JID: JTICE

ARTICLE IN PRESS [m5G;November 22, 2017;9:38]

Journal of the Taiwan Institute of Chemical Engineers 0 0 0 (2017) 1–5

Contents lists available at ScienceDirect

Journal of the Taiwan Institute of Chemical Engineers

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

Dehydration of waste cutting oil using a pervaporation process

Tai-Hsiang Chen, Yao-Hui Huang∗
Department of Chemical Engineering, National Cheng Kung University, Tainan 70101, Taiwan, ROC

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

Article history: A commercial polyvinyl alcohol (PVA)/polyacrylonitrile (PAN) hybrid membrane (B3 membrane) was
Received 13 May 2017 tested to evaluate its ability to dehydrate waste cutting oil by pervaporation (PV). The major material and
Revised 12 August 2017
the water content of the waste cutting oil were analyzed by TGA, NMR, and Karl Fisher titrator. A prelim-
Accepted 19 August 2017
inary PV experiment that involved treatment with 15 wt% aqueous diethylene glycol (DEG) was carried
Available online xxx
out at various temperatures, and 85 °C was found to be the optimal temperature. The separation factor
Keywords: increased from around 530 to 1140 and the total flux decreased from approximately 0.88 to 0.01 kg/m2 /h
Pervaporation as the water concentration in the feed solution decreased. Then, a PV experiment that involved treating
Waste cutting oil the waste cutting oil was carried out at 85 °C and the obtained results were almost the same as those in
Diethylene glycol the preliminary experiment, indicating that PV is an effective method for recycling waste cutting oil.
Dehydration © 2017 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction the filtrates to 160 °C at a pressure of 700 mmHg to yield pure

The solar energy and semiconductor industries are requiring
more and better wafers. The general cutting process cannot sat-
isfy these requirements, so the wire saw process was developed.
In the wire saw process, a material called cutting oil is added to
the wire saw instrument. The cutting oil is composed an abrasive,
such as SiC particles, solvent, and additives. The solvent must have
two features: it is easily washed away using water and viscosity
must cause the abrasive to be suspended in the solvent. Ethylene
glycol (EG), diethylene glycol (DEG), and polyethylene glycol (PEG)
are frequently used as solvents for cutting oil.
DEG, (HOCH2 CH2 )2 O, is a co-product with ethylene glycol and
triethylene glycol. It is a colorless, sticky, and practically odorless
liquid that is miscible in water, alcohol, and ether [1]. Therefore,
DEG exhibits the necessary characteristics requirements of cutting
oil and it is the most commonly used solvent in Taiwan. However,
the boiling point of DEG is 245.3 °C. This fact and its miscibility
with water increase difficulty of its recovery. Tons of waste cutting
oil is generated by wafer slicing. The waste cutting oil consists of
the original cutting oil, water, and Si particles. SiC and Si parti-
cles will be recovered as filter cake by solid-liquid separation. The
filtrates cannot be returned as the solvent and can be burned as
fuel because of their high water content. The only way to return
them as the solvent is to dehydrate them. The most commonly
used method of dehydration is distillation. However, even the use
of vacuum distillation to dehydrate the filtrates requires heating of
∗
Corresponding author.
E-mail address: yhhuang@mail.ncku.edu.tw (Y.-H. Huang).
DEG. A new process to recover the filtrates must be developed.
In our previous research [2], membrane distillation (MD) has been
tested to dehydrate DEG solution and the water content of DEG
can be reduced from 15 wt% to 2.26 wt%. Although the result is
lower than 5 wt% which is the maximal water content to recy-
cle as the cutting oil, MD still cannot make the water content less
than 1 wt%. It still needs to find another process to make the wa-
ter content less than 1 wt%. There are two papers mentioned with
the dehydration of DEG by pervaporation [3,4]. Therefore, pervapo-
ration will be tested to treat the waste cutting oil.
Pervaporation (PV) is a membrane method involves permeation
and evaporation. It is an effective alternative process for the sepa-
ration of aqueous-organic mixtures, and especially azeotropic mix-
tures, heat-sensitive products, and mixtures of substances with
similar boiling points [5]. Unlike conventional membrane-based
techniques, such as reverse osmosis (RO) and ultrafiltration (UF),
PV uses a compact membrane or a hybrid membrane. PV exploits
the affinity between the membrane and the liquids to separate the
mixture [6]. The separation mechanism of PV mainly occurs on the
surface of the membrane. The liquids are easily absorbed by the
membrane if the affinity between the membrane and the liquids is
very strong [7]. For example, the membrane used for the dehydra-
tion in PV is hydrophilicity for which the polar groups in the mem-
brane matrix are responsible. The polar groups attract water and
act as the fixed carriers to let water quickly diffuse through the
membrane [6,8,9]. The other organic molecules diffuse through the
membrane by the free volumes between the polymer chains due to
the dense membrane. This phenomenon leads organic molecules
to be almost blocked on the feed side. Therefore, the water and

http://dx.doi.org/10.1016/j.jtice.2017.08.028
1876-1070/© 2017 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Please cite this article as: T.-H. Chen, Y.-H. Huang, Dehydration of waste cutting oil using a pervaporation process, Journal of the Taiwan
Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.08.028
JID: JTICE
ARTICLE IN PRESS [m5G;November 22, 2017;9:38]

2 T.-H. Chen, Y.-H. Huang / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–5

the organic solution can be separated on the surface of the mem-

brane. The transport processes in the PV membrane can be de-
scribed by solution-diffusion model [10,11]. In PV, the upstream,
delimited by the membrane, is maintained at atmospheric pressure
and the downstream is kept in a vacuum. The pressure drop across
the membrane causes the liquids to diffuse through the membrane
and separate. PV has been widely investigated. Kober [12] was the
first one to observe the phenomenon of pervaporation, but the first
known quantitative work on the topic was published by Heisler
et al. [13], who separated water/ethanol mixtures using a cellu-
lose membrane [14]. PV has three broad applications, which are
the dehydration of organic solution [15–20], the removal of or-
ganic matter from dilute aqueous solution [21–23], and the sepa-
ration of organic/organic mixtures (such as methyl tert-butyl ether
(MTBE)/methanol and thiophenes/FCC gasoline) [24]. If the mem- Fig. 1. PV apparatus.
brane is made of hydrophilic material, such as chitosan [25,26] or
polyvinyl alcohol (PVA) [27–29], it can be used to dehydrate or-
ganic solutions, such as alcohols, ethers, esters, and acids. However, 2.3. Pervaporation experiment
a hydrophobic membrane is suitable for removing organic matter
from dilute aqueous solution, such as is carried out in the removal Fig. 1 schematically depicts the PV apparatus. The effective area
of volatile organic compounds. There are already two articles dis- of the membrane was 19.6 cm2 and the downstream pressure was
cussing the dehydration of DEG solution by pervaporation, the sys- 0.04 to 0.1 Torr. The system ran for one hour to establish a steady
tems are different between this study and the other two papers. state. Both feed and permeate samples were obtained at 0, 1, 3, 5,
The membrane used in this study is flat polymer membrane and 7, 11, 19 and 24 h The permeate was collected in a tube that was
those used in the other two articles are hollow fiber membrane cooled by liquid nitrogen. The water content of both the feed and
[3] and HybSi ceramic membrane [4]. Another different part is the the permeate solution were measured using a Karl Fischer titra-
target of treatment. The other two articles force on the DEG solu- tor. The water contents of the permeates that were collected at
tion used in gas complex processing plants at gas fields and this 1, 19, and 24 h were not measured because their volumes were
study forces on the waste cutting oil. too small. The experiment was divided into two parts: the first
In this study, the target of treatment is the filtrates after solid– involved synthetic wastewater and the second involved the real
liquid separation. First, the filtrates are analyzed by NMR and TGA waste cutting oil. First, synthetic wastewater that was composed
to confirm what the main component is. According to the result, of water and DEG was used to determine the effects of tempera-
a synthetic wastewater that comprises the main component and ture on the dehydration of aqueous DEG. The initial feed solution
water (binary mixture) was tested to determine the optimal treat- was 85 wt% aqueous DEG (75 g). The temperature was controlled
ment temperature. Real waste cutting oil was tested under the op- between 55 °C and 95 °C. Afterwards, the real waste cutting oil
timal conditions and compared with synthetic wastewater to con- was tested under the optimal temperature and compared with the
firm that PV is an effective process for dehydrating waste cutting synthetic wastewater. The membranes before and after experiment
oil. were detected using a high resolution field emission scanning elec-
tron microscopy (SU8010, Hitachi).
To determine the performance of the PV system, the total flux
(F, kg/m2 /h), separation factor (α ), and pervaporation separation
2. Material and methods
index (PSI), defined by Eqs. (1), (2), and (3), were used. Where W
(kg) is the total amount of the permeate that was collected at ex-
2.1. Material
perimental time t (h); A is the effective area of the membrane,
and Xwater and Ywater are the weight fractions of the water in the
In this study, a commercial hybrid PVA/PAN pervaporation
feed and permeate solutions, respectively.
membrane, crosslinked by maleic anhydride, is as described in pre-
viously published papers [16]. DEG with a purity of 99% was pur- F = W/At (1)
chased from Alfa Aesar. Real waste cutting oil after solid-liquid
separation was obtained from a recycling plant.
α = [Ywater /(1 − Ywater )]/[Xwater /(1 − Xwater )] (2)

2.2. Characterization of waste cutting oil PSI =F ×(α − 1 ) (3)

The vaporization and solid content of the real waste cut-
ting oil were measured using a thermal gravimetric analyzer
(TGAQ50, Perkin Elmer). Nuclear magnetic resonance (Bruker
Avance: 600 MHz NMR) was used to determine the composition of
the real waste cutting oil. 1 H NMR spectra of the real waste cut-
ting oil was compared with those of pure DEG to prove that DEG
was the main component of the real waste cutting oil. In the NMR
analysis, D2 O was used as the solvent. The solid content was an-
alyzed by TGA. In TGA, the sample was heated from 50 to 400 °C
at a rate of 10 °C/min in N2 gas. The water content of the real
waste cutting oil was measured using a Karl Fischer titrator (787
KF Titrino, Metrohm).
3. Results and discussion
3.1. Characterization of waste cutting oil
The waste cutting oil was a yellow, sticky and clear liquid. Fig. 2
shows the 1 H NMR results. The peak at 4.7 ppm was the solvent
peak. The two longer peaks at 3.6 and 3.7 ppm were attributable
to the CH2 group. Not only the positions of the two peaks from the
DEG and the waste cutting oil, but also the splitting of these two
peaks were almost the same. The results demonstrate that the ma-
jor component of waste cutting oil is DEG. In the waste cutting oil
result, there was a little peak at about 3.35 ppm. It belonged to the

Please cite this article as: T.-H. Chen, Y.-H. Huang, Dehydration of waste cutting oil using a pervaporation process, Journal of the Taiwan
Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.08.028
JID: JTICE
ARTICLE IN PRESS [m5G;November 22, 2017;9:38]

T.-H. Chen, Y.-H. Huang / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–5 3

Fig. 2. NMR spectrum of pure DEG and waste cutting oil.

Fig. 4. Effect of temperature on water content of feed solution.

Fig. 3. TGA curve of waste cutting oil. Fig. 5. Effect of temperature on total flux and separation factor.

Table 1
other components like the additives of the cutting oil and/or some- Flux at various temperatures and Xwater , calculated using Eqs. (4)–(7).
thing added during the solid-liquid separation. This result indicates Temperature (°C) F (kg/m2 /h)
the waste cutting oil is a multicomponent mixture. Fig. 3 presents
10 wt% 9 wt% 8 wt% 7 wt% 6 wt% 5 wt%
the results of TGA. The first weight loss (50–130 °C) was caused
mainly by the evaporation of water. Above around 130 °C, the main 55 0.28 0.25 0.22 0.18 0.15 0.11
reason for the continued decline in weight was the evaporation of 70 0.74 0.66 0.57 0.48 0.39 0.30
85 0.82 0.73 0.64 0.55 0.47 0.38
DEG. Comparing with the pure DEG, the weight loss of pure DEG 95 1.01 0.90 0.79 0.68 0.57 0.46
was more rapid than the waste cutting oil. The phenomenon might
be caused by the impurities of the waste cutting oil. The solid con-
tent was 0.3 wt%. Finally the weight fell 0.3 wt%. The water content
of the waste cutting oil was about 12 wt%.
F = 0.089 Xwater − 0.069 at 85 ◦ C (6)

3.2. Synthetic waste cutting oil pervaporation experiment

Based on the water content of real waste cutting oil, the initial
water content of the synthetic waste cutting oil was set at 15 wt%.
Figs. 4 and 5 present the results of the synthetic waste cutting oil
pervaporation experiment. Initially, Xwater declined rapidly, becom-
ing stable after 11 h, except in at 55 °C. The final Xwater decreased
as the temperature increased. Fig. 5 shows a very clear linear re-
lationship between F and Xwater . Eqs. (4)–(7) are the results after
line fitting,
F = 0.034 Xwater − 0.057 at 55 ◦ C
F = 0.088 Xwater − 0.137 at 70 ◦ C
F = 0.111 Xwater − 0.099 at 95 ◦ C, (7)
All adjusted R-squares of the above equations exceed 0.99. The
flux at different values of Xwater and temperatures can be calcu-
lated according to those equations, and are presented in Table 1.
Clearly, F increases with both temperature and Xwater . It is be-
cause that the increase in temperature leads to an increase in
the mobility of the polymer chain. The free volumes of the poly-
mer also increase as the mobility of the polymer chain increase.
This phenomenon leads the water and DEG molecules to diffuse
(4) through the membrane more quickly. The relationship between
F and temperature can be described the Arrhenius equation (ln
F = −Ea/RT + ln A). The curve of ln F vs. 1/T at different Xwater can
(5) be obtained from combining the Arrhenius equation with the data

Please cite this article as: T.-H. Chen, Y.-H. Huang, Dehydration of waste cutting oil using a pervaporation process, Journal of the Taiwan
Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.08.028
JID: JTICE
ARTICLE IN PRESS [m5G;November 22, 2017;9:38]

4 T.-H. Chen, Y.-H. Huang / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–5

Fig. 6. Water content of waste cutting oil and synthetic waste water.

Fig. 8. The SEM pictures of the membrane (a) before and (b) after PV experiment.

Fig. 7. Flux in waste cutting oil and synthetic waste water.

3.3. Real waste cutting oil pervaporation experiment

In the synthetic waste cutting oil experiment, PV with B3 mem-

Table 2
PSI in different periods and at various temperatures. brane can reduce Xwater to less than 1 wt% at 85 °C and 95 °C. The
final Xwater at 85 °C and 95 °C were almost the same, so the real
Temperature (°C) PSI (kg/m2 /h)
waste cutting oil experiment was performed at the lower temper-
1–3 h 3–5 h 5–7 h 7–9 h 9–11 h ature, 85 °C. Figs. 6 and 7 compare the results obtained using real
55 370 110 80 70 30 waste cutting oil comparing with those obtained using the syn-
70 420 230 140 110 120 thetic oil. Based on the results of the synthetic waste cutting oil
85 460 240 170 120 130 experiment, the real waste cutting oil was treated to an Xwater of
95 560 290 200 140 130
less than 1 wt%. The curve of F vs. Xwater in the real waste cut-
ting oil experiment was almost the same as that in the synthetic
waste cutting oil experiment, even if the real waste cutting oil was
in Table 1. After linear regression, the values of −Ea/R are −3636, a multicomponent mixture. Xwater reduced from 12 wt% to 1.18 wt%
−3628, −3632, −3791, −3854, and −4164 as Xwater are 10, 9, 8, after 11 h of treatment and 0.58 wt% after 24 h of treatment. The
7, 6, and 5 wt%. Therefore, Ea increases as Xwater falls. The results permeates are colorless.
indicate that the effect of temperature becomes stronger as Xwater Fig. 8 shows the SEM pictures of the membrane before and af-
decreases. In other words, the dehydration of DEG by PV process ter PV experiment. From the pictures it is clear that the membrane
needs to operate at a higher temperature as Xwater is lower. α re- contains three layers: a compact layer, a porous layer, and a fiber
mains almost constant as Xwater increases above 6 wt%. As Xwater layer. The compact layer is made of PVA crosslinked by maleic an-
declines under 6 wt%, both the value and the rate of increase of hydride and has the separating ability. Comparing the two pictures
α increase. The most large value of α is 1500 as Xwater is 1.75 at showed in Fig. 8, the membrane did not change significantly after
95 °C. Table 2 presents the PSI in different periods and at various 24 h of PV treatment. However, there are some differences in the
temperatures. PSI increases with temperature and/or Xwater . The EDS results of the compact layer. The EDS results show that there
values of PSI at the two temperatures exceeded 100 kg/m2 /h when are only O and C in the compact layer before PV experiment. For
Xwater exceeded 2 wt%. the membrane after PV experiment, not only O and C but also N,

Please cite this article as: T.-H. Chen, Y.-H. Huang, Dehydration of waste cutting oil using a pervaporation process, Journal of the Taiwan
Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.08.028
JID: JTICE
ARTICLE IN PRESS
T.-H. Chen, Y.-H. Huang / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–5
Table 3
The comparison of distillation and pervaporation.
Condition
Distillation Pressure: 700 mmHg
Temperature: part I – 130 °C
part II – 160 °C
Pervaporation Pressure: feed side – normal pressure
permeate side – 10−2 mmHg
Temperature: 85 °C
S, Cl, and Cu are detected. Obviously, the impurities of real waste
cutting oil also diffuse through the membrane. This phenomenon
lead to that Yw of the real waste cutting oil experiment is lower
than that of the synthetic waste cutting oil experiment.
The current technique to recycle the waste cutting oil is dis-
tillation. Table 3 lists the comparison of distillation and pervapo-
ration. Obviously, temperature requirement of PV is much lower
than distillation. Even PV need to be operate under medium vac-
uum (25–10−3 Torr), the energy consumption of PV is much lower
than distillation. All results indicate that PV is an effective method
for recovering real waste cutting oil.
4. Conclusion
NMR enabled the major component of the waste cutting oil to
be identified as DEG and it was a multicomponent mixture. In the
synthetic wastewater experiment, Xwater was reduced to less than
1 wt% at 85 °C and 95 °C. The values of PSI at the two temper-
atures exceeded 100 kg/m2 /h when Xwater exceeded 2 wt%. When
the energy consumption is taken into account, 85 °C is the opti-
mal temperature. The PV process reduced the Xwater of real waste
cutting oil to less than 1 wt%, indicating that PV is an effective
method for recycling the real waste cutting oil.
References
[1] Schep LJ, Slaughter RJ, Temple WA, Beasley DMG. Diethylene glycol poisoning.
Clin Toxicol 2009;47:525–35.
[2] Chen T-H, Huang Y-H. Dehydration of diethylene glycol using a vacuum mem-
brane distillation process. J Taiwan Inst Chem Eng 2017;74:233–7.
[3] Yim DW, Kong SH. Pervaporative dehydration of diethylene glycol through a
hollow fiber membrane. J Appl Polym Sci 2013;129:499–506.
[4] Akberov RR, Fazlyev AR, Klinov AV, Malygin AV, Farakhov MI, Maryakhina VA,
et al. Dehydration of diethylene glycol by pervaporation using HybSi ceramic
membranes. Theor Found Chem Eng 2014;48:650–5.
[5] Shaban HI. Pervaporation separation of water from organic mixtures. Sep Purif
Technol 1997;11:119–26.
[6] Shao P, Huang RYM. Polymeric membrane pervaporation. J Membr Sci
2007;287:162–79.
[7] Okada T, Matsuura T. A new transport model for pervaporation. J Membr Sci
1991;59:133–50.
[8] Yoshikawa M, Yokoi H, Sanui K, Ogata N, Shimidzu T. Polymer membrane as a
reaction field. 2. Effect of membrane environment on permselectivity for water
ethanol binary-mixtures. Polym J 1984;16:653–6.
[9] Yoshikawa M, Ogata N, Shimidzu T. Polymer membrane as a reaction field. 3.
Effect of membrane polarity on selective separation of a water ethanol binary
mixture through synthetic-polymer membranes. J Membr Sci 1986;26:107–13.
Please cite this article as: T.-H. Chen, Y.-H. Huang, Dehydration of waste cutting oil using a pervaporation process, Journal of the Taiwan
Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.08.028
[m5G;November 22, 2017;9:38]
5
Advantage
Mature technology
Easy to operate
Energy saving
Small facilities
[10] Heintz A, Stephan W. A generalized solution diffusion-model of the pervapo-
ration process through composite membranes. 2. Concentration polarization,
coupled diffusion and the influence of the porous support layer. J Membr Sci
1994;89:153–69.
[11] Wijmans JG, Baker RW. The solution-diffusion model – a review. J Membr Sci
1995;107:1–21.
[12] Kober PA. Pervaporation, perstillation and percrystallization. J Am Chem Soc
1917;39:944–8.
[13] Heisler EG, Hunter AS, Siciliano J, Treadway RH. Solute and temperature effects
in the pervaporation of aqueous alcoholic solutions. Science 1956;124:77–9.
[14] Feng XS, Huang RYM. Liquid separation by membrane pervaporation: a review.
Ind Eng Chem Res 1997;36:1048–66.
[15] Li SY, Srivastava R, Parnas RS. Separation of 1-butanol by pervaporation using
a novel tri-layer PDMS composite membrane. J Membr Sci 2010;363:287–94.
[16] Chen T-C, Chen T-H, Wu R-Y, Huang Y-H. The novel of high efficiency recovery
technology for dehydration of alcohol solutions by a pervaporation process.
Sustain Environ Res 2013;23:171–7.
[17] Chen T-C, Chen T-H, Lyu J-Y, Huang Y-H. The dehydration and recovery of
wasted TFP solution using the pervaporation process. Sustain Environ Res
2013;23:325–31.
[18] Burshe MC, Netke SA, Sawant SB, Joshi JB, Pangarkar VG. Pervaporative dehy-
dration of organic solvents. Sep Sci Technol 1997;32:1335–49.
[19] Rezakazemi M, Shahverdi M, Shirazian S, Mohammadi T, Pak A. CFD simulation
of water removal from water/ethylene glycol mixtures by pervaporation. Chem
Eng J 2011;168:60–7.
[20] Zhang SY, Zou Y, Wei TY, Mu CX, Liu XJ, Tong ZF. Pervaporation dehydration
of binary and ternary mixtures of n-butyl acetate, n-butanol and water using
PVA-CS blended membranes. Sep Purif Technol 2017;173:314–22.
[21] Li L, Xiao ZY, Tan SJ, Liang P, Zhang ZB. Composite PDMS membrane with high
flux for the separation of organics from water by pervaporation. J Membr Sci
2004;243:177–87.
[22] Zhang YL, Benes NE, Lammertink RGH. Performance study of pervaporation
in a microfluidic system for the removal of acetone from water. Chem Eng J
2016;284:1342–7.
[23] Rom A, Friedl A. Investigation of pervaporation performance of POMS mem-
brane during separation of butanol from water and the effect of added acetone
and ethanol. Sep Purif Technol 2016;170:40–8.
[24] Jain M, Attarde D, Gupta SK. Removal of thiophenes from FCC gasoline by using
a hollow fiber pervaporation module: modeling, validation, and influence of
module dimensions and flow directions. Chem Eng J 2017;308:632–48.
[25] Sun H, Lu L, Chen X, Jiang Z. Surface-modified zeolite-filled chitosan
membranes for pervaporation dehydration of ethanol. Appl Surf Sci
2008;254:5367–74.
[26] Sajjan AM, Premakshi HG, Kariduraganavar MY. Synthesis and characterization
of GTMAC grafted chitosan membranes for the dehydration of low water con-
tent isopropanol by pervaporation. J Ind Eng Chem 2015;25:151–61.
[27] Li BB, Xu ZL, Qusay FA, Li R. Chitosan-poly (vinyl alcohol)/poly (acryloni-
trile) (CS-PVA/PAN) composite pervaporation membranes for the separation of
ethanol-water solutions. Desalination 2006;193:171–81.
[28] Peters TA, Benes NE, Keurentjes JTF. Hybrid ceramic-supported thin PVA per-
vaporation membranes: long-term performance and thermal stability in the
dehydration of alcohols. J Membr Sci 2008;311:7–11.
[29] Shameli A, Ameri E. Synthesis of cross-linked PVA membranes embedded with
multi-wall carbon nanotubes and their application to esterification of acetic
acid with methanol. Chem Eng J 2017;309:381–96.