chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 316322

Contents lists available at ScienceDirect

Chemical Engineering Research and Design

journal homepage: www.elsevier.com/locate/cherd

Improvement of the Membrane Distillation

performance through the integration of different
congurations

A. Criscuoli
Institute on Membrane Technology (ITM-CNR), via P. Bucci 17/C, Rende (CS) 87030, Italy

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

Article history: In this work, the integration of two Membrane Distillation (MD) congurations was investi-
Received 25 March 2016 gated as possible means to reduce the specic thermal energy consumption. Tests of Direct
Received in revised form 12 May Contact Membrane Distillation (DCMD) and Air Gap Membrane Distillation (AGMD) were
2016 carried out on 40 cm2 lab-scale modules equipped with a commercial at polypropylene
Accepted 20 May 2016 membrane of 0.2 m, by sending distilled water as feed. The performance of integrated
Available online 27 May 2016 schemes with the feed exiting from the DCMD module sent as coolant stream in the AGMD
module where it is heated by the permeating vapor, before being recycled back to the DCMD
Keywords: unit was analyzed. When compared to the single DCMD units, the integrated DCMD-AGMD
Membrane Distillation systems led to lower specic thermal energy consumption, as well as higher Gained Output
Integrated congurations Ratio (GOR) and permeate production.
Specic thermal energy 2016 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
consumption

1. Introduction progress is addressed to the use of renewable energy, like

In Membrane Distillation volatile species and water evap-
orate from an aqueous stream (feed) through microporous
hydrophobic membranes and are collected, after condensa-
tion, at the permeate side. In this way, it is possible to recover
puried water from a variety of polluted streams. Many are
the elds where MD can nd application, like the treatment
of water and wastewaters, the seawater and brackish water
desalination, the production of ultrapure water for semicon-
ductor industry and, more recently, the dehydration of solid
particles (El-Bourawi et al., 2006; Alkhudhiri et al., 2012; Wang
and Chung, 2015; Drioli et al., 2014; Criscuoli et al., 2016).
Despite the efforts made in last years for the production of
membranes and modules specically designed for MD, its
implementation at industrial scale is still limited. Main crit-
ical points are the lack of long-term investigations that are
essential for ensuring a good performance in time, and the
high specic thermal energy consumption of the process that
affects the cost. Concerning this last issue, the research in
the solar one (Wang et al., 2009; Koschikowski et al., 2009;
Vega-Beltran et al., 2010; Mericq et al., 2011; Saffarini et al.,
2012; Cipollina et al., 2012; Surez et al., 2015), and to the
enhancement of the internal heat recovery. Among the MD
congurations, Direct Contact Membrane Distillation and Air
Gap Membrane Distillation are those more investigated. In
particular, DCMD is the most studied conguration at lab scale,
due to its simplicity, while AGMD is the preferred congu-
ration for the design and production of modules to be used
at pilot scale, mainly because of the possibility to make the
internal heat recovery inside the module-self. Specically, in
DCMD one side of the membrane is in contact with the aque-
ous feed (hot) to be treated while the other side is in contact
with the distillate (cold). Both liquid streams cannot pene-
trate inside the membrane pores, thanks to the hydrophobic
character of the membrane and, then, a difference of vapor
pressure is established across the membrane, due to the tem-
perature difference of the two streams. The distillation occurs
by evaporation of the water/volatile species at the feed side,

Tel.: +39 0984 492118; fax: +39 0984 402103.
E-mail address: a.criscuoli@itm.cnr.it
http://dx.doi.org/10.1016/j.cherd.2016.05.020
0263-8762/ 2016 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 316322 317

a) a) 25C
Cooling
Evaporation Condensation

Migration Membrane Exchange

20C 30C surface

Feed Distillate

DCMD Heat
Unit Exchanger
Condensing
surface
b)
85C
Evaporation
90C 80C
Condensation
Migration

Feed C o l d Heating
stream

Distillate b) Feed Heating

T= 5-10C
Fig. 1 Scheme of the distillation process in DCMD (a) and
AGMD (b).

the migration of water vapor/volatile molecules through the

dry-pores and the condensation of the vapor/volatiles in the
cold distillate (Fig. 1a). Therefore, in the membrane module the
feed stream is cooled down while the distillate is heated up.
In AGMD, the aqueous feed (hot) is still in contact with one
membrane surface while at the other side an air gap is cre-
ated between the membrane and a condensing surface. The
condensing surface is in contact with a cold stream. In this
conguration, the water vapor/volatile species move from the
feed side through the dry-pores, the air gap and, nally, con-
dense on the condensing surface (Fig. 1b). Inside the module,
the feed stream is cooled down while the cooling stream is
heated up due to vapor/volatiles condensation. In both con-
gurations there is, then, the need to supply heat to the feed
and to remove heat from the cold stream (the distillate and
the cooling stream for DCMD and AGMD, respectively).
Depending on the specic MD conguration, the internal
heat recovery can be obtained in different ways:
- in DCMD, by coupling the membrane module to a heat
exchanger where the feed and distillate streams are sent for
a pre-heating and pre-cooling, respectively. The two streams
are, then, heated and cooled to the desired temperatures
in other heat exchangers, before being recycled back to the
membrane module (see Fig. 2a);
- in AGMD, by using the feed as both cold and hot stream
(Winter et al., 2011; Guillen-Burrieza et al., 2011; Jansen et al.,
2013). In this case an external heat source is needed to
provide the necessary difference of temperature across the
membrane (see Fig. 2b).
Recently, a Vacuum Multi-Effect Membrane Distillation
(VMEMD) unit has been designed, where the AGMD occurs in
different stages, each one working at lower temperature and
pressure than the previous one (Heinzl et al., 2012; Zhao et al.,
2013; Memsys, 2016). The system is equipped with a steam
raiser/heating unit and a condenser.
All mentioned internal heat recovery systems need to work
at high feed temperatures (5090 C) in order to obtain accept-
able uxes/heat recoveries, and they cannot be efciently used
at lower feed temperatures (e.g., 40 C). The aim of this work
Membrane Condensing
surface
Feed OUT Feed IN
(hot stream) (cold stream)
Distillate
Fig. 2 Internal heat recovery in DCMD (a) and AGMD (b).
was to evaluate the potential of the integration of DCMD and
AGMD as an alternative means to reduce the specic thermal
energy consumption of DCMD. The idea behind was to heat the
feed stream exiting the DCMD module by using it as a coolant
stream in an AGMD unit.
2. Materials and methods
2.1. Membrane and membrane modules
DCMD and AGMD modules were equipped with a 0.2 m
at polypropylene membrane (Membrana, Germany) having a
thickness of 91 m and 70% of average porosity. In both cases
the membrane area was of 40 cm2 (4 10), a typical lab-module
size. In the AGMD module, the condensing foil (30 m thick)
was in polypropylene and the air gap was of 3 mm.
2.2. Experimental set-up and procedure
Modules and all lines of the experimental set-up were well
insulated to minimize the heat loss toward the environment.
The temperatures of the involved streams were read at the
module inlets and outlets by four thermocouples, whose
correct functioning was previously conrmed by comparing
their measures to those of a galinstan thermometer (res-
olution 0.1 C). The DCMD distillate ux was calculated by
318 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 316322

registering the increase of the distillate tank weight, and by

diving the measured mass by the operating time and the
membrane area. For the AGMD, the distillate was collected
from the module-self at the end of the test, weighted, and
the ux calculated as for DCMD. Distilled water was used
for the feed, the distillate and the coolant streams. In both
modules, the involved streams (the feed and the distillate
for the DCMD; the feed and the coolant for the AGMD)
were sent in counter-current ow mode. Experiments were
carried out by sending the feed at different temperatures
(40 C50 C60 C). For the DCMD conguration, different
feed ow rates (30 L/h50 L/h100 L/h) were also investigated,
whereas the operative feed ow rate for the AGMD module
was xed at 100 L/h. Each test was repeated three times and
a variation in ux values lower than 3% was registered.
2.3. GOR calculation
A way often used to measure the thermal efciency of Mem-
brane Distillation is to calculate the Gain Output Ratio of
the process. This indicator gives, in fact, information on how
much of the thermal energy supplied to the system is usefully
employed to produce the permeate. The GOR of an MD plant
can be calculated by the formula:
(Qp Hv )
GOR = (1)
QH
where Qp is the permeate ow rate (g/h), Hv the enthalpy of
vaporization (J/g), and QH is the thermal energy requirement
of the MD plant (J/h).
The thermal energy requirement of the MD plant was cal-
culated as the sum of the single thermal energies supplied to
the involved streams, each one obtained by:
QHi = Qfi cpi Ti (2)
where QHi is the thermal energy requirement of stream i
(J/h), Qfi the ow rate of stream i (g/h), cpi the specic heat
capacity of stream i (J/g K), and Ti is the difference of tem-
perature of stream i inside the plant (K).
GOR higher than 1 should be achieved in order to well
utilize the heat supplied. Depending on the membrane and
module properties, on the operating conditions used and on
the heat recovery systems adopted, GOR in MD can signi-
cantly vary. Experiments carried out at lab scale lead often to
GOR lower than 1 (Summers et al., 2012).
Fig. 3 DCMD tests: effect of the feed ow rate and
temperature on the trans-membrane ux.
of the reduced heat transfer resistance, due to the reduction
of the boundary layer at the feed side. This effect is, however,
less pronounced than that of the temperature and reduces at
higher feed ow rates, indicating a decreasing control of the
uidynamic on the ux.
The variation of the feed temperature along the module
(Tf = Tf inlet Tf outlet ) as function of the feed temperature and
ow rate is reported in Fig. 4. It can be noticed that higher feed
temperatures lead to higher Tf . This result can be linked to
the higher evaporative uxes achievable at high temperatures,
that more signicantly cool down the feed during the DCMD
test. Concerning the effect of the feed ow rate, low values
lead to a more evident Tf , because of the higher residence
time of the feed inside the module. This is especially valid
for the small membrane area of the module used during the
tests, in which low feed ow rates are needed to appreciate the
difference of temperature along the module, with the highest
Tf obtained at 30 L/h.
This feed temperature decay during DCMD is the main
responsible of the thermal energy consumption of the pro-
cess. In this work, a new way for heating the feed stream is
proposed: the coupling of the DCMD unit with an AGMD mod-
ule, by sending the feed stream exiting the DCMD module as
a coolant stream in the AGMD one (see Fig. 5). To apply the
concept is essential to verify that the temperature increase of
the coolant stream (Tc ) in the AGMD module is enough to
cover the Tf registered in DCMD. As for the DCMD unit, the
membrane area of the AGMD module used was quite small
and, therefore, low coolant ow rates should be preferred to

3. Results and discussion

In this section, the results obtained with the single DCMD and
AGMD units are rst reported and discussed. Then, the results
of the integrated DCMD-AGMD system are presented and its
performance analyzed for two DCMD feed ow rates.

3.1. DCMD experiments

DCMD tests were carried out at different feed ow rates and

temperatures, by xing the distillate ow rate and tempera-
ture at 100 L/h and 18 C, respectively. Fig. 3 shows the effect
of both parameters on the trans-membrane ux. As expected,
the ux increases with the feed temperature, because of the Fig. 4 DCMD tests: variation of the feed temperature along
higher water vapor pressure and, therefore, the higher driving the module as function of the feed temperature and ow
force. Similarly, it also increases with the ow rate, because rate.
 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 316322
Qf DCMD=Qc, 30 L/h
?
Tf DCMD = Tc AGMD Tf AGMD
Qf, 100 L/h
Fig. 5 The cooling of the feed (Tf AGMD ) and the heating of
the coolant stream (Tc ) in an AGMD module.
work with high residence time and, then, to achieve high Tc .
Based on this consideration, AGMD tests were carried out by
sending the coolant stream at 30 L/h. The obtained results are
presented in the following section.
3.2. AGMD experiments
Tests were carried out by sending the feed (100 L/h) at different
temperatures, while the coolant stream ow rate and temper-
ature were xed at 30 L/h and 18 C, respectively. Fig. 6 shows
an increase of the trans-membrane ux with the feed temper-
ature. If compared with the DCMD performance (also reported
in the Figure), it can be noticed that the AGMD module leads
to lower uxes, due to the higher transfer resistance at the
permeate side.
The effect of the feed temperature on the Tc is reported in
Fig. 7. Moving from 40 C to 60 C the coolant stream becomes
warmer, because of the higher uxes and, then, the higher
heat transferred by vapor condensation. In particular, the
obtained Tc ranged from 0.6 to 1.6 C.
3.3. The DCMDAGMD system
If we look back at Fig. 4, the lowest Tf during DCMD tests were
obtained at a feed temperature of 40 C, varying from 1.3 to
0.6 C for feed ow rates of 30 and 100 L/h, respectively. Then,
by taking into account that the highest Tc in the AGMD mod-
ule was achieved for a feed temperature of 60 C, a rst test
by sending in the AGMD unit the feed stream at 100 L/h and
Fig. 6 AGMD and DCMD tests: effect of the feed
temperature on the trans-membrane ux (AGMD: Qf ,
100 L/h; Qc , 30 L/h; Tc , 18 C; DCMD: Qf , 100 L/h; Qd , 100 L/h;
Td , 18 C).
319
Fig. 7 AGMD tests: heating of the coolant stream as
function of the feed temperature.
Qf DCMD=Qc, 30 L/h
38.7C
AGMD
Tc, 0.5C 2 Tf , 0.3C
40 cm
Qf, 100 L/h
60C
Fig. 8 The heating of the exiting DCMD feed in a single
AGMD module.
60 C, while the coolant one (the exiting DCMD feed stream)
owed counter-currently at 30 L/h and 38.7 C, was carried out
(see Fig. 8). The aim was to check the ability of the AGMD unit
to give back to the DCMD feed stream the degrees lost dur-
ing the DCMD process (1.3 C, for the selected conditions). As
reported in Fig. 8, the Tc was, in this case, quite lower than
the desired value (0.5 C vs 1.3 C). Therefore, the behavior of
an AGMD module of bigger membrane area was analyzed by
performing on the 40 cm2 lab module different tests in series:
the feed and coolant streams exiting from the module (test 1)
were sent as inlet streams (test 2) and so on, until the desired
Tc (1.3 C) was obtained (see Fig. 9).
From these experiments, it resulted that it was possible to
reach the target with an AGMD module of 104 cm2 membrane
area (4 26). The corresponding Tf and average ux were
0.78 C and 6 kg/hm2 , respectively. Based on these results, a
comparison between the performance of the DCMD unit alone
and of the DCMD unit integrated with the AGMD one was car-
ried out in terms of specic thermal energy consumption, GOR
and productivity. In particular, the calculations were made for
the lowest and highest DCMD feed ow rate values investi-
gated, 30 and 100 kg/h, sent at 40 C. In both case studies the
coolant and the feed stream were sent to the AGMD units at
30 kg/h and 100 kg/h, respectively. The analyzed ow sheets
with the information on the involved streams are reported in
Figs. 10 and 11 and the corresponding results of the compari-
son are summarized in Tables 1 and 2, respectively. It has to be
noticed that for the 100 kg/h DCMD feed ow rate, only 90 kg/h
were heated by means of AGMD units, while the remaining
10 kg/h were sent to an heat exchanger. In particular, based on
320 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 316322

Fig. 9 The heating of the exiting DCMD feed in a series of AGMD modules.

QdOUT
QdOUT
D
D 40 cm2
40 cm2 C
C 30 kg/h
30 kg/h M M
40C
40C D
D
29977.8 g/h 62.4 g/h 62.4 g/h
29977.8 g/h QdIN 59.22C 25C
QdIN 38.7C
38.7C
99937.6g/h
A QEXT1
Q1 30 kg/h 22.2 g/h QEXT 22.2 g/h 59.22C
38.7C 38.7C 25C G 104 cm2
M
D 100 kg/h
59.22C
22.2 g/h 22.2 g/h 29977.8 g/h 100 kg/h
25C 40 C 40C 60C

QEXT2 Q1

Fig. 10 DCMD and DCMDAGMD ow sheets for a DCMD feed ow rate of 30 kg/h sent at 40 C.

QdOUT
QdOUT
D
D 40 cm2
40 cm2 C
C
100 kg/h M
100 kg/h M 40C
40C D
D 32.4 g/h 9967.6 g/h
99967.6 g/h QdIN 99967.6 g/h
40C 40 C
QdIN 39.4C 39.4C
9967.6 g/h 30 kg/h 30 kg/h 30 kg/h
32.4 g/h QEXT1
Q1 100 kg/h 32.4 g/h QEXT 32.4 g/h Q1 39.4C 39.4C 39.4C 39.4C
39.4C 39.4C 25C 25C
A A A
G G G
90 kg/h M M M
40C D 48 cm2 D 48 cm2 D 48 cm2

99913.6g/h
58.92C
100 kg/h 100 kg/h
60C 58.92C
86.4 g/h
Q2 58.92C
QEXT2

86.4 g/h
25C

Fig. 11 DCMD and DCMDAGMD ow sheets for a DCMD feed ow rate of 100 kg/h sent at 40 C.

the experimental results obtained by the tests in series, three in the single AGMD unit equal to 0.36 C, the overall Tf was
AGMD modules of 48 cm2 membrane area (4 12) each were of 1.08 C and, therefore, there was not a signicant varia-
needed to give back to the stream the 0.6 C lost in the DCMD tion in the average AGMD ux with respect to the other case
module (see Fig. 11). Furthermore, being the experimental Tf study.
 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 316322
Table 1 Performance comparison between the single DCMD unit and the integrated DCMDAGMD system. Qf DCMD ,
30 kg/h, Tf , 40 C.
Conguration Membrane area (cm2 ) Permeate production (g/h)
DCMD 40 22.2
DCMDAGMD 144 84.6
Table 2 Performance comparison between the single DCMD unit and the integrated DCMD-AGMD system. Qf DCMD ,
100 kg/h, Tf , 40 C.
Conguration Membrane area (cm2 ) Permeate production (g/h)
DCMD 40 32.4
DCMDAGMD 184 118.8
Based on the obtained values, a rst comment is that the
low GORs achieved in both single and integrated ow sheets
are due to the low membrane areas considered in the analysis.
It is, in fact, well known that GOR increases with the mem-
brane area (Cipollina et al., 2012; Summers et al., 2012; Guan
et al., 2015) and, therefore, higher values are expected when
moving from lab to pilot scale.
The integrated DCMD-AGMD system is more complex than
the single DCMD system and needs higher costs for the AGMD
units, pumps and heat exchangers. Nevertheless, for both case
studies, it leads to a reduction of the specic thermal energy
consumption (calculated as the ratio between the thermal
energy needed in the plant and the permeate produced) of
around 40%, an increase of GOR and an about 4-fold increase
of the permeate production (plant productivity) with respect
to the DCMD unit alone. The increase of the permeate produc-
tion corresponds to the increase of the overall membrane area
for the 30 kg/h DCMD feed ow rate, due to the similar average
uxes of the DCMD and AGMD modules. A different situation
is observed for the 100 kg/h DCMD feed ow rate, for which a
4-fold increase of the permeate production is obtained with a
5-fold increase of the overall membrane surface. This result
depends on the higher ux of the DCMD unit with respect to
the AGMD ones (around 8 kg/hm2 and 6 kg/hm2 , respectively).
However, in both cases, if compared to the integration of the
DCMD unit with an heat exchanger, the integrated MD system
is able to couple the reduction of the specic thermal energy
consumption to an increase of the overall permeate produc-
tion. Another advantage to mention is that by coupling the
DCMD to the AGMD unit, it is possible not only to treat the
same feed at two different temperatures, but also two types of
feed, each one at the needed operating temperature. Finally,
the DCMD-AGMD system is able to efciently treat feeds at low
temperatures. All these positive ndings make the proposed
integrated MD system highly exible and able to well respond
to the process intensication strategy. The analysis did not
include the electric energy needed for pumping the involved
streams, because Membrane Distillation works at atmospheric
pressure and pumps are requested to just overcome the pres-
sure drops of the plant. Moreover, it is well known that the
main contribution to the overall energy consumption in MD is
the thermal one.
4. Conclusions
It has been demonstrated that the coupling of DCMD and
AGMD can be an option to consider to improve the perfor-
mance of DCMD. The integrated DCMD-AGMD, in fact, allows
321
Specic thermal energy consumption (W/g/h) GOR
2.06 0.31
1.21 0.51
Specic thermal energy consumption (W/g/h) GOR
2.17 0.29
1.25 0.49
to reduce the specic thermal energy consumption while
increasing the GOR and the permeate production, with respect
to the DCMD unit alone. Moreover, the integrated MD system
can be efciently applied to low feed temperatures (e.g., 40 C),
while existing internal heat recovery systems need to work at
high feed temperature values (5090 C) to obtain acceptable
uxes/heat recoveries. By coupling the DCMD to the AGMD
unit, it is also possible to simultaneously treat two differ-
ent feeds, each one at the needed operating temperature.
Although in the performed study the operating temperatures
were 40 C and 60 C for the DCMD and the AGMD unit, respec-
tively, the integrated system can be operated also at higher
temperatures (e.g., 50 C60 C for the DCMD and 70 C80 C
for the AGMD), covering a broad range of applications.
References
Alkhudhiri, A., Darwish, N., Hilal, N., 2012. Membrane distillation:
a comprehensive review. Desalination 287, 218.
Cipollina, A., Di Sparti, M.G., Tamburini, A., Micale, G., 2012.
Development of a Membrane Distillation module for solar
energy seawater desalination. Chem. Eng. Res. Des. 90,
21012121.
Criscuoli, A., Carnevale, M.C., Drioli, E., 2016. Study of the
performance of a membrane-based vacuum drying process.
Sep. Purif. Techn. 158, 259265.
Drioli, E., Carnevale, M.C., Figoli, A., Criscuoli, A., 2014. Vacuum
Membrane Dryer (VMDr) for the recovery of solid
microparticles from aqueous solutions. J. Membr. Sci. 472,
6776.
El-Bourawi, M.S., Ding, Z., Ma, R., Khayet, M., 2006. A framework
for better understanding membrane distillation separation
process. J. Membr. Sci. 285, 429.
Guan, G., Yang, X., Wang, R., Fane, A.G., 2015. Evaluation of heat
utilization in membrane distillation desalination system
integrated with heat recovery. Desalination 366, 8093.
Guillen-Burrieza, E., Blanco, J., Zaragoza, G., Alarcon, D.,
Palenzuela, P., Ibarra, M., Gernjak, W., 2011. Experimental
analysis of an air gap membrane distillation solar
desalination pilot system. J. Membr. Sci. 379, 386396.
Heinzl, W., Bttner, S., Lange, G., 2012. Industrialized modules for
MED desalination with polymer surfaces. Desal. Water Treat.
42, 177180.
Jansen, A.E., Assink, J.W., Hanemaaijer, J.H., van Medevoort, J.,
van Sonsbeek, E., 2013. Development and pilot testing of
full-scale membrane distillation modules for deployment of
waste heat. Desalination 323, 5565.
Koschikowski, J., Wieghaus, M., Rommel, M., Subiela Ortin, V.,
Penate Suarez, B., Betancort Rodriguez, J.R., 2009.
Experimental investigations on solar driven stand alone
membrane distillation systems for remote areas. Desalination
248, 125131.
322 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 316322
www.memsys.eu, Accessed on March 2016.
Mericq, J.P., Laborie, S., Cabassud, C., 2011. Evaluation of systems
coupling vacuum membrane distillation and solar energy for
seawater desalination. Chem. Eng. J. 166, 596606.
Saffarini, R.B., Summers, E.K., Arafat, H.A., Leinhard, J.H.V., 2012.
Technical evaluation of stand-alone solar powered membrane
distillation systems. Desalination 286, 332341.
Surez, F., Ruskowitz, J.A., Tyler, S.W., Childress, A.E., 2015.
Renewable water: direct contact membrane distillation
coupled with solar ponds. Appl. Energy 158, 532539.
Summers, E.K., Arafat, H.A., Lienhard, J.H.V., 2012. Energy
efciency in comparison of single-stage membrane
distillation (MD) desalination cycles in different
congurations. Desalination 290, 5466.
Vega-Beltran, J.C., Garca-Rodrguez, L., Martin-Mateos, I.,
Blanco-Glvez, J., 2010. Solar membrane distillation:
theoretical assessment of multi-stage concept. Desal. Water
Treat. 18, 133138.
Wang, P., Chung, T.-S., 2015. Recent advances in membrane
distillation processes: membrane development, conguration
design and application exploring. J. Membr. Sci. 474,
3956.
Wang, X., Zhang, L., Yang, H., Chen, H., 2009. Feasibility research
of potable water production via solar-heated hollow ber
membrane distillation system. Desalination 247, 403411.
Winter, D., Koschikowski, J., Wieghaus, M., 2011. Desalination
using membrane distillation: experimental studies on full
scale spiral wound modules. J. Membr. Sci. 375, 104112.
Zhao, K., Heinzl, W., Wenzel, M., Buttner, S., Bollen, F., Lange, G.,
Heinzl, S., Sarda, N., 2013. Experimental study of the memsys
vacuum-multi-effect-membrane-distillation (V-MEMD)
module. Desalination 323, 150160.