- Email: [email protected]
Titanium dioxide nanotubes embedded mixed matrix PES membranes characterization and membrane performance
Accepted Manuscript Title: Titanium dioxide nanotubes embedded mixed matrix PES membranescharacterization and membrane performance Author: Mohamed Shaban Heba AbdAllah Lamiaa Said Hany S. Hamdy Ahmed Abdel khalek PII: DOI: Reference:
S0263-8762(14)00485-7 http://dx.doi.org/doi:10.1016/j.cherd.2014.11.008 CHERD 1739
To appear in: Received date: Revised date: Accepted date:
29-3-2014 9-11-2014 15-11-2014
Please cite this article as: Shaban, M., AbdAllah, H., Said, L., Hamdy, H.S., khalek, A.A.,Titanium dioxide nanotubes embedded mixed matrix PES membranescharacterization and membrane performance, Chemical Engineering Research and Design (2014), http://dx.doi.org/10.1016/j.cherd.2014.11.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
abstract
Titanium dioxide nanotubes embedded mixed matrix PES membranes characterization and membrane performance
1
ip t
Mohamed Shaban1, *Heba AbdAllah2, Lamiaa Said1,3, Hany S. Hamdy1, Ahmed Abdel khalek3 Nanophotonics and Applications (NPA)Lab, Department of Physics, Faculty of Science, Beni-
Engineering Research Division, Chemical Engineering & Pilot Plant Department, National
us
2
cr
Suef University, Beni-Suef 62514, Egypt
Research Center (NRC), Dokki 12311, Cairo, Egypt
Department of Chemistry, Faculty of Science, Beni-Suef University, Beni-Suef 62111, Egypt
an
3
M
*Corresponding author e-mail: [email protected]
d
ABSTRACT
te
This study describes the preparation, characterization, and evaluation of performance of blend polyethersulfone / Titanium dioxide nanotubes (PES/TiO2NTs) membranes. TiO2NTs were
Ac ce p
synthesized via hydrothermal process and used in preparation of blend PES/TiO2NTs membranes by phase inversion process. The effects of embedding TiO2 nanotubes on membrane morphology, mechanical properties and performance were presented. The scanning electron microscopy (SEM) images displayed a typical asymmetric membrane structure with a dense top layer, a porous sublayer, and fully developed macropores at the bottom of PES/TiO2NTs membranes due to the migration of TiO2NTs to membrane surface during the phase inversion process. Contact angle measurements indicated that the hydrophilicity of the membrane was improved by adding TiO2NTs. Using vacuum membrane distillation (VMD) application, the results indicated that the membrane performance was improved, particularly, permeate flux and
Page 1 of 38
salt rejection %. The mathematical model of VMD was applied to determine the highest performance of the membrane at the optimum conditions. The experimental results indicated that using 0.53 % TiO2NTs in membrane preparation solution; it was possible to obtain NF/RO blend
ip t
membrane with high performance, where the salt rejection % increased to 97% with permeate
Titanium
dioxide
nanotubes;
Blend
membrane;
Membrane
performance;
us
Keywords:
cr
flux 18.2 Kg/m2.h. These results agree well the results of the VMD model.
Ac ce p
te
d
M
an
Mathematical model.
Page 2 of 38
Graphical Abstract (for review)
Titanium dioxide nanotubes embedded mixed matrix PES membranes characterization and membrane performance Mohamed Shaban1, *Heba AbdAllah2, Lamiaa Said1,3, Hany S. Hamdy1, Ahmed Abdel khalek3 Nanophotonics and Applications (NPA)Lab, Department of Physics, Faculty of Science, BeniSuef University, Beni-Suef 62514, Egypt 2 Engineering Research Division, Chemical Engineering & Pilot Plant Department, National Research Center (NRC), Dokki 12311, Cairo, Egypt 3 Department of Chemistry, Faculty of Science, Beni-Suef University, Beni-Suef 62111, Egypt *Corresponding author e-mail: [email protected]
us
cr
ip t
1
Ac ce p
te
d
M
an
Graphical abstract
Page 3 of 38
*Research Highlights
Highlights
te
d
M
an
us
cr
ip t
Synthesis of TiO2NTs via hydrothermal process. Preparations of PES/ TiO2NTs blend membrane using the wet phase inversion method. The membrane morphological, structural, mechanical, and hydrophilic properties are studied. PES/TiO2NTs blend membrane provided a good membrane performance using VMD system. The predicted results from mathematical VMD model agree well with the experimental results.
Ac ce p
1. 2. 3. 4. 5.
Page 4 of 38
*Manuscript
Titanium dioxide nanotubes embedded mixed matrix PES
ip t
membranescharacterization and membrane performance
Mohamed Shaban1, *Heba AbdAllah2, Lamiaa Said1,3, Hany S. Hamdy1, Ahmed Abdel khalek3 1
University, Beni-Suef 62514, Egypt
Engineering Research Division, Chemical Engineering& Pilot Plant Department, National
us
2
cr
Nanophotonics and Applications (NPA)Lab, Department of Physics, Faculty of Science, Beni-Suef
an
Research Center (NRC), Dokki 12311, Cairo, Egypt 3
Department of Chemistry, Faculty of Science, Beni-Suef University, Beni-Suef 62111, Egypt author e-mail:[email protected]
M
*Corresponding
te
d
1. Introduction
Ac ce p
Nanomaterials have been used in membrane fabrication to produce membranes with desirable structure due to their large specific surface area [1, 2]. Using nanomaterials, the membrane performance can be improved and lead to high selectivity and permeability. In addition, the hydrophilicity, strength, stiffness, water permeability and antifouling properties of the membrane can be enhanced by introducing inorganic nanomaterials into membrane matrix [3-7]. Titanium dioxide (TiO2) is one of the most important semiconductors in our daily life for its unique properties including superior photo-reactivity, superhydrophilicity, nontoxicity, longterm stability, and low cost [8, 9]. In the recent years, many researchers focused on the preparation of TiO2 nano-inorganic mixed matrix membranes to improve membrane performance by reducing fouling and increasing hydrophilicity [10–24]. 1 Page 5 of 38
Blending of polymers is a technical way for providing polymers with set of desired properties at the lowest price such as a combination of strength and toughness, impact strength or solvent resistance. Blending also benefits the industrialist by offering improved process ability,
ip t
rapid formulation changes, deposit flexibility, product uniformity, and high yield [25].
Phase inversion method is one of the most important and popular techniques for
cr
manufacturing many functional polymeric materials that are widely used in engineering
us
applications [26, 27]. In addition, the antifouling properties for the membrane can be modified by blending with inorganic nanoparticles such as Al2O3, SiO2, TiO2, Fe3O4, and carbon
an
nanotubes [28–31]. In recent years, TiO2 nanoparticles blend membranes have been fabricated by
with polymeric casting solution [14, 33].
M
precipitating the nanoparticles on the surface of the porous membranes [3, 32] or blending them
d
The addition of nanotubes such as carbon (CNTs) or titanium dioxide (TiO2) in
te
preparation of polymeric membrane can provide high hydrophilicity, good chemical stability, innocuity, and high-surface area [34].
Because of nanotube materials can reinforce the
Ac ce p
polymeric membrane as a result of their high-aspect ratio and high in axis strength. In addition, titanium dioxide nanotubes (TiO2NTs) can be synthesized by a solution chemical method to form unique open-end nanotubular morphology with typically 3-5 nm and 10-13 nm in inner and outer diameters, respectively [35]. As a consequence of the physical-chemical characteristics, mutual and synergy combination of it slow-dimensional nanostructure, membrane performances can be enhanced by the addition of optimized ratio of TiO2NTs during the preparation process [35]. However, there are no detailed studies for utilizing titanium dioxide nanotubes TiO2NTs to modify the membrane properties and performances, which could solve the problems of fouling and increase the lifetime of the membrane [7-13]. In addition, TiO2NTs could be used to produce
2 Page 6 of 38
the next generation of membranes with high selectivity, reasonable flux, low fouling, and enhanced mechanical properties due to remarkable physical and chemical properties. So TiO2NTs could play a great role for improving the membrane performance due to the excellent
ip t
properties such as porosity, thermal stability, high mechanical strength and high-surface area [7, 12, 13, 20 ].
cr
The hydrophobic porous membranes are preferable in membrane distillation systems,
us
which will resist wetting and oppose liquid water from entering the membrane pores. Then the hydraulic pressure of the bulk water on either sides of the membrane can overcome the inherent
an
surface tension and allow feed water to pass through the membrane .Also, the salt rejection increases by increasing the hydrophobicity of the membrane [36]. But, the fouling by proteins
M
and other organic matter is commonly attributed to the hydrophobic nature of membrane
d
materials that leads to a high interfacial energy with water-rich media, which is decreased upon
te
biomolecules adsorption. So, the suggested structure of the membrane that used in this research is different; it isn’t hydrophobic porous membrane, but it is asymmetric dense membrane with
Ac ce p
adopted hydrophilic properties. The hydrophilic properties of suggested membrane will be adopted to improve the membrane antifouling properties by introducing the titanium dioxide (TiO2) nanotubes at different percentage into the membrane matrix. This is highly expected to improve the strength and stiffness, the water permeability, and the antifouling properties of membrane. Also, the salt rejection % is expected to increase due to the dense layer which is the separation layer [12, 13]. In this study, the fabrications of PES/ TiO2NTs blend membrane by mixing of the polymer with different percentage of TiO2 nanotubes have been investigated. The membranes were fabricated by the wet phase inversion method. The membrane structure and properties were
3 Page 7 of 38
characterized using SEM, mechanical testing system, and water contact angle measurements setup. Desalination of brackish water was applied using the prepared blend membrane by vacuum membrane distillation process. Moreover, simple theoretical model of VMD was used to
ip t
determine the optimum parameters that can effect on the prepared membrane performance.
us
cr
2. Experimental
an
2.1 Materials
M
Analytical grade N, methyl-2-pyrrolidone (NMP) as a solvent and polyethersulfone (PES ultrason E6020P with MW= 58,000 g/mol) were supplied by BASF Company (Germany).
d
Sodium dodecyl sulphate (SDS) was obtained from Merck, Germany. Commercial TiO2 anatase
te
powder and HCl were purchased from Merck and NaOH pellets were purchased from oxford. All chemicals used in the experiments were of reagent grade. Commercial NaCl was used in
Ac ce p
desalination experiments.
2.2 Preparation of TiO2 nanotubes by hydrothermal method Hydrothermal synthesis of titanium dioxide nanotubes was carried out using a 1L autoclave to provide closed reaction environment and the slightly higher pressure during the process. 4 g of commercial anatase TiO2 powder was added to 400 ml of 10 M NaOH. The mixture was stirred at ambient temperature for 30 min. Thereafter, the mixture was heated in a 1 L Teflon-lined autoclave at 150°C [35]. The white product was filtered and washed with 1N HCl and deionized water. Then the sample was dried at 80oC for 4h, after that the produced TiO2NTs 4 Page 8 of 38
post-treated at 500°C for 3 h [35]. Fig.1 illustrates a schematic diagram of the hydrothermal
2.3 Fabrication of asymmetric PES/TiO2NTsblend membranes
ip t
synthesis method of titanium dioxide nanotubes.
The asymmetric PES /TiO2NTs blend membranes were fabricated by phase inversion
cr
induced by immersion precipitation method using casting solutions containing PES (20 wt.%),
us
0.5% sodium dodecyl sulphate (SDS) and the proper amount of TiO2NTs powder in N,methyl-2pyrrolidone (NMP) as solvent. The compositions of casting solutions for all membranes are
an
depicted in Table 1.
The specific percentages of TiO2NTs (0.18, 0.35, 0.53 and0.85%) were added into
M
solution of 0.5% sodium dodecyl sulphate (SDS) and the solution was stirred for 3 h. The casting
d
solution was prepared by dispersing 0.9% solution of TiO2NTs with sodium dodecyl sulphate
te
into NMP solvent with respect to casting solution. After dispersing TiO2NTs /SDS solution in a solvent, PES was dissolved in the dope solution by continuous stirring for 6 h and the polymer
Ac ce p
mixture solution (casting solution) left in the refrigerator for 24 h to remove air bubbles. After that, the casting solution was casted onto a clean glass plate with 150 µm thickness. Subsequently, the glass plate was horizontally immersed into distilled water at room temperature. After primary phase separation and membrane solidification, the membranes were stored in freshly distilled water for 24 h to guarantee the complete phase inversion. Afterward, the membranes were sandwiched between two filters papers for 24 h at room temperature for drying.
2.4 Membrane characterization
5 Page 9 of 38
2.4.1 Membrane morphology Scanning electron microscopy (SEM) was used to observe the morphology of
ip t
PES/TiO2NTs blend membranes, where as the samples for cross-sectional view were coated with gold to provide electrical conductivity. The cross-sectional snapshots of membrane were taken
cr
on a JEOL 5410 SEM and conducted at 10 kV.
us
2.4.2 Mechanical properties
an
Mechanical properties of PES and PES/TiO2NTsblend membraneswere investigated to determine the appropriate percentage of TiO2NTs(wt%) which can be used without affecting the
M
membrane strength. The membrane tensile strength and elongation were measured using mechanical testing system (INSTRON-5500R). The gauge length and width of dumbbell tensile
d
specimens were 6.2 and 0.16 mm, respectively. The specimen of membrane was placed between
te
the grips of the testing machine, and then the tensile strength and elongation were calculated.
Ac ce p
The measurement accuracy is within ± 5%.
2.4.3 Membrane porosity and contact anglemeasurements Membrane porosity plays an important role in describing its performance. In order to evaluate porosity of the PES/TiO2NTs blend membranes, they were initially impregnated with water then weighed after wiping superficial water with filter papers. After that, the wet membranes were placed in an air-circulating oven at 80°C for 24h to be completely dried. Finally, the dry membranes were weighed and the porosity of membranes was calculated using the following equation [37, 38]:
6 Page 10 of 38
W0 W1 V
(1)
Where ɛ is the membrane porosity; WO and W1 are the weights of wet and dry membranes in
ip t
gram, respectively; V= A.δ where A is the membrane surface area in m2 and δ is the membrane thickness in m. In order to minimize the experimental errors, the membrane porosity of each
cr
sample was measured several times, at least 3 times, and the results were reported in average (the
us
error of porosity ±2%). The average pores diameters and the total surface area of the pores were measured using Hg porosimeter (model type No.9810). Distilled water was used for the contact
an
angle (θ) measurement by the sessile drop method. The measurements were carried out four
M
times for each membrane sample and the average values were reported.
2.5 Membrane performance measurements using membrane distillation technique
d
The experiments were carried out on vacuum membrane distillation (VMD) system as
te
shown in Fig. 2. This system contains flat sheet membrane module of three opening for vacuum
Ac ce p
pressure, feeding, and recycling. The feed was continuously flowing to the membrane module from an open feeding tank (glass flask) by peristaltic feeding pump. Vacuum pressure was applied using a vacuum pump. The feed mixture was heated using a hot plate and the temperature was controlled by thermostat of the hot plate and recorded by thermometer. Water condenser was used as a cooling system by recycling cold water using circulation cooling water bath through inlet and outlet in the condenser jacket. Product was collected from downstream of the condenser. The prepared PES/TiO2NTs blend membranes with different TiO2NTspercentages (wt%) were located in stainless steel plate module of 47 mm in diameter. In all experiments, the aqueous feed solution of about 8000 ppm of NaCl in DI water were prepared and continuously fed to the membrane module where the other MD parameterѕ are remained constant at feed flow 7 Page 11 of 38
rate of 12 ml/s, vacuum pressure of
200 mbar and temperature of 65°C. Effect of feed
temperature in the range from 25°C to 65°C and effect of vacuum pressure on permeate side in the range from 200 mbar to 800 mbar were investigated and compared with the predicted results
ip t
from the theoretical VMD model.
V . A.t
(2)
us
Jw
cr
For this system, the water flux, Jw (Kg/m2 h), is given by the following equation:
Where V is the volume of the pure water permeate (m3), A is the effective area of the
an
membrane (m2), which was circular membrane with area 0.0017 m2, ρ is the water density (Kg/m3) and t is the permeation time (h).
M
The concentration of permeate collected was measured using a conductivity meter. In
d
addition, the salt rejection (SR %) was conducted in triplicates for each membrane and the
C f Cp Cf
*100
(3)
Ac ce p
SR%
te
average result was calculated using the following equation:
Where, Cf and Cp are concentrations (mg/l) at feed bulk and permeate, respectively.
2.6 Mathematical model for VMD using PES/TiO2NTs blend membrane A VMD model was developed in order to simulate the VMD desalination process for TiO2NTs /PES blend flat sheet membranes. It allows us to calculate permeate vapor flow through the membrane and to determine the temperature profiles alongside the membrane (in feed side or close to the membrane).VMD process is motivated by both the total pressure difference and the temperature difference between two sides of the membrane. The total pressure difference rises from the vacuum dragged by a vacuum pump in the cold chamber of the membrane module. The 8 Page 12 of 38
permeate vapor is sucked out in the cold condenser by the vacuum pump and condensed [36, 39, 40]. If the hot water flows in the direction tangent over the membrane, the temperature difference at
ip t
two sides of the membrane can provide water vapor pressure gradient within the membrane pores. This driving force produces a mass flux through the membrane. According to this case, the
cr
Knudsen and molecular diffusion are involved in VMD process; the mass transfer resistance
us
caused by molecule-molecule collision can be neglected, and diffusion process within the pores is dominated by Knudsen diffusion. The mass flux related to this mechanism (JK) is given by
8r 3
M Pv 2RT
(4)
M
JK
an
equation (4):
d
Where r is the pore radius, τ is tortuosity factor, M is molecular weight of feeding water,
te
R is Universal gas constant, T is feeding temperature, and ∆Pv is the water vapor pressure
Ac ce p
gradient within the membrane pores. Tortuosity (τ) factor was calculated using the following equation [41]:
(2 ) 2
(5)
The total pressure gradient causes a convective mass transport through the pores, which is known as Poiseuille flow. Because almost no air exists within membrane pores, water vapor in the pores is considered pure gas and thus the total pressure difference acting as the driving force for Poiseuille flow. So, the mass flux related to this mechanism (Jp) is given by equation (6) as follows:
9 Page 13 of 38
Jp
r 2 Pv M Pv 8RTm
(6)
ip t
Where, Pv is water vapor pressure, η is water viscosity, and Tm is the mean temperature. In
(J) can be calculated by summation of two mechanisms [36,39, 40].
us
J=JK+JP
Pv
(7)
(8)
M
M r 2 Pv M 2RT 8RTm
an
Then, 8 r J 3
cr
VMD, these two mechanisms contribute in parallel to the mass transfer, and the total mass flux
M r 2 Pv M 2RT 8RTm
Pfm Ppm
(9)
Ac ce p
8 r J 3
te
membrane is given by equation (9):
d
By integrating equation (8) relative to membrane thickness, the total mass flux through the
Where, Pfm is the water vapor pressure at the membrane surface of the feed side, and Ppm is the pressure of the permeate side which is considered to be equal to the vacuum pressure of the condenser. The non-linear equation (9) was solved using MATLAB SIMULINK software (The Mathworks, Release 2009b). The water vapor pressure at the liquid/vapor interface in Pascal is related with the temperature in K according to the Antoine’s equation [39]:
Pv (T ) exp( 23.1964
3816.44 ) T 46.13
(10)
10 Page 14 of 38
The temperature polarization phenomenon occurs as a result of a temperature difference between the bulk feed temperature (Tb), interfacial temperature (Tim), and the temperature in the
ip t
vacuum membrane side (Tv). This temperature gradient is due to the heat flux through the liquid layer, which is needed to provide the required heat for evaporation at the membrane interface.
cr
The internal temperatures were measured using lab Infrared thermometer TM-939 (made in
us
Taiwan) was purchased from Lutron electronic enterprise company. The temperature variation across the membrane can be described by the temperature polarization factor that is defined as
Tb Tim Tb Tv
te
d
3. Results and discussion
(11)
M
an
[39]:
Ac ce p
3.1 Characterization of PES/TiO2NTs blend membrane 3.1.1 Membrane morphology
Fig. 3 shows cross –sectional SEM images of PES membranes blended with TiO2NTsof different percentages. All membranes showed a typical asymmetric membrane structure with a dense top layer, a porous sublayer, and fully developed macropores at the bottom. Nevertheless, the formation of macropores was suppressed by the addition of TiO2NTs into the membrane structure as shown in Fig. 3 (a-d). Also, using TiO2NTscould adjust the pores distribution according to TiO2NTs%, where the structure was well-designed finger-like sub-layer with dense top layer.
11 Page 15 of 38
These results might be described by the fact that by increasing the amount of TiO2NTs in the casting solution to 0.53 %, the porosity of the synthesized membrane increased due to the enhanced phase separation with TiO2NTs. Further increase in TiO2NTs amount leads to a denser
ip t
structure in the sub-layer due to the delayed phase separation as the viscosity increased, see Fig. 3(c and d) [42-44]. Whereas the blank PES membrane in Fig.3 (e) shows spongy- porous
us
cr
structure of the membrane without any appears of porous finger-like in sub-layer.
3.1.2 Mechanical properties
an
The blend membrane tensile strength and elongation were measured using mechanical testing system. Fig. 4 shows the relation between tensile strength, elongation, and TiO2NTs
M
percentage. This figure presents that the maximum tensile strength (47.5 kg/cm2) and the
d
maximum elongation% (6%) are reached when the membrane blended with 0.53% of TiO2NTs.
te
The delay phase separation is related to solid-liquid de-mixing, where the viscosity of polymer solution is increased by increasing the nanomaterials (TiO2NTs) percentage, which could delay
Ac ce p
the exchange of solvent and nonsolvent. This delay time in the solid-liquid de-mixing during membrane casting leads to initiate a number of microcracks by inorganic fillers (TiO2NTs) [7, 44]. Accordingly, the increasing in TiO2NTs to 0.85% leads to decrease in mechanical properties due to the delayed phase separation during blend membrane casting as a result of high viscosity of polymer solution.
3.1.3 Membrane porosity and contact angle The overall porosity information of the prepared mixed matrix PES/TiO2NTs blend membranes was presented in Table 2. The results of the porosity measurement revealed that all 12 Page 16 of 38
the prepared PES/TiO2NTs blend membranes possessed a good porosity in the range of 61– 91%compared to 27.2% for pure PES membrane. As shown in this Table, the porosity of blend membranes increased as the percentage of TiO2NTs increased. This means that, the use of
ip t
TiO2NTs as a blend material increases the porosity of the membrane internal structures, which also improves the properties of the membrane and may result in an increase in lateral flow rates
cr
through the membranes [45].
us
Contact angle measurement of membranes is considered to be an important parameter for membrane characterization and indirect indication of the hydrophilicity and flux behavior. The
an
hydrophilicity of the PES/TiO2NTs blend membranes was determined based on water contact angle using the sessile drop method. The contact angles were measured several times and then
M
average values were reported. Table 2 illustrates the water contact angles of the mixed matrix
d
PES/TiO2NTs blend membranes and membrane wettability with different percentage of
te
TiO2NTs. As shown in this table, the bare PES membrane had the highest contact angle (85± 2°), corresponding to the lowest hydrophilicity. In the case of blend membranes, lower contact angle
Ac ce p
was observed in comparison with bare PES membrane. In addition, the contact angle decreases from 75 ± 2° to 45 ± 2° as TiO2NTs percentage increased from 0.18 to 0.85 wt.% due to increase of hydrophilicity. Therefore, it was expected that the blend membranes absorb more permeate into the membrane and thus enhance the flux rate [45]. Also, Table (2) illustrates the average pores diameter and total pores area on the membrane surface. The average pores diameters are decreased from 0.0035 to 0.002 µm for the blend membranes as TiO2NTs percentage increased from 0.18 to 0.85 wt% compared with 0.032 µm of bare PES. But, the measured total pores areas of the blend membranes are higher than that of the bare PES. This may be ascribed to the geometrical shape of TiO2NTs, a cylindrical tubular shape as shown in figure (1), which can provide the organization, re-arrangement, and generation of new 13 Page 17 of 38
pores on the membrane surface [21, 26]. Instead of the decrease of the pores sizes, the pores density, total pores area, and membrane porosity are increased on the surface of the blend membrane. This may
explain why the flux of bare PES is higher than the blend membranes whereas the salt rejection
ip t
% of blend membranes is higher than that of the bare PES.
cr
3.2 Membrane performance measurements using membrane distillation technique
us
Average water permeation flux and average salt rejection % of PES/TiO2NTs blend membranes with different percentages of TiO2NTs were measured using vacuum membrane
an
distillation process. The effects of TiO2NTs percentages on water flux and salt rejection % of prepared membranes are shown in Figs. 5 and 6, respectively. The average flux of blend
M
membranes reached a minimum when 0.18wt% TiO2NTs was used, then increased as TiO2NTs%
te
preparation.
d
increased and reached a maximum when 0.85wt% of TiO2NTs was used in the membrane
Figure 5 also indicates that the bare PES provides the highest permeate flux while Fig. 6 rejection percent
Ac ce p
indicates the lowest
of bare PES
compared with
blend PES/
TiO2NTsmembrane. This may be ascribed to the large pores size and the average pore diameter of bare PES (0.032 µm) as shown in Table (2). The flux of the blend membranes decreased after blending with TiO2NTs due to the formation of a dense layer on the top of the membrane, which leads to the decrease of the permeate flux and increase in the salt rejection % compared with bare PES. Fig.6 indicates that the average salt rejection of membranes increased to 97% by addition of TiO2NTs in the casting solution to prepare blend membranes. The variations in hydrophilicity and morphology of blend membrane affect its performance in two different ways. The membrane hydrophilicity increases the flux due to enhance of membrane hydrophilicity and porosity, where 14 Page 18 of 38
the membrane morphology decreases the flux due to a decrease in membrane pore size and an increase in skin layer thickness [46]. Table 3 illustrates a comparison of the performance of different membranes and the
ip t
prepared membranes in this study based on permeate flux and salt rejection. This table indicated that the prepared membrane using TiO2NTs exhibits better membrane performance based on
us
cr
permeate flux and salt rejection at feed concentration of 8000 ppm NaCl.
an
3.3 Mathematical model for VMD using TiO2NTs /PES blend membrane Validation of the VMD model was performed by comparing the experimental and
M
computed values of the permeate flux under the optimum operating conditions for PES/ 0.53% TiO2NTs blend membrane using the synthetic seawater 8000 mg/l.
d
Feed temperature is one of the important parameters that effect on the permeate flux.
te
Comparison between experimental results and predicted results of permeate flux for synthetic
Ac ce p
solution at different feed temperatures is presented in Fig. 7. The experiments were conducted at pressure of 200 mbar, feed temperatures between 25 and 65°C, and feed flow rate of 12 ml/s. This figure shows good agreement between experimental and predicted results. Whereas the increasing of feed temperature leads to increase in permeate flux as a consequence of the rise in vapor pressure of vapor-liquid interface on liquid feed side, which causes a simultaneous increase in the driving force of mass transfer [47]. Fig. 8 shows a comparison between experimental and theoretical results for different permeate pressure from 200 to 800 mbar, where the feed temperature was kept at 65°C. It can be seen that a high-permeate flux with a low-permeate pressure and a high-feed temperature. When permeate pressure varies from 200 to 800 mbar, the permeate flux is more than doubled whereas 15 Page 19 of 38
the specific energy requirement is nearly constant, where the energy required to maintain vacuum pressure is only a small portion of the total energy requirement (less than 2%) [48]. The theoretical predictions from VMD model agreed well with experimental measurements of
ip t
permeate flux.
Temperature polarization coefficient (θ) is the ratio between the temperature close to the
cr
membrane surface calculated by simulations using equation (11) and the experimental
us
temperature measured at the feed side [48-49]. It indicates the durable contribution of the boundary layer on the total heat transfer resistance of the VMD system. If θ ≈ zero for a system,
an
the process is controlled by mass transfer through the membrane and resistance in the liquid phase is negligible but if the value of θ approaches to one, the process is restricted by heat
M
transfer to the membrane surface and the resistance in the membrane is negligible [50]. The
d
calculated values of θ as a function of bulk feed temperature are shown in Fig. 9. The results
te
indicate that θ decreases by increasing bulk feed temperature (Tb), according to that θ decreases from 0.33 at feed temperature 25 ºC to 0.16 at feed temperature 65ºC as the membrane
Ac ce p
permeability and permeate flux increase.
4. Conclusion
The asymmetric PES/TiO2NTs blend membranes were successfully fabricated by phase inversion method, where titanium oxide nanotubes (TiO2NTs) were synthesized by hydrothermal method. The addition of TiO2NTs resulted in higher porosity, lower contact angle, and better hydrophilicity than bare PES. Also, the performance of prepared blend membranes were improved due to the addition of TiO2NTs, where the permeate flux was improved at using 0.85%
16 Page 20 of 38
due to increase of membrane hydrophilicity and the salt rejection % increased to 97% compared with the bare PES which has the lowest salt rejection %. The prepared blend membranes were applied on vacuum membrane distillation system.
ip t
The experiments showed that PES/TiO2NTsblend membranes provided average accumulated flux of 18.2 Kg/m2.h and average salt rejection % reached 97% when 0.85 wt% of TiO2NTs is
cr
used in the membrane preparation. This means that TiO2NTs are good modifier for formation of
us
NF/RO membrane with high reproducibility. Moreover, VMD model was developed in order to simulate the optimum conditions for PES/TiO2NTs blend flat sheet membranes. The predicted
an
results from the model agreed well with the experimental results for PES/0.53 wt%
M
TiO2NTsblend membrane at 200 mbar, 65°C, and flow rateof 12 ml/s.
te
d
Nomenclature
A: Effective area of the membrane (m2)
Ac ce p
Cf: Concentration at feed bulk (mg/L)
Cp: Concentration at permeate (mg/L) J: Mass vapor flux (kg/m2 h)
Mw: Molecular weight of water (kg/kmol) P: Total pressure (Pa)
Pfm: Water vapor pressure at the feed membrane surface (Pa) Ppm: Water vapor pressure at the permeate surface (Pa) ΔPv: Vaporpressuredifference at membranesides (Pa) Pv: Vapor pressure at feed side(Pa) R: Universal gas constant (m3 Pa/mol K) r: Membrane pore radius (m) Tm: Meantemperature(K)
17 Page 21 of 38
Tf: Water feed temperature (K) Tfb: Temperature at the feed bulk side (K) Tpb: Temperature at the permeate bulk side(K)
ip t
Tfm: Temperature at the feed membrane surface(K) Tpm: Temperature at the bulk membrane surface (K) Tim: Interfacial temperature at the membrane surface (k)
cr
t : Permeation time (h).
us
V:Volume of the pure water permeate (m3) Dimensionless Numbers
an
θ: Temperature polarization coefficient (TPC).
M
τ: Tortuosity factor
Greek Letters δ: Membrane thickness (m)
Ac ce p
ρ :Water density (Kg/m3)
te
η : Water vapour viscosity Pa.sec
d
ε: Membraneporosity
Subscripts b: Bulk f : Feed
m: Membrane
VMD: Vacuum Membrane distillation.
18 Page 22 of 38
References 1. J. Kim, S. H. R. Davies, M. J. Baumann, V. V. Tarabara, S. J. Masten, Effect of ozone dosage and hydrodynamic conditions on the permeate flux in a hybrid ozonation-ceramic
2.
ip t
ultrafiltration system treating natural waters, J. Membr. Sci. 311 (2008) 165-172.
L. Brunet , D. Y. Lyon, K. Zodrow, J. C. Rouch, B.Cassat, P. Serp, J. C. Remigy, M. R.
cr
Wiesner, P. J. J. Alvarez, Properties of Membranes Containing Semi-dispersed Carbon
us
Nanotubes, Environ Eng. Sci. 25 (2008) 565-576.
3. V. Vatanpour, S.S. Madaeni, A.R. Khataee, E. Salehi, S. Zinadini, H. A. Monfared, TiO2
an
embedded mixed matrix PES nanocomposite membranes: Influence of different sizes and types of nanoparticles on antifouling and performance, Desalination 292 (2012) 19–29.
M
4. J. Kim, B. Van der Bruggen, The use of nanoparticles in polymeric and ceramic
d
membrane structures: review of manufacturing procedures and performance improvement
5. J.B.
Li,
J.W.
te
for water treatment, Environ. Pollut. 158 (2010) 2335–2349. Zhu,
M.S.
Zheng,
Morphologies
and
properties
of
Ac ce p
poly(phthalazinoneethersulfoneketone) matrix ultrafiltration membranes with entrapped TiO2 nanoparticles, J. Appl. Polym. Sci. 103 (2007) 3623–3629. 6. Z.L. Xu, L.Y. Yu, L.F. Han, Polymer-nanoinorganic particles composite membranes: a brief overview, Front. Chem. Eng. Chin. 3 (2009) 318–329. 7. A. Razmjou, J. Mansouri, V. Chen, The effects of mechanical and chemical modification of TiO2 nanoparticles on the surface chemistry, structure and fouling performance of PES ultrafiltration membranes, J. Membr. Sci. 378 (2011) 73–84. 8. Y. Mansourpanah, S.S. Madaeni, A. Rahimpour, A. Farhadian, A.H. Taheri, Formation of appropriate sites on nanofiltration membrane surface for binding TiO2 photo-catalyst:
19 Page 23 of 38
performance, characterization and fouling-resistant capability, J. Membr. Sci. 330 (2009) 297–306. 9. S. S. Madaeni, N. Ghaemi, A. Alizadeh, M. Joshaghani, Influence of photo-induced
ultrafiltration membranes, Appl. Surf. Sci. 257 (2011) 6175–6180.
ip t
superhydrophilicity of titanium dioxide nanoparticles on the anti-fouling performance of
cr
10. A.R. Khataee, V. Vatanpour, A.R. Amani Ghadim, Decolorization of C.I. Acid Blue 9
us
solution by UV/Nano-TiO2, Fenton, Fenton-like, electro-Fenton and electrocoagulation processes: a comparative study, J. Hazard. Mater. 161 (2009) 1225–1233.
an
11. Y. H. Teow, B. S. Ooi, A. L. Ahmad, J. K. Lim, Mixed-Matrix Membrane for Humic Acid Removal: Influence of Different Types of TiO2 on Membrane Morphology and
M
Performance, IJCEA 3 (2012) 374-379.
d
12. J.F. Li, Z.L. Xu, H. Yang, L.Y. Yu, Effect of TiO2 Nanoparticles on the Surface
4725–4732.
te
Morphology and Performance of Microporous PES Membrane, Appl Surf Sci 255 (2009)
Ac ce p
13. H. Abdallah, A.F. Moustafa, A. A. AlAnezi , H. E. M. El-Sayed, Performance of a newly developed titanium oxide nanotubes/polyethersulfone blend membrane for water desalination using vacuum membrane distillation, Desalination 346 (2014) 30–36. 14. H. Cong, M. Radosz, B. F. Towler, Y. Shen, Polymer-inorganic nanocomposite membranes for gas separation, Sep. Purif. Technol. 55 (2007) 281–291. 15. T. S. Chung, L.Y. Jiang, Y. Li, S. Kulprathipanja, Mixed matrix membranes (MMMs) comprising organic polymers with dispersed inorganic fillers for gas separation, Prog. Polym. Sci. 32 (2007) 483–507.
20 Page 24 of 38
16. S. Y. Kwak, S. H. Kim, Hybrid organic/inorganic reverse osmosis (RO) membrane for bactericidal anti-fouling. 1. Preparation and characterization of TiO2 nanoparticle selfassembled aromatic polyamide thin-film-composite (TFC) membrane, Environ. Sci.
ip t
Technol. 35 (2001) 2388–2394.
17. S. H. Kim, S. Y. Kwak, B. H. Sohn, T. H. Park, Design of TiO2 nanoparticle self
cr
assembled aromatic polyamide thin-film-composite (TFC) membrane as an approach to
us
solve biofouling problem, J. Membr. Sci. 211 (2003) 157–165.
18. M. L. Luo, J.Q. Zhao, W. Tang, C.S. Pu, Hydrophilic modification of poly(ether sulfone)
an
ultrafiltration membrane surface by self-assembly of TiO2 nanoparticles, Appl. Surf. Sci. 249 (2005) 76–84.
M
19. T. H. Bae, T.M. Tak, Effect of TiO2 nanoparticles on fouling mitigation of ultrafiltration
d
membranes for activated sludge filtration, J. Membr. Sci. 249 (2005) 1–8.
te
20. T. H. Bae, T.M. Tak, Preparation of TiO2 self-assembled polymeric nanocomposite membranes and examination of their fouling mitigation effects in a membrane bioreactor
Ac ce p
system, J. Membr. Sci. 266 (2005) 1–5. 21. T. H. Bae, I.C. Kim, T.M. Tak, Preparation and characterization of fouling-resistant TiO2 self-assembled nanocomposite membranes, J. Membr. Sci. 275 (2006) 1–5. 22. J. Mo, S.H. Son, J. Jegal, J. Kim, Y.H. Lee, Preparation and characterization of polyamide nanofiltration composite membranes with TiO2 layers chemically connected to the membrane surface, J. Appl. Polym. Sci. 105 (2007) 1267–1274. 23. S. S. Madaeni, N. Ghaemi, Characterization of self-cleaning RO membranes coated with TiO2 particles under UV irradiation, J. Membr. Sci. 303 (2007) 221–233.
21 Page 25 of 38
24. J. H. Li, Y.Y. Xu, L. P. Zhu, J. -H. Wang, C. -H. Du, Fabrication and characterization of a novel TiO2 nanoparticle self-assembly membrane with improved fouling resistance, J. Membr. Sci. 326 (2009) 659–666.
ip t
25. X. L. Wang, H. J. Qian, L. J. Chen, Z. Yuan Lu, Z. Sheng Li, Dissipative particle dynamics simulation on the polymer membrane formation by immersion precipitation, J.
cr
Membr. Sci. 311 (2008) 251–258.
us
26. K. Wa, D. Leea, P. K. Chanb, X. Feng, Morphology development and characterization of the phase-separated structure resulting from the thermal-induced phase separation
an
phenomenon in polymer solutions under a temperature gradient, Chem. Eng. Sci. 59 (2004) 1491 – 1504.
d
I.RE.CH.E. 4 (2012) 455-465.
M
27. H. Abdallah, S. S. Ali, Themodynamic modeling of PES/CA Blend Membrane Prepration,
te
28. Z. Q. Huang, K. Chen, S.N. Li, X.T. Yin, Z. Zhang, H.T. Xu, Effect of ferrosoferric oxide content on the performances of polysulfone–ferrosoferric oxide ultrafiltration membranes,
Ac ce p
J. Membr. Sci. 315 (2008) 164–171.
29. J. H. Choi, J. Jegal, W. N. Kim, Fabrication and characterization of multi-walled carbon nanotubes/polymer blend membranes, J. Membr. Sci. 284 (2006) 406–415. 30. E. Celik, H. Park, H. Choi, H. Choi, Carbon nanotube blended polyethersulfone membranes for fouling control in water treatment, Water Res. 45 (2011) 274–282. 31. V. Vatanpour, S. S. Madaeni, R. Moradian, S. Zinadini, B. Astinchap, Fabrication and characterization of novel antifouling nanofiltration membrane prepared from oxidized multiwalled carbon nanotube/Polyethersulfonenanocomposite, J. Membr. Sci. 375 (2011) 284–294.
22 Page 26 of 38
32. S. S. Madaeni, S. Zinadini, V. Vatanpour, A new approach to improve antifouling property of PVDF membrane using in situ polymerization of PAA functionalized TiO2 nanoparticles, J. Membr. Sci. 380 (2011) 155–162.
ip t
33. K. Ebert, D. Fritsch, J. Koll, C. Tjahjawiguna, Influence of inorganic fillers on the compaction behavior of porous polymer based membranes, J. Membr. Sci. 233 (2004) 71–
cr
78.
us
34. T. H. Bae, T. M. Tak, Effect of TiO2 nanoparticles on fouling mitigation of ultrafiltration membranes for activated sludge filtration, J. Membr. Sci. 249 (2005) 1–8.
an
35. M. Shaban, H. AbdAllah, L. Said, H. S. Hamdy, A. Abdel khalek,Fabrication of PES/ TiO2 nanotubes reverse osmosis (RO) membranes, J. Chem. Acta. 2 (2013) 59-61.
M
36. M. S. El-Bourawi, Z. Ding, R. Ma, and M. Khayet, A framework for better understanding
d
membrane distillation separation process, J. Membr. Sci. l (2006) 4–29.
te
37. Y. Lai, R.Y. Chen, Preparation and properties of vinyl acetate-grafted nylon 6 membranes by using homografting method, J. Membr. Sci. 66 (1992) 169–178.
Ac ce p
38. Q. Zheng, P. Wang, Y. Yang, D. Cui, The relationship between porosity and kinetics parameter of membrane formation in PSF ultrafiltration membrane, J. Membr. Sci. 286 (2006) 7–11.
39. L. P. Bhausaheb, B. P. Saroj, M. A. Rajendra, and G. Mahendra, The heat and mass transfer phenomena in vacuum membrane distillation for desalination, Int. J. Chem and Biom. Eng. 3 (2010) 33–38. 40. J. I. Mengual, M. Khayet, and M. P. Godino, Heat and mass transfer in vacuum membrane distillation, Int. J. Heat Mass Transfer. 47(2004) 865–875.
23 Page 27 of 38
41. A. Alkhudhiri, N. Darwish , N. Hilal, Membrane distillation: A comprehensive review, Desalination. 287 (2012) 2–18. 42. A. Burgoyne and M. M. Vahdati, Permeate flux modeling of membrane distillation,
ip t
Filtrat. Separ. 36 (1999) 49–53.
43. M. J. Han, S.T. Nam, Thermodynamic and rheological variation in polysulfone solution
cr
by PVP and its effect in the preparation of phase inversion membrane. J. Membr. Sci. 202
us
(2002) 55-61.
44. M. Amirilargani, M. Sadrzadeh, T. Mohammadi, Synthesis and characterization of
an
polyethersulfone membranes. J. Polym. Res. 17 (2010) 363-377. 45. V. Vatanpour , S. S. Madaeni , R. Moradian , S.
Zinadini , B. Astinchap, Novel
M
antibifoulingnanofiltrationpolyethersulfone membrane fabricated from embedding TiO2
d
coated multiwalled carbon nanotubes, Sep. Purif. Technol. 90 (2012) 69–82.
te
46. J. P. Mericq, S. Laborie, C. Cabassud, Vacuum membrane distillation for an integrated seawater desalination process, Desalin. Water. Treat. 9 (2009) 293–302.
Ac ce p
47. M. Padakia, A. M. Isloora, J. Fernandesb, K. N. Prabhuc, New polypropylene supported chitosan NF-membrane for desalination application, Desalination, 280 (2011) 419-423. 48. M. Padakia, A.M. Isloora, K.K. Nagarajab, H.S. Nagarajab, M. Pattabic, Conversion of microfiltration membrane into nanofiltration membrane by vapour phase deposition of aluminum for desalination application. Desalination. 274 (2011) 177-181. 49. J. P. Mericq, S. Laborie, C. Cabassud, Vacuum membrane distillation for an integrated seawater desalination process, Desalin. Water. Treat. 9 (2009) 293–302. 50. J. P.Mericq, S Laborie, C. Cabassud, Vacuum membrane distillation of sea water reverse osmosis brines, water res, 44 (18),(2010) 5260-5273.
24 Page 28 of 38
Figure captain
Figure captain
Fig.1 . Schematic diagram of the hydrothermal synthesis method of TiO2NTs.
ip t
Fig 2. Schematic of MD experimental laboratory setup
compositions.
us
Fig. 4. Mechanical properties of PES/TiO2NTs blend membranes
cr
Fig.3. The cross-sectional SEM images of membranes with different blend PES/TiO2NTs
an
Fig. 5. Average flux of various blend TiO2NTs /PES membranes during VMD process. Fig. 6. Average salt rejection% of various blends PES/TiO2NTs membranes during VMD
M
process.
Fig. 7. Comparison between experimental and modeled permeate flux for synthetic seawater
d
with different feed temperature using 0.53% PES/TiO2NTs blend membrane.
te
Fig. 8. Comparison between experimental and modeled permeate flux for synthetic seawater
Ac ce p
with different vacuum pressure using 0.53% PES/TiO2NTs blend membranes Fig. 9. Temperature polarization coefficient at feed salt concentration 8000 mg/l, feed flow rate 12 ml/s and vacuum pressure 200 mbar.
Page 29 of 38
Figure
d
M
an
us
cr
ip t
Figures
Ac ce p
te
Fig.1 . Schematic diagram of the hydrothermal synthesis method of TiO 2NTs.
Fig 2. Schematic of MD experimental laboratory setup
Page 30 of 38
ip t cr us an M d
Ac ce p
te
Fig.3. The cross-sectional SEM images of membranes with different blend PES/TiO2NTs compositions.
Page 31 of 38
ip t cr us an
M
Fig. 4. Mechanical properties of PES/TiO2NTs blend membranes
69
d te
63
Ac ce p
2
Flux (Kg/m .h)
66
TiO2NTs 0.18% TiO2NTs 0.36% TiO2NTs 0.53% TiO2NTs 0.85% bare PES
15 10 5
0
50
100
150
200
250
300
Time (min)
Fig. 5. Average flux of various blend TiO2NTs /PES membranes during VMD process.
Page 32 of 38
97
ip t
95 4.8
cr
TiO2NTs 0.18% TiO2NTs 0.36% TiO2NTs 0.53% TiO2NTs 0.85% bare PES
4.4 4.0 3.6 50
100
150
Time (min)
200
250
300
an
0
us
Salt Rejection %
96
Ac ce p
te
d
M
Fig. 6. Average salt rejection% of various blends PES/TiO2NTs membranes during VMD process.
Fig. 7. Comparison between experimental and modeled permeate flux for synthetic seawater with different feed temperature using 0.53% PES/TiO2NTs blend membrane.
Page 33 of 38
ip t cr us an M
Fig. 8. Comparison between experimental and modeled permeate flux for synthetic seawater with different vacuum
Ac ce p
te
d
pressure using 0.53% PES/TiO2NTs blend membranes
Fig. 9. Temperature polarization coefficient at feed salt concentration 8000 mg/l, feed flow rate 12 ml/s and vacuum pressure 200 mbar.
Page 34 of 38
Page 35 of 38
d
te
Ac ce p us
an
M
cr
ip t
Table
Tables
Table 1
ip t
Compositions of casting solution for the preparation of nanocomposite membranes PES (wt%)
SDS (wt%)
TiO2NTs (wt%)
NMP (wt%)
Pure PES
20
-
-
80
Blend PES/
20
0.5
20
us
79.1
0.5
0.35
79.1
0.5
0.53
79.1
0.5
0.85
79.1
M
Blend PES/
0.18
an
TiO2NTs (0.18%)
cr
Membrane
Blend PES/
20
Ac ce p
TiO2NTs (0.53%)
te
d
TiO2NTs (0.35%)
Blend PES/
20
TiO2NTs (0.85%)
Page 36 of 38
Table 2 Membrane porosity and contact angle of bare PES and PES/TiO2NTs blend membranes Average pore
Total pore
Contact
diameter (µm)
area m2/gm
angle 75°
61
0.0035
455.7
0.35%
73
0.0031
644.2
0.53%
79.4
0.0023
0.85%
91
0.00201
an
0.18%
65°
790.4
55°
920.1
45°
298.13
85°
M
photos
bare PES
Ac ce p
te
d
Membrane wettability
ip t
Porosity (%)
cr
%
us
TiO2NTs
27.7
0.032
Page 37 of 38
Table 3 Comparison of the performance of different membranes Salt
Fluxing
Synthetic NaCl
rejection
Kg/m2.h
solution (ppm)
%
CA-CNT membrane
1000
69.4%
PS/TFC membrane
2000
98.8%
Polypropylene/chitosan
3500
d
3500
8000
Ac ce p
PES/ TiO2 nanotubes
cr us 89.25
[44]
35
[45]
42.22%
16.4
[46]
97%
18.2
This work
38.42 %
te
PS/Al metal membrane
[43]
M
membrane
19.57
References
ip t
Feed conc.
an
*Membrane
membrane
*Where CA-CNT is Carbon nanotubes-cellulose acetate nanocomposite membrane, PS/TFC membrane is thin film composite polysulfone membrane, PS/Al metal is composite membrane of vapor deposition of aluminum metal on polysulfone membrane.
Page 38 of 38
S0263-8762(14)00485-7 http://dx.doi.org/doi:10.1016/j.cherd.2014.11.008 CHERD 1739
To appear in: Received date: Revised date: Accepted date:
29-3-2014 9-11-2014 15-11-2014
Please cite this article as: Shaban, M., AbdAllah, H., Said, L., Hamdy, H.S., khalek, A.A.,Titanium dioxide nanotubes embedded mixed matrix PES membranescharacterization and membrane performance, Chemical Engineering Research and Design (2014), http://dx.doi.org/10.1016/j.cherd.2014.11.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
abstract
Titanium dioxide nanotubes embedded mixed matrix PES membranes characterization and membrane performance
1
ip t
Mohamed Shaban1, *Heba AbdAllah2, Lamiaa Said1,3, Hany S. Hamdy1, Ahmed Abdel khalek3 Nanophotonics and Applications (NPA)Lab, Department of Physics, Faculty of Science, Beni-
Engineering Research Division, Chemical Engineering & Pilot Plant Department, National
us
2
cr
Suef University, Beni-Suef 62514, Egypt
Research Center (NRC), Dokki 12311, Cairo, Egypt
Department of Chemistry, Faculty of Science, Beni-Suef University, Beni-Suef 62111, Egypt
an
3
M
*Corresponding author e-mail: [email protected]
d
ABSTRACT
te
This study describes the preparation, characterization, and evaluation of performance of blend polyethersulfone / Titanium dioxide nanotubes (PES/TiO2NTs) membranes. TiO2NTs were
Ac ce p
synthesized via hydrothermal process and used in preparation of blend PES/TiO2NTs membranes by phase inversion process. The effects of embedding TiO2 nanotubes on membrane morphology, mechanical properties and performance were presented. The scanning electron microscopy (SEM) images displayed a typical asymmetric membrane structure with a dense top layer, a porous sublayer, and fully developed macropores at the bottom of PES/TiO2NTs membranes due to the migration of TiO2NTs to membrane surface during the phase inversion process. Contact angle measurements indicated that the hydrophilicity of the membrane was improved by adding TiO2NTs. Using vacuum membrane distillation (VMD) application, the results indicated that the membrane performance was improved, particularly, permeate flux and
Page 1 of 38
salt rejection %. The mathematical model of VMD was applied to determine the highest performance of the membrane at the optimum conditions. The experimental results indicated that using 0.53 % TiO2NTs in membrane preparation solution; it was possible to obtain NF/RO blend
ip t
membrane with high performance, where the salt rejection % increased to 97% with permeate
Titanium
dioxide
nanotubes;
Blend
membrane;
Membrane
performance;
us
Keywords:
cr
flux 18.2 Kg/m2.h. These results agree well the results of the VMD model.
Ac ce p
te
d
M
an
Mathematical model.
Page 2 of 38
Graphical Abstract (for review)
Titanium dioxide nanotubes embedded mixed matrix PES membranes characterization and membrane performance Mohamed Shaban1, *Heba AbdAllah2, Lamiaa Said1,3, Hany S. Hamdy1, Ahmed Abdel khalek3 Nanophotonics and Applications (NPA)Lab, Department of Physics, Faculty of Science, BeniSuef University, Beni-Suef 62514, Egypt 2 Engineering Research Division, Chemical Engineering & Pilot Plant Department, National Research Center (NRC), Dokki 12311, Cairo, Egypt 3 Department of Chemistry, Faculty of Science, Beni-Suef University, Beni-Suef 62111, Egypt *Corresponding author e-mail: [email protected]
us
cr
ip t
1
Ac ce p
te
d
M
an
Graphical abstract
Page 3 of 38
*Research Highlights
Highlights
te
d
M
an
us
cr
ip t
Synthesis of TiO2NTs via hydrothermal process. Preparations of PES/ TiO2NTs blend membrane using the wet phase inversion method. The membrane morphological, structural, mechanical, and hydrophilic properties are studied. PES/TiO2NTs blend membrane provided a good membrane performance using VMD system. The predicted results from mathematical VMD model agree well with the experimental results.
Ac ce p
1. 2. 3. 4. 5.
Page 4 of 38
*Manuscript
Titanium dioxide nanotubes embedded mixed matrix PES
ip t
membranescharacterization and membrane performance
Mohamed Shaban1, *Heba AbdAllah2, Lamiaa Said1,3, Hany S. Hamdy1, Ahmed Abdel khalek3 1
University, Beni-Suef 62514, Egypt
Engineering Research Division, Chemical Engineering& Pilot Plant Department, National
us
2
cr
Nanophotonics and Applications (NPA)Lab, Department of Physics, Faculty of Science, Beni-Suef
an
Research Center (NRC), Dokki 12311, Cairo, Egypt 3
Department of Chemistry, Faculty of Science, Beni-Suef University, Beni-Suef 62111, Egypt author e-mail:[email protected]
M
*Corresponding
te
d
1. Introduction
Ac ce p
Nanomaterials have been used in membrane fabrication to produce membranes with desirable structure due to their large specific surface area [1, 2]. Using nanomaterials, the membrane performance can be improved and lead to high selectivity and permeability. In addition, the hydrophilicity, strength, stiffness, water permeability and antifouling properties of the membrane can be enhanced by introducing inorganic nanomaterials into membrane matrix [3-7]. Titanium dioxide (TiO2) is one of the most important semiconductors in our daily life for its unique properties including superior photo-reactivity, superhydrophilicity, nontoxicity, longterm stability, and low cost [8, 9]. In the recent years, many researchers focused on the preparation of TiO2 nano-inorganic mixed matrix membranes to improve membrane performance by reducing fouling and increasing hydrophilicity [10–24]. 1 Page 5 of 38
Blending of polymers is a technical way for providing polymers with set of desired properties at the lowest price such as a combination of strength and toughness, impact strength or solvent resistance. Blending also benefits the industrialist by offering improved process ability,
ip t
rapid formulation changes, deposit flexibility, product uniformity, and high yield [25].
Phase inversion method is one of the most important and popular techniques for
cr
manufacturing many functional polymeric materials that are widely used in engineering
us
applications [26, 27]. In addition, the antifouling properties for the membrane can be modified by blending with inorganic nanoparticles such as Al2O3, SiO2, TiO2, Fe3O4, and carbon
an
nanotubes [28–31]. In recent years, TiO2 nanoparticles blend membranes have been fabricated by
with polymeric casting solution [14, 33].
M
precipitating the nanoparticles on the surface of the porous membranes [3, 32] or blending them
d
The addition of nanotubes such as carbon (CNTs) or titanium dioxide (TiO2) in
te
preparation of polymeric membrane can provide high hydrophilicity, good chemical stability, innocuity, and high-surface area [34].
Because of nanotube materials can reinforce the
Ac ce p
polymeric membrane as a result of their high-aspect ratio and high in axis strength. In addition, titanium dioxide nanotubes (TiO2NTs) can be synthesized by a solution chemical method to form unique open-end nanotubular morphology with typically 3-5 nm and 10-13 nm in inner and outer diameters, respectively [35]. As a consequence of the physical-chemical characteristics, mutual and synergy combination of it slow-dimensional nanostructure, membrane performances can be enhanced by the addition of optimized ratio of TiO2NTs during the preparation process [35]. However, there are no detailed studies for utilizing titanium dioxide nanotubes TiO2NTs to modify the membrane properties and performances, which could solve the problems of fouling and increase the lifetime of the membrane [7-13]. In addition, TiO2NTs could be used to produce
2 Page 6 of 38
the next generation of membranes with high selectivity, reasonable flux, low fouling, and enhanced mechanical properties due to remarkable physical and chemical properties. So TiO2NTs could play a great role for improving the membrane performance due to the excellent
ip t
properties such as porosity, thermal stability, high mechanical strength and high-surface area [7, 12, 13, 20 ].
cr
The hydrophobic porous membranes are preferable in membrane distillation systems,
us
which will resist wetting and oppose liquid water from entering the membrane pores. Then the hydraulic pressure of the bulk water on either sides of the membrane can overcome the inherent
an
surface tension and allow feed water to pass through the membrane .Also, the salt rejection increases by increasing the hydrophobicity of the membrane [36]. But, the fouling by proteins
M
and other organic matter is commonly attributed to the hydrophobic nature of membrane
d
materials that leads to a high interfacial energy with water-rich media, which is decreased upon
te
biomolecules adsorption. So, the suggested structure of the membrane that used in this research is different; it isn’t hydrophobic porous membrane, but it is asymmetric dense membrane with
Ac ce p
adopted hydrophilic properties. The hydrophilic properties of suggested membrane will be adopted to improve the membrane antifouling properties by introducing the titanium dioxide (TiO2) nanotubes at different percentage into the membrane matrix. This is highly expected to improve the strength and stiffness, the water permeability, and the antifouling properties of membrane. Also, the salt rejection % is expected to increase due to the dense layer which is the separation layer [12, 13]. In this study, the fabrications of PES/ TiO2NTs blend membrane by mixing of the polymer with different percentage of TiO2 nanotubes have been investigated. The membranes were fabricated by the wet phase inversion method. The membrane structure and properties were
3 Page 7 of 38
characterized using SEM, mechanical testing system, and water contact angle measurements setup. Desalination of brackish water was applied using the prepared blend membrane by vacuum membrane distillation process. Moreover, simple theoretical model of VMD was used to
ip t
determine the optimum parameters that can effect on the prepared membrane performance.
us
cr
2. Experimental
an
2.1 Materials
M
Analytical grade N, methyl-2-pyrrolidone (NMP) as a solvent and polyethersulfone (PES ultrason E6020P with MW= 58,000 g/mol) were supplied by BASF Company (Germany).
d
Sodium dodecyl sulphate (SDS) was obtained from Merck, Germany. Commercial TiO2 anatase
te
powder and HCl were purchased from Merck and NaOH pellets were purchased from oxford. All chemicals used in the experiments were of reagent grade. Commercial NaCl was used in
Ac ce p
desalination experiments.
2.2 Preparation of TiO2 nanotubes by hydrothermal method Hydrothermal synthesis of titanium dioxide nanotubes was carried out using a 1L autoclave to provide closed reaction environment and the slightly higher pressure during the process. 4 g of commercial anatase TiO2 powder was added to 400 ml of 10 M NaOH. The mixture was stirred at ambient temperature for 30 min. Thereafter, the mixture was heated in a 1 L Teflon-lined autoclave at 150°C [35]. The white product was filtered and washed with 1N HCl and deionized water. Then the sample was dried at 80oC for 4h, after that the produced TiO2NTs 4 Page 8 of 38
post-treated at 500°C for 3 h [35]. Fig.1 illustrates a schematic diagram of the hydrothermal
2.3 Fabrication of asymmetric PES/TiO2NTsblend membranes
ip t
synthesis method of titanium dioxide nanotubes.
The asymmetric PES /TiO2NTs blend membranes were fabricated by phase inversion
cr
induced by immersion precipitation method using casting solutions containing PES (20 wt.%),
us
0.5% sodium dodecyl sulphate (SDS) and the proper amount of TiO2NTs powder in N,methyl-2pyrrolidone (NMP) as solvent. The compositions of casting solutions for all membranes are
an
depicted in Table 1.
The specific percentages of TiO2NTs (0.18, 0.35, 0.53 and0.85%) were added into
M
solution of 0.5% sodium dodecyl sulphate (SDS) and the solution was stirred for 3 h. The casting
d
solution was prepared by dispersing 0.9% solution of TiO2NTs with sodium dodecyl sulphate
te
into NMP solvent with respect to casting solution. After dispersing TiO2NTs /SDS solution in a solvent, PES was dissolved in the dope solution by continuous stirring for 6 h and the polymer
Ac ce p
mixture solution (casting solution) left in the refrigerator for 24 h to remove air bubbles. After that, the casting solution was casted onto a clean glass plate with 150 µm thickness. Subsequently, the glass plate was horizontally immersed into distilled water at room temperature. After primary phase separation and membrane solidification, the membranes were stored in freshly distilled water for 24 h to guarantee the complete phase inversion. Afterward, the membranes were sandwiched between two filters papers for 24 h at room temperature for drying.
2.4 Membrane characterization
5 Page 9 of 38
2.4.1 Membrane morphology Scanning electron microscopy (SEM) was used to observe the morphology of
ip t
PES/TiO2NTs blend membranes, where as the samples for cross-sectional view were coated with gold to provide electrical conductivity. The cross-sectional snapshots of membrane were taken
cr
on a JEOL 5410 SEM and conducted at 10 kV.
us
2.4.2 Mechanical properties
an
Mechanical properties of PES and PES/TiO2NTsblend membraneswere investigated to determine the appropriate percentage of TiO2NTs(wt%) which can be used without affecting the
M
membrane strength. The membrane tensile strength and elongation were measured using mechanical testing system (INSTRON-5500R). The gauge length and width of dumbbell tensile
d
specimens were 6.2 and 0.16 mm, respectively. The specimen of membrane was placed between
te
the grips of the testing machine, and then the tensile strength and elongation were calculated.
Ac ce p
The measurement accuracy is within ± 5%.
2.4.3 Membrane porosity and contact anglemeasurements Membrane porosity plays an important role in describing its performance. In order to evaluate porosity of the PES/TiO2NTs blend membranes, they were initially impregnated with water then weighed after wiping superficial water with filter papers. After that, the wet membranes were placed in an air-circulating oven at 80°C for 24h to be completely dried. Finally, the dry membranes were weighed and the porosity of membranes was calculated using the following equation [37, 38]:
6 Page 10 of 38
W0 W1 V
(1)
Where ɛ is the membrane porosity; WO and W1 are the weights of wet and dry membranes in
ip t
gram, respectively; V= A.δ where A is the membrane surface area in m2 and δ is the membrane thickness in m. In order to minimize the experimental errors, the membrane porosity of each
cr
sample was measured several times, at least 3 times, and the results were reported in average (the
us
error of porosity ±2%). The average pores diameters and the total surface area of the pores were measured using Hg porosimeter (model type No.9810). Distilled water was used for the contact
an
angle (θ) measurement by the sessile drop method. The measurements were carried out four
M
times for each membrane sample and the average values were reported.
2.5 Membrane performance measurements using membrane distillation technique
d
The experiments were carried out on vacuum membrane distillation (VMD) system as
te
shown in Fig. 2. This system contains flat sheet membrane module of three opening for vacuum
Ac ce p
pressure, feeding, and recycling. The feed was continuously flowing to the membrane module from an open feeding tank (glass flask) by peristaltic feeding pump. Vacuum pressure was applied using a vacuum pump. The feed mixture was heated using a hot plate and the temperature was controlled by thermostat of the hot plate and recorded by thermometer. Water condenser was used as a cooling system by recycling cold water using circulation cooling water bath through inlet and outlet in the condenser jacket. Product was collected from downstream of the condenser. The prepared PES/TiO2NTs blend membranes with different TiO2NTspercentages (wt%) were located in stainless steel plate module of 47 mm in diameter. In all experiments, the aqueous feed solution of about 8000 ppm of NaCl in DI water were prepared and continuously fed to the membrane module where the other MD parameterѕ are remained constant at feed flow 7 Page 11 of 38
rate of 12 ml/s, vacuum pressure of
200 mbar and temperature of 65°C. Effect of feed
temperature in the range from 25°C to 65°C and effect of vacuum pressure on permeate side in the range from 200 mbar to 800 mbar were investigated and compared with the predicted results
ip t
from the theoretical VMD model.
V . A.t
(2)
us
Jw
cr
For this system, the water flux, Jw (Kg/m2 h), is given by the following equation:
Where V is the volume of the pure water permeate (m3), A is the effective area of the
an
membrane (m2), which was circular membrane with area 0.0017 m2, ρ is the water density (Kg/m3) and t is the permeation time (h).
M
The concentration of permeate collected was measured using a conductivity meter. In
d
addition, the salt rejection (SR %) was conducted in triplicates for each membrane and the
C f Cp Cf
*100
(3)
Ac ce p
SR%
te
average result was calculated using the following equation:
Where, Cf and Cp are concentrations (mg/l) at feed bulk and permeate, respectively.
2.6 Mathematical model for VMD using PES/TiO2NTs blend membrane A VMD model was developed in order to simulate the VMD desalination process for TiO2NTs /PES blend flat sheet membranes. It allows us to calculate permeate vapor flow through the membrane and to determine the temperature profiles alongside the membrane (in feed side or close to the membrane).VMD process is motivated by both the total pressure difference and the temperature difference between two sides of the membrane. The total pressure difference rises from the vacuum dragged by a vacuum pump in the cold chamber of the membrane module. The 8 Page 12 of 38
permeate vapor is sucked out in the cold condenser by the vacuum pump and condensed [36, 39, 40]. If the hot water flows in the direction tangent over the membrane, the temperature difference at
ip t
two sides of the membrane can provide water vapor pressure gradient within the membrane pores. This driving force produces a mass flux through the membrane. According to this case, the
cr
Knudsen and molecular diffusion are involved in VMD process; the mass transfer resistance
us
caused by molecule-molecule collision can be neglected, and diffusion process within the pores is dominated by Knudsen diffusion. The mass flux related to this mechanism (JK) is given by
8r 3
M Pv 2RT
(4)
M
JK
an
equation (4):
d
Where r is the pore radius, τ is tortuosity factor, M is molecular weight of feeding water,
te
R is Universal gas constant, T is feeding temperature, and ∆Pv is the water vapor pressure
Ac ce p
gradient within the membrane pores. Tortuosity (τ) factor was calculated using the following equation [41]:
(2 ) 2
(5)
The total pressure gradient causes a convective mass transport through the pores, which is known as Poiseuille flow. Because almost no air exists within membrane pores, water vapor in the pores is considered pure gas and thus the total pressure difference acting as the driving force for Poiseuille flow. So, the mass flux related to this mechanism (Jp) is given by equation (6) as follows:
9 Page 13 of 38
Jp
r 2 Pv M Pv 8RTm
(6)
ip t
Where, Pv is water vapor pressure, η is water viscosity, and Tm is the mean temperature. In
(J) can be calculated by summation of two mechanisms [36,39, 40].
us
J=JK+JP
Pv
(7)
(8)
M
M r 2 Pv M 2RT 8RTm
an
Then, 8 r J 3
cr
VMD, these two mechanisms contribute in parallel to the mass transfer, and the total mass flux
M r 2 Pv M 2RT 8RTm
Pfm Ppm
(9)
Ac ce p
8 r J 3
te
membrane is given by equation (9):
d
By integrating equation (8) relative to membrane thickness, the total mass flux through the
Where, Pfm is the water vapor pressure at the membrane surface of the feed side, and Ppm is the pressure of the permeate side which is considered to be equal to the vacuum pressure of the condenser. The non-linear equation (9) was solved using MATLAB SIMULINK software (The Mathworks, Release 2009b). The water vapor pressure at the liquid/vapor interface in Pascal is related with the temperature in K according to the Antoine’s equation [39]:
Pv (T ) exp( 23.1964
3816.44 ) T 46.13
(10)
10 Page 14 of 38
The temperature polarization phenomenon occurs as a result of a temperature difference between the bulk feed temperature (Tb), interfacial temperature (Tim), and the temperature in the
ip t
vacuum membrane side (Tv). This temperature gradient is due to the heat flux through the liquid layer, which is needed to provide the required heat for evaporation at the membrane interface.
cr
The internal temperatures were measured using lab Infrared thermometer TM-939 (made in
us
Taiwan) was purchased from Lutron electronic enterprise company. The temperature variation across the membrane can be described by the temperature polarization factor that is defined as
Tb Tim Tb Tv
te
d
3. Results and discussion
(11)
M
an
[39]:
Ac ce p
3.1 Characterization of PES/TiO2NTs blend membrane 3.1.1 Membrane morphology
Fig. 3 shows cross –sectional SEM images of PES membranes blended with TiO2NTsof different percentages. All membranes showed a typical asymmetric membrane structure with a dense top layer, a porous sublayer, and fully developed macropores at the bottom. Nevertheless, the formation of macropores was suppressed by the addition of TiO2NTs into the membrane structure as shown in Fig. 3 (a-d). Also, using TiO2NTscould adjust the pores distribution according to TiO2NTs%, where the structure was well-designed finger-like sub-layer with dense top layer.
11 Page 15 of 38
These results might be described by the fact that by increasing the amount of TiO2NTs in the casting solution to 0.53 %, the porosity of the synthesized membrane increased due to the enhanced phase separation with TiO2NTs. Further increase in TiO2NTs amount leads to a denser
ip t
structure in the sub-layer due to the delayed phase separation as the viscosity increased, see Fig. 3(c and d) [42-44]. Whereas the blank PES membrane in Fig.3 (e) shows spongy- porous
us
cr
structure of the membrane without any appears of porous finger-like in sub-layer.
3.1.2 Mechanical properties
an
The blend membrane tensile strength and elongation were measured using mechanical testing system. Fig. 4 shows the relation between tensile strength, elongation, and TiO2NTs
M
percentage. This figure presents that the maximum tensile strength (47.5 kg/cm2) and the
d
maximum elongation% (6%) are reached when the membrane blended with 0.53% of TiO2NTs.
te
The delay phase separation is related to solid-liquid de-mixing, where the viscosity of polymer solution is increased by increasing the nanomaterials (TiO2NTs) percentage, which could delay
Ac ce p
the exchange of solvent and nonsolvent. This delay time in the solid-liquid de-mixing during membrane casting leads to initiate a number of microcracks by inorganic fillers (TiO2NTs) [7, 44]. Accordingly, the increasing in TiO2NTs to 0.85% leads to decrease in mechanical properties due to the delayed phase separation during blend membrane casting as a result of high viscosity of polymer solution.
3.1.3 Membrane porosity and contact angle The overall porosity information of the prepared mixed matrix PES/TiO2NTs blend membranes was presented in Table 2. The results of the porosity measurement revealed that all 12 Page 16 of 38
the prepared PES/TiO2NTs blend membranes possessed a good porosity in the range of 61– 91%compared to 27.2% for pure PES membrane. As shown in this Table, the porosity of blend membranes increased as the percentage of TiO2NTs increased. This means that, the use of
ip t
TiO2NTs as a blend material increases the porosity of the membrane internal structures, which also improves the properties of the membrane and may result in an increase in lateral flow rates
cr
through the membranes [45].
us
Contact angle measurement of membranes is considered to be an important parameter for membrane characterization and indirect indication of the hydrophilicity and flux behavior. The
an
hydrophilicity of the PES/TiO2NTs blend membranes was determined based on water contact angle using the sessile drop method. The contact angles were measured several times and then
M
average values were reported. Table 2 illustrates the water contact angles of the mixed matrix
d
PES/TiO2NTs blend membranes and membrane wettability with different percentage of
te
TiO2NTs. As shown in this table, the bare PES membrane had the highest contact angle (85± 2°), corresponding to the lowest hydrophilicity. In the case of blend membranes, lower contact angle
Ac ce p
was observed in comparison with bare PES membrane. In addition, the contact angle decreases from 75 ± 2° to 45 ± 2° as TiO2NTs percentage increased from 0.18 to 0.85 wt.% due to increase of hydrophilicity. Therefore, it was expected that the blend membranes absorb more permeate into the membrane and thus enhance the flux rate [45]. Also, Table (2) illustrates the average pores diameter and total pores area on the membrane surface. The average pores diameters are decreased from 0.0035 to 0.002 µm for the blend membranes as TiO2NTs percentage increased from 0.18 to 0.85 wt% compared with 0.032 µm of bare PES. But, the measured total pores areas of the blend membranes are higher than that of the bare PES. This may be ascribed to the geometrical shape of TiO2NTs, a cylindrical tubular shape as shown in figure (1), which can provide the organization, re-arrangement, and generation of new 13 Page 17 of 38
pores on the membrane surface [21, 26]. Instead of the decrease of the pores sizes, the pores density, total pores area, and membrane porosity are increased on the surface of the blend membrane. This may
explain why the flux of bare PES is higher than the blend membranes whereas the salt rejection
ip t
% of blend membranes is higher than that of the bare PES.
cr
3.2 Membrane performance measurements using membrane distillation technique
us
Average water permeation flux and average salt rejection % of PES/TiO2NTs blend membranes with different percentages of TiO2NTs were measured using vacuum membrane
an
distillation process. The effects of TiO2NTs percentages on water flux and salt rejection % of prepared membranes are shown in Figs. 5 and 6, respectively. The average flux of blend
M
membranes reached a minimum when 0.18wt% TiO2NTs was used, then increased as TiO2NTs%
te
preparation.
d
increased and reached a maximum when 0.85wt% of TiO2NTs was used in the membrane
Figure 5 also indicates that the bare PES provides the highest permeate flux while Fig. 6 rejection percent
Ac ce p
indicates the lowest
of bare PES
compared with
blend PES/
TiO2NTsmembrane. This may be ascribed to the large pores size and the average pore diameter of bare PES (0.032 µm) as shown in Table (2). The flux of the blend membranes decreased after blending with TiO2NTs due to the formation of a dense layer on the top of the membrane, which leads to the decrease of the permeate flux and increase in the salt rejection % compared with bare PES. Fig.6 indicates that the average salt rejection of membranes increased to 97% by addition of TiO2NTs in the casting solution to prepare blend membranes. The variations in hydrophilicity and morphology of blend membrane affect its performance in two different ways. The membrane hydrophilicity increases the flux due to enhance of membrane hydrophilicity and porosity, where 14 Page 18 of 38
the membrane morphology decreases the flux due to a decrease in membrane pore size and an increase in skin layer thickness [46]. Table 3 illustrates a comparison of the performance of different membranes and the
ip t
prepared membranes in this study based on permeate flux and salt rejection. This table indicated that the prepared membrane using TiO2NTs exhibits better membrane performance based on
us
cr
permeate flux and salt rejection at feed concentration of 8000 ppm NaCl.
an
3.3 Mathematical model for VMD using TiO2NTs /PES blend membrane Validation of the VMD model was performed by comparing the experimental and
M
computed values of the permeate flux under the optimum operating conditions for PES/ 0.53% TiO2NTs blend membrane using the synthetic seawater 8000 mg/l.
d
Feed temperature is one of the important parameters that effect on the permeate flux.
te
Comparison between experimental results and predicted results of permeate flux for synthetic
Ac ce p
solution at different feed temperatures is presented in Fig. 7. The experiments were conducted at pressure of 200 mbar, feed temperatures between 25 and 65°C, and feed flow rate of 12 ml/s. This figure shows good agreement between experimental and predicted results. Whereas the increasing of feed temperature leads to increase in permeate flux as a consequence of the rise in vapor pressure of vapor-liquid interface on liquid feed side, which causes a simultaneous increase in the driving force of mass transfer [47]. Fig. 8 shows a comparison between experimental and theoretical results for different permeate pressure from 200 to 800 mbar, where the feed temperature was kept at 65°C. It can be seen that a high-permeate flux with a low-permeate pressure and a high-feed temperature. When permeate pressure varies from 200 to 800 mbar, the permeate flux is more than doubled whereas 15 Page 19 of 38
the specific energy requirement is nearly constant, where the energy required to maintain vacuum pressure is only a small portion of the total energy requirement (less than 2%) [48]. The theoretical predictions from VMD model agreed well with experimental measurements of
ip t
permeate flux.
Temperature polarization coefficient (θ) is the ratio between the temperature close to the
cr
membrane surface calculated by simulations using equation (11) and the experimental
us
temperature measured at the feed side [48-49]. It indicates the durable contribution of the boundary layer on the total heat transfer resistance of the VMD system. If θ ≈ zero for a system,
an
the process is controlled by mass transfer through the membrane and resistance in the liquid phase is negligible but if the value of θ approaches to one, the process is restricted by heat
M
transfer to the membrane surface and the resistance in the membrane is negligible [50]. The
d
calculated values of θ as a function of bulk feed temperature are shown in Fig. 9. The results
te
indicate that θ decreases by increasing bulk feed temperature (Tb), according to that θ decreases from 0.33 at feed temperature 25 ºC to 0.16 at feed temperature 65ºC as the membrane
Ac ce p
permeability and permeate flux increase.
4. Conclusion
The asymmetric PES/TiO2NTs blend membranes were successfully fabricated by phase inversion method, where titanium oxide nanotubes (TiO2NTs) were synthesized by hydrothermal method. The addition of TiO2NTs resulted in higher porosity, lower contact angle, and better hydrophilicity than bare PES. Also, the performance of prepared blend membranes were improved due to the addition of TiO2NTs, where the permeate flux was improved at using 0.85%
16 Page 20 of 38
due to increase of membrane hydrophilicity and the salt rejection % increased to 97% compared with the bare PES which has the lowest salt rejection %. The prepared blend membranes were applied on vacuum membrane distillation system.
ip t
The experiments showed that PES/TiO2NTsblend membranes provided average accumulated flux of 18.2 Kg/m2.h and average salt rejection % reached 97% when 0.85 wt% of TiO2NTs is
cr
used in the membrane preparation. This means that TiO2NTs are good modifier for formation of
us
NF/RO membrane with high reproducibility. Moreover, VMD model was developed in order to simulate the optimum conditions for PES/TiO2NTs blend flat sheet membranes. The predicted
an
results from the model agreed well with the experimental results for PES/0.53 wt%
M
TiO2NTsblend membrane at 200 mbar, 65°C, and flow rateof 12 ml/s.
te
d
Nomenclature
A: Effective area of the membrane (m2)
Ac ce p
Cf: Concentration at feed bulk (mg/L)
Cp: Concentration at permeate (mg/L) J: Mass vapor flux (kg/m2 h)
Mw: Molecular weight of water (kg/kmol) P: Total pressure (Pa)
Pfm: Water vapor pressure at the feed membrane surface (Pa) Ppm: Water vapor pressure at the permeate surface (Pa) ΔPv: Vaporpressuredifference at membranesides (Pa) Pv: Vapor pressure at feed side(Pa) R: Universal gas constant (m3 Pa/mol K) r: Membrane pore radius (m) Tm: Meantemperature(K)
17 Page 21 of 38
Tf: Water feed temperature (K) Tfb: Temperature at the feed bulk side (K) Tpb: Temperature at the permeate bulk side(K)
ip t
Tfm: Temperature at the feed membrane surface(K) Tpm: Temperature at the bulk membrane surface (K) Tim: Interfacial temperature at the membrane surface (k)
cr
t : Permeation time (h).
us
V:Volume of the pure water permeate (m3) Dimensionless Numbers
an
θ: Temperature polarization coefficient (TPC).
M
τ: Tortuosity factor
Greek Letters δ: Membrane thickness (m)
Ac ce p
ρ :Water density (Kg/m3)
te
η : Water vapour viscosity Pa.sec
d
ε: Membraneporosity
Subscripts b: Bulk f : Feed
m: Membrane
VMD: Vacuum Membrane distillation.
18 Page 22 of 38
References 1. J. Kim, S. H. R. Davies, M. J. Baumann, V. V. Tarabara, S. J. Masten, Effect of ozone dosage and hydrodynamic conditions on the permeate flux in a hybrid ozonation-ceramic
2.
ip t
ultrafiltration system treating natural waters, J. Membr. Sci. 311 (2008) 165-172.
L. Brunet , D. Y. Lyon, K. Zodrow, J. C. Rouch, B.Cassat, P. Serp, J. C. Remigy, M. R.
cr
Wiesner, P. J. J. Alvarez, Properties of Membranes Containing Semi-dispersed Carbon
us
Nanotubes, Environ Eng. Sci. 25 (2008) 565-576.
3. V. Vatanpour, S.S. Madaeni, A.R. Khataee, E. Salehi, S. Zinadini, H. A. Monfared, TiO2
an
embedded mixed matrix PES nanocomposite membranes: Influence of different sizes and types of nanoparticles on antifouling and performance, Desalination 292 (2012) 19–29.
M
4. J. Kim, B. Van der Bruggen, The use of nanoparticles in polymeric and ceramic
d
membrane structures: review of manufacturing procedures and performance improvement
5. J.B.
Li,
J.W.
te
for water treatment, Environ. Pollut. 158 (2010) 2335–2349. Zhu,
M.S.
Zheng,
Morphologies
and
properties
of
Ac ce p
poly(phthalazinoneethersulfoneketone) matrix ultrafiltration membranes with entrapped TiO2 nanoparticles, J. Appl. Polym. Sci. 103 (2007) 3623–3629. 6. Z.L. Xu, L.Y. Yu, L.F. Han, Polymer-nanoinorganic particles composite membranes: a brief overview, Front. Chem. Eng. Chin. 3 (2009) 318–329. 7. A. Razmjou, J. Mansouri, V. Chen, The effects of mechanical and chemical modification of TiO2 nanoparticles on the surface chemistry, structure and fouling performance of PES ultrafiltration membranes, J. Membr. Sci. 378 (2011) 73–84. 8. Y. Mansourpanah, S.S. Madaeni, A. Rahimpour, A. Farhadian, A.H. Taheri, Formation of appropriate sites on nanofiltration membrane surface for binding TiO2 photo-catalyst:
19 Page 23 of 38
performance, characterization and fouling-resistant capability, J. Membr. Sci. 330 (2009) 297–306. 9. S. S. Madaeni, N. Ghaemi, A. Alizadeh, M. Joshaghani, Influence of photo-induced
ultrafiltration membranes, Appl. Surf. Sci. 257 (2011) 6175–6180.
ip t
superhydrophilicity of titanium dioxide nanoparticles on the anti-fouling performance of
cr
10. A.R. Khataee, V. Vatanpour, A.R. Amani Ghadim, Decolorization of C.I. Acid Blue 9
us
solution by UV/Nano-TiO2, Fenton, Fenton-like, electro-Fenton and electrocoagulation processes: a comparative study, J. Hazard. Mater. 161 (2009) 1225–1233.
an
11. Y. H. Teow, B. S. Ooi, A. L. Ahmad, J. K. Lim, Mixed-Matrix Membrane for Humic Acid Removal: Influence of Different Types of TiO2 on Membrane Morphology and
M
Performance, IJCEA 3 (2012) 374-379.
d
12. J.F. Li, Z.L. Xu, H. Yang, L.Y. Yu, Effect of TiO2 Nanoparticles on the Surface
4725–4732.
te
Morphology and Performance of Microporous PES Membrane, Appl Surf Sci 255 (2009)
Ac ce p
13. H. Abdallah, A.F. Moustafa, A. A. AlAnezi , H. E. M. El-Sayed, Performance of a newly developed titanium oxide nanotubes/polyethersulfone blend membrane for water desalination using vacuum membrane distillation, Desalination 346 (2014) 30–36. 14. H. Cong, M. Radosz, B. F. Towler, Y. Shen, Polymer-inorganic nanocomposite membranes for gas separation, Sep. Purif. Technol. 55 (2007) 281–291. 15. T. S. Chung, L.Y. Jiang, Y. Li, S. Kulprathipanja, Mixed matrix membranes (MMMs) comprising organic polymers with dispersed inorganic fillers for gas separation, Prog. Polym. Sci. 32 (2007) 483–507.
20 Page 24 of 38
16. S. Y. Kwak, S. H. Kim, Hybrid organic/inorganic reverse osmosis (RO) membrane for bactericidal anti-fouling. 1. Preparation and characterization of TiO2 nanoparticle selfassembled aromatic polyamide thin-film-composite (TFC) membrane, Environ. Sci.
ip t
Technol. 35 (2001) 2388–2394.
17. S. H. Kim, S. Y. Kwak, B. H. Sohn, T. H. Park, Design of TiO2 nanoparticle self
cr
assembled aromatic polyamide thin-film-composite (TFC) membrane as an approach to
us
solve biofouling problem, J. Membr. Sci. 211 (2003) 157–165.
18. M. L. Luo, J.Q. Zhao, W. Tang, C.S. Pu, Hydrophilic modification of poly(ether sulfone)
an
ultrafiltration membrane surface by self-assembly of TiO2 nanoparticles, Appl. Surf. Sci. 249 (2005) 76–84.
M
19. T. H. Bae, T.M. Tak, Effect of TiO2 nanoparticles on fouling mitigation of ultrafiltration
d
membranes for activated sludge filtration, J. Membr. Sci. 249 (2005) 1–8.
te
20. T. H. Bae, T.M. Tak, Preparation of TiO2 self-assembled polymeric nanocomposite membranes and examination of their fouling mitigation effects in a membrane bioreactor
Ac ce p
system, J. Membr. Sci. 266 (2005) 1–5. 21. T. H. Bae, I.C. Kim, T.M. Tak, Preparation and characterization of fouling-resistant TiO2 self-assembled nanocomposite membranes, J. Membr. Sci. 275 (2006) 1–5. 22. J. Mo, S.H. Son, J. Jegal, J. Kim, Y.H. Lee, Preparation and characterization of polyamide nanofiltration composite membranes with TiO2 layers chemically connected to the membrane surface, J. Appl. Polym. Sci. 105 (2007) 1267–1274. 23. S. S. Madaeni, N. Ghaemi, Characterization of self-cleaning RO membranes coated with TiO2 particles under UV irradiation, J. Membr. Sci. 303 (2007) 221–233.
21 Page 25 of 38
24. J. H. Li, Y.Y. Xu, L. P. Zhu, J. -H. Wang, C. -H. Du, Fabrication and characterization of a novel TiO2 nanoparticle self-assembly membrane with improved fouling resistance, J. Membr. Sci. 326 (2009) 659–666.
ip t
25. X. L. Wang, H. J. Qian, L. J. Chen, Z. Yuan Lu, Z. Sheng Li, Dissipative particle dynamics simulation on the polymer membrane formation by immersion precipitation, J.
cr
Membr. Sci. 311 (2008) 251–258.
us
26. K. Wa, D. Leea, P. K. Chanb, X. Feng, Morphology development and characterization of the phase-separated structure resulting from the thermal-induced phase separation
an
phenomenon in polymer solutions under a temperature gradient, Chem. Eng. Sci. 59 (2004) 1491 – 1504.
d
I.RE.CH.E. 4 (2012) 455-465.
M
27. H. Abdallah, S. S. Ali, Themodynamic modeling of PES/CA Blend Membrane Prepration,
te
28. Z. Q. Huang, K. Chen, S.N. Li, X.T. Yin, Z. Zhang, H.T. Xu, Effect of ferrosoferric oxide content on the performances of polysulfone–ferrosoferric oxide ultrafiltration membranes,
Ac ce p
J. Membr. Sci. 315 (2008) 164–171.
29. J. H. Choi, J. Jegal, W. N. Kim, Fabrication and characterization of multi-walled carbon nanotubes/polymer blend membranes, J. Membr. Sci. 284 (2006) 406–415. 30. E. Celik, H. Park, H. Choi, H. Choi, Carbon nanotube blended polyethersulfone membranes for fouling control in water treatment, Water Res. 45 (2011) 274–282. 31. V. Vatanpour, S. S. Madaeni, R. Moradian, S. Zinadini, B. Astinchap, Fabrication and characterization of novel antifouling nanofiltration membrane prepared from oxidized multiwalled carbon nanotube/Polyethersulfonenanocomposite, J. Membr. Sci. 375 (2011) 284–294.
22 Page 26 of 38
32. S. S. Madaeni, S. Zinadini, V. Vatanpour, A new approach to improve antifouling property of PVDF membrane using in situ polymerization of PAA functionalized TiO2 nanoparticles, J. Membr. Sci. 380 (2011) 155–162.
ip t
33. K. Ebert, D. Fritsch, J. Koll, C. Tjahjawiguna, Influence of inorganic fillers on the compaction behavior of porous polymer based membranes, J. Membr. Sci. 233 (2004) 71–
cr
78.
us
34. T. H. Bae, T. M. Tak, Effect of TiO2 nanoparticles on fouling mitigation of ultrafiltration membranes for activated sludge filtration, J. Membr. Sci. 249 (2005) 1–8.
an
35. M. Shaban, H. AbdAllah, L. Said, H. S. Hamdy, A. Abdel khalek,Fabrication of PES/ TiO2 nanotubes reverse osmosis (RO) membranes, J. Chem. Acta. 2 (2013) 59-61.
M
36. M. S. El-Bourawi, Z. Ding, R. Ma, and M. Khayet, A framework for better understanding
d
membrane distillation separation process, J. Membr. Sci. l (2006) 4–29.
te
37. Y. Lai, R.Y. Chen, Preparation and properties of vinyl acetate-grafted nylon 6 membranes by using homografting method, J. Membr. Sci. 66 (1992) 169–178.
Ac ce p
38. Q. Zheng, P. Wang, Y. Yang, D. Cui, The relationship between porosity and kinetics parameter of membrane formation in PSF ultrafiltration membrane, J. Membr. Sci. 286 (2006) 7–11.
39. L. P. Bhausaheb, B. P. Saroj, M. A. Rajendra, and G. Mahendra, The heat and mass transfer phenomena in vacuum membrane distillation for desalination, Int. J. Chem and Biom. Eng. 3 (2010) 33–38. 40. J. I. Mengual, M. Khayet, and M. P. Godino, Heat and mass transfer in vacuum membrane distillation, Int. J. Heat Mass Transfer. 47(2004) 865–875.
23 Page 27 of 38
41. A. Alkhudhiri, N. Darwish , N. Hilal, Membrane distillation: A comprehensive review, Desalination. 287 (2012) 2–18. 42. A. Burgoyne and M. M. Vahdati, Permeate flux modeling of membrane distillation,
ip t
Filtrat. Separ. 36 (1999) 49–53.
43. M. J. Han, S.T. Nam, Thermodynamic and rheological variation in polysulfone solution
cr
by PVP and its effect in the preparation of phase inversion membrane. J. Membr. Sci. 202
us
(2002) 55-61.
44. M. Amirilargani, M. Sadrzadeh, T. Mohammadi, Synthesis and characterization of
an
polyethersulfone membranes. J. Polym. Res. 17 (2010) 363-377. 45. V. Vatanpour , S. S. Madaeni , R. Moradian , S.
Zinadini , B. Astinchap, Novel
M
antibifoulingnanofiltrationpolyethersulfone membrane fabricated from embedding TiO2
d
coated multiwalled carbon nanotubes, Sep. Purif. Technol. 90 (2012) 69–82.
te
46. J. P. Mericq, S. Laborie, C. Cabassud, Vacuum membrane distillation for an integrated seawater desalination process, Desalin. Water. Treat. 9 (2009) 293–302.
Ac ce p
47. M. Padakia, A. M. Isloora, J. Fernandesb, K. N. Prabhuc, New polypropylene supported chitosan NF-membrane for desalination application, Desalination, 280 (2011) 419-423. 48. M. Padakia, A.M. Isloora, K.K. Nagarajab, H.S. Nagarajab, M. Pattabic, Conversion of microfiltration membrane into nanofiltration membrane by vapour phase deposition of aluminum for desalination application. Desalination. 274 (2011) 177-181. 49. J. P. Mericq, S. Laborie, C. Cabassud, Vacuum membrane distillation for an integrated seawater desalination process, Desalin. Water. Treat. 9 (2009) 293–302. 50. J. P.Mericq, S Laborie, C. Cabassud, Vacuum membrane distillation of sea water reverse osmosis brines, water res, 44 (18),(2010) 5260-5273.
24 Page 28 of 38
Figure captain
Figure captain
Fig.1 . Schematic diagram of the hydrothermal synthesis method of TiO2NTs.
ip t
Fig 2. Schematic of MD experimental laboratory setup
compositions.
us
Fig. 4. Mechanical properties of PES/TiO2NTs blend membranes
cr
Fig.3. The cross-sectional SEM images of membranes with different blend PES/TiO2NTs
an
Fig. 5. Average flux of various blend TiO2NTs /PES membranes during VMD process. Fig. 6. Average salt rejection% of various blends PES/TiO2NTs membranes during VMD
M
process.
Fig. 7. Comparison between experimental and modeled permeate flux for synthetic seawater
d
with different feed temperature using 0.53% PES/TiO2NTs blend membrane.
te
Fig. 8. Comparison between experimental and modeled permeate flux for synthetic seawater
Ac ce p
with different vacuum pressure using 0.53% PES/TiO2NTs blend membranes Fig. 9. Temperature polarization coefficient at feed salt concentration 8000 mg/l, feed flow rate 12 ml/s and vacuum pressure 200 mbar.
Page 29 of 38
Figure
d
M
an
us
cr
ip t
Figures
Ac ce p
te
Fig.1 . Schematic diagram of the hydrothermal synthesis method of TiO 2NTs.
Fig 2. Schematic of MD experimental laboratory setup
Page 30 of 38
ip t cr us an M d
Ac ce p
te
Fig.3. The cross-sectional SEM images of membranes with different blend PES/TiO2NTs compositions.
Page 31 of 38
ip t cr us an
M
Fig. 4. Mechanical properties of PES/TiO2NTs blend membranes
69
d te
63
Ac ce p
2
Flux (Kg/m .h)
66
TiO2NTs 0.18% TiO2NTs 0.36% TiO2NTs 0.53% TiO2NTs 0.85% bare PES
15 10 5
0
50
100
150
200
250
300
Time (min)
Fig. 5. Average flux of various blend TiO2NTs /PES membranes during VMD process.
Page 32 of 38
97
ip t
95 4.8
cr
TiO2NTs 0.18% TiO2NTs 0.36% TiO2NTs 0.53% TiO2NTs 0.85% bare PES
4.4 4.0 3.6 50
100
150
Time (min)
200
250
300
an
0
us
Salt Rejection %
96
Ac ce p
te
d
M
Fig. 6. Average salt rejection% of various blends PES/TiO2NTs membranes during VMD process.
Fig. 7. Comparison between experimental and modeled permeate flux for synthetic seawater with different feed temperature using 0.53% PES/TiO2NTs blend membrane.
Page 33 of 38
ip t cr us an M
Fig. 8. Comparison between experimental and modeled permeate flux for synthetic seawater with different vacuum
Ac ce p
te
d
pressure using 0.53% PES/TiO2NTs blend membranes
Fig. 9. Temperature polarization coefficient at feed salt concentration 8000 mg/l, feed flow rate 12 ml/s and vacuum pressure 200 mbar.
Page 34 of 38
Page 35 of 38
d
te
Ac ce p us
an
M
cr
ip t
Table
Tables
Table 1
ip t
Compositions of casting solution for the preparation of nanocomposite membranes PES (wt%)
SDS (wt%)
TiO2NTs (wt%)
NMP (wt%)
Pure PES
20
-
-
80
Blend PES/
20
0.5
20
us
79.1
0.5
0.35
79.1
0.5
0.53
79.1
0.5
0.85
79.1
M
Blend PES/
0.18
an
TiO2NTs (0.18%)
cr
Membrane
Blend PES/
20
Ac ce p
TiO2NTs (0.53%)
te
d
TiO2NTs (0.35%)
Blend PES/
20
TiO2NTs (0.85%)
Page 36 of 38
Table 2 Membrane porosity and contact angle of bare PES and PES/TiO2NTs blend membranes Average pore
Total pore
Contact
diameter (µm)
area m2/gm
angle 75°
61
0.0035
455.7
0.35%
73
0.0031
644.2
0.53%
79.4
0.0023
0.85%
91
0.00201
an
0.18%
65°
790.4
55°
920.1
45°
298.13
85°
M
photos
bare PES
Ac ce p
te
d
Membrane wettability
ip t
Porosity (%)
cr
%
us
TiO2NTs
27.7
0.032
Page 37 of 38
Table 3 Comparison of the performance of different membranes Salt
Fluxing
Synthetic NaCl
rejection
Kg/m2.h
solution (ppm)
%
CA-CNT membrane
1000
69.4%
PS/TFC membrane
2000
98.8%
Polypropylene/chitosan
3500
d
3500
8000
Ac ce p
PES/ TiO2 nanotubes
cr us 89.25
[44]
35
[45]
42.22%
16.4
[46]
97%
18.2
This work
38.42 %
te
PS/Al metal membrane
[43]
M
membrane
19.57
References
ip t
Feed conc.
an
*Membrane
membrane
*Where CA-CNT is Carbon nanotubes-cellulose acetate nanocomposite membrane, PS/TFC membrane is thin film composite polysulfone membrane, PS/Al metal is composite membrane of vapor deposition of aluminum metal on polysulfone membrane.
Page 38 of 38