food and bioproducts processing 9 4 ( 2 0 1 5 ) 180–186

Contents lists available at ScienceDirect

Food and Bioproducts Processing

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

Beer dealcoholization using non-porous

membrane distillation

M. Purwasasmita, D. Kurnia, F.C. Mandias, Khoiruddin, I.G. Wenten ∗

Department of Chemical Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia

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

Article history: In this present work, performance of non-porous membrane distillation method is inves-
Received 11 June 2014 tigated during beer dealcoholization process. A thin-film composite polyamide membrane
Received in revised form 6 is used as membrane module while a commercial beer product with 5%-volume ethanol
December 2014 is used as feed. The results indicate that non-porous membrane distillation can be used
Accepted 2 March 2015 to dealcoholize beer without losing the other nutrients and flavoring components such as
Available online 6 March 2015 maltose and glycerol. The increase of feed pressure and vacuum pressure can improve mem-
brane flux due to higher permeability. However, membrane selectivity is decreased with the
Keywords: increase of vacuum pressure. The membrane flux and ethanol concentration in perme-
Dealcoholization ate are 0.15–0.76 L/m2 h and 3.66–4.64%-vol., respectively. Meanwhile, there are no specific
Beer sequences on the maltose concentration in the effect of operating conditions. The slight
Non-porous membrane loss of maltose in the dealcoholized beer can be attributed to adsorption phenomena in
Membrane distillation membrane surface thus membrane flushing may be conducted to recover it. The glycerol
Non-alcoholic beer behavior in dealcoholization process is similar to maltose. Some glycerol compounds are
Polyamide found in the permeate stream but all of them are less than 0.005%-vol. At 3000 mbar feed
pressure and 580 mbar vacuum pressure, the flux of membrane is 0.69 L/m2 h with 3.70%-vol.
and 4.60%-vol. of ethanol concentration in dealcoholized beer and permeates side respec-
tively, no maltose and only 0.001%-vol. glycerol in permeate side. A long run operation for
beer dealcoholization using these operating conditions can reduce the alcohol content from
5%-vol. to 2.45%-vol. in 6 h.
© 2015 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction Non-alcoholic beers can be produced by interrupting the

The market of non-alcoholic beer keeps increasing in the last
few years because there are regulations concerning health and
religion issues (Catarino and Mendes, 2011a,b). Non-alcoholic
beer is a beer with very low (<0.1%) or no alcohol content
(Perpète and Collin, 2000). In most of the EU countries beers
with low alcohol content are divided into alcohol free beers
(AFB) containing ≤0.5% alcohol by volume (ABV), and to low-
alcoholic beers (LAB) with no more than 1.2% ABV (Montanari
et al., 2009). AFB and LAB have been produced and distributed
widely in America and some European countries. Both AFB
and LAB are valued with higher price than normal beers.
fermentation process of beers (Kunze, 1999). Another alter-
native process for low-alcoholic beverages is ethanol removal
from a completely fermented beverage. The alcoholic contents
removed from the beer can be used for various uses and
not considered as waste. One of the usage for the alcoholic
contents is for bioethanol fuel. However, most of these meth-
ods produce beers which lack nutrient and flavoring content
compared to normal beers.
In order to dealcoholize alcoholic beverages without reduc-
ing the aroma and flavor contents, researchers consider the
usage of membrane methods which are permeable only for
alcohol such as osmotic membrane distillation (Varavuth

∗
Corresponding author. Tel.: +62 818620014; fax: +62 22 2511404.
E-mail address: igw@che.itb.ac.id (I.G. Wenten).
http://dx.doi.org/10.1016/j.fbp.2015.03.001
0960-3085/© 2015 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
 food and bioproducts processing 9 4 ( 2 0 1 5 ) 180–186
et al., 2009; Gostoli, 1999; Liguori et al., 2013a,b), dialysis mem-
brane (Branyik et al., 2012), pervaporation membrane (Tan
et al., 2005; Takacs et al., 2007; Catarino and Mendes, 2011a,b;
del Olmo et al., 2014), nanofiltration membrane (Labanda et al.,
2009; Catarino and Mendes, 2011a,b), reverse osmosis (Branyik
et al., 2012; Catarino et al., 2006; Gil et al., 2013; Pilipovik and
Riverol, 2005), and membrane contactor (Liguori et al., 2010;
Diban et al., 2008, 2013).
Membrane distillation method is a separation process
that is energy efficient compared to other methods such as
reverse osmosis or distillation. Membrane distillation has var-
ious advantages such as: 100% theoretic rejection of ions,
colloids, biological cells, and non-volatile components; low
operation temperature, low operation pressure, less chemi-
cal interactions between membrane and process fluids, less
vapor gap than distillation, and simple membrane mechanical
characteristics. Even though membrane distillation has many
advantages, this system also has many limitations, such as
the feed solution must be dilute enough to avoid membrane
wetting (Lawson and Lloyd, 1997).
Membrane distillation is one of thermally based membrane
processes in which the membrane is not directly involved in
separation and acts as barrier between two phases. Selectiv-
ity is determined by the vapor–liquid equilibrium thus the
component with the highest partial pressure has the high-
est permeation rate (Mulder, 1996). For example, in the case
of an ethanol/water mixture where the membrane is not wet-
ted, both components are transported through the membrane.
However, since the ethanol has higher vapor pressure, the per-
meation rate of ethanol is always relatively higher.
The membrane used in this study is a non-porous mem-
brane thus has similar mechanism with pervaporation. The
permeate is removed as a vapor due to low vapor pressure
existing on the permeate side by applying a vacuum pump.
Since the membrane has dense structure, membrane wetting
is not a drawback of the process. In addition the dense struc-
ture provides another advantages for beer dealcoholization
which avoid the permeation of other components in the beer.
In this study, vacuum membrane distillation method with
non-porous membrane is investigated during dealcoholiza-
tion of beer. The vacuum condition is introduced to create a
driving force for the mass transfer of ethanol which will shift
the vapor–liquid equilibrium of ethanol where the evapora-
tion of ethanol happen followed by ethanol diffusion through
the membrane. Non-porous membrane can avoid the loss
of important components from beer because of diffusion
through the membrane.
2. Materials and methods
2.1. Materials
The beer used in this research is Anker Bir from PT Delta
Djakarta Tbk. The beer main compositions are described in
Table 1. There are some peaks detected in the HPLC analysis
results which can be seen in Fig. 1. However the size area of
the peak is too small to be considered major component in
beer. Therefore, it was analyzed only two nutritious content
of the beer, maltose and glycerol. The other trace compo-
nents are not included in the table because it is undetected
in the HPLC analysis, thus they can be ignored. A commer-
cial spiral wound non-porous membrane, TW30-1812-75 from
DOW Filmtec (Dow Chemical Company) made from polyamide
181
Table 1 – Anker Bir main compositions.
Compounds Concentration Retention time (min)
(%-vol.) (g/L)
Ethanol 5.00 12.06 21.965
Maltose 1.01 9.43 8.932
Glycerol 0.02 0.23 13.659
Water 93.97 936.9 19.426
thin-film composite with 0.37 m2 surface area is used. The
maximum operation temperature and pressure for this mem-
brane is 45 ◦ C and 10 bar, respectively.
2.2. Methods
The membrane distillation system is described in Fig. 2. This
system is operated in batch mode operation. The beer feed
(3.1 L) is contained in a feed tank which is circulated by a
pumping system into the membrane module. The stream is
split into permeate stream that permeates through the mem-
brane and contains mostly alcohol and water and the retentate
stream that contains dealcoholized beer with reduced alco-
hol contents. The retentate stream is then recirculated into
the feed tank and the permeate stream is then flowed onto a
chiller by the help of vacuum pumping system. The alcohol
contents are condensed in the chiller system.
Before the experiment is started, it is necessary to wash the
membrane distillation system with demineralized water or
fresh tap water by pouring the water in the feed tank and tur-
ning on the circulation pump. The washing is done for about
20–30 min and then the membrane distillation system is dried
until all of the water pours out.
The process is conducted by varying operation conditions
both feed pressure in the feed side and vacuum pressure in
the permeate side. In the feed side, the feed tank is at room
temperature. Beer is pumped into the membrane module by
varying the pressure from 2 to 3 bar (gauge). Meanwhile, in
the permeate side, permeate stream is at vacuum condition
by varying vacuum pressure from 490 to 660 mbar. The vapor
outlet is then condensed in a chiller system (4 ◦ C). Because
water is also permeated through the membrane, the other
components of beer will increase in concentration. By assum-
ing no other beer component but ethanol permeates through
the membrane, we must add some makeup water to the feed
tank until the initial beer volume is reached in order to main-
tain the concentration of other components in dealcoholized
beer.
In order to identify and analyze the concentrations of var-
ious components in both permeate and retentate stream, a
high performance liquid chromatography (HPLC) method is
used with HPLC Waters system and Aminex HPX-87H col-
umn. The eluent used for HPLC analytical method is sulfuric
acid (H2 SO4 ) with 0.6 mL/min flow rate. The operation pres-
sure for HPLC is 988 psi, and the internal and external (heater)
HPLC temperature are 40 ◦ C and 60 ◦ C respectively. Quantita-
tive data were obtained by comparing the peak areas of the
compounds with those of standards of known concentrations.
The experimental samples (both permeate streams and reten-
tate streams) were collected for about 10 mL on the end of each
run, stored in a refrigerator with cool condition (±4 ◦ C) and
were analyzed using HPLC method.
182 food and bioproducts processing 9 4 ( 2 0 1 5 ) 180–186

Fig. 1 – Peaks detected in HPLC analysis of fresh Anker Bir.

3. Results and discussion
3.1. Results
All of the experimental results are summarized in Table 2. It is
shown from Table 2 that ethanol concentrations are reduced
in all experiment compared to their initial concentration in
fresh beer (see Table 1). On the other wise, ethanol are found
in permeate stream with higher concentration than those
in retentate. Meanwhile, concentration of other components,
namely maltose and glycerol are slightly reduced in retentate
and very small concentration even undetected in permeate
(for maltose). The results indicate that the non-porous mem-
brane distillation is effective to remove ethanol from beer
while retains its nutrients (maltose, glycerol, and other unde-
tected sugars, organic acids, esters, etc.). The operational
condition we wanted to get is a condition which yields high
flux and low nutritious content in the permeate stream, and
also low alcohol and high nutritious content in the retentate
stream. From Table 2, it is shown that ethanol concentration
in the retentate stream mostly increases with increase vac-
uum pressure which means that the dealcoholization process
is less successful.
In Fig. 3, the permeate flux and ethanol concentration in
ethanol stream for each experimental run are shown. We can
see that the permeate flux is proportional with both feed pres-
sure and vacuum pressure.
The maltose concentration in retentate stream is shown in
Fig. 4. It is shown that there are no specific sequences on the
maltose concentration in the effect of operating conditions.
However, most of the concentration is lower compared to the
fresh beer (1.01%-vol.) and there are no maltose detected in
the permeate stream. The relationship between glycerol con-
centration (%-vol.) in the retentate stream with feed pressure
and vacuum pressure is depicted in Fig. 4. The figure shows
that the plot of glycerol concentration and maltose concen-
tration in retentate stream are the same. It is found that the
glycerol behavior in this dealcoholization process is similar
to maltose. But there are some glycerol compounds found in
the permeate stream but all of them are less than 0.005%-
vol.During the long run, the sample from the retentate stream
is taken. The concentration profile of ethanol in the retentate
stream in relationship with time is shown in Fig. 5.
3.2. Discussion
At higher vacuum level, volatile components can evaporate
and migrate through membrane easily. Therefore, it reduces
membrane selectivity. However, in the 2500 mbar feed pres-
sure condition we can see that the ethanol concentration in

Fig. 2 – Membrane distillation system diagram.

 food and bioproducts processing 9 4 ( 2 0 1 5 ) 180–186 183

Fig. 3 – Effect of feed pressure and vacuum pressure on permeate flux and ethanol concentration in permeate stream
(%-vol.).

Fig. 4 – Maltose and glycerol concentration in retentate.

Fig. 5 – Profile of ethanol concentration (%-vol.) in retentate and permeate stream.

184 food and bioproducts processing 9 4 ( 2 0 1 5 ) 180–186

retentate stream decreases with increasing vacuum pressure

0.06
0.09
0.10
0.09
0.08
0.10
0.12
0.13
0.14
0.15
0.19
0.24
0.23
0.22
which is different from the other feed pressure conditions.
P·S

–
It possibly implies that feed pressure contributes to volatile
components permeation (or diffusion) from bulk phase into
membrane phase while the evaporation effect is provided by
Selectivity, S*

vacuum pressure thus combination of both parameter should

1.06
1.02
0.95
0.80
1.09
1.12
1.16
1.16
1.16
1.20
1.22
1.24
1.12
1.02
be adjusted appropriately. Non-porous membrane distillation
–

has similar process to pervaporation. In pervaporation, there

are three steps in sequence which involves in the process:
(i) selective sorption into the membrane on the feed side, (ii)
selective diffusion through the membrane, and (iii) desorp-
tion into a vapor phase on the permeate side (Mulder, 1996).
P (L/m2 h bar)
Permeability,

The selectivity is determined by membrane characteristics

0.00
0.06
0.09
0.10
0.11
0.07
0.09
0.11
0.11
0.12
0.12
0.16
0.19
0.21
0.22
and operation conditions that suitable for components separa-
tion. It is known that vapor pressure of each components are
different that makes them separated from the others which
is controlled by the vacuum pressure. Hence, the vacuum
pressure should be adjusted properly to obtain high selectiv-
ity. It is also shown that the bigger the feed pressure, there
4.56
4.46
4.26
3.66
4.52
4.58
4.64
4.59
4.54
4.41
4.50
4.60
4.42
4.23

will be lower ethanol concentration in the retentate stream,

E

which shows that feed pressure increases the permeability

Permeate (%-vol.)

of ethanol. The lowest concentration of ethanol in retentate

stream is achieved by 3000 mbar feed pressure and 660 mbar
0.001
0.002
0.002

0.002
0.002
0.004
0.001
0.001
0.001

vacuum pressure (3.67%-vol.) which is closely followed by the

G

–
–

–
–

–
–

same feed pressure with 630 mbar vacuum pressure (3.69%-

vol.).
The increase on both feed and vacuum pressure also
increases the permeability of both ethanol and water through
M

–
–
–
–
–
–
–
–
–
–
–
–
–
–
–

the membrane thus improves the flux. On the experimen-

tal run with 2000 mbar feed pressure and 660 mbar vacuum
pressure, the permeate flux is zero because there are no fluid
on the permeate stream. It is shown that this condition is
4.29
4.39
4.50
4.60
4.15
4.09
4.00
3.97
3.93
3.67
3.69
3.70
3.93
4.16
E

insufficient for ethanol and water to permeate through the

–

membrane. Meanwhile, the highest permeate flux is achieved

Retentate (%-vol.)

in 3000 mbar feed pressure and 490 mbar vacuum pressure

(0.76 L/m2 h) since it provides the highest driving force – high-
0.018
0.018
0.020
0.022
0.017
0.019
0.022
0.023
0.024
0.010
0.016
0.022
0.019
0.015

est pressure difference – for permeation across the membrane.

G
Table 2 – Membrane dealcoholization: experimental conditions and results.

The maltose concentrations in the retentate are lower com-

pared to the fresh beer (<1%-vol.) and there are no maltose
detected in the permeate streams. The loss of maltose is
0.594
0.694
0.837
0.979
0.669
0.761
0.853
0.865
0.877
0.582
0.741
0.899
0.722
0.545

probably attributed to adsorption phenomenon of maltose on

M

membrane surface. In membrane filtration, some particles

–

could interact with the membrane by adsorption or locating

in pores which can be both surface and internal interactions
(Chen et al., 1997). Since the membrane used in this study is
Flux (L/m2 h)

a non-porous membrane, the results possibly indicate there

is interaction between maltose and membrane surface which
0.15
0.24
0.26
0.28
0.23
0.28
0.33
0.34
0.35
0.45
0.57
0.69
0.73
0.76
0

then adsorbed maltose onto membrane. Therefore, in order

to recover maltose contents in dealcoholized beer, membrane
flushing should be conducted. Generally, the increase of vac-
uum pressure does not influence the maltose concentration
Vacuum pressure

in retentate stream since there are no maltose content in per-

meate stream as indication. However, the increase of maltose
(mbar)

M, maltose; G, glycerol; E, ethanol.

concentration as shown in Fig. 4 is mostly affected by perme-

660
630
580
530
490
660
630
580
530
490
660
630
580
530
490

ation of other components. Even though glycerol is considered

as non-volatile component, some of the glycerol still can
permeate through the membrane. This phenomenon can be
S = Epermeate /Eretentate

associated to the size of glycerol molecule (MW = 92.09 g/mol)

which is slightly different than ethanol (MW = 46.07 g/mol)
Feed pressure

thus it still has the possibility to permeate through the non-

(mbar gauge)

2500 (P2.5)
2500 (P2.5)
2500 (P2.5)
2500 (P2.5)
2500 (P2.5)

porous membrane.
2000 (P2)
2000 (P2)
2000 (P2)
2000 (P2)
2000 (P2)

3000 (P3)
3000 (P3)
3000 (P3)
3000 (P3)
3000 (P3)

From all of the experimental runs, it is found that the

optimum condition is 3000 mbar feed pressure and 580 mbar
∗

vacuum pressure. This condition gives a high flux (0.69 L/m2 h)

 food and bioproducts processing 9 4 ( 2 0 1 5 ) 180–186
Fig. 6 – Comparison between short run (2.5 h) and long run (6 h) of permeate and retentate concentration (first letter:
M = maltose, G = glycerol, E = ethanol; second letter: R = retentate, P = permeate).
and also good ethanol separation which can achieve 3.70%-
vol. and 4.60%-vol. ethanol concentration in the retentate and
permeate stream, respectively. As indicated by Table 2, this
condition provides the highest permselectivity, P.S, perme-
ability times selectivity. Performance comparation of long run
(6 h) and short run (2.5 h) operation are shown in Fig. 6. From
Fig. 6 we can see that there are no significant difference on
the flux (which are around 0.69 L/m2 h) as well as for ethanol
and glycerol concentration in permeate stream. The differ-
ence of the long run and the short run is the concentration
in retentate stream, wherein all of the components concen-
tration decrease. During the long run operation, the ethanol
concentration in retentate stream is lower than those encoun-
tered in short run operation due to long permeation time. A
similar trend is also found for maltose concentration. How-
ever, concentration reduction of maltose in retentate stream
during long run operation is associated with adsorption as
previously explained.
The concentration profile of ethanol in the retentate stream
can be approached with exponential regression as follows:
CER = 0.0499 · exp(−0.119t) (1)
where CER is the ethanol in retentate stream (%-vol.) and t
is time (h). The ethanol concentration in both retentate and
permeate stream gradually decreased with time. During the
6 h operation, there is significant decrement of the ethanol
concentration in retentate stream. The ethanol concentration
in the beer after 6 h run reaches 2.45%-vol. In the permeate
stream, the ethanol concentration does not decrease signifi-
cantly. The ethanol concentration in permeate stream is down
to 4.52%-vol. Ethanol content in the beer decreases within
time as well as the ethanol permeation which explains the
decrement of ethanol concentration in the permeate.
The results of some similar researches focusing on the
dealcoholization process of alcoholic beverages, such as wine
and beer are as follows. Ethanol content in wine was reduced
from 12%-vol. to 5%-vol. using nanofiltration and reverse
osmosis membrane (Catarino and Mendes, 2011a). Aroma of
the wine changes, but aroma compound recovery was done
using pervaporation membrane (Catarino and Mendes, 2011a).
However, high pressure is required for nanofiltration and
185
reverse osmosis – high energy consumption. Dealcoholization
of beer was done using pervaporation membrane and spinning
cone column distillation to obtain ethanol content less than
0.5%-vol, then the aroma of the beer was recovered to obtain
dealcoholized beer with good flavor profile (Catarino and
Mendes, 2011b). This combination process could be complex
since involving two different processes. Osmotic distillation
technique was used to reduce ethanol content in wine from
13%-vol. to 8.71%-vol. and 13.2%-vol. to 8.1%-vol. (Varavuth
et al., 2009). Aroma components were significantly lost dur-
ing the operation of osmotic distillation. The lost of aroma
components could be attributed to porous membrane used.
Polypropylene hollow fiber membrane contactor (Liqui-Cel,
Extra-Flow 4×28, Celgard X50) was used for dealcoholiza-
tion of Aglianico wine from 12.8%-vol. to less than 0.5%-vol.
ethanol content through five cycles of process (Liguori et al.,
2010). There are major changes in the beer aroma (Liguori
et al., 2010). Similar process using polypropylene (PP) hollow
fiber membrane contactor (Liqui-Cel® Extra-Flow 2.5 in. × 8
in., 1.4 m2 , Celgard) was used for reducing ethanol content
in wine (10–13%-vol. ethanol content) to 2%-vol. and no sig-
nificant changes of beer aroma occurred (Diban et al., 2008).
Two red wines with 14.8%-vol. and 16.2%-vol. ethanol content
were partially dealcoholized by reverse osmosis, reducing the
ethanol content as much as 1%-vol. and 2%-vol. (Gil et al.,
2013). No statistically significant differences were found in
pH, color intensity, total phenolic index, proanthocyanidin
concentration (Gil et al., 2013). The results indicate that mem-
brane distillation has some advantages compared to some of
other technologies since there are almost no nutrients and
flavor components of the beer permeated through the mem-
brane. However, in the range of operation condition during this
study, the permeate flux is still low. Thus, further development
for dealcoholization using non-porous membrane distillation
is needed to improve membrane permeability and selectivity.
4. Conclusions
It was demonstrated that non-porous membrane distillation
can be used to dealcoholize beer without losing the other
nutritive and flavoring components in beer such as maltose
and glycerol. The results indicate that non-porous membrane
186 food and bioproducts processing 9 4 ( 2 0 1 5 ) 180–186
distillation can be used to dealcoholize beer without losing
the other nutrients and flavoring components such as maltose
and glycerol. The increase of feed pressure and vacuum pres-
sure can improve membrane flux due to higher permeability.
However, membrane selectivity is decreased with the increase
of vacuum pressure. The membrane flux and ethanol concen-
tration in permeate are 0.15–0.76 L/m2 h and 3.66–4.64%-vol.,
respectively. Meanwhile, there are no specific sequences on
the maltose concentration in the effect of operating condi-
tions. The slight loss of maltose in the dealcoholized beer
can be attributed to adsorption phenomena in membrane sur-
face thus membrane flushing may be conducted to recover
it. The glycerol behavior in dealcoholization process is sim-
ilar to maltose. Some glycerol compounds are found in the
permeate stream but all of them are less than 0.005%-vol.
The most optimum conditions obtained from the experiment
are 3000 mbar feed pressure and 580 mbar vacuum pressure,
resulting 0.69 L/m2 h membrane flux, 3.70%-vol. and 4.60%-vol.
ethanol concentration in beer and permeate side respectively,
no maltose and only 0.001%-vol. glycerol in permeate side. A
long run operation for beer dealcoholization using these oper-
ating conditions can reduce the alcohol content from 5%-vol.
to 2.45%-vol. in 6 h. Analysis of concentration of some other
components in the beer is essential to ensure there are not any
significant changes in the beer composition. Sensory charac-
teristics for the dealcoholized beer also have to be tested.
References
Branyik, T., Silva, D.P., Baszczynski, M., Lehnert, R., de Silva, J.B.A.,
2012. A review of methods of low alcohol and alcohol-free
beer production. J. Food Eng. 108, 493–506, http://dx.doi.org/10.
1016/j.jfoodeng.2011.09.020.
Catarino, M., Mendes, A., Madeira, L., Ferreira, A., 2006. Beer
dealcoholization by reverse osmosis. Desalination 200,
397–399, http://dx.doi.org/10.1016/j.desal.2006.03.346.
Catarino, M., Mendes, A., 2011a. Dealcoholizing wine by
membrane separation processes. Innov. Food Sci. Emerg.
Technol. 12, 330–337, http://dx.doi.org/10.1016/j.ifset.
2011.03.006.
Catarino, M., Mendes, A., 2011b. Non-alcoholic beer – a new
industrial process. Sep. Purif. Technol. 79, 342–351,
http://dx.doi.org/10.1016/j.seppur.2011.03.020.
Chen, V., Fane, A.G., Madaeni, S., Wenten, I.G., 1997. Particle
deposition during membrane filtration of colloids: transition
between concentration polarization and cake formation. J.
Membr. Sci. 125, 109–122, http://dx.doi.org/10.1016/S0376-
7388(96)00187-1.
del Olmo, A., Blanco, C.A., Palacio, L., Prádanos, P., Hernández, A.,
2014. Pervaporation methodology for improving alcohol-free
beer quality through aroma recovery. J. Food Eng. 133, 1–8,
http://dx.doi.org/10.1016/j.jfoodeng.2014.02.014.
Diban, N., Athes, V., Bes, M., Souchon, I., 2008. Ethanol and aroma
compounds transfer study for partial dealcoholization of wine
using membrane contactor. J. Membr. Sci. 311, 136–146,
http://dx.doi.org/10.1016/j.memsci.2007.12.004.
Diban, N., Arruti, A., Barceló, A., Puxeu, M., Urtiaga, A., Ortiz, I.,
2013. Membrane dealcoholization of different wine varieties
reducing aroma losses. Modeling and experimental
validation. Innov. Food Sci. Emerg. Technol. 20, 259–268,
http://dx.doi.org/10.1016/j.ifset.2013.05.011.
Gil, M., Estevez, S., Kontoudakis, N., Fort, F., Canals, J.M., Zamora,
F., 2013. Influence of partial dealcoholization by reverse
osmosis on red wine composition and sensory characteristics.
Eur. Food Res. Technol. 237, 481–488, http://dx.doi.org/10.
1007/s00217-013-2018-6.
Gostoli, C., 1999. Thermal effects in osmotic distillation. J.
Membr. Sci. 163, 75–91, http://dx.doi.org/10.1016/S0376-
7388(99)00157-X.
Kunze, W., 1999. Technology Brewing and Malting, 2nd ed. VLB,
Berlin.
Labanda, J., Vichi, S., Llorens, J., López-Tamames, E., 2009.
Membrane separation technology for the reduction of
alcoholic degree of a white model wine. LWT – Food Sci.
Technol. 42, 1390–1395, http://dx.doi.org/10.1016/j.lwt.
2009.03.008.
Lawson, K.W., Lloyd, D.R., 1997. Membrane distillation. J. Membr.
Sci. 124, 1–25, http://dx.doi.org/10.1016/S0376-7388(96)00236-0.
Liguori, L., Attanasio, G., Albanese, D., Di Matteo, M., 2010.
Aglianico wine dealcoholization tests. In: Elsevier 20th
European Symposium on Computer Aided Process
Engineering, pp. 325–330.
Liguori, L., Russo, P., Albanesea, D., Di Matteo, M., 2013a.
Evolution of quality parameters during red wine
dealcoholization by osmotic distillation. Food Chem. 140,
68–75, http://dx.doi.org/10.1016/j.foodchem.2013.02.059.
Liguori, L., Russo, P., Albanese, D., Di Matteo, M., 2013b. Effect of
process parameters on partial dealcoholization of wine by
osmotic distillation. Food Bioprocess Technol. 6, 2514–2524,
http://dx.doi.org/10.1007/s11947-012-0856-z.
Montanari, L., Marconi, O., Mayer, H., Fantozzi, P., 2009. In: Preedy,
V.R. (Ed.), Production of Alcohol-Free Beer. Beer in Health and
Disease Prevention. Elsevier Inc., Burlington, MA, pp. 61–75.
Mulder, M., 1996. Basic Principles of Membrane Technology, 2nd
ed. Kluwer Academic Publishers, Netherlands.
Perpète, P., Collin, S., 2000. Influence of beer ethanol content on
the wort flavour perception. Food Chem. 17, 379–385,
http://dx.doi.org/10.1016/S0308-8146(00)00179-5.
Pilipovik, M.V., Riverol, C., 2005. Assessing dealcoholization
systems based on reverse osmosis. J. Food Eng. 69, 437–441,
http://dx.doi.org/10.1016/j.jfoodeng.2004.08.035.
Takacs, L., Vatai, G., Korany, K., 2007. Production of alcohol free
wine by pervaporation. J. Food Eng. 78, 118–125,
http://dx.doi.org/10.1016/j.jfoodeng.2005.09.005.
Tan, S., Li, L., Xiao, Z., Wu, Y., Zhang, Z., 2005. Pervaporation of
alcoholic beverages – the coupling effects between ethanol
and aroma compounds. J. Membr. Sci. 264, 129–136,
http://dx.doi.org/10.1016/j.memsci.2005.04.028.
Varavuth, S., Jiraratananon, R., Atchariyawut, S., 2009.
Experimental study on dealcoholization of wine by osmotic
distillation process. Sep. Purif. Technol. 66, 313–321,
http://dx.doi.org/10.1016/j.seppur.2008.12.011.