization, this effect intensifies near the surface of the membrane,
being able to reach the limit of the salts solubility present in the solution causing
precipitation; the most common are calcium carbonate, calcium sulfate, complexes
of silica, barium sulphate, strontium sulfate and calcium phosphate.
Biofouling occurs due to the accumulation of organic material on the membrane
surface, including cellular fragments, extracellular polymer substance and
microorganisms, which result in the formation of biofilms, being equally detri-
mental to the membrane lifetime [39].
During membrane filtration, the formation of the polarized layer and incrusta-
tions result in declining the membrane performance with respect to quality and
flow, reducing its reliability, since in applications with membranes, fouling control
is essential [40].

1.3 Technological Applications of Ceramic Membranes

Due to its versatility, uniformity in permeate quality, chemical inertia, biological

stability, high temperature resistance and long shelf life, ceramic membranes have
the most varied technological applications, present in several industrial fields, such
as the textile, paper and cellulose, oil, food and water treatment.
120 H. L. F. Magalhães et al.

In pulp and paper industry, ceramic membranes are commonly used as the basis
of advanced treatment processes because of their hydrophilic characteristics, high
durability, high permeate fluxes, and thermal, chemical and mechanical stabilities.
According to Ebrahimi et al. [41], paper processing produces substantial volumes of
wastewater, requiring efficient recovery of waste, impurities and by-products prior
to disposal into environment. Thus, membrane filtration is an effective alternative,
which acts both in wastewater treatment and in the recovery of added value
effluents.
Unlu et al. [42], in his work about the use of membranes for the reuse of waste
water, report that in face of the various methods applicable to the dye wastewater
treatment/recovery in textile effluents, membrane filtration technology is considered
one of the most promising. Silva et al. [43] point out ceramic membranes as an
important tool for the treatment of water from this sector. Meksi et al. [44] notice
that textile wastewater has high staining, biochemical oxygen demand, chemical
oxygen demand and salinity, which makes its treatment essential before its reuse or
disposal in the environment, stressing out the membranes importance.
Lee et al. [45] report the use of ceramic membranes for water treatment. These
authors argue that ceramic membranes are still less used in the treatment of
wastewater to obtain potable water because of a historical conception about the high
manufacturing cost of these membranes in comparison to the polymer membranes
already consolidated in the market for this purpose. However, because the practice
of sustainable development linked to the manufacture of ceramic membranes, it was
possible to make membranes with better ratio cost-benefit, enabling the production
of higher performance and more compact systems, especially in forms tubular,
monolithic and in-plane-aligned, which currently present superior than performance
the hollow fiber membranes and present a large permeation fluxes.
Kumar et al. [46] conducted a research on the treatment of milk wastewater
using a low-cost tubular ceramic membrane. According to the authors, the dairy
industry produces an enormous quantity of wastewater due to the different opera-
tions present in the unit that demand this resource, generating approximately 2.5 L
of wastewater for each 1 L of processed milk. This residue contains a high con-
centration of nutrients, as well as high chemical and biological oxygen demand, as
well as high total suspended solids. Therefore, it is necessary to carry out an
appropriate treatment prior to disposal in the environment. In this context, mem-
branes, especially the ceramic, appear as promising alternative technology, when
compared to conventional treatment techniques, being a more economical process
and requiring less space for operation. Further, we can cite its resistance to cor-
rosive chemical products even in high temperatures.
In oil industry, membrane filtration technology is used in the processes of water
treatment produced from oil and gas exploration and production activities.
According to Jamaly et al. [47] oily wastewater is contaminated water at various
concentrations of oil, and may contain fats, hydrocarbons and petroleum fractions,
such as diesel oil, gasoline and kerosene, thus requiring adequate treatment for
subsequent disposal. According to the specifications regulated by the
Environmental Control Brazilian Agency, the content of oils and grease in effluents
Ceramic Membranes: Theory and Engineering Applications 121

cannot exceed 20 mg/L [48]. Despite of the various technologies applied to the
separation processes, the ceramic membranes have been distinguished by their high
permeate index, acting as one of the final stages of this treatment process.
Damak et al. [36] point out the separation processes using membranes as one of
the most significant developments in chemical and biological process engineering,
evidencing their application in a wide range of industrial operations, in addition to
those previously mentioned, such as: separation of organic molecules, pharma-
ceutical, biological macromolecules, colloids, ions and solvents.

2 Liquid Filtration with Membranes

2.1 Laminar Flow Through Porous Membrane

For laminar flow mechanisms in membranes, the hydrodynamic theory is consid-

ered simple. However, it is necessary to know information about distinct parame-
ters, such as: average pore diameter, capillary mean length, tortuosity, pore size
distribution, particle diameter, porosity, etc. Therefore, a phenomenological dis-
cussion of hydrodynamic flow through porous systems is necessary [35].

2.1.1 Darcy’s Law

Considered as the law that describes the fluid flow through a porous system,
Darcy’s law demonstrates that the fluid flow rate is directly proportional to the
pressure gradient, as follows:

VJ K DP
¼ ð7Þ
S
t l e

where VJ is the volume, S represent the area, t is the time, l viscosity, ΔP is the
pressure drop, e is the membrane thickness and K correspond to Darcy’s
permeability.
Hwang and Kammermeyer [35] report that Darcy’s law denotes that the flow
resistance is due to viscous drag, and that the permeability parameter, K, contains all
the properties of the porous system. Fluid viscosity is defined as the internal friction
between fluid sheets flowing at different velocities, where for a fluid in laminar flow,
such friction produces shear forces. When a fluid comes into contact with a solid
surface, it adheres to the surface, resulting in a zero velocity, and as a consequence of
the viscosity and fluid adhesion property, the solid surface experiences a drag force;
the viscous resistance is, thus, a contrary force to the drag force.
Damak et al. [36, 49], Cunha [50], Pak et al. [51], De Souza [14], and others, used
in their research, mathematical models whose the Navier–Stokes equations numer-
ical solutions are connected to Darcy’s law, written as a series resistance model.
122 H. L. F. Magalhães et al.

2.1.2 Kozeny–Carman’s Equation

By considering a porous membrane constituted by a capillary bundle with

non-circular cross-section, that the fluid flow path well be tortuous, and using the
concept of hydraulic radius, the following Darcy’s law was derived:

VJ e3p DP
¼  2 2 ð8Þ
S
t k 1  ep S l e
0
0

where S0 is the surface area per particle volume and k′ is a non-dimensional

constant, dependent on the pore structure. When comparing the Eqs. (7) and (8), the
Darcy’s permeability, K, becomes:

e3p
K¼  2 ð9Þ
k0 1  ep S20

According to Hwang and Kammermeyer [35], modifications on the Kozeny’s

theory are numerous. The Kozeny–Carman equation is widely used for the case of
laminar flow in macroscopic scale, especially in filtration problems.

2.1.3 Capillarity Model (Hagen–Poiseuille Equation)

When a porous membrane consists of straight cylindrical capillaries of equal size,

the Hagen–Poiseuille equation can be applied directly to describe the fluid flow
rate. It is defined as follows:

VJ nSpr4 DP
¼ ð10Þ
t 8l e

where n is the number of capillary tubes per area unit, and r is the capillary tube
radius.

2.1.4 Damak’s Model

Occurring when the fluid moves in a well-defined trajectory, the laminar flow
exhibits layers that individually preserve the characteristics of the fluid, in which
the viscosity is responsible for weakening the tendency for turbulences to appear.
Thus, Damak et al. [36] report a model in which the permeation rate is expressed
as a function of the pressure gradient, viscosity and flow resistance, as follows:
Ceramic Membranes: Theory and Engineering Applications 123

DP
Uw ¼   ð11Þ
l Rm þ Rp

where Rm is the membrane resistance. This parameter is a function of the membrane

thickness and the porous media permeability. Rp is the resistance due to polarization
by concentration defined, as follows:

Rp ¼ rp dp ð12Þ

where dp is the concentration layer thickness and rp is the resistance to the

homogeneous concentration layers, determined by the Carmen–Kozeny equation
(Eq. 8).
 2
1  ep
rp ¼ 180 ð13Þ
d2p e3p

In the Eq. (13), dp is the solute particles average diameter, and ep is the porosity
of the polarization layer per concentration. Equation (13) is valid for disperse
spherical particles (non-deformable) with porosity varying in the range
0.35  ep  0.75.
Damak et al. [36], when studying a separation process using laminar tubular
membrane, mentioned that the concentration polarization layer can be quantified by
the empirical correlation for boundary layer thickness of concentration, dp,
(Eq. 14), valid for Reynolds number varying between 300 < Re < 1000 and
Schmidt number ranging between 600 < Sc < 3200.

0:33
dp z   
¼2 ðRe ScÞ0:33 Re0:3
w 1  0:4377 Sc0:0018 Re0:1551
w ð14Þ
di di

where Re is the Reynolds number, Rew represent the permeate Reynolds number, di
is the internal diameter, Sc correspond to Schmidt number, and z is the axial
coordinate.
The Reynolds permeation number is given by Eq. (15), as follows:

qUw Deq
Rew ¼ ð15Þ
l

where the equivalent diameter can be given by:


u
Deq ¼ dp ð16Þ
1u
124 H. L. F. Magalhães et al.

where u is the membrane porosity, dp is the average diameter of the particles that
constitute the porous material, considering as spherical particles, and Uw is the
permeation velocity.
By using the correlation of the boundary layer thickness it is possible to con-
clude that increasing the number of Reynolds leads to a reduction of the boundary
layer thickness and an increase in the permeate flux, and that an increase in per-
meation Reynolds number and Schmidt number lead to increasing polarization by
concentration [50].

2.2 Turbulent Flow Through Membranes

Early discussion is limited to laminar flow through porous media at low Reynolds
number (typically Re < 10) for which a proportionality validity relationships exists
between the fluid volumetric flow rate and pressure gradient.
Unlike the laminar flow regime, turbulent flow (flow at large Reynolds number)
occurs when the fluid does not move along a well-defined path, producing
momentum transfer. In this case, the deviation of the Darcy’s law is clear.
Due to the existence of turbulent flows, the mass conservation and linear
momentum equations only cannot adequately predict the oscillations resulting from
this phenomenon. Therefore, it is necessary to consider other terms to the equation
that predicts the fluid flow behavior. There are several models to predict turbulent
flow, e.g., RNG k-e model.
The RNG (Renormalization Group Theory) model emerged as an alternative of
the k-e standard model to flows with large Reynolds number, since various authors
defend the idea that the k-e turbulent model is inadequate to predict physical
situations where there is rotational flux. For this condition, the model overestimates
the kinetic energy dissipation, which results in a central recirculation region,
smaller than that experimentally observed. Thus, the turbulent kinetic energy and
kinetic energy dissipation rate equations (Eqs. 17 and 18) are defined by:
 *

 
@ lT
ðqkÞ þ r
qUk ¼ r
lþ rk þ Pk  qe ð17Þ
@t rkRNG

and
 *

 
@ lT e
ðqeÞ þ r
qUe ¼ r
lþ re þ ðCe1RNG Pk  Ce2RNG qeÞ ð18Þ
@t reRNG k

where k is the turbulent kinetic energy, e is the turbulent kinetic dissipation rate, l
is the dynamic viscosity, q is the density and lT is the turbulent viscosity, given by:
Ceramic Membranes: Theory and Engineering Applications 125

Table 2 Typical values of b

Ce Cl rk re
the parameters in the RNG k-e
model 1.9200 0.0850 0.0120 0.7179 0.7179
Source Ref. [52]

k2
lT ¼ Cl q ð19Þ
e

where Cl is an experimental calibration constant.

The coefficient Ce1RNG is given by:
 
g 1  4:38
g
Ce1RNG ¼ 1:42  ð20Þ
ð1 þ g3 bRNG Þ

where
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Pk
g¼ ð21Þ
qeClRNG

In Eq. (21), the parameter Pk is the production term by the shear effect given by:


** *T
Pk ¼ lT rU
rU þ rU þ Pkb ð22Þ

where PKb is the fluctuation production term (Eq. 23) determined as follows:
lT *
Pkb ¼  g
rq ð23Þ
qrq
*
where g is the gravity acceleration vector, q the fluid specific mass, lT is the
turbulent viscosity and rl is the turbulent Prandtl number.
The values for the model constants, Ce2RNG, ClRNG, bRNG, are specified on the
Table 2.

3 CFD Applications: Separation Process by Porous

Membranes

3.1 The Physical Problem

Herein, a theoretical study of a two-dimensional water/oil separation process using

a tubular ceramic membrane is performed. Figure 5 illustrate the physical problem
to be considered. It consists in a ceramic tubular membrane with 0.03 m diameter,
126 H. L. F. Magalhães et al.

and 3 m length, according to the dimensions used in the work of Damak et al. [53],
being, therefore, subjected to a tangential flow of oily water. As the membrane has
angular symmetry, only one cross section was used in the YZ plane.
For the numerical simulation, was generated a computational domain 2D, using
ICEM CFD software, as shown in Fig. 6. The mesh was created with a total of
having a total of 205,056 elements and 180,000 nodes, with concentration of the
elements near the surface of the membrane. Figure 7 shows the boundaries con-
ditions used in the modeling: inlet, concentrate outlet and permeate outlet.

Input

PlanYZ

Exit
(Concentrated)

Fig. 5 Geometric representation: tubular membrane 2D

Fig. 6 Two-dimensional mesh of the cross-section of the tubular membrane

Permeate outlet
Ceramic
membrane

Input Outlet of
concentrate
Axis of symmetry

Fig. 7 Representation of membrane boundaries

Ceramic Membranes: Theory and Engineering Applications 127

3.2 Mathematical Modeling

For this purpose, the following considerations were assumed:

– Viscosity and density of the fluids are constant and equal to those of the pure
solvent;
– The diffusion coefficient is considered constant;
– Laminar and permanent flow;
– Incompressible and isotropic porous system;
– The local permeation velocity is determined according to the in series resistance
theory;
– Concentration layer is assumed to be homogenous and the Carmen–Kozeny
expression is valid;
– Gravitational effect is neglected;
– Axial velocity profile is parabolic, according to laminar and fully developed
flow at the tubular membrane inlet cross section;
– Symmetry in relation to the angular coordinate;
– The porous medium obstruction by the solute is neglected;
– Solute adsorption on the membrane contact surface along with possible reac-
tions are neglected;
– The resistance resulting from concentration polarization, which is due to the
concentration layer formed at the membrane-fluid interface was considered;
– Wall roughness was disregarded.

3.2.1 Governing Equations

From the considerations, the mass and linear momentum conservation equations
can be written as:
 *
r
qU ¼ 0 ð24Þ

*
where q and U are, respectively, fluid density and velocity vector.
• Momentum equation


 *
 
* * * T
r
qU  U  r
lrU ¼ rp þ r
l rU ð25Þ

where p is pressure and l is the solution viscosity.

128 H. L. F. Magalhães et al.

• Energy conservation equation

@  *
ðqHÞ þ r
qU H  r
ðCe rTÞ ¼ SH ð26Þ
@t

where Ce is the effective thermal diffusivity, H is enthalpy and SH is the heat source
term. The first term of Eq. (26) is responsible by the energy accumulation; the
second relates to heat transfer by convection and the third one is responsible for
heat transfer by diffusion.
• Mass transfer equation

*
U
rC ¼ DAB r2 C ð27Þ

where C, is the solution (oil) concentration, DAB is the mass diffusivity, considered
constant for each Schmidt number (Sc), as follows:
l
DAB ¼ ð28Þ
Scq

3.2.2 Boundary Conditions

(a) Inlet conditions (mixture, z = 0)

Fully developed flow is assumed at the tubular membrane inlet section. Thus,
radial velocity is null and axial velocity is given as follows:

 y 2 
Uz ð0; yÞ ¼ 2Uz 1  ð29Þ
R

where y is the radial coordinate, R is the tubular membrane internal radius and Uz is
the average velocity, determined by:

Re l
Uz ¼ ð30Þ
qR

where Re is Reynolds number.

Initial oil concentration in the mixture, C0, will be given as follows:
Ceramic Membranes: Theory and Engineering Applications 129

C ¼ C0 ð31Þ

and inlet temperature; will be as follows:

T ¼ T0 ð32Þ

(b) Outlet conditions (Concentrated, z = L)

In the outlet section, we consider:

P ¼ P0 ð33Þ

@C
¼0 ð34Þ
@z
@Uz
¼0 ð35Þ
@y

rT ¼ 0 ð36Þ

(c) Symmetry condition (y = 0)

The following symmetry conditions were assumed at the central axis of the tube:

@Uz
¼0 ð37Þ
@y

@C
¼0 ð38Þ
@y

Uy ¼ 0 ð39Þ

@T
¼0 ð40Þ
@y

(d) Transverse plane

Due to the presented symmetry condition by the membrane, it was admitted the
following conditions for the transverse planes:

@Uz
¼0 ð41Þ
@x
@C
¼0 ð42Þ
@x
130 H. L. F. Magalhães et al.

@T
¼0 ð43Þ
@x

(e) Porous wall tube (y = R)

In the porous wall, zero axial velocity was assumed (non-slip condition) and a
temperature gradient null. Then, we can write:

Uz ¼ 0 ð44Þ

rT ¼ 0 ð45Þ

Through the porous wall, radial velocity Uy is equal to the permeation velocity
Uw, as follows:

DP
Uy ¼ Uw ¼   ð46Þ
l Rm þ Rp

where Rm and Rp are, respectively, the membrane resistance and the resistance from
concentration polarization, and ΔP is the transmembrane pressure.
The equation responsible for the mass transport (Eq. 47) [54] was included in the
model as a source term. This, term is given by:

@C
Uw Rr C ¼ DAB ð47Þ
@y

where Rr is the solute intrinsic retention by the membrane that can be assumed to be
constant for a membrane-solute system.
The transmembrane pressure (Eq. 48) is defined as being the difference between
the average permeate pressure, Pp , and the external pressure to the membrane Pex,
as follows:

DP ¼ Pp  Pex ð48Þ

The membrane resistance, Rm, the resistance from the polarization by concen-
tration, Rp, the concentration layer thickness, dp, and the resistance, rp, determined
when the concentration layer is considered homogenous, are given by Eqs. (2),
(12), (13) and (14).
Damak et al. [49] used Eq. 14 to determine local variation of the boundary layer
thickness by polarization, dp, so that the equilibrium between the convective and the
diffusive flows was reached when [(C − C0)/C0] < 0.001.
According to Damak et al. [49], the parameters (Re, Sc, Rew and z/d) from
Eq. (14) correspond to a membrane separation system for liquid ultrafiltration, low
particle concentration and laminar flow in the wall of the porous tube.
Ceramic Membranes: Theory and Engineering Applications 131

Table 3 Fluid and Parameters

membrane parameters used in
the simulations Fluid Density, q (kg/m3) 997
Viscosity, l (cP) 50
Molecular weight, MM (kg/ 18.02
kmol)
Membrane Porosity, ep (–) 0.35
Permeability, K (m2) 3.33  10−11
Thickness, e (m) 0.01

Table 4 Parameters used in Permeation Reynolds number Rew 0.1

the simulations
Initial concentration (kg m−3) C0 1
Temperature (°C) T0 25
Intrinsic solute retention by the membrane Rr 1
Oil droplet average diameter dp 51
Schmidt number Sc 1000
Atmospheric pressure (atm) P0 1

Table 5 Conditions used in Case Re (–)

the two-dimensional
01 1000
simulation with tubular
membrane 02 600
03 300

3.2.3 Fluid and Membrane Data and Simulated Cases

Important parameters used in the study of the 2D flow correspond to the thermal
and physico-chemical properties of the fluid and the membrane. The parameters are
given in Table 3.
In the simulations carried out, (Ansys CFX 15 software), the permeation
Reynolds number, the average diameter of the oil droplet, dp, Schmidt number,
initial concentration, permeability, temperature, and intrinsic solute retention by the
membrane, Rr, were kept constant according to Table 4, varying the axial Reynolds
number, Re, as shown in Table 5. All the cases were simulated in isothermal
conditions.

3.2.4 Results Analysis

(a) Concentration profile

In Fig. 8, the concentration profiles in different axial positions are represented, z/L
(0.25, 0.50, 0.75 and 1.00), where L = 100d (tubular membrane length). It is
132 H. L. F. Magalhães et al.

Fig. 8 Concentration profile

inside the membrane at
different axial positions
(case 1)

verified that the concentration increases with axial position and that the behavior of
the simulated results is similar to that one reported in the literature.
In Fig. 9, concentration profiles for different Reynolds numbers on the axial
position z = 0.50L are illustrated. It is expected that as the permeation velocity
increases a larger number of particles will be conducted convectively for near the
membrane surface. Further, according to the resistance model, the permeate flux
increases with transmembrane pressure. Therefore, higher Reynolds number pro-
vokes an increase in the transmembrane pressure, and consequently, concentration
in the membrane surface also will increase. From this figure, we can see that results
agreed well with those obtained in the literature.

(b) Thickness of the concentration boundary layer

Figure 10 illustrate the behavior of the thickness of the concentration boundary
layer as a function of the non-dimensional distance z/d, for Reynolds numbers 300,
600 and 1000. It is observed that the boundary layer grows with increasing the
non-dimensional distance z/d, and its thickness drops when Reynolds number
increases. This behavior provokes an increase in the transmembrane pressure, and
thus, we have an increase in the transport by convection of a higher number of
particles to near oh the membrane surface, as observed by Damak et al. [36].
A reduction of 0.035 in the non-dimensional boundary layer thickness, dp/d, was
observed when the Reynolds number ranged from 300 to 1000.
Ceramic Membranes: Theory and Engineering Applications 133

Fig. 9 Reynolds number effect on the concentration profile inside the membrane (z = 0.50L)
(cases 1, 2 and 3)

Fig. 10 Thickness of the concentration boundary layer versus the non-dimensional variable z/d
for different Reynolds numbers

4 Final Considerations

Ceramic membranes are considered in the literature as one of the most promising
technologies in the present day. They are always present mainly in the stages of
separation and effluents treatment from the most varied industrial sectors. They can
be produced in order to assume various configurations and modules, adapting
134 H. L. F. Magalhães et al.

continuously the processes needs. Membrane ceramic present diverse characteris-

tics such as chemical inertia, biological stability, manufacturing flexibility, high
permeability index and high temperature and temperature resistance, which pro-
vides high applicability, this versatility combined to cost-benefit ratio makes this
equipment very competitive in comparison to many others already consolidated in
the market.
In general, the topics presented in this chapter discuss the universe of application
in which the ceramic membranes are inserted. The growing applications concern
with sustainable industrial progress being the driving force for the continuous
development of new research and technologies in this field of study. Besides,
information about different models applied to filtration process in porous mem-
branes are discussed, and CFD applications are related.
From the simulated results, we can conclude that the mathematical model was
able to predict the behavior of the fluid in the water/oil separation process through
the tubular membrane, and that, concentration increases with the axial position and
with increasing Reynolds number. The increase in the Reynolds number leads to a
reduction in the thickness of the concentration boundary layer.

Acknowledgements The authors are grateful for financial support provided by CNPq, CAPES
and FINEP (Brazilian Research Agencies). We also acknowledge scientific support from the
authors mentioned along this chapter.

References

1. Habert, A.C., Borges, C.P., Nobrega, R.: Membrane separation processes. E-papers, Brazil,
Rio de Janeiro (2006) (in Portuguese)
2. Timoteo, J.F.J.: Anodizing for the production of ceramic membranes, p. 69. Masters
dissertation, Mechanical engineering, UFRN, Rio Grande do Norte, Brazil (2007) (in
Portuguese)
3. Park, J.Y., Jin Choi, S., Reum Park, B.: Effect of N2-back-flushing in multichannels ceramic
microfiltration system for paper wastewater treatment. Desalination 202(1–3), 207–214
(2007)
4. Ma, W., Guo, Z., Zhao, J., Yu, Q., Wang, F., Han, J., Pan, H., Yao, J., Zhang, Q., Samal, S.
K., De Smedt, S.C., Huang, C.: Polyimide/cellulose acetate core/shell electrospun fibrous
membranes for oil-water separation. Sep. Purif. Technol. 177, 71–85 (2017)
5. Jiang, Y., Hou, J., Xu, J., Shan, B.: Switchable oil/water separation with efficient and robust
Janus nanofiber membranes. Carbon 115, 477–485 (2017)
6. Cheng, Q., Ye, D., Chang, C., Zhang, L.: Facile fabrication of superhydrophilic membranes
consisted of fibrous tunicate cellulose nanocrystals for highly efficient oil/water separation.
J. Membr. Sci. 525, 1–8 (2017)
7. Matos, M., Gutiérrez, G., Lobo, A., Coca, J., Pazos, C., Benito, J.M.: Surfactant effect on the
ultrafiltration of oil-in-water emulsions using ceramic membranes. J. Membr. Sci. 520, 749–
759 (2016)
8. Zhu, X., Dudchenko, A., Gu, X., Jassby, D.: Surfactant-stabilized oil separation from water
using ultrafiltration and nanofiltration. J. Membr. Sci. 529, 159–169 (2017)
9. Darcovich, K., Dal-Cin, M.M., Ballevre, S., Wavelet, J.P.: CFD assisted thin channel
membrane characterization module. J. Membr. Sci. 124(2), 181–193 (1997)
Ceramic Membranes: Theory and Engineering Applications 135

10. Geraldes, V., Semião, V., Pinho, M.N.: Numerical modeling of mass transfer in slits with
semi-permeable membrane walls. Eng. Comput. 17(3), 192–217 (2000)
11. Serra, C.A., Wiesner, M.R., Laîné, J.M.: Rotating membrane disk filters: design evaluation
using computational fluid dynamics. Chem. Eng. J. 72(1), 1–17 (1999)
12. Serra, C.A., Wiesner, M.R.: A comparison of rotating and stationary membrane disk filters
using computational fluid dynamics. J. Membr. Sci. 165(1), 19–29 (2000)
13. Bellhouse, B.J., Costigan, G., Abhinava, K., Merry, A.: The performance of helical
screw-thread inserts in tubular membranes. Sep. Purif. Technol. 22–23, 89–113 (2001)
14. De Souza, J.S.: Theoretical study of the microfiltration process in ceramic membranes, p. 134.
Doctoral Thesis, Process Engineering, UFCG, Paraíba, Brazil (2014) (in Portuguese)
15. Magalhães, H.L.F.: Study of the thermofluidodynamics of the treatment of effluents using
ceramic membranes: modeling and simulation, p. 102. Masters dissertation in Mechanical
Engineering, Federal University of Campina Grande, Campina Grande, Brazil (2017) (in
Portuguese)
16. Taketa, T.B., Ferreira, M.Z., Gomes, M.C.S., Curvelo, N.: Production of biodiesel by ethyl
transesterification of vegetable oils and their separation and purification by ceramic
membranes. In: VIII Brazilian Congress of Chemical Engineering in Scientific Initiation,
Uberlândia, Minas Gerais, Brasil (2009) (in Portuguese)
17. De Carvalho, R.B., Cristiano, P.B., Nobrega, R.: Formation of double-spreading cellulosic flat
membranes for nanofiltration and reverse osmosis processes. Polymers 11(2), 65–75 (2001)
(in Portuguese)
18. Lopes, A.C.: Degradation study of polymeric membranes of commercial nanofiltration by
sodium hypochlorite, p. 92. Master dissertation, Federal University of Rio de Janeiro, Rio de
Janeiro, Brazil (2006) (in Portuguese)
19. Mierzwa, J.C., Da Silva, M.C.C., Rodrigues, L.D.B., Hespanhol, I.: Water treatment for
public supply by ultrafiltration: comparative evaluation through the direct costs of
implantation and operation with the conventional and conventional systems with activated
carbon. Sanitary Environ. Eng. 13(1), 78–87 (2008) (in Portuguese)
20. Rosa, D.S., Salvini, V.R., Pandolfelli, V.C.: Processing and evaluation of the properties of
porous ceramic tubes for microfiltration of emulsions. Ceramics 52(322), 167–171 (2006) (in
Portuguese)
21. Makabe, R., Akamatsu, K., Nakao, S.: Classification and diafiltration of polydispersed
particles using cross-flow microfiltration under high flow rate. J. Membr. Sci. 523, 8–14
(2017)
22. Yang, X., Zhou, S., Li, M., Wang, R., Zhao, Y.: Purification of cellulase fermentation broth
via low cost ceramic microfiltration membranes with nanofibers-like attapulgite separation
layers. Sep. Purif. Technol. 175, 435–442 (2017)
23. Wang, X., Wang, C., Tang, C.Y., Hu, T., Li, X., Ren, Y.: Development of a novel anaerobic
membrane bioreactor simultaneously integrating microfiltration and forward osmosis
membranes for low strength wastewater treatment. J. Membr. Sci. 527, 1–7 (2017)
24. Suresh, K., Pugazhenthi, G., Uppaluri, R.: Fly ash based ceramic microfiltration membranes
for oil-water emulsion treatment: parametric optimization using response surface methodol-
ogy. J. Water Process Eng. 13, 27–43 (2016)
25. Suresh, K., Pugazhenthi, G.: Cross flow microfiltration of oil-water emulsions using clay
based ceramic membrane support and TiO2 composite membrane. Egypt. J. Petrol. 14(1), 1–
10 (2016)
26. Becker, C.M.: Obtaining and characterizing sulphonated polyelectrolytes based on styrenic
copolymers for polymer membranes, p. 96. Masters dissertation, Federal University of Rio
Grande do Sul, Porto Alegre, Brazil (2007) (in Portuguese)
27. Leite, A.M.D., Ito, E.N., Araújo, E.M., Lira, H. De L., Barbosa, R.: Obtaining microporous
membranes from nanocomposites of polyamide 6/national clay. Part 1: influence of clay
presence on membrane morphology. Polymers 19(4), 271–277 (2009) (in Portuguese)
136 H. L. F. Magalhães et al.

28. Perles, C.E.: Physical and chemical properties related to the development of Nafion®
membranes for PEMFC fuel cell applications. Polymers 18(4), 281–288 (2008) (in
Portuguese)
29. Silva, A.A., Melo, K.S., Maia, J.B.N.: Study of the water/oil separation potential of alumina
tubular ceramic membranes through analysis of flow and turbidity measurements. In: 2º
Brazilian Congress of P & D in Oil & Gas. Rio de Janeiro: UFRJ, Rio de Janeiro, Brazil
(2003) (in Portuguese)
30. Cui, J., Zhang, X., Liu, H., Liu, S., Yeung, K.L.: Preparation and application of zeolite/
ceramic microfiltration membranes for treatment of oil contaminated water. J. Membr. Sci.
325(1), 420–426 (2008)
31. Silva, F.A., Lira, H.L.: Preparation and characterization of cordierite ceramic membranes.
Ceramics 52(324), 276–282 (2006) (in Portuguese)
32. Zhu, L., Chen, M., Dong, Y., Tang, C.Y., Huang, A., Li, L.: A low-cost mullite-titania
composite ceramic hollow fiber microfiltration membrane for highly efficient separation of
oil-in water emulsion. Water Res. 90, 277–285 (2016)
33. Maia, D.F.: Development of ceramic membranes for oil/water separation, p. 111. Doctorate
Thesis in Process Engineering, Federal University of Campina Grande, Campina Grande,
Brazil (2006) (in Portuguese)
34. Chen, H., Jia, X., Wei, M., Wang, Y.: Ceramic tubular nanofiltration membranes with tunable
performances by atomic layer deposition and calcination. J. Membr. Sci. 528, 95–102 (2017)
35. Hwang, S.T., Kammermeyer, K.: Membranes in Separation. Wiley, Canada (1975)
36. Damak, K., Ayadi, A., Schmitz, P., Zeghmati, B.: Modeling of cross-flow membrane
separation processes under laminar flow conditions in tubular membrane. Desalination 168,
231–239 (2004)
37. Mulder, M.: Basic Principles of Membrane Technology, 1st edn. Kluwer Academic
Publishers, Netherlands (1996)
38. Adams, M.C., Barbano, D.M.: Effect of ceramic membrane channel diameter on limiting
retentate protein concentration during skim milk microfiltration. J. Dairy Sci. 99(1), 167–182
(2016)
39. Oliveira, D.R.: Pre-treatment of the reverse osmosis process using microfiltration and
investigation of membrane cleaning and recovery techniques, p. 129. Master dissertation,
Federal University of Rio de Janeiro, Rio de Janeiro, Brazil (2007) (in Portuguese)
40. Ahmed, S., Seraji, M.T., Jahedi, J., Hashib, M.A.: Application of CFD for simulation of a
baffled tubular membrane. Chem. Eng. Res. Des. 90(5), 600–608 (2012)
41. Ebrahimi, M., Busse, N., Kerker, S., Schmitz, O., Hilpert, M., Czermak, P.: Treatment of the
bleaching effluent from sulfite pulp production by ceramic membrane filtration. MDPI
Membr. 7, 1–15 (2015)
42. Unlu, M., Yukseler, H., Yetis, U.: Indigo dyeing wastewater reclamation by membrane-based
filtration and coagulation processes. Desalination 240, 178–185 (2009)
43. Silva, M.C., Oliveira, R.C., Lira, H.L., Freitas, N.L.: Obtaining a ceramic membrane to treat
effluent from the textile industry. Electron. J. Mater. Process. 9, 81–85 (2014) (in Portuguese)
44. Meksi, N., Ben Ticha, M., Kechida, M., Mhenni, M.F.: Using of ecofriendly
a-hydroxycarbonyls as reducing agents to replace sodium dithionite in indigo dyeing
processes. J. Clean. Prod. 24, 149–158 (2012)
45. Lee, M., Wu, Z., Li, K.: Advances in membrane technologies for water treatment: materials,
processes and applications. In: Advances in Ceramic Membranes for Water Treatment, vol. 1,
pp. 43–82, 1st edn. Woodhead Publishing, England (2015)
46. Kumar, R. V., Goswami, L., Pakshirajanb, K., Pugazhenthi, G.: Dairy wastewater treatment
using a novel low cost tubular ceramic membrane and membrane fouling mechanism using
pore blocking models. J. Water Process Eng. 13, 168–175 (2016)
47. Jamaly, S., Giwa, A., Hasan, S.W.: Recent improvements in oily wastewater treatment:
progress, challenges, and future opportunities. J. Environ. Sci. 37, 15–30 (2015)
48. CONAMA No20/ART.21, RE. Standard CONAMA n°20, de June 18, Brazil (1986) (in
Portuguese)
Ceramic Membranes: Theory and Engineering Applications 137

49. Damak, K., Ayadi, A., Zeghmati, B., Schmitz, P.: Concentration polarisation in tubular
membranes—a numerical approach. Desalination 171(2), 139–153 (2004)
50. Cunha, A.L.: Treatment of effluents from the petroleum industry via ceramic membranes—
modeling and simulation, p. 201. Doctorate Thesis in Process Engineering, Federal University
of Campina Grande, Campina Grande, Brazil (2014) (in Portuguese)
51. Pak, A., Mohammad, T., Hosseinalipour, S.M., Allahdinib, V.: CFD modeling of porous
membranes. Desalination 222(1–3), 482–488 (2008)
52. Ansys: CFX 15, Solver Theory Guide. Ansys, Japan (2015)
53. Damak, K., Ayadi, A., Zeghmati, B., Schmitz, P.: New Navier-Stokes and Darcy’s law
combined model for fluid flow in cross-flow filtration tubular membranes. Desalination 161
(1), 67–77 (2004)
54. Minnikanti, V.S., Dasgupta, S., De, S.: Prediction of mass transfer coefficient with suction for
turbulent flow in cross flow ultrafiltration. J. Membr. Sci. 154(2), 227–239 (1999)