- Email: [email protected]
Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs
ARTICLE IN PRESS
JID: JTICE
[m5G;April 17, 2017;16:1]
Journal of the Taiwan Institute of Chemical Engineers 0 0 0 (2017) 1–10
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs Mohammad Reza Mahdavi a, Mohammad Delnavaz a, Vahid Vatanpour b,∗ a b
Faculty of Engineering, Civil Engineering Department, Kharazmi University, 15719-14911 Tehran, Iran Faculty of Chemistry, Kharazmi University, 15719-14911 Tehran, Iran
a r t i c l e
i n f o
Article history: Received 1 December 2016 Revised 16 March 2017 Accepted 28 March 2017 Available online xxx Keywords: Nanofiltration Nanocomposite Carbon nanotubes Thin film composite Desalination
a b s t r a c t Multiwalled carbon nanotubes (MWCNTs) were exposed to acid solution containing HNO3 and H2 SO4 to synthesis of oxidized MWCNTs and used for preparation of piperazine-based polyamide thin film nanocomposite nanofiltration membrane. Both raw and oxidized MWCNTs were applied in the fabrication of the membranes with four different concentrations of 0.0 01, 0.0 02, 0.0 05, and 0.01 wt. % in the piperazine solution. Salt rejection, permeation, and antifouling properties of unfilled, raw and oxidized MWCNTs embedded membranes were investigated. Water flux for 0.005 wt. % oxidized MWCNTs significantly increased due to this fact that membrane hydrophilicity improved as a result of functionalization of MWCNTs. Contact angle measurements confirmed the improvement of hydrophilicity by adding oxidized MWCNTs to the membranes. Surface SEM and AFM images illustrated that the MWCNTs made surface of the membranes smoother and macro-voids enlargement leads to water flux enhancement. The antifouling properties were investigated using bovine serum albumin (BSA)/salt solution. The results showed that the membranes with a smoother surface had a better resistance against the fouling. The salt rejection performance exhibited that by embedding of the raw MWCNTs and oxidized MWCNTs, improvement in rejecting of Na2 SO4 salt can be observed. It could conclude that the addition of both raw and oxidized MWCNTs could improve the desalination performance of the piperazine–polyamide NF membranes. © 2017 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
1. Introduction For the last two decades, membrane processes have become one of the most important approaches to enhance water and wastewater treatment in commercial scales. This mostly comes from the high stability, great efficiency and low energy requirement of the membrane processes [1,2]. Based on this technology, separation and concentration have technically become possible due to their efficiency. Among all the membrane processes such as reverse osmosis (RO), nanofiltration (NF), ultrafiltration (UF), microfiltration (MF), and so on, the NF membranes have shown significant features in energy consumption and ease of operation. The NF membranes separation characteristics are between RO and UF. The NF membranes have the advantage of retaining low molecular weight organics, divalent ions and relatively large monovalent ions
∗
Corresponding author. E-mail addresses: [email protected], [email protected] (V. Vatanpour).
from water solution due to the separation mechanisms, in which this retention is higher than UF membrane and lesser than RO membranes [3]. Most of NF membranes are thin film composite (TFC) membranes prepared by interfacial polymerization (IP) technique [4]. The TFC nanofiltration membrane usually consists of three components: a polyester non-woven fabric base, an asymmetric porous support (usually polysulfone UF membrane) casted onto the base, and an ultra-thin selective skin layer (piperazine-amide) formed by the IP in-situ on the surface of the support [4,5]. Each layer is fit to specific requirements for maximum performance. The selective skin layer of TFC membrane is principally responsible for permeability and selectivity [6]. A great number of TFC membranes have been successfully developed to give two key parameters to NF membranes, high permeation flux and moderate solute rejection for evaluating their performance. Nowadays, usage of nanocomposite membranes is increasingly growing, where a nanofiller material will disperse in a polymer matrix [7]. Mixed matrix/nanocomposite membranes have shown satisfactory characteristics in sorption capacity, reaction, and sepa-
http://dx.doi.org/10.1016/j.jtice.2017.03.039 1876-1070/© 2017 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
JID: JTICE 2
ARTICLE IN PRESS
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
ration [8-10]. Carbon nanotubes (CNTs), as an excellent choice for fabricating mixed matrix membranes, have been found extremely helpful to achieve maximum flux and solute rejection with reduced operating pressure in order to solve the problems which address this issue [11,12]. Wu et al. prepared MWCNTs/polyester thin film nanocomposite (TFN) membranes by IP of triethanolamine (TEOA) and trimesoyl chloride (TMC) on a polysulfone supporting membrane. They found that water permeability would go up if the carboxylated MWCNTs concentration in aqueous phase increases up to 0.5 mg/mL. Additionally, the resulted membrane showed a fine long-term stability [13]. Majeed et al. have used three concentrations of hydroxyl functionalized MWCNTs in polyacrylonitrile (PAN), 0.5, 1 and 2 wt. % whose results showed an increase in water flux by 63% at 0.5 wt. % loading compared to the neat PAN membranes. Furthermore, comparing the MWCNTs embedded membranes with the neat ones in case of resistance against compaction have exhibited that loading of 2 wt. % MWCNTs caused to 36% improvement as well as an increase over 97% in tensile strength of the membranes [14]. Shah and Murthy reported synthesizing a functionalized MWCNT/PSf composite membrane by the phase inversion method using DMF as solvent, in which the amount of composite membranes had a crucial role in morphology, and permeation properties of the membranes. Rejection of heavy metal showed a great result with increasing the amount of MWNTs and the interesting results were seen at a pressure of 0.49 MPa [15]. Stretz and co-workers investigated the fabrication of ultrafiltration membranes applying polysulfone/nano-TiO2 /multiwalled carbon nanotube with variable ratios of TiO2 /MWCNTs to watch effects on membrane pore size, fouling, permeation, and rejection of humid acid (HA). They reached the conclusion that an equivalent mixture of both NPs provided the possibility to make properties of a single membrane better under both types of fouling conditions [16]. In a study, through self-polymerization of musselinspired dopamine (DA), Cheng et al. successfully coated a layer on the polyethylene glycol (PEG) based NF membrane. After being coated with mussel-inspired polydopamine, the hydrophilic PEG based membranes showed an increase in the salt rejection. Their results exhibited that in spite of increase in feed concentration up to 800 ppm, the rejections to all of antibiotics remained high [17]. Xu et al. fabricated a novel nanocomposite organic solvent nanofiltration (OSN) membrane by one-step deposition of mussel-inspired octaammonium polyhedral oligomeric silsesquioxane onto supports. The prepared membrane showed remarkable performance for dyes removal which may encourage the design of advanced nanocomposite membranes for environmental application [18]. The functionalized MWCNTs have been used for preparation of polyamide TFC membranes for reverse osmosis and forward osmosis purpose [19-21]. For polyamide nanofiltration performance, to our knowledge, only amine-functionalized MWCNTs have been applied [22]. Since MWCNTs show a significant improvement in mechanical, chemical, and thermal in membrane properties as well as permeation, rejection, and antifouling properties, the amount of MWCNTs to reach this goal is important. To address this issue, investigation in oxidized and non-oxidized MWCNTs is an appropriate field that seems a lot of work can be done. To our knowledge, the raw and oxidized MWCNTs are not used in preparation of piperazineamide nanofiltration membranes. In this study, the performance of unfilled NF membrane, raw MWCNTs and functionalized MWCNTs embedded membranes in polyamide layer was investigated. To emphasize the impact of concentration on different aspects of the membranes, all of these membranes were synthesized with different concentrations of MWCNTs varying between 0.001 to 0.01 wt. % and the resulted membranes were tested by permeation, salt rejection and BSA fouling experiments.
2. Materials and methods 2.1. Materials The pristine MWCNTs (length 10–30 μm, diameter 20–30 nm and inner diameter 5–10 nm, purity 95%) were manufactured from US Research Nanomaterials, Inc. Polysulfone (PSf), used for preparation of the supports, was obtained from BASF Company (Germany). Piperazine (PIP, 99.5%), trimesoyl chloride (TMC, 99%), solvent of N-methyl-2-pyrroldiinone (NMP, 99%), sulfuric acid (98%), nitric acid (65%) and n-hexane (99%) were purchased from Merck chemical Company. Hollytex 3329 non-woven fabric was used as a polyester support for preparation of PSf layer. 2.2. Functionalization of MWCNTs To functionalize pristine MWCNTs, they were mixed with an acid mixture of a 3:1 mixture of concentrated HNO3 and H2 SO4 , following 30 min sonication and after that, they were refluxed for 16 h at 90 °C. This step is supposed to purify the synthesized raw MWCNTs from metal catalysts and graphite and other impurities and inserting hydrophilic functional groups on the CNTs surface. Then, the solution was diluted with distilled water to reach neutral solution with a pH 5–6. This was followed by filtering the MWCNTs using PVDF ultrafiltration membranes and put those in an oven for 24 h at 80 °C in order to dry those [22]. 2.3. Membrane fabrication The UF support membranes were fabricated by the non-solvent induced phase inversion technique, which is a widely used method for the fabrication of polymeric membranes. Polyester non-woven fabric is used for membrane strength. To prepare the membranes, first dried polysulfone granules were dissolved in NMP solvent with concentration of 19 wt. %. Then with a constant speed (400 rpm) and temperature fixed at 50 °C, the solution was mixed on a stirrer. Eventually the prepared solution was put in a dark place to reduce the bubbles dissolved in it. Afterward the solution was casted on the surface of non-woven fabrics using Doctor Blade to make uniform films with thickness of 175 μm. Next, immediately, they were immersed in water bath (25 °C) to consolidate and form supports. Table 1 presented the characteristics of the prepared wet PSf support. More characterization of the used support was presented in previous work [24]. To convert these support membranes to NF membranes, first they were immersed in a piperazine aqueous solution bath (2 wt. %) for 10 min, followed by immersing these membranes in another bath which contained TMC solution (0.4 wt. %) in n-hexane for 2 min. After that the membranes were washed with n-hexane and put in an oven for 10 min in 70 °C to form a dense polyamide layer. The process of synthesizing was over and the membranes were kept in distilled water. To fabricate nanocomposite membranes, the PIP solutions were mixed with the MWCNTs and sonicated for 30 min, and then the supports were immersed in these solutions. The rest was similar to previous procedure. 2.4. Characterization of membrane Scanning of the surface and cross-sectional morphology of the membranes was conducted using scanning electron microscope (SEM) with a VEGA (TESCAN, Czech Republic) one. In order to prevent the sample charging, the dried samples were sputtered by a very thin layer of gold in an electro-plating device. After sputtering with gold, they were viewed with the microscope at 20 kV.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
3
Table 1 Characteristics of the used polysulfone support for preparation of PA nanofiltration membrane. Support
Pure water flux (L/m2 h) in 4 bar
Molecular weight cut-off (Da)
Contact angle (°)
Mean pore size (nm)
19 wt. % PSf/NMP
312 ± 24
4200 ± 300
56.2 ± 4.2
28.1 ± 3.2
Fig. 1. FTIR spectra of raw and oxidized MWCNTs.
Atomic force microscopy (AFM) (Dual Scope C-26, DME Corp., Denmark) analysis was utilized to study the surface roughness of the membranes and calculate the roughness parameters. FTIR analysis (ABB Bomem FTIR spectrometer (MB-104)) was performed to approve the formation of oxidized functional groups onto the raw MWCNTs. For hydrophilicity investigation, the contact angle measurement was conducted by means of goniometer (G10, Kruss, Germany) at 25 °C and a relative humidity of 45%. The droplet of water was applied as a probe liquid and the average of five random places was stated as the contact angle to diminish the experimental errors. The fabricated membranes were tested in a cross-flow testing setup, which consists of a high-pressure pump, stainless steel feed tank, and three cells where membranes were put therein and water circulated on the surface of the membranes and came back to the tank. The amount of water passing through membrane was collected in volume measurement containers via a tube, which was in the end of the cell. In all of tests, the operation pressure was 10 bar and feed flow was 120 L/h. The permeate flux calculated as Eq. (1):
J=
Vp A ∗ t (h )
(1)
where J is the permeate flux (L/m2 h), Vp (L) is the permeate volume, A (m2 ) is the membrane area, and t (h) is the filtration time. Salt rejection of the nanofiltration membranes were evaluated with a multivalent ion (Na2 SO4 ) and a conductivity/TDS meter was used to measure the salt concentration in feed and permeate. The rejection ratios were calculated with the following Eq. (2):
R (% ) =
Cp 1− Cf
× 100
(2)
where Cp and Cf are the conductivity of permeate and feed, respectively. For antifouling experiments, 100 mg/L BSA solution containing 20 0 0 mg/L Na2 SO4 were passed through both the bare and the MWCNTs embedded membranes at 10 bar and feed flow of 120 L/h. The filtration experiments were carried out for 24 h. Estimation of fouling extent of the prepared nanocomposite membranes was based on the flux dropped during BSA/Na2 SO4 filtration and normalized flux. The normalized permeate flux of the foulant during 24 h filtration was calculated with the following Eq. (3):
Jt = J/J0 ∗ 100%
(3)
where J0 and J are the water flux of initial and time t in the filtration process, respectively. The membrane surface charge was measured by streaming potential method using an EKA electro kinetic analyzer instrument (Anton Paar, Austria). In the streaming potential method, movement of the electrolyte solution through a capillary system produces a Zeta streaming potential where the relation with the zeta potential of the capillaries is given by Smoluchowski–Helmholtz approach [25,26]:
ζ=
dU dp
η L ε × ε◦ Q. R
(4)
where ζ is zeta potential, dU/dp is slope of streaming potential versus pressure, η is the electrolyte viscosity, ε is dielectric constant of electrolyte, ε o is permittivity, L is the length of the capillary system, Q is cross-sectional area of the capillary system and R is AC
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE 4
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
resistance of the cell by an electrolyte solution. In accord with Fairbrother and Mastin approach [27], for electrolyte amounts larger than 10−3 M, the ratio L/Q×R in Eq. (4) can be substituted by KB, which is the specific electrical conductivity of the electrolyte solution outside the capillary system. Before zeta potential measurements, the membranes were washed by circulating deionized water across the membrane surface for 20 min. A clamping cell was applied to conduct the Zeta potential measurements. Schematic of the clamping cell and the geometric parameters defining a single capillary was the same as accessible in Walker et al. [28]. The measurements were made with 0.001 M KCl electrolyte solution. During the measurements, pH of the solutions was changed in the range of 2–10 by adding 0.1 N solutions of HCl and NaOH. Due to construction design of clamping cell, the measurements were performed by firmly pressing the test surface against a PMMA grooved spacer. Therefore, the zeta potential of the sample was calculated according to the following equation:
ζs = 2ζtot − ζPMMA
(5)
where ζ tot is the zeta potential of sample plus PMMA, ζ PMMA is the zeta potential of PMMA, and ζ s is the net zeta potential of the sample.
3. Results and discussion 3.1. Characterization of MWCNT-OH Fig. 1 illustrates the Fourier transform infrared spectrum (FTIR) of raw and oxidized MWCNTs. New peaks are exhibited for oxidized MWCNTs that means the surface of oxidized MWCNTs is successfully introduced by many functional groups in the process of acid treatment. The emergence of this new peaks at 1674 and 1384 cm−1 are assigned to carbonyl group (C=O) and carboxyl (–COOH), respectively [29]. Also, the intensity of peaks at around 3300 cm−1 increased, showing the increasing hydroxyl groups (O–H) on the surface of oxidized MWCNTs [30]. The presence of these functional groups enhances the dispersion of oxidized MWCNTs due to their high hydrophilic properties, and consequently it leads to good dispersion of oxidized MWCNTs in PIP aqueous solution and producing more hydrophilic membranes.
3.2. Characterization of the membranes For investigating both surface and cross-sectional structure of the membranes, the SEM images were used. Fig. 2a shows the cross-sectional morphology of a membrane containing 0.005 wt. % oxidized MWCNTs. The image shows that the PSf support has the porous sub-layer supporting the top dense skin layer. This proves the existence of finger-like macro-pores in asymmetric structure. The finger-like porous support was fabricated onto the nonwoven fabric with a thickness of around 80 μm. According to Wang et al. to experience a better permeate flux of the membranes, less closed cells in the membranes is needed [1]. The hydrophilic MWCNTs provide this desirable property [14,31], therefore by looking at the result of this study, it is noticeable that the embedded membranes show a better permeate fluxes rather than the bare one. The thickness of prepared PA layer on the support was less than 200 nm. The surface morphology image is a way in which some of membrane characteristics can be explained. In Fig. 2, the surface SEM images of polyamide nanocomposite membranes containing various amounts of raw MWCNTs and oxidized MWCNTs were
Fig. 2. SEM images of the bare, oxidized and non-oxidized embedded membranes with various amounts of the MWCNTs. (a) Cross-sections of membrane with 0.005% oxidized MWCNTs.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
5
Table 2 Surface roughness parameters of the prepared piperazineamide NF membranes resulted from analyzing three randomly chosen AFM images (2 μm × 2 μm). Membrane
Bare PA NF 0.001 Raw MWCNTs 0.002 Raw MWCNTs 0.005 Raw MWCNTs 0.01 Raw MWCNTs 0.001 Oxidized MWCNTs 0.002 Oxidized MWCNTs 0.005 Oxidized MWCNTs 0.01 Oxidized MWCNTs
Roughness parameters Sa (nm)
Sq (nm)
Sy (nm)
33 21 17 22 27 14 15 16 23
44 29 26 31 37 23 24 27 30
350 202 166 217 284 130 143 166 243
presented. Because of the low concentration of MWCNTs, the SEM image could not show the MWCNTs embedded in PA layer. Attention to the surface images showed that polymerization reaction caused formation of an active rough layer covering the surface of the membranes. The higher roughness of the bare membrane is an intrinsic quality. However, by looking at images of the embedded membranes, it is obvious that this intrinsic property can be overcome by adding the MWCNTs. By comparing the raw and oxidized MWCNTs, it can be seen that by adding the oxidized MWCNTs, the surface becomes smoother. With increasing the loading of the MWCNTs, the peaks and nodules form more cross-linking areas and by enlarging these areas, decreasing in roughness happened [32-34]. This reduction in membrane surface roughness can be obviously seen in AFM images as presented in Fig. 3. Regarding the surface of the 0.001 wt. % raw and oxidized MWCNTs membranes, the change in the roughness is considerable. Table 2 presents the surface roughness parameters including average roughness (Sa ), root mean square of the Z data (Sq ) and the height difference between the highest peak and the lowest valley (Sy ). As presented, the surface of the bare PA NF membrane had the highest roughness and it was influenced significantly by blending the MWCNTs. The surface roughness was decreased by increasing the content of the MWCNTs up to 0.002 wt. % and the smoothest surface was observed in the case of 0.001 oxidized MWCNTs membrane. This roughness decrease can be related to the hydrogen bonding between the polyamide layer and the functional groups of oxidized MWCNTs [4]. The hydroxyl groups present on MWCNTs underwent hydrogen bonding, resulting in a more compact chain structure. Similar results have been reported by other authors [19,35]. In the case of SEM images of the 0.001 wt. % raw and oxidized MWCNTs membranes, the same change can be observed only with this difference that the nodules are more linked which leads to better permeate flux. Although improvement of the roughness occurred in this amount of embedding, but adding more MWCNTs up to 0.005 wt. % resulted in an even more linked nodules, in which cross-linking reaction could lead to smooth surface [36] and hydrogen bonding may cause nodular structure to become smaller and thus the total surface roughness decreases [37-40]. The hydrophilicity evaluation of the fabricated nanofiltration membranes was conducted by contact angle measurement, which is presented in Fig. 4. Not having the advantage of embedding oxidized MWCNTs, the bare membrane showed a relatively high contact angle; however, compared to raw MWCNTs membrane its contact angle was lower. The high contact angle of raw MWCNTs membranes can be explained by natural carbon-based properties of the MWCNTs. The polyamide layer have carbonyl and N–H groups that could have better hydrophilic properties rather than
Fig. 3. Three-dimensional AFM images of the prepared PA NF nanocomposite membranes (the Z-scaling: 2 has been applied for all images for better illustration).
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
JID: JTICE 6
ARTICLE IN PRESS
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
Fig. 4. Surface contact angles of the MWCNTs embedded NF membranes.
the C–C structure of the MWCNTs. Using of oxidized MWCNTs, the contact angle decreased significantly, as a result hydrophilicity increased due to this fact [41]. It can be seen that contact angle for membrane with 0.01 wt. % raw MWCNTs was 61.8° and for the same amount of oxidized MWCNTs was 51.4°. This might be reasonable by considering the presence of –COOH and –OH functional groups, which induce their hydrophilicity to the membrane surface [35]. 3.3. Nanofiltration performance The evaluation of nanofiltration performance can be measured by parameters like permeate flux and salt rejection. Porosity of active layer and surface hydrophilicity lead to determination of pure water flux [43,44]. Mostly, the pore size of nanofiltration membranes is about 1 nm, and the diameter of water molecule is 0.27 nm, so then water molecule could easily permeate the membrane [45]. Fig. 5 shows the saline aqueous solution flux of the membranes. All raw and oxidized MWCNTs embedded membranes presented higher water flux compared to the bare PA NF membrane. This confirms the effective act of blending MWCNTs based nanofillers on producing highly porous membranes [14]. However, this increment is higher for oxidized MWCNTs compared to raw MWCNTs in the same concentrations of the nanofiller. A key factor, which plays an important role in membrane permeability, is hydrophilicity. Fig. 4 shows that presence of oxidized MWCNTs in the polyamide layer caused an effect on water contact angle and therefore, higher hydrophilicity. The oxidized MWCNTs contents had more effect on the water flux of the membranes. By looking at both permeate flux figures of raw and oxidized MWCNTs, it is obvious that on the whole permeate flux in oxidized MWCNTs membranes was much higher than the bare one for all of the additive percentages and higher than the ones with raw MWCNTs. As can be seen in Fig. 5, the permeate flux increased with the increase of oxidized MWCNTs contents, however, the flux increasing stopped in 0.005 wt. % and after this point, with addition of more oxidized MWCNTs, the flux rate decreased. This could result from consequence of more compact structures by MWCNTs agglomeration that results in blocking the pores and reduction in flux [22,23].
In order to test the rejection performance of the prepared membranes, Na2 SO4 feed solution was used. The pH was 6.5 during all tests. The rejection factor for the bare membrane was well above average of some other works with 95.70% [46-48], however, addition of both raw and oxidized MWCNTs showed a growth in rejection. Fig. 6 shows the rejection of the membranes for three kinds of membranes (bare, raw MWCNT with 0.0 01, 0.0 02, 0.0 05, and 0.01 wt. % concentration, and oxidized MWCNTs embedded membranes with the same concentrations as the raw MWCNT ones). Regarding raw MWCNTs, it can be seen that increasing the amount of MWCNTs up to 0.002 wt. % was effective in enhancing the rejection capability, however, addition of more MWCNTs after this point leads to decrease in rejection. With respect to the oxidized MWCNTs, the trend was almost similar to previous one with a slight difference that the most efficient dosage of oxidized MWCNTs membranes was 0.005 wt. % which is the best dosage among all membranes. The reason of this high rejection property of the nanofiltration membranes lies on the negative charges on the membrane surface which can be explained by Donnan exclusion effects and steric exclusion [42, 49]. Fig. 7 shows the Zeta potential measurements of the Bare, 0.005 wt. % raw and 0.005 wt. % oxidized MWCNTs membranes. As shown, by addition of the oxidized MWCNTs to the polyamide layer, due to presence of functional groups in the MWCNTs structure, the negative charge can be induced to the membrane surface. The addition of oxidized MWCNTs to the PA membrane not only declined its contact angle, but also induced a surface negative charge. Moreover, at higher pH, the membrane with oxidized MWCNTs content presented a more negative surface charge due to the existence of carboxylic groups and their subsequent deprotonation. However, the raw MWCNTs have no significant effect on the surface charge. Generally it could be said that the addition of the MWCNTs improved the permeability without reduction in the rejection of the prepared nanocomposite membranes. Table 3 compares the flux, salt retention and water contact angle results in current work with the results of reported nanofiltration membranes in the literatures. As can be seen, the rejection property of our prepared raw and oxidized MWCNTs is the highest and the flux is suitable.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
7
Fig. 5. Saline aqueous solution flux of the prepared PA NF membranes; (a) raw MWCNTs and (b) oxidized MWCNTs embedded membranes.
Table 3 Comparison of results of current study with the literature MWCNTs embedded membranes. Membrane
Optimum flux (L/m2 h)
Na2 SO4 rejection (%)
Contact angle (°)
Ref.
0.5 wt. % carboxylated MWNTs/polyester 0.005 NH2 -MWCNTs in PIP Polymethyl methacrylate-MWCNTs in TMC 0.04 wt. % oxidized MWCNTs/PES 0.001 wt. % sulfonated MWCNTs in PIP 0.002 wt. % raw MWCNTs
≈ 22 (6 bar) 55.4 (10 bar) 69.8 (10 bar) 9.1 (4 bar) 79.4 (6 bar) 58.8 (10 bar)
≈ 80 95.7 ≈ 99 ≈ 70 96.8 97.7
23.2 59.3 49 63.3 38 60.9
[13] [23] [37] [46] [50] This study
0.005 wt. % oxidized MWCNTs
65.7 (10 bar)
97.9
57.8
This study
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE 8
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
Fig. 6. Na2 SO4 rejection of the membranes embedded with raw and oxidized MWCNTs.
30 0.005 Raw MWCNTs Bare NF 0.005 Oxidized MWCNTs
Zeta potenal (mV)
20 10 0 1
3
5
7
9
pH 11
-10 -20 -30
In regards to the oxidized MWCNTs, they made impressive impacts on the fouling, for example, 0.002 wt. % and 0.005 wt. % oxidized MWCNTs are the two best fouling resistance membranes. Although introducing MWCNTs made great differences but the amount of addition is also important. With introducing 0.01 wt. % or 0.001 wt. % oxidized MWCNTs, the fouling properties was not that significant as what is expected whereas the fouling of 0.01 wt. % of raw MWCNTs membrane was as high as the bare one. Pore blocking and agglomeration phenomenon of nanofiller can be the reason of this reduction in fouling properties [23,53]. In addition, comparing fouling trend with surface roughness parameters (Table 2) shows that the smoother surface has lower fouling tendency.
-40 -50
4. Conclusion
Fig. 7. Surface charge of the prepared PA nanofiltration membranes over pH range.
3.4. Antifouling evaluation In order to evaluate fouling resistance capability of the membranes, BSA/Na2 SO4 solution was used. Fig. 8 shows the flux of the membranes, which were exposed to the foulant solution for 24 h. As can be seen, the permeate flux for all the membranes is reduced during filtration. However, this reduction is lower for the MWCNTs embedded membranes. As a general expectation, the bare membrane has the lowest flux (see Fig. 5) as well as rejection (see Fig. 6) showed the most fouling. The oxidized MWCNTs with the most fluxes among the whole membranes experienced the minimum fouling, although it can be seen that raw MWCNTs show a considerable resistance but not as significant as the oxidized ones. This might be due to the fact that oxidized MWCNTs membranes have more hydrophilic surface because of expanded surface coverage of hydrophilic groups [51,52]. By looking back to Fig. 4 (contact angle measurement), it can be noticed that generally membranes with low contact angle have minimum fouling. In Fig. 9, normalized fluxes of all membranes are illustrated for better possibility of tracing the trend of the fouling. The flux of the bare membrane showed a dramatic drop by adding the BSA/Na2 SO4 solution to the feed. However, by embedding MWCNTs to the membranes the trend of the fouling changed significantly.
In this study, performance of three different kinds of membranes was evaluated in three subjects of salt rejection, flux, and antifouling capability. Bare membrane was the one membrane, which made it possible to compare the raw MWCNTs membranes with four concentrations and oxidized MWCNTs membranes with the same amounts. Both raw and oxidized MWCNTs had an impact on increasing the flux, however, the impact of oxidized MWCNTs was more significant so that 0.005 wt. % oxidized MWCNTs embedded membrane had the most permeation. From SEM and AFM images, it can be conceived that by embedding oxidized MWCNTs, the surface of the membranes becomes smoother; hence, the surface is less vulnerable to BSA, therefore the fouling rate decreased. Furthermore, addition of MWCNTs improved the antifouling properties until 0.002 wt. %, nevertheless further increasing ended up in reduction of this quality, due to roughness increasing by addition of more MWCNTs. Na2 SO4 rejection in all kind was acceptable, but again a range of increase was observed in oxidized MWCNTs as well as raw ones.
Acknowledgments The authors are thankful to the partial support from Iranian Nano Technology Initiative Council, Iran National Science Foundation and Kharazmi University.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
JID: JTICE
ARTICLE IN PRESS
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
Fig. 8. Foulant solution flux of the membranes during 24 h filtration (BSA conc. = 100 ppm, Na2 SO4 conc. = 20 0 0 ppm, 10 bar).
Fig. 9. Normalized permeate flux of the foulant solution during 24 h filtration.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
9
JID: JTICE 10
ARTICLE IN PRESS
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
References [1] Wang L, Song X, Wang T, Wang S, Wang Z, Gao C. Fabrication and characterization of polyethersulfone/carbon nanotubes (PES/CNTs) based mixed matrix membranes (MMMs) for nanofiltration application. Appl Surf Sci 2015;330:118–25. [2] Zhu J, Tian M, Zhang Y, Zhang H, Liu J. Fabrication of a novel “loose” nanofiltration membrane by facile blending with Chitosan–Montmorillonite nanosheets for dyes purification. Chem Eng J 2015;265:184–93. [3] Xu YC, Wang ZX, Cheng XQ, Xiao YC, Shao L. Positively charged nanofiltration membranes via economically mussel-substance-simulated co-deposition for textile wastewater treatment. Chem Eng J 2016;303:555–64. [4] Safarpour M, Vatanpour V, Khataee A, Esmaeili M. Development of a novel high flux and fouling-resistant thin film composite nanofiltration membrane by embedding reduced graphene oxide/TiO2 . Sep Purif Technol 2015;154:96–107. [5] Liu Y, Zhang S, Zhou Z, Ren J, Geng Z, Luan J, et al. Novel sulfonated thin-film composite nanofiltration membranes with improved water flux for treatment of dye solutions. J Membr Sci 2012;394:218–29. [6] Nunes S, Peinemann KV. Presently available membranes for liquid separation. Membrane Technology: In the Chemical Industry. 2nd Revised and Extended ed; 2006. p. 15–38. [7] Jhaveri JH, Murthy Z. A comprehensive review on anti-fouling nanocomposite membranes for pressure driven membrane separation processes. Desalination 2016;379:137–54. [8] Jeong BH, Hoek EM, Yan Y, Subramani A, Huang X, Hurwitz G, et al. Interfacial polymerization of thin film nanocomposites: a new concept for reverse osmosis membranes. J Membr Sci 2007;294:1–7. [9] Jadav GL, Singh PS. Synthesis of novel silica-polyamide nanocomposite membrane with enhanced properties. J Membr Sci 2009;328:257–67. [10] Kim ES, Deng B. Fabrication of polyamide thin-film nano-composite (PA-TFN) membrane with hydrophilized ordered mesoporous carbon (H-OMC) for water purifications. J Membr Sci 2011;375:46–54. [11] Liang S, Qi G, Xiao K, Sun J, Giannelis EP, Huang X, et al. Organic fouling behavior of superhydrophilic polyvinylidene fluoride (PVDF) ultrafiltration membranes functionalized with surface-tailored nanoparticles: Implications for organic fouling in membrane bioreactors. J Membr Sci 2014;463:94–101. [12] Roy S, Ntim SA, Mitra S, Sirkar KK. Facile fabrication of superior nanofiltration membranes from interfacially polymerized CNT-polymer composites. J Membr Sci 2011;375:81–7. [13] Wu H, Tang B, Wu P. Optimization, characterization and nanofiltration properties test of MWNTs/polyester thin film nanocomposite membrane. J Membr Sci 2013;428:425–33. [14] Majeed S, Fierro D, Buhr K, Wind J, Du B, Boschetti-de-Fierro A, et al. Multi-walled carbon nanotubes (MWCNTs) mixed polyacrylonitrile (PAN) ultrafiltration membranes. J Membr Sci 2012;403:101–9. [15] Shah P, Murthy C. Studies on the porosity control of MWCNT/polysulfone composite membrane and its effect on metal removal. J Membr Sci 2013;437:90–8. [16] Esfahani MR, Tyler JL, Stretz HA, Wells MJ. Effects of a dual nanofiller, nano-TiO2 and MWCNT, for polysulfone-based nanocomposite membranes for water purification. Desalination 2015;372:47–56. [17] Park J, Choi W, Kim SH, Chun BH, Bang J, Lee KB. Enhancement of chlorine resistance in carbon nanotube based nanocomposite reverse osmosis membrane. Desalin. Water Treat. 2010;15:198–204. [18] Cheng XQ, Zhang C, Wang ZX, Shao L. Tailoring nanofiltration membrane performance for highly-efficient antibiotics removal by mussel-inspired modification. J Membr Sci 2016;499:326–534. [19] Zhao H, Qiu S, Wu L, Zhang L, Chen H, Gao C. Improving the performance of polyamide reverse osmosis membrane by incorporation of modified multi-walled carbon nanotubes. J Membr Sci 2014;450:249–56. [20] Xu YC, Tang YP, Liu LF, Guo ZH, Shao L. Nanocomposite organic solvent nanofiltration membranes by a highly-efficient mussel-inspired co-deposition strategy. J Membr Sci 2017;526:532–42. [21] Amini M, Jahanshahi M, Rahimpour A. Synthesis of novel thin film nanocomposite (TFN) forward osmosis membranes using functionalized multi-walled carbon nanotubes. J Membr Sci 2013;435:233–41. [22] Farahbaksh J, Delnavaz M, Vatanpour V. Investigation of raw and oxidized multiwalled carbon nanotubes in fabrication of reverse osmosis polyamide membranes for improvement in desalination and antifouling properties. Desalination 2017;410:1–9. [23] Zarrabi H, Yekavalangi ME, Vatanpour V, Shockravi A, Safarpour M. Improvement in desalination performance of thin film nanocomposite nanofiltration membrane using amine-functionalized multiwalled carbon nanotube. Desalination 2016;394:83–90. [24] Yakavalangi ME, Rimaz S, Vatanpour V. Effect of surface properties of polysulfone support on the performance of thin film composite polyamide reverse osmosis membranes. J Appl Polym Sci 2017;134:44444. [25] Jacobasch HJ, Schurz J. Characterization of polymer surfaces by means of electrokinetic measurements. Prog Colloid Polym Sci 1988;77:40–8. [26] Mehdipour S, Vatanpour V, Kariminia H-R. Influence of ion interaction on lead removal by a polyamide nanofiltration membrane. Desalination 2015;362:84–92. [27] Fairbrother F, Mastin H. Studies in electroendosmosis. J Chem Soc 1924;125:2319–30.
[28] Walker SL, Bhattacharjee S, Hoek EMV, Elimelech M. A novel asymmetric clamping cell for measuring streaming potential of flat surfaces. Langmuir 2002;18:2193–8. [29] Yang S, Li J, Shao D, Hu J, Wang X. Adsorption of Ni (II) on oxidized multi-walled carbon nanotubes: effect of contact time, pH, foreign ions and PAA. J Hazard Mater 2009;166:109–16. [30] Madaeni SS, Derakhshandeh K, Ahmadi S, Vatanpour V, Zinadini S. Effect of modified multi-walled carbon nanotubes on release characteristics of indomethacin from symmetric membrane coated tablets. J Membr Sci 2012;389:110–16. [31] Daraei P, Madaeni SS, Ghaemi N, Khadivi MA, Astinchap B, Moradian R. Enhancing antifouling capability of PES membrane via mixing with various types of polymer modified multi-walled carbon nanotube. J Membr Sci 2013;444:184–91. [32] Kwak SY, Jung SG, Yoon YS, Ihm DW. Details of surface features in aromatic polyamide reverse osmosis membranes characterized by scanning electron and atomic force microscopy. J Polym Sci Part B Polym Phys 1999;37:1429–40. [33] Rajaeian B, Rahimpour A, Tade MO, Liu S. Fabrication and characterization of polyamide thin film nanocomposite (TFN) nanofiltration membrane impregnated with TiO2 nanoparticles. Desalination 2013;313:176–88. [34] Jin L, Yu S, Shi W, Yi X, Sun N, Ge Y, et al. Synthesis of a novel composite nanofiltration membrane incorporated SiO2 nanoparticles for oily wastewater desalination. Polymer 2012;53:5295–303. [35] Bano S, Mahmood A, Kim S-J, Lee K-H. Graphene oxide modified polyamide nanofiltration membrane with improved flux and antifouling properties. J Mater Chem A 2015;3:2065–71. [36] Huang SH, Hsu CJ, Liaw DJ, Hu CC, Lee KR, Lai JY. Effect of chemical structures of amines on physicochemical properties of active layers and dehydration of isopropanol through interfacially polymerized thin-film composite membranes. J Membr Sci 2008;307:73–81. [37] Shen J n, Yu C c, Ruan H m, Gao C j, Van der Bruggen B. Preparation and characterization of thin-film nanocomposite membranes embedded with poly (methyl methacrylate) hydrophobic modified multiwalled carbon nanotubes by interfacial polymerization. J Membr Sci 2013;442:18–26. [38] Wang Z, Yu H, Xia J, Zhang F, Li F, Xia Y, et al. Novel GO-blended PVDF ultrafiltration membranes. Desalination 2012;299:50–4. [39] Sun H, Liu S, Zhou G, Ang HM, Tade´ MO, Wang S. Reduced graphene oxide for catalytic oxidation of aqueous organic pollutants. ACS Appl Mater Interface 2012;4:5466–71. [40] Lee SY, Kim HJ, Patel R, Im SJ, Kim JH, Min BR. Silver nanoparticles immobilized on thin film composite polyamide membrane: characterization, nanofiltration, antifouling properties. Polym Adv Technol 2007;18:562–8. [41] Sun M, Su Y, Mu C, Jiang Z. Improved antifouling property of PES ultrafiltration membranes using additive of silica−PVP nanocomposite. Ind Eng Chem Res 2009;49:790–6. [42] Choi JH, Jegal J, Kim WN. Fabrication and characterization of multi-walled carbon nanotubes/polymer blend membranes. J Membr Sci 2006;284:406–15. [43] Yu HY, Xu ZK, Yang Q, Hu MX, Wang SY. Improvement of the antifouling characteristics for polypropylene microporous membranes by the sequential photoinduced graft polymerization of acrylic acid. J Membr Sci 2006;281:658–65. [44] Zinadini S, Zinatizadeh A, Rahimi M, Vatanpour V, Zangeneh H, Beygzadeh M. Novel high flux antifouling nanofiltration membranes for dye removal containing carboxymethyl chitosan coated Fe3 O4 nanoparticles. Desalination 2014;349:145–54. [45] Jin Jb, Liu Dq, Zhang Dd, Yin Yh, Zhao Xy, Zhang Yf. Preparation of thin-film composite nanofiltration membranes with improved antifouling property and flux using 2, 2 -oxybis-ethylamine. Desalination 2015;355:141–6. [46] Vatanpour V, Madaeni SS, Moradian R, Zinadini S, Astinchap B. Fabrication and characterization of novel antifouling nanofiltration membrane prepared from oxidized multiwalled carbon nanotube/polyethersulfone nanocomposite. J Membr Sci 2011;375:284–94. [47] Zhao FY, An QF, Ji YL, Gao CJ. A novel type of polyelectrolyte complex/MWCNT hybrid nanofiltration membranes for water softening. J Membr Sci 2015;492:412–21. [48] Madaeni SS, Zinadini S, Vatanpour V. Preparation of superhydrophobic nanofiltration membrane by embedding multiwalled carbon nanotube and polydimethylsiloxane in pores of microfiltration membrane. Sep Purif Technol 2013;111:98–107. [49] Liu M, Yao G, Cheng Q, Ma M, Yu S, Gao C. Acid stable thin– film composite membrane for nanofiltration prepared from naphthalene-1,3,6-trisulfonylchloride (NTSC) and piperazine (PIP). J Membr Sci 2012;415:122–31. [50] Zheng J, Li M, Yu K, Hu J, Zhang X, Wang L. Sulfonated multiwall carbon nanotubes assisted thin-film nanocomposite membrane with enhanced water flux and anti-fouling property. J Membr Sci 2017;524:344–53. [51] Wei Y, Chu HQ, Dong BZ, Li X, Xia SJ, Qiang ZM. Effect of TiO2 nanowire addition on PVDF ultrafiltration membrane performance. Desalination 2011;272:90–7. [52] Zhang M, Nguyen QT, Ping Z. Hydrophilic modification of poly (vinylidene fluoride) microporous membrane. J Membr Sci 2009;327:78–86. [53] Qiu S, Wu L, Pan X, Zhang L, Chen H, Gao C. Preparation and properties of functionalized carbon nanotube/PSF blend ultrafiltration membranes. J Membr Sci 2009;342:165–72.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
JID: JTICE
[m5G;April 17, 2017;16:1]
Journal of the Taiwan Institute of Chemical Engineers 0 0 0 (2017) 1–10
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs Mohammad Reza Mahdavi a, Mohammad Delnavaz a, Vahid Vatanpour b,∗ a b
Faculty of Engineering, Civil Engineering Department, Kharazmi University, 15719-14911 Tehran, Iran Faculty of Chemistry, Kharazmi University, 15719-14911 Tehran, Iran
a r t i c l e
i n f o
Article history: Received 1 December 2016 Revised 16 March 2017 Accepted 28 March 2017 Available online xxx Keywords: Nanofiltration Nanocomposite Carbon nanotubes Thin film composite Desalination
a b s t r a c t Multiwalled carbon nanotubes (MWCNTs) were exposed to acid solution containing HNO3 and H2 SO4 to synthesis of oxidized MWCNTs and used for preparation of piperazine-based polyamide thin film nanocomposite nanofiltration membrane. Both raw and oxidized MWCNTs were applied in the fabrication of the membranes with four different concentrations of 0.0 01, 0.0 02, 0.0 05, and 0.01 wt. % in the piperazine solution. Salt rejection, permeation, and antifouling properties of unfilled, raw and oxidized MWCNTs embedded membranes were investigated. Water flux for 0.005 wt. % oxidized MWCNTs significantly increased due to this fact that membrane hydrophilicity improved as a result of functionalization of MWCNTs. Contact angle measurements confirmed the improvement of hydrophilicity by adding oxidized MWCNTs to the membranes. Surface SEM and AFM images illustrated that the MWCNTs made surface of the membranes smoother and macro-voids enlargement leads to water flux enhancement. The antifouling properties were investigated using bovine serum albumin (BSA)/salt solution. The results showed that the membranes with a smoother surface had a better resistance against the fouling. The salt rejection performance exhibited that by embedding of the raw MWCNTs and oxidized MWCNTs, improvement in rejecting of Na2 SO4 salt can be observed. It could conclude that the addition of both raw and oxidized MWCNTs could improve the desalination performance of the piperazine–polyamide NF membranes. © 2017 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
1. Introduction For the last two decades, membrane processes have become one of the most important approaches to enhance water and wastewater treatment in commercial scales. This mostly comes from the high stability, great efficiency and low energy requirement of the membrane processes [1,2]. Based on this technology, separation and concentration have technically become possible due to their efficiency. Among all the membrane processes such as reverse osmosis (RO), nanofiltration (NF), ultrafiltration (UF), microfiltration (MF), and so on, the NF membranes have shown significant features in energy consumption and ease of operation. The NF membranes separation characteristics are between RO and UF. The NF membranes have the advantage of retaining low molecular weight organics, divalent ions and relatively large monovalent ions
∗
Corresponding author. E-mail addresses: [email protected], [email protected] (V. Vatanpour).
from water solution due to the separation mechanisms, in which this retention is higher than UF membrane and lesser than RO membranes [3]. Most of NF membranes are thin film composite (TFC) membranes prepared by interfacial polymerization (IP) technique [4]. The TFC nanofiltration membrane usually consists of three components: a polyester non-woven fabric base, an asymmetric porous support (usually polysulfone UF membrane) casted onto the base, and an ultra-thin selective skin layer (piperazine-amide) formed by the IP in-situ on the surface of the support [4,5]. Each layer is fit to specific requirements for maximum performance. The selective skin layer of TFC membrane is principally responsible for permeability and selectivity [6]. A great number of TFC membranes have been successfully developed to give two key parameters to NF membranes, high permeation flux and moderate solute rejection for evaluating their performance. Nowadays, usage of nanocomposite membranes is increasingly growing, where a nanofiller material will disperse in a polymer matrix [7]. Mixed matrix/nanocomposite membranes have shown satisfactory characteristics in sorption capacity, reaction, and sepa-
http://dx.doi.org/10.1016/j.jtice.2017.03.039 1876-1070/© 2017 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
JID: JTICE 2
ARTICLE IN PRESS
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
ration [8-10]. Carbon nanotubes (CNTs), as an excellent choice for fabricating mixed matrix membranes, have been found extremely helpful to achieve maximum flux and solute rejection with reduced operating pressure in order to solve the problems which address this issue [11,12]. Wu et al. prepared MWCNTs/polyester thin film nanocomposite (TFN) membranes by IP of triethanolamine (TEOA) and trimesoyl chloride (TMC) on a polysulfone supporting membrane. They found that water permeability would go up if the carboxylated MWCNTs concentration in aqueous phase increases up to 0.5 mg/mL. Additionally, the resulted membrane showed a fine long-term stability [13]. Majeed et al. have used three concentrations of hydroxyl functionalized MWCNTs in polyacrylonitrile (PAN), 0.5, 1 and 2 wt. % whose results showed an increase in water flux by 63% at 0.5 wt. % loading compared to the neat PAN membranes. Furthermore, comparing the MWCNTs embedded membranes with the neat ones in case of resistance against compaction have exhibited that loading of 2 wt. % MWCNTs caused to 36% improvement as well as an increase over 97% in tensile strength of the membranes [14]. Shah and Murthy reported synthesizing a functionalized MWCNT/PSf composite membrane by the phase inversion method using DMF as solvent, in which the amount of composite membranes had a crucial role in morphology, and permeation properties of the membranes. Rejection of heavy metal showed a great result with increasing the amount of MWNTs and the interesting results were seen at a pressure of 0.49 MPa [15]. Stretz and co-workers investigated the fabrication of ultrafiltration membranes applying polysulfone/nano-TiO2 /multiwalled carbon nanotube with variable ratios of TiO2 /MWCNTs to watch effects on membrane pore size, fouling, permeation, and rejection of humid acid (HA). They reached the conclusion that an equivalent mixture of both NPs provided the possibility to make properties of a single membrane better under both types of fouling conditions [16]. In a study, through self-polymerization of musselinspired dopamine (DA), Cheng et al. successfully coated a layer on the polyethylene glycol (PEG) based NF membrane. After being coated with mussel-inspired polydopamine, the hydrophilic PEG based membranes showed an increase in the salt rejection. Their results exhibited that in spite of increase in feed concentration up to 800 ppm, the rejections to all of antibiotics remained high [17]. Xu et al. fabricated a novel nanocomposite organic solvent nanofiltration (OSN) membrane by one-step deposition of mussel-inspired octaammonium polyhedral oligomeric silsesquioxane onto supports. The prepared membrane showed remarkable performance for dyes removal which may encourage the design of advanced nanocomposite membranes for environmental application [18]. The functionalized MWCNTs have been used for preparation of polyamide TFC membranes for reverse osmosis and forward osmosis purpose [19-21]. For polyamide nanofiltration performance, to our knowledge, only amine-functionalized MWCNTs have been applied [22]. Since MWCNTs show a significant improvement in mechanical, chemical, and thermal in membrane properties as well as permeation, rejection, and antifouling properties, the amount of MWCNTs to reach this goal is important. To address this issue, investigation in oxidized and non-oxidized MWCNTs is an appropriate field that seems a lot of work can be done. To our knowledge, the raw and oxidized MWCNTs are not used in preparation of piperazineamide nanofiltration membranes. In this study, the performance of unfilled NF membrane, raw MWCNTs and functionalized MWCNTs embedded membranes in polyamide layer was investigated. To emphasize the impact of concentration on different aspects of the membranes, all of these membranes were synthesized with different concentrations of MWCNTs varying between 0.001 to 0.01 wt. % and the resulted membranes were tested by permeation, salt rejection and BSA fouling experiments.
2. Materials and methods 2.1. Materials The pristine MWCNTs (length 10–30 μm, diameter 20–30 nm and inner diameter 5–10 nm, purity 95%) were manufactured from US Research Nanomaterials, Inc. Polysulfone (PSf), used for preparation of the supports, was obtained from BASF Company (Germany). Piperazine (PIP, 99.5%), trimesoyl chloride (TMC, 99%), solvent of N-methyl-2-pyrroldiinone (NMP, 99%), sulfuric acid (98%), nitric acid (65%) and n-hexane (99%) were purchased from Merck chemical Company. Hollytex 3329 non-woven fabric was used as a polyester support for preparation of PSf layer. 2.2. Functionalization of MWCNTs To functionalize pristine MWCNTs, they were mixed with an acid mixture of a 3:1 mixture of concentrated HNO3 and H2 SO4 , following 30 min sonication and after that, they were refluxed for 16 h at 90 °C. This step is supposed to purify the synthesized raw MWCNTs from metal catalysts and graphite and other impurities and inserting hydrophilic functional groups on the CNTs surface. Then, the solution was diluted with distilled water to reach neutral solution with a pH 5–6. This was followed by filtering the MWCNTs using PVDF ultrafiltration membranes and put those in an oven for 24 h at 80 °C in order to dry those [22]. 2.3. Membrane fabrication The UF support membranes were fabricated by the non-solvent induced phase inversion technique, which is a widely used method for the fabrication of polymeric membranes. Polyester non-woven fabric is used for membrane strength. To prepare the membranes, first dried polysulfone granules were dissolved in NMP solvent with concentration of 19 wt. %. Then with a constant speed (400 rpm) and temperature fixed at 50 °C, the solution was mixed on a stirrer. Eventually the prepared solution was put in a dark place to reduce the bubbles dissolved in it. Afterward the solution was casted on the surface of non-woven fabrics using Doctor Blade to make uniform films with thickness of 175 μm. Next, immediately, they were immersed in water bath (25 °C) to consolidate and form supports. Table 1 presented the characteristics of the prepared wet PSf support. More characterization of the used support was presented in previous work [24]. To convert these support membranes to NF membranes, first they were immersed in a piperazine aqueous solution bath (2 wt. %) for 10 min, followed by immersing these membranes in another bath which contained TMC solution (0.4 wt. %) in n-hexane for 2 min. After that the membranes were washed with n-hexane and put in an oven for 10 min in 70 °C to form a dense polyamide layer. The process of synthesizing was over and the membranes were kept in distilled water. To fabricate nanocomposite membranes, the PIP solutions were mixed with the MWCNTs and sonicated for 30 min, and then the supports were immersed in these solutions. The rest was similar to previous procedure. 2.4. Characterization of membrane Scanning of the surface and cross-sectional morphology of the membranes was conducted using scanning electron microscope (SEM) with a VEGA (TESCAN, Czech Republic) one. In order to prevent the sample charging, the dried samples were sputtered by a very thin layer of gold in an electro-plating device. After sputtering with gold, they were viewed with the microscope at 20 kV.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
3
Table 1 Characteristics of the used polysulfone support for preparation of PA nanofiltration membrane. Support
Pure water flux (L/m2 h) in 4 bar
Molecular weight cut-off (Da)
Contact angle (°)
Mean pore size (nm)
19 wt. % PSf/NMP
312 ± 24
4200 ± 300
56.2 ± 4.2
28.1 ± 3.2
Fig. 1. FTIR spectra of raw and oxidized MWCNTs.
Atomic force microscopy (AFM) (Dual Scope C-26, DME Corp., Denmark) analysis was utilized to study the surface roughness of the membranes and calculate the roughness parameters. FTIR analysis (ABB Bomem FTIR spectrometer (MB-104)) was performed to approve the formation of oxidized functional groups onto the raw MWCNTs. For hydrophilicity investigation, the contact angle measurement was conducted by means of goniometer (G10, Kruss, Germany) at 25 °C and a relative humidity of 45%. The droplet of water was applied as a probe liquid and the average of five random places was stated as the contact angle to diminish the experimental errors. The fabricated membranes were tested in a cross-flow testing setup, which consists of a high-pressure pump, stainless steel feed tank, and three cells where membranes were put therein and water circulated on the surface of the membranes and came back to the tank. The amount of water passing through membrane was collected in volume measurement containers via a tube, which was in the end of the cell. In all of tests, the operation pressure was 10 bar and feed flow was 120 L/h. The permeate flux calculated as Eq. (1):
J=
Vp A ∗ t (h )
(1)
where J is the permeate flux (L/m2 h), Vp (L) is the permeate volume, A (m2 ) is the membrane area, and t (h) is the filtration time. Salt rejection of the nanofiltration membranes were evaluated with a multivalent ion (Na2 SO4 ) and a conductivity/TDS meter was used to measure the salt concentration in feed and permeate. The rejection ratios were calculated with the following Eq. (2):
R (% ) =
Cp 1− Cf
× 100
(2)
where Cp and Cf are the conductivity of permeate and feed, respectively. For antifouling experiments, 100 mg/L BSA solution containing 20 0 0 mg/L Na2 SO4 were passed through both the bare and the MWCNTs embedded membranes at 10 bar and feed flow of 120 L/h. The filtration experiments were carried out for 24 h. Estimation of fouling extent of the prepared nanocomposite membranes was based on the flux dropped during BSA/Na2 SO4 filtration and normalized flux. The normalized permeate flux of the foulant during 24 h filtration was calculated with the following Eq. (3):
Jt = J/J0 ∗ 100%
(3)
where J0 and J are the water flux of initial and time t in the filtration process, respectively. The membrane surface charge was measured by streaming potential method using an EKA electro kinetic analyzer instrument (Anton Paar, Austria). In the streaming potential method, movement of the electrolyte solution through a capillary system produces a Zeta streaming potential where the relation with the zeta potential of the capillaries is given by Smoluchowski–Helmholtz approach [25,26]:
ζ=
dU dp
η L ε × ε◦ Q. R
(4)
where ζ is zeta potential, dU/dp is slope of streaming potential versus pressure, η is the electrolyte viscosity, ε is dielectric constant of electrolyte, ε o is permittivity, L is the length of the capillary system, Q is cross-sectional area of the capillary system and R is AC
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE 4
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
resistance of the cell by an electrolyte solution. In accord with Fairbrother and Mastin approach [27], for electrolyte amounts larger than 10−3 M, the ratio L/Q×R in Eq. (4) can be substituted by KB, which is the specific electrical conductivity of the electrolyte solution outside the capillary system. Before zeta potential measurements, the membranes were washed by circulating deionized water across the membrane surface for 20 min. A clamping cell was applied to conduct the Zeta potential measurements. Schematic of the clamping cell and the geometric parameters defining a single capillary was the same as accessible in Walker et al. [28]. The measurements were made with 0.001 M KCl electrolyte solution. During the measurements, pH of the solutions was changed in the range of 2–10 by adding 0.1 N solutions of HCl and NaOH. Due to construction design of clamping cell, the measurements were performed by firmly pressing the test surface against a PMMA grooved spacer. Therefore, the zeta potential of the sample was calculated according to the following equation:
ζs = 2ζtot − ζPMMA
(5)
where ζ tot is the zeta potential of sample plus PMMA, ζ PMMA is the zeta potential of PMMA, and ζ s is the net zeta potential of the sample.
3. Results and discussion 3.1. Characterization of MWCNT-OH Fig. 1 illustrates the Fourier transform infrared spectrum (FTIR) of raw and oxidized MWCNTs. New peaks are exhibited for oxidized MWCNTs that means the surface of oxidized MWCNTs is successfully introduced by many functional groups in the process of acid treatment. The emergence of this new peaks at 1674 and 1384 cm−1 are assigned to carbonyl group (C=O) and carboxyl (–COOH), respectively [29]. Also, the intensity of peaks at around 3300 cm−1 increased, showing the increasing hydroxyl groups (O–H) on the surface of oxidized MWCNTs [30]. The presence of these functional groups enhances the dispersion of oxidized MWCNTs due to their high hydrophilic properties, and consequently it leads to good dispersion of oxidized MWCNTs in PIP aqueous solution and producing more hydrophilic membranes.
3.2. Characterization of the membranes For investigating both surface and cross-sectional structure of the membranes, the SEM images were used. Fig. 2a shows the cross-sectional morphology of a membrane containing 0.005 wt. % oxidized MWCNTs. The image shows that the PSf support has the porous sub-layer supporting the top dense skin layer. This proves the existence of finger-like macro-pores in asymmetric structure. The finger-like porous support was fabricated onto the nonwoven fabric with a thickness of around 80 μm. According to Wang et al. to experience a better permeate flux of the membranes, less closed cells in the membranes is needed [1]. The hydrophilic MWCNTs provide this desirable property [14,31], therefore by looking at the result of this study, it is noticeable that the embedded membranes show a better permeate fluxes rather than the bare one. The thickness of prepared PA layer on the support was less than 200 nm. The surface morphology image is a way in which some of membrane characteristics can be explained. In Fig. 2, the surface SEM images of polyamide nanocomposite membranes containing various amounts of raw MWCNTs and oxidized MWCNTs were
Fig. 2. SEM images of the bare, oxidized and non-oxidized embedded membranes with various amounts of the MWCNTs. (a) Cross-sections of membrane with 0.005% oxidized MWCNTs.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
5
Table 2 Surface roughness parameters of the prepared piperazineamide NF membranes resulted from analyzing three randomly chosen AFM images (2 μm × 2 μm). Membrane
Bare PA NF 0.001 Raw MWCNTs 0.002 Raw MWCNTs 0.005 Raw MWCNTs 0.01 Raw MWCNTs 0.001 Oxidized MWCNTs 0.002 Oxidized MWCNTs 0.005 Oxidized MWCNTs 0.01 Oxidized MWCNTs
Roughness parameters Sa (nm)
Sq (nm)
Sy (nm)
33 21 17 22 27 14 15 16 23
44 29 26 31 37 23 24 27 30
350 202 166 217 284 130 143 166 243
presented. Because of the low concentration of MWCNTs, the SEM image could not show the MWCNTs embedded in PA layer. Attention to the surface images showed that polymerization reaction caused formation of an active rough layer covering the surface of the membranes. The higher roughness of the bare membrane is an intrinsic quality. However, by looking at images of the embedded membranes, it is obvious that this intrinsic property can be overcome by adding the MWCNTs. By comparing the raw and oxidized MWCNTs, it can be seen that by adding the oxidized MWCNTs, the surface becomes smoother. With increasing the loading of the MWCNTs, the peaks and nodules form more cross-linking areas and by enlarging these areas, decreasing in roughness happened [32-34]. This reduction in membrane surface roughness can be obviously seen in AFM images as presented in Fig. 3. Regarding the surface of the 0.001 wt. % raw and oxidized MWCNTs membranes, the change in the roughness is considerable. Table 2 presents the surface roughness parameters including average roughness (Sa ), root mean square of the Z data (Sq ) and the height difference between the highest peak and the lowest valley (Sy ). As presented, the surface of the bare PA NF membrane had the highest roughness and it was influenced significantly by blending the MWCNTs. The surface roughness was decreased by increasing the content of the MWCNTs up to 0.002 wt. % and the smoothest surface was observed in the case of 0.001 oxidized MWCNTs membrane. This roughness decrease can be related to the hydrogen bonding between the polyamide layer and the functional groups of oxidized MWCNTs [4]. The hydroxyl groups present on MWCNTs underwent hydrogen bonding, resulting in a more compact chain structure. Similar results have been reported by other authors [19,35]. In the case of SEM images of the 0.001 wt. % raw and oxidized MWCNTs membranes, the same change can be observed only with this difference that the nodules are more linked which leads to better permeate flux. Although improvement of the roughness occurred in this amount of embedding, but adding more MWCNTs up to 0.005 wt. % resulted in an even more linked nodules, in which cross-linking reaction could lead to smooth surface [36] and hydrogen bonding may cause nodular structure to become smaller and thus the total surface roughness decreases [37-40]. The hydrophilicity evaluation of the fabricated nanofiltration membranes was conducted by contact angle measurement, which is presented in Fig. 4. Not having the advantage of embedding oxidized MWCNTs, the bare membrane showed a relatively high contact angle; however, compared to raw MWCNTs membrane its contact angle was lower. The high contact angle of raw MWCNTs membranes can be explained by natural carbon-based properties of the MWCNTs. The polyamide layer have carbonyl and N–H groups that could have better hydrophilic properties rather than
Fig. 3. Three-dimensional AFM images of the prepared PA NF nanocomposite membranes (the Z-scaling: 2 has been applied for all images for better illustration).
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
JID: JTICE 6
ARTICLE IN PRESS
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
Fig. 4. Surface contact angles of the MWCNTs embedded NF membranes.
the C–C structure of the MWCNTs. Using of oxidized MWCNTs, the contact angle decreased significantly, as a result hydrophilicity increased due to this fact [41]. It can be seen that contact angle for membrane with 0.01 wt. % raw MWCNTs was 61.8° and for the same amount of oxidized MWCNTs was 51.4°. This might be reasonable by considering the presence of –COOH and –OH functional groups, which induce their hydrophilicity to the membrane surface [35]. 3.3. Nanofiltration performance The evaluation of nanofiltration performance can be measured by parameters like permeate flux and salt rejection. Porosity of active layer and surface hydrophilicity lead to determination of pure water flux [43,44]. Mostly, the pore size of nanofiltration membranes is about 1 nm, and the diameter of water molecule is 0.27 nm, so then water molecule could easily permeate the membrane [45]. Fig. 5 shows the saline aqueous solution flux of the membranes. All raw and oxidized MWCNTs embedded membranes presented higher water flux compared to the bare PA NF membrane. This confirms the effective act of blending MWCNTs based nanofillers on producing highly porous membranes [14]. However, this increment is higher for oxidized MWCNTs compared to raw MWCNTs in the same concentrations of the nanofiller. A key factor, which plays an important role in membrane permeability, is hydrophilicity. Fig. 4 shows that presence of oxidized MWCNTs in the polyamide layer caused an effect on water contact angle and therefore, higher hydrophilicity. The oxidized MWCNTs contents had more effect on the water flux of the membranes. By looking at both permeate flux figures of raw and oxidized MWCNTs, it is obvious that on the whole permeate flux in oxidized MWCNTs membranes was much higher than the bare one for all of the additive percentages and higher than the ones with raw MWCNTs. As can be seen in Fig. 5, the permeate flux increased with the increase of oxidized MWCNTs contents, however, the flux increasing stopped in 0.005 wt. % and after this point, with addition of more oxidized MWCNTs, the flux rate decreased. This could result from consequence of more compact structures by MWCNTs agglomeration that results in blocking the pores and reduction in flux [22,23].
In order to test the rejection performance of the prepared membranes, Na2 SO4 feed solution was used. The pH was 6.5 during all tests. The rejection factor for the bare membrane was well above average of some other works with 95.70% [46-48], however, addition of both raw and oxidized MWCNTs showed a growth in rejection. Fig. 6 shows the rejection of the membranes for three kinds of membranes (bare, raw MWCNT with 0.0 01, 0.0 02, 0.0 05, and 0.01 wt. % concentration, and oxidized MWCNTs embedded membranes with the same concentrations as the raw MWCNT ones). Regarding raw MWCNTs, it can be seen that increasing the amount of MWCNTs up to 0.002 wt. % was effective in enhancing the rejection capability, however, addition of more MWCNTs after this point leads to decrease in rejection. With respect to the oxidized MWCNTs, the trend was almost similar to previous one with a slight difference that the most efficient dosage of oxidized MWCNTs membranes was 0.005 wt. % which is the best dosage among all membranes. The reason of this high rejection property of the nanofiltration membranes lies on the negative charges on the membrane surface which can be explained by Donnan exclusion effects and steric exclusion [42, 49]. Fig. 7 shows the Zeta potential measurements of the Bare, 0.005 wt. % raw and 0.005 wt. % oxidized MWCNTs membranes. As shown, by addition of the oxidized MWCNTs to the polyamide layer, due to presence of functional groups in the MWCNTs structure, the negative charge can be induced to the membrane surface. The addition of oxidized MWCNTs to the PA membrane not only declined its contact angle, but also induced a surface negative charge. Moreover, at higher pH, the membrane with oxidized MWCNTs content presented a more negative surface charge due to the existence of carboxylic groups and their subsequent deprotonation. However, the raw MWCNTs have no significant effect on the surface charge. Generally it could be said that the addition of the MWCNTs improved the permeability without reduction in the rejection of the prepared nanocomposite membranes. Table 3 compares the flux, salt retention and water contact angle results in current work with the results of reported nanofiltration membranes in the literatures. As can be seen, the rejection property of our prepared raw and oxidized MWCNTs is the highest and the flux is suitable.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
7
Fig. 5. Saline aqueous solution flux of the prepared PA NF membranes; (a) raw MWCNTs and (b) oxidized MWCNTs embedded membranes.
Table 3 Comparison of results of current study with the literature MWCNTs embedded membranes. Membrane
Optimum flux (L/m2 h)
Na2 SO4 rejection (%)
Contact angle (°)
Ref.
0.5 wt. % carboxylated MWNTs/polyester 0.005 NH2 -MWCNTs in PIP Polymethyl methacrylate-MWCNTs in TMC 0.04 wt. % oxidized MWCNTs/PES 0.001 wt. % sulfonated MWCNTs in PIP 0.002 wt. % raw MWCNTs
≈ 22 (6 bar) 55.4 (10 bar) 69.8 (10 bar) 9.1 (4 bar) 79.4 (6 bar) 58.8 (10 bar)
≈ 80 95.7 ≈ 99 ≈ 70 96.8 97.7
23.2 59.3 49 63.3 38 60.9
[13] [23] [37] [46] [50] This study
0.005 wt. % oxidized MWCNTs
65.7 (10 bar)
97.9
57.8
This study
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
ARTICLE IN PRESS
JID: JTICE 8
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
Fig. 6. Na2 SO4 rejection of the membranes embedded with raw and oxidized MWCNTs.
30 0.005 Raw MWCNTs Bare NF 0.005 Oxidized MWCNTs
Zeta potenal (mV)
20 10 0 1
3
5
7
9
pH 11
-10 -20 -30
In regards to the oxidized MWCNTs, they made impressive impacts on the fouling, for example, 0.002 wt. % and 0.005 wt. % oxidized MWCNTs are the two best fouling resistance membranes. Although introducing MWCNTs made great differences but the amount of addition is also important. With introducing 0.01 wt. % or 0.001 wt. % oxidized MWCNTs, the fouling properties was not that significant as what is expected whereas the fouling of 0.01 wt. % of raw MWCNTs membrane was as high as the bare one. Pore blocking and agglomeration phenomenon of nanofiller can be the reason of this reduction in fouling properties [23,53]. In addition, comparing fouling trend with surface roughness parameters (Table 2) shows that the smoother surface has lower fouling tendency.
-40 -50
4. Conclusion
Fig. 7. Surface charge of the prepared PA nanofiltration membranes over pH range.
3.4. Antifouling evaluation In order to evaluate fouling resistance capability of the membranes, BSA/Na2 SO4 solution was used. Fig. 8 shows the flux of the membranes, which were exposed to the foulant solution for 24 h. As can be seen, the permeate flux for all the membranes is reduced during filtration. However, this reduction is lower for the MWCNTs embedded membranes. As a general expectation, the bare membrane has the lowest flux (see Fig. 5) as well as rejection (see Fig. 6) showed the most fouling. The oxidized MWCNTs with the most fluxes among the whole membranes experienced the minimum fouling, although it can be seen that raw MWCNTs show a considerable resistance but not as significant as the oxidized ones. This might be due to the fact that oxidized MWCNTs membranes have more hydrophilic surface because of expanded surface coverage of hydrophilic groups [51,52]. By looking back to Fig. 4 (contact angle measurement), it can be noticed that generally membranes with low contact angle have minimum fouling. In Fig. 9, normalized fluxes of all membranes are illustrated for better possibility of tracing the trend of the fouling. The flux of the bare membrane showed a dramatic drop by adding the BSA/Na2 SO4 solution to the feed. However, by embedding MWCNTs to the membranes the trend of the fouling changed significantly.
In this study, performance of three different kinds of membranes was evaluated in three subjects of salt rejection, flux, and antifouling capability. Bare membrane was the one membrane, which made it possible to compare the raw MWCNTs membranes with four concentrations and oxidized MWCNTs membranes with the same amounts. Both raw and oxidized MWCNTs had an impact on increasing the flux, however, the impact of oxidized MWCNTs was more significant so that 0.005 wt. % oxidized MWCNTs embedded membrane had the most permeation. From SEM and AFM images, it can be conceived that by embedding oxidized MWCNTs, the surface of the membranes becomes smoother; hence, the surface is less vulnerable to BSA, therefore the fouling rate decreased. Furthermore, addition of MWCNTs improved the antifouling properties until 0.002 wt. %, nevertheless further increasing ended up in reduction of this quality, due to roughness increasing by addition of more MWCNTs. Na2 SO4 rejection in all kind was acceptable, but again a range of increase was observed in oxidized MWCNTs as well as raw ones.
Acknowledgments The authors are thankful to the partial support from Iranian Nano Technology Initiative Council, Iran National Science Foundation and Kharazmi University.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
JID: JTICE
ARTICLE IN PRESS
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
Fig. 8. Foulant solution flux of the membranes during 24 h filtration (BSA conc. = 100 ppm, Na2 SO4 conc. = 20 0 0 ppm, 10 bar).
Fig. 9. Normalized permeate flux of the foulant solution during 24 h filtration.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039
9
JID: JTICE 10
ARTICLE IN PRESS
[m5G;April 17, 2017;16:1]
M.R. Mahdavi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–10
References [1] Wang L, Song X, Wang T, Wang S, Wang Z, Gao C. Fabrication and characterization of polyethersulfone/carbon nanotubes (PES/CNTs) based mixed matrix membranes (MMMs) for nanofiltration application. Appl Surf Sci 2015;330:118–25. [2] Zhu J, Tian M, Zhang Y, Zhang H, Liu J. Fabrication of a novel “loose” nanofiltration membrane by facile blending with Chitosan–Montmorillonite nanosheets for dyes purification. Chem Eng J 2015;265:184–93. [3] Xu YC, Wang ZX, Cheng XQ, Xiao YC, Shao L. Positively charged nanofiltration membranes via economically mussel-substance-simulated co-deposition for textile wastewater treatment. Chem Eng J 2016;303:555–64. [4] Safarpour M, Vatanpour V, Khataee A, Esmaeili M. Development of a novel high flux and fouling-resistant thin film composite nanofiltration membrane by embedding reduced graphene oxide/TiO2 . Sep Purif Technol 2015;154:96–107. [5] Liu Y, Zhang S, Zhou Z, Ren J, Geng Z, Luan J, et al. Novel sulfonated thin-film composite nanofiltration membranes with improved water flux for treatment of dye solutions. J Membr Sci 2012;394:218–29. [6] Nunes S, Peinemann KV. Presently available membranes for liquid separation. Membrane Technology: In the Chemical Industry. 2nd Revised and Extended ed; 2006. p. 15–38. [7] Jhaveri JH, Murthy Z. A comprehensive review on anti-fouling nanocomposite membranes for pressure driven membrane separation processes. Desalination 2016;379:137–54. [8] Jeong BH, Hoek EM, Yan Y, Subramani A, Huang X, Hurwitz G, et al. Interfacial polymerization of thin film nanocomposites: a new concept for reverse osmosis membranes. J Membr Sci 2007;294:1–7. [9] Jadav GL, Singh PS. Synthesis of novel silica-polyamide nanocomposite membrane with enhanced properties. J Membr Sci 2009;328:257–67. [10] Kim ES, Deng B. Fabrication of polyamide thin-film nano-composite (PA-TFN) membrane with hydrophilized ordered mesoporous carbon (H-OMC) for water purifications. J Membr Sci 2011;375:46–54. [11] Liang S, Qi G, Xiao K, Sun J, Giannelis EP, Huang X, et al. Organic fouling behavior of superhydrophilic polyvinylidene fluoride (PVDF) ultrafiltration membranes functionalized with surface-tailored nanoparticles: Implications for organic fouling in membrane bioreactors. J Membr Sci 2014;463:94–101. [12] Roy S, Ntim SA, Mitra S, Sirkar KK. Facile fabrication of superior nanofiltration membranes from interfacially polymerized CNT-polymer composites. J Membr Sci 2011;375:81–7. [13] Wu H, Tang B, Wu P. Optimization, characterization and nanofiltration properties test of MWNTs/polyester thin film nanocomposite membrane. J Membr Sci 2013;428:425–33. [14] Majeed S, Fierro D, Buhr K, Wind J, Du B, Boschetti-de-Fierro A, et al. Multi-walled carbon nanotubes (MWCNTs) mixed polyacrylonitrile (PAN) ultrafiltration membranes. J Membr Sci 2012;403:101–9. [15] Shah P, Murthy C. Studies on the porosity control of MWCNT/polysulfone composite membrane and its effect on metal removal. J Membr Sci 2013;437:90–8. [16] Esfahani MR, Tyler JL, Stretz HA, Wells MJ. Effects of a dual nanofiller, nano-TiO2 and MWCNT, for polysulfone-based nanocomposite membranes for water purification. Desalination 2015;372:47–56. [17] Park J, Choi W, Kim SH, Chun BH, Bang J, Lee KB. Enhancement of chlorine resistance in carbon nanotube based nanocomposite reverse osmosis membrane. Desalin. Water Treat. 2010;15:198–204. [18] Cheng XQ, Zhang C, Wang ZX, Shao L. Tailoring nanofiltration membrane performance for highly-efficient antibiotics removal by mussel-inspired modification. J Membr Sci 2016;499:326–534. [19] Zhao H, Qiu S, Wu L, Zhang L, Chen H, Gao C. Improving the performance of polyamide reverse osmosis membrane by incorporation of modified multi-walled carbon nanotubes. J Membr Sci 2014;450:249–56. [20] Xu YC, Tang YP, Liu LF, Guo ZH, Shao L. Nanocomposite organic solvent nanofiltration membranes by a highly-efficient mussel-inspired co-deposition strategy. J Membr Sci 2017;526:532–42. [21] Amini M, Jahanshahi M, Rahimpour A. Synthesis of novel thin film nanocomposite (TFN) forward osmosis membranes using functionalized multi-walled carbon nanotubes. J Membr Sci 2013;435:233–41. [22] Farahbaksh J, Delnavaz M, Vatanpour V. Investigation of raw and oxidized multiwalled carbon nanotubes in fabrication of reverse osmosis polyamide membranes for improvement in desalination and antifouling properties. Desalination 2017;410:1–9. [23] Zarrabi H, Yekavalangi ME, Vatanpour V, Shockravi A, Safarpour M. Improvement in desalination performance of thin film nanocomposite nanofiltration membrane using amine-functionalized multiwalled carbon nanotube. Desalination 2016;394:83–90. [24] Yakavalangi ME, Rimaz S, Vatanpour V. Effect of surface properties of polysulfone support on the performance of thin film composite polyamide reverse osmosis membranes. J Appl Polym Sci 2017;134:44444. [25] Jacobasch HJ, Schurz J. Characterization of polymer surfaces by means of electrokinetic measurements. Prog Colloid Polym Sci 1988;77:40–8. [26] Mehdipour S, Vatanpour V, Kariminia H-R. Influence of ion interaction on lead removal by a polyamide nanofiltration membrane. Desalination 2015;362:84–92. [27] Fairbrother F, Mastin H. Studies in electroendosmosis. J Chem Soc 1924;125:2319–30.
[28] Walker SL, Bhattacharjee S, Hoek EMV, Elimelech M. A novel asymmetric clamping cell for measuring streaming potential of flat surfaces. Langmuir 2002;18:2193–8. [29] Yang S, Li J, Shao D, Hu J, Wang X. Adsorption of Ni (II) on oxidized multi-walled carbon nanotubes: effect of contact time, pH, foreign ions and PAA. J Hazard Mater 2009;166:109–16. [30] Madaeni SS, Derakhshandeh K, Ahmadi S, Vatanpour V, Zinadini S. Effect of modified multi-walled carbon nanotubes on release characteristics of indomethacin from symmetric membrane coated tablets. J Membr Sci 2012;389:110–16. [31] Daraei P, Madaeni SS, Ghaemi N, Khadivi MA, Astinchap B, Moradian R. Enhancing antifouling capability of PES membrane via mixing with various types of polymer modified multi-walled carbon nanotube. J Membr Sci 2013;444:184–91. [32] Kwak SY, Jung SG, Yoon YS, Ihm DW. Details of surface features in aromatic polyamide reverse osmosis membranes characterized by scanning electron and atomic force microscopy. J Polym Sci Part B Polym Phys 1999;37:1429–40. [33] Rajaeian B, Rahimpour A, Tade MO, Liu S. Fabrication and characterization of polyamide thin film nanocomposite (TFN) nanofiltration membrane impregnated with TiO2 nanoparticles. Desalination 2013;313:176–88. [34] Jin L, Yu S, Shi W, Yi X, Sun N, Ge Y, et al. Synthesis of a novel composite nanofiltration membrane incorporated SiO2 nanoparticles for oily wastewater desalination. Polymer 2012;53:5295–303. [35] Bano S, Mahmood A, Kim S-J, Lee K-H. Graphene oxide modified polyamide nanofiltration membrane with improved flux and antifouling properties. J Mater Chem A 2015;3:2065–71. [36] Huang SH, Hsu CJ, Liaw DJ, Hu CC, Lee KR, Lai JY. Effect of chemical structures of amines on physicochemical properties of active layers and dehydration of isopropanol through interfacially polymerized thin-film composite membranes. J Membr Sci 2008;307:73–81. [37] Shen J n, Yu C c, Ruan H m, Gao C j, Van der Bruggen B. Preparation and characterization of thin-film nanocomposite membranes embedded with poly (methyl methacrylate) hydrophobic modified multiwalled carbon nanotubes by interfacial polymerization. J Membr Sci 2013;442:18–26. [38] Wang Z, Yu H, Xia J, Zhang F, Li F, Xia Y, et al. Novel GO-blended PVDF ultrafiltration membranes. Desalination 2012;299:50–4. [39] Sun H, Liu S, Zhou G, Ang HM, Tade´ MO, Wang S. Reduced graphene oxide for catalytic oxidation of aqueous organic pollutants. ACS Appl Mater Interface 2012;4:5466–71. [40] Lee SY, Kim HJ, Patel R, Im SJ, Kim JH, Min BR. Silver nanoparticles immobilized on thin film composite polyamide membrane: characterization, nanofiltration, antifouling properties. Polym Adv Technol 2007;18:562–8. [41] Sun M, Su Y, Mu C, Jiang Z. Improved antifouling property of PES ultrafiltration membranes using additive of silica−PVP nanocomposite. Ind Eng Chem Res 2009;49:790–6. [42] Choi JH, Jegal J, Kim WN. Fabrication and characterization of multi-walled carbon nanotubes/polymer blend membranes. J Membr Sci 2006;284:406–15. [43] Yu HY, Xu ZK, Yang Q, Hu MX, Wang SY. Improvement of the antifouling characteristics for polypropylene microporous membranes by the sequential photoinduced graft polymerization of acrylic acid. J Membr Sci 2006;281:658–65. [44] Zinadini S, Zinatizadeh A, Rahimi M, Vatanpour V, Zangeneh H, Beygzadeh M. Novel high flux antifouling nanofiltration membranes for dye removal containing carboxymethyl chitosan coated Fe3 O4 nanoparticles. Desalination 2014;349:145–54. [45] Jin Jb, Liu Dq, Zhang Dd, Yin Yh, Zhao Xy, Zhang Yf. Preparation of thin-film composite nanofiltration membranes with improved antifouling property and flux using 2, 2 -oxybis-ethylamine. Desalination 2015;355:141–6. [46] Vatanpour V, Madaeni SS, Moradian R, Zinadini S, Astinchap B. Fabrication and characterization of novel antifouling nanofiltration membrane prepared from oxidized multiwalled carbon nanotube/polyethersulfone nanocomposite. J Membr Sci 2011;375:284–94. [47] Zhao FY, An QF, Ji YL, Gao CJ. A novel type of polyelectrolyte complex/MWCNT hybrid nanofiltration membranes for water softening. J Membr Sci 2015;492:412–21. [48] Madaeni SS, Zinadini S, Vatanpour V. Preparation of superhydrophobic nanofiltration membrane by embedding multiwalled carbon nanotube and polydimethylsiloxane in pores of microfiltration membrane. Sep Purif Technol 2013;111:98–107. [49] Liu M, Yao G, Cheng Q, Ma M, Yu S, Gao C. Acid stable thin– film composite membrane for nanofiltration prepared from naphthalene-1,3,6-trisulfonylchloride (NTSC) and piperazine (PIP). J Membr Sci 2012;415:122–31. [50] Zheng J, Li M, Yu K, Hu J, Zhang X, Wang L. Sulfonated multiwall carbon nanotubes assisted thin-film nanocomposite membrane with enhanced water flux and anti-fouling property. J Membr Sci 2017;524:344–53. [51] Wei Y, Chu HQ, Dong BZ, Li X, Xia SJ, Qiang ZM. Effect of TiO2 nanowire addition on PVDF ultrafiltration membrane performance. Desalination 2011;272:90–7. [52] Zhang M, Nguyen QT, Ping Z. Hydrophilic modification of poly (vinylidene fluoride) microporous membrane. J Membr Sci 2009;327:78–86. [53] Qiu S, Wu L, Pan X, Zhang L, Chen H, Gao C. Preparation and properties of functionalized carbon nanotube/PSF blend ultrafiltration membranes. J Membr Sci 2009;342:165–72.
Please cite this article as: M.R. Mahdavi et al., Fabrication and water desalination performance of piperazine–polyamide nanocomposite nanofiltration membranes embedded with raw and oxidized MWCNTs, Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2017.03.039