TECHNICAL PAPER

Ma
No n
t f ey
or Pu
Di bl
st ish
rib in
ut g
io
n

Energy and exergy analysis and modeling

temperature distribution in a membrane
distillation feed channel for VMD and DCM
Malik M. A. Fakron*

This research is a comparative study between vacuum membrane distillation and direct contact
membrane distillation technologies. The first part focuses on an energy and exergy analysis of
vacuum membrane distillation and direct contact membrane distillation configurations. The results
of the energy analysis show the vacuum membrane distillation has lower energy losses across the
membrane compared to direct contact membrane distillation. The results of the exergy analysis
show that direct contact membrane distillation requires less work compared to vacuum membrane distillation. The second part of this research focuses on modeling of temperature distribution in feed channel for DCMD and VMD. The results of the numerical solution show that the
temperature polarization effort in VMD is smaller than in DCMD. The membrane temperature in
VMD approach constant temperature case and this reduces dissipated energy in the membrane
distillation process.
Keywords: Vacuum, Membrane distillation, Direct contact membrane distillation, Energy, Exergy, Temperature

Nomenclature
A:
COP:
CP:
D .:
Dequ.:
EN:

EX:
EX 9:
f:
K:
h:
hc:
J:
LW:
_
m:
p:
Pext :
PV FEED :
PV PERMEATE :
PP:
Q:
T:
T W ; m:
W:
u, v:
v:

membrane area
coefficient of performance for the process
specific heat
diameter
diameter of air gap on membrane surface
energy input for feed stream or permeate
stream
exergy
final exergy
friction factor
thermal conductivity
enthalpy
heat transfer coefficient
flux in kg/m2.hr
lost work
mass flow rate
pressure
working pressure
vapor pressure at feed side
vapor pressure at permeate side
pumping power for feed stream or permeate
stream
heat transfer
temperature in Cu
temperature of membrane surface
work of expansion
are the velocity components in X, Y
specific volume

University of Calabria, Italy

*Corresponding author, email: [email protected]

2015 International Desalination Association

Received 31 October 2014; accepted 3 December 2014
DOI 10.1179/2051645214Y.0000000031

vvFEED :
vvpermeate :
nsuction :
Ma:
Nu:
Re:
Ra:
Pr:
m:
g:
d:
dT :

specific volume at feed side

specific volume at permeate side
suction velocity
Malik number
Nusslet number
Reynolds number
Rayleigh number
Prandlt number
dynamics viscosity
exergy efficiency
Thickness of hydrodynamics boundary layer
Thickness of Thermal boundary layer

Introduction

Membrane distillation for water treatment is a thermally

driven process. This process has various advantages
compared to traditional distillation and pressure driven
membrane processes such as reverse osmosis (RO).
Membrane distillation technology has attracted interest
from the research community due to its simplicity and
ease of implementation, the potential for using waste
heat, and its wide range of applications. Research has
focused on the use of membrane distillation in desalination, waste water treatment, water treatment and
concentration of aqueous solutions (Phattaranawik
et al., 2003). Membrane distillation can also be used to
concentrate thermally sensitive liquid feeds, where the
removal of water is the main concern, and has been
shown to achieve high solid contents at low operating
temperatures (Criscuoli et al., 2008; Lawson and Lloyd
1996, 1997).

IDA Journal of Desalination and Water Reuse

2014

VOL

NO

3-4

121

Fakron

Energy and exergy analysis and modeling temperature distribution

Membrane distillation can be configured as; direct

contact membrane distillation (DCMD), air gap membrane distillation (AGMD), sweeping gas membrane
distillation (SGMD) or vacuum membrane distillation
(VMD) (Imdakm et al., 2007; Imdakm and Matsuura,
2004; Drioli, 2005). The main problem encountered
during membrane distillation is low flux, which may be
due to various reasons including, temperature polarization phenomena and low energy efficiency for membrane distillation (Ding et al., 2008; Khayet et al., 2004;
Ratkovich et al., 2009).
This paper is split into two parts. The first gives an
energy and exergy analysis of two types of membrane
distillation, DCMD and VMD. The second models the
temperature distribution along a membrane distillation
feed channel for DCMD and VMD.

whilst on the permeate side they are given by

_ pout hpout 2 m
_ pinlet hpinlet
EN m

PP Qin Pin 2 Qout Pout

Ma
No n
t f ey
or Pu
Di bl
st ish
rib in
ut g
io
n

The process performance is therefore given by

 X

Xpermeate
permeate
PP
EN
COP W
feed
Feed

Exergy analysis of the membrane distillation

process

The exergy analysis is divided into two systems, the cold

permeate system and the hot feed system.

Energy analysis for membrane

distillation

Analysis of feed side for a DCMD cell

In the thermal analysis for membrane distillation process the process was divided into two systems; a feed
system and a permeate system. The interaction between
the feed stream system and the permeate stream system
included heat transfer and work. Figure 2 shows the
systems and heat and work were transferred across the
membrane boundary.
The exergy equations are

The first assumption is to consider the vapor transport

across the membrane as the system and the effect desired
in this process is the expanding work of vapor across the
membrane, from saturated vapor at the feed side temperature, to saturated vapor at permeates side temperature (Criscuoli et al., 2008).

Performance of direct contact membrane distillation

Q1 A hc T 1 2 T membrane

12

EX h 2 T SURRONDING S

13

This process is not isotropic as heat transfer occurs by

conduction to the vapor inside. As the vapor expands
from the specific volume of saturated vapor at feed vapor
pressure to the specific volume of saturated vapor at the
permeate vapor pressure, Equations (1) to (2) become
3

_ AJ
m

The energy and pumping power input on the feed side of

membrane are given by
_ fout hfout
_ finlet hfinlet 2 m
EN m

PP Qin Pin 2 Qout Pout

1 Vapor system across membrane

IDA Journal of Desalination and Water Reuse

EX OUT 2 EX IN

LW

15

T 5 T 3 T 4 =2

16

_ 1 C p T 1 2 T 0 2 C P T 0 ln T 1 =T 0
EX 1 m

17

_12m
_ 3 C p T 2 2 T 0
EX 19 m
2 C P T 0 ln T 2 =T 0

18

2 Thermal analysis for MD process

2014

VOL

14




T SURROUNDING
EX Q 1 2
W SYSTEM
T SYSTEM
in




X
T SURROUNDING
EX Q 1 2
2
WS
T SYSTEM
OUT

_
W PV FEED 2 PV PERMEATE vvpermeate 2 vvFEED m

X

V1

122

Q2 Awall hc T 1 2 T wall

W MIN

Pext PV Feed 2 PV PERMEATE

11

PdV

V1

V2

Pext dV

10

V2

Figure 1 shows the vapor system inside the membrane

pores for the membrane distillation process.
Let us consider the work done by a specific volume of
a saturated vapor system at feed temperature and vapor
pressure, expanding to reach the specific volume of
permeate at permeate vapor pressure, and then start
condensation. The expansion work of the vapor system
inside the pore can be modeled by:
W 2

NO

3-4

Energy and exergy analysis and modeling temperature distribution

Fakron

_ 2 C p T 3 2 T 0 2 C P T 0 ln T 3 =T 0
EX 2 m

19

_2m
_ 3 C p T 4 2 T 0
EX 29 m
2 C P T 0 ln T 4 =T 0

20

_ 3 C p T 5 2 T 0 2 C P T 0 ln T 5 =T 0
EX 3 m

21

W min EX 19 2 EX 1 EX 3

22

Ma
No n
t f ey
or Pu
Di bl
st ish
rib in
ut g
io
n



T5
LW EX 1 2 EX 19 2 W 2 Q1 1 2
T1


T0
2 EX 3
2 Q2 1 2
T1

23

giving an exergy efficiency on the feed side of

gfeed

W min
LW W min

24

2.2.2 Analysis for permeate side of DCMD cell

The exergy equations are

W min EX 29 2 EX 2 2 EX 3



T1
LW EX 2 2 EX 92 W Q1 1 2
T5
EX 3

25

26

This is the exergy giving an efficiency on the permeate

side of

gpermeate

W min
LW W min

27

For VMD, the same equations and analysis were used

for the feed side but the difference in permeate because
the negative pressure in permeate side in the case VMD
permeate side.
EX 2 0:0; EX 29 EX 3 ;

28

Then, W min < 0:0, g < 0:0

The experimental data (from Criscuoli et al., 2008),
calculations and analysis were based on the thermal
analysis for membrane distillation process for two systems. Between these two systems there is heat transfer,
mass transfer and work done for transfer the matter.

3 Velocity profile in membrane distillation process

The suction velocity has an important impact on the

thickness of the boundary layer. Due to the increase of
the rate of deformation of the fluid elements in the
region near to the lower plate, the lower boundary layer
thickness is decreased as shown in Fig. 3
The flow regime is fully developed laminar flow in
duct with variable t velocity profile along the module
length in order to find the solution for the temperature
profile. In reality the velocity and temperature profiles
vary along the length of the membrane distillation cell
due to the changing suction velocity and membrane
temperatures (caused, in part, by non-constant friction
factors). To facilitate the modeling of the membrane
distillation process a new term (called the MALIK
number) is introduced by modification of the Weber
number for the membrane distillation case, with new
heat transfer correlations.
Dequ A 2 porosity*A*4=P1=2 28;
rn0 Dequ
Ma
s
rVD
Re
m
Pr

31

32

mC p
31; Nu a Re b Pr c Ma d
K

33

The continuity equation is

Lu Lv
0:0
Lx Ly

34

While the momentum equation is

Theoretical modelling for temperature

distribution in membrane distillation
feed channel

Lu
Lu
1 LP
L2 u L2 u
n
u v 2

Lx
Ly
r Lx
Lx 2 Ly 2

The systems to be considered in this research are flat

sheet DCMD modules and VMD modules. This study
concentrates on modeling of the temperature distribution in the feed channel. The feed channel is modelled
as laminar flow forced convection between two parallel
plates the length of the cell, with suction from the lower
plate (at y 0.0 in Figs. 36). For the analysis a uniform
suction velocity along the cell length shall be assumed,
such that the suction velocity is given by

suction velocity flux area=density porosity area

29

In reality, the suction velocity along the cell length is not

constant, declining along the cell membrane length due
to the change in the process driving force, such that

nsuction F x; T m1 ; T m2 ; f ; porosity

30

35

And the energy equation

LT
LT
n L2 T
u
v

Lx
Ly pr Ly 2

36

The boundary conditions are

Lu
LT
0:0; v 0:0;
0:0
Ly
Ly
y 0:0; u 0:0; v cons; T W cons
y d;

37a

The first step puts the continuity, momentum and energy

equations in the dimensionless form, to use these
dimensionless parameters. The boundary-layer thickness
is then calculated, if the thickness is less than the cell
height, then takes it as the second point of boundary
conditions and if higher than cell height take the half of
the cell as the second point boundary conditions and this

IDA Journal of Desalination and Water Reuse

2014

VOL

NO

3-4

123

Energy and exergy analysis and modeling temperature distribution

Ma
No n
t f ey
or Pu
Di bl
st ish
rib in
ut g
io
n

Fakron

4 Temperature profile for DCM

mean, the suction very low and there is no significant

effect on the rate of deformation in the fluid boundarylayer thickness. To find the center line for velocity
component in x-direction, which is the maximum value
for the velocity in x-direction which it can be calculated
from the perturbation solution for these equations
(Schofield et al., 1987; Oosthuizen et al., 1999).
y d;

Lu
LT
0:0; v 0:0;
0:0
Ly
Ly

This case considers the flow between two parallel plates

with suction to the lower plate. To solve this problem
the finite difference method applies, the finite difference
explicit scheme converts the partial differential
equations sets of algebraic equations then generated the
matrix of coefficients, solved by using algorithm themes
solver (Oosthuizen et al., 1999).

Method of solution: finite difference explicit

scheme discretization

37b

y 0:0; u 0:0; v cons; T W cons:

37c

p
u
x
; X ; V u Re L =ucc ;
ucc
L
p
Y y Re =L; P P 2 P1 =ru 2cc ;
LU LV

0:0
h T 2 T 1 =T W 2 T 1 ;
LX LY
U

37d

LU
LU
dp L2 U
V
2
LX
LY
dx LY

38

Lh
Lh
1 L2 h
V

LX
LY
Pr LY 2

39

IDA Journal of Desalination and Water Reuse

2014

VOL

40

U i;j1 U i;j21 2 2U i;j

L2 U

LY 2
i;j
DY 2

41

U i;j 2 U i21;j
LU

LX
i;j
DX

42

As shown in Figure 7.
Substituted in the momentum equation to get the algebraic equation:
Aj U i;j Bj U i;j1 C j U i;j21 Dj

5 Temperature profile for VMD

124

U i;j1 2 U i;j21
LU

LY
i;j
2DY

NO

3-4

43

Energy and exergy analysis and modeling temperature distribution

Ma
No n
t f ey
or Pu
Di bl
st ish
rib in
ut g
io
n

Fakron

Temperature profile for DCMD

Temperature profile for VMD

6 Temperature profile for DCM & VMD

7 Finite difference scheme for flow field

Aj

Bj

U i21;j
DX

V i21;j
2DY

DY 2


1
2
DY 2

44



 
V i21;j
1
Cj 2
2
2DY
DY 2

Dn

U 2i21;j
DX

dp
dx

45

46

U i;1 0:0

A2 U i;2 B2 U i;3 C 2 U i;1 D2 ;

A3 U i;3 B3 U i;4 C 3 U i;2 D3

47

48

49

AN21 U i;N21 BN21 U i;N C N21 U i;N22 DN21

50

U i;N U 1

51

32 U 3 2 0 3
i:1
7 6
7
6
76
7 6
7
6 C 2 A2 B2 ::::::::::::::::::::::::::0 76
7 6
7
6
76
7
7
6
6
6
76
7
7
6
6
76
7 6
7
60
76
7
7
6
6
76
7 6
7
6
76
7
7
6
6:
76
7
6
7 52
6
76
76
7
6
76
7
6
7
6
76
7 6
7
6:
76
7
6
7
6
76
7
6
7
6
76
7 6
7
6 0:::::::::::::C
7
6
7
6
7
6
N21 AN21 BN21 76
7 6
7
4
54
5 4
5
U i;N
U1
0::::::::::::::::::::::::::::::::::::1
2

1::::::::::::::::::::::::::::::::::::0

QU U i;j RU

IDA Journal of Desalination and Water Reuse

53

2014

VOL

NO

3-4

125

Fakron

Energy and exergy analysis and modeling temperature distribution

Then

Where Qu is a tridiagonal matrix this equation can be

solved using the standard traditional matrix solver
algorithm, which is often termed the Thomas algorithm
(Oosthuizen et al., 1999)
A1
D1
Bi
; c1
; ai
;
C1
A1
Ai 2 ai21 C i
Di 2 C i ci21
ci
Ai 2 C i ai21

54

Then X n cn 54;

55

70

hi ci 2 ai hI1

71

The Thomas algorithm was used in a case study:

y 0:0; u 0:0; v cons; T W cons

72

Ma
No n
t f ey
or Pu
Di bl
st ish
rib in
ut g
io
n

a1

hn cn

X i ci 2 ai X I1

For the energy equation:

hi;j 2 hi21;j
Lh
U i21;j
U
LX
DX




hi;j1 2 hi;j21
Lh
V i21;j
V
LY i;j
2DY

The treatment of the flow in MD process as flow

between two parallel plates with suction at the lower
plate considers the first part which is straight from the
membrane surface to d and the second part start from
the upper surface of the channel minus d, the case is fully
developed. The membrane distillation cell consists of a
membrane surface area of 40 cm2, a membrane width of
4 cm, a membrane length of 10 cm, a channel width of
4 cm and a channel height of 0.5 cm.
Case study for DCMD:

56

57

L2 h
hi;j1 hi;j21 2 2hi;j

2
LY
DY 2

58

J 25:8 L=hr:m 2 ; T fin 60C 0 ; T fout 59:4C 0 ;

T din 17:4C 0 ; T dout 18C 0 ; Qf 230 L=hr;

Qd 200 L=hr

Substitute into the energy equation to convert the energy

equation into a set of algebraic equations:
E j hi;j F j hi;j1 G j hi;j21 H j

 

U i21;j
2

Ej
DX
Pr DY 2


 
V i21;j
1
Fj
2
DY
Pr DY 2

59

E 3 hi;3 F 3 hi;4 G 3 hi;2 H 3



 
V i21;j
1
Fj 2
2
DY
Pr DY 2


U i21;j hi21;j
Hj
DX

62

Case study for VMD:

J 43:7kg=hr:m 2 ;T fin 59:3C 0 ;

Vacuum pressure 60mbar:
T fout 58:7C 0 ;Qf 200L=hr;

60
61

Results and discussion

Figures 4, 5 and 6 show the temperature distribution in

DCMD where the temperature at the membrane surface is
around 53uC, the temperature at the center line of the flow
regime fluctuated between 53uC and 59uC. The cell wall
temperature was founded at 37.5uC in VMD and the
membrane surface temperature fluctuated between 56uC to
55uC as shown in Fig. 5. The temperature at the center line
of the flow regime fluctuated between 59uC to 58uC (Fig. 5).
Fig. 6 shows the effect of temperature polarization
phenomena, this was found in direct contact membrane
distillation configurations greater than the vacuum
membrane distillation configuration. This is due to the
temperature gradient in the membrane surface because
the thermal boundary layer is thicker in direct contact
membrane distillation configuration than vacuum membrane distillation configuration and due to the characteristics of the permeate stream. Further information is
provided in Tables 1 and 2.

63

64

hi;1 hw ; E 2 hi;2 F 2 hi;3 G 2 hi;1 H 2

65

E N21 hi;N21 F N21 hi;N G N21 hi;N22 H N21

2
3 2
3
hw
2
3 hi;1
7 6
7
1::::::::::::::::::::::::::::::::::::0 6
6
7 6
7
6
76
7 6
7
6 G2 E 2 F 2
76
7 6
7
6
76
7 6
7
60
76
7 6
7
6
76
7 6
7
6
76
7 6
7
6:
76
7 6
7
6
76
76
7
6
76
7 6
7
6:
76
7 6
7
6
76
7 6
7
6
76
7 6
7
6 0:::::::::::::G N21 E N21 F N21 76
7 6
7
4
56
7 6
7
6
7 6
7
4
5
4
5
0::::::::::::::::::::::::::::::::::::1
hi;N
hi;N

66

QT hi;j RT

68

67

Table 1 Thermal analysis data for direct contact membrane

distillation (Criscuoli et al., 2008)
Membrane area
Feed temperature
Permeate temperature
Water flux
Feed vapor pressure
Permeate vapor pressure
Permeate saturated vapor
Feed saturated vapor
Mass flow rate
Work
Total energy consumption
Energy efficiency
Exergy efficiency for feed stream
Exergy efficiency for permeate stream

Where the QT is a tridiagonal matrix this equation

solved by the standard traditional matrix solver algorithm which is often termed the Thomas algorithm
(Imdakm et al., 2007; Oosthuizen et al., 1999).
A1
D1
Bi
; c1
; ai
;
C1
A1
Ai 2 ai21 C i
Di 2 C i ci21
ci
Ai 2 C i ai21

a1

126

IDA Journal of Desalination and Water Reuse

69

2014

VOL

NO

3-4

40
59
13.4
25.4
19040.7
1598.94
82.7981
8.00932
2.82222E-05
36.81441996
360.3
10.21771301
25.87899082
7.917178774

cm2
C0
C0
kg/m2hr
Pa
Pa
m3/kg
m3/kg
kg/sec
Watt
Watt
%
%
%

Fakron

Table 2 Thermal analysis data for vacuum membrane

distillation process (Criscuoli et al., 2008)
40
59
19040.7
8.00932
1000
129.183
48.8
2484.38
389.5
118.5329231
22.61178511
18.02284908
0.0

cm2
C0
Pa
m3/kg
Pa
m3/kg
kg/m2hr
kJ/kg
Watt
Watt
%
%
%

distillation configuration has higher exergy efficiency

than the vacuum membrane distillation configurations, these results suggest that enhancement in the
permeate side will increase the overall efficiency of the
process.

References

Ma
No n
t f ey
or Pu
Di bl
st ish
rib in
ut g
io
n

Membrane area
Feed temperature
Feed vapor pressure
Feed saturated vapor
Vacuum pressure
Saturated vapor to permeate
Water flux
Specific enthalpy of vaporization
Energy consumption
Work
Energy performance
Exergy efficiency for feed stream
Exergy efficiency for permeate stream

Energy and exergy analysis and modeling temperature distribution

Conclusions

heat loss due to conduction across the membrane

N The
in direct contact membrane distillation configurations

N
N
N
N

are higher than the vacuum membrane distillation

configuration.
The temperature polarization phenomena is appears
to have greater effect in direct contact membrane
distillation configuration then vacuum membrane
distillation configuration.
The development optimum membrane materials and
optimum membrane morphology for enforcement the
flux and reduce the heat losses across the membrane.
This membrane makes the membrane distillation
process affordable as desalination process.
Different types of geometry membrane surface to
produce voracities to enforcement mass transfer in
the boundary-layer.
The exergy analysis of feed side shows that the direct
contact membrane distillation configuration has a
higher exergy efficiency than the vacuum membrane
distillation configuration and the exergy analysis for
permeate side shows the direct contact membrane

Criscuoli, A., Carnevale, M. C. and Drioli, E. 2008. Evaluation of

energy requirements in membrane distillation, Chem. Eng.
Process., 47, (7), pp. 10981105.
Ding, Z., Liu, L., Yu, J., Ma, R. and Yang, Z. 2008. Concentrating the
extract of traditional Chinese medicine by direct contact
membrane distillation, J. Membr. Sci., 310, pp. 539549.
Drioli, E. 2005. Membrane distillation and related operations a
review, J. Separation Pur. Rev., 34, pp. 3586.
Imdakm, A. O., Khayet, M. and Matsuura, T. 2007. A Monte Carlo
simulation model for vacuum membrane distillation process,
J. Membr. Sci., 306, (12), pp. 341348.
Imdakm, A. O. and Matsuura, T. 2004. A Monte Carlo simulation
model for membrane distillation processes: direct contact (MD),
J. Membr. Sci., 237, (12), pp. 5159.
Khayet, M., Velazquez, A. and Mengual, J. I. 2004. Modelling mass
transport through a porous partition: effect of pore size
distribution, J. Non Equilibr. Thermodyn., 29, (3), pp. 279299.
Lawson, K. W. and Lloyd, D. R. 1996. Membrane distillation I module
design and performance evaluation using vacuum membrane
distillation, J. Membr. Sci., 120, pp. 111121.
Lawson, K. W. and Lloyd, D. R. 1997. Membrane distillation,
J. Membr. Sci., 124, pp. 125.
Oosthuizen, P. H. and Naylor, D. 1999. Introduction to convective heat
transfer analysis, McGraw-Hill.
Phattaranawik, J., Jiraratananon, R. and Fane, A. G. 2003. Heat
transport and membrane distillation coefficients in direct
contact membrane distillation, J. Membr. Sci., 212, (12), pp.
177193.
Ratkovich, N., Chan, C. C. V., Berube, P. R. and Nopens, I. 2009.
Experimental study and CFD modelling of a two-phase slug
flow for an airlift tubular membrane, Chem. Eng. Sci., 64,
pp. 357635845.
Schofield, R. W., Fane, A. G. and Fell, C. J. D. 1987. Heat and masstransfer in membrane distillation, J. Membr. Sci., 33, (3),
pp. 299313.

IDA Journal of Desalination and Water Reuse

2014

VOL

NO

3-4

127