- Email: [email protected]
Novel antifouling nano-enhanced thin-film composite membrane containing cross-linkable acrylate-alumoxane nanoparticles for water softening
Journal of Colloid and Interface Science 485 (2017) 81–90
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis
Novel antifouling nano-enhanced thin-film composite membrane containing cross-linkable acrylate-alumoxane nanoparticles for water softening Negin Ghaemi Department of Chemical Engineering, Kermanshah University of Technology, 67178 Kermanshah, Iran
g r a p h i c a l a b s t r a c t
Mg2+ rejection tests
a r t i c l e
i n f o
Article history: Received 10 June 2016 Revised 26 August 2016 Accepted 15 September 2016 Available online 15 September 2016 Keywords: Thin-film nanocomposite membrane Cross-linkable nanoparticles Polypyrrole Acrylate-alumoxane Water softening Antifouling
a b s t r a c t A novel thin-film composite (TFC) nanofiltration membrane was prepared using polymerization of pyrrole monomers on the PES ultrafiltration membrane. To improve the characteristics of hydrophobic polypyrrole (PPy) thin-film layer, cross-linkable acrylate-functionalized alumoxane nanoparticles with different concentrations were embedded into the thin-film during polymerization process, and thinfilm nanocomposite (TFNC) membranes were prepared. The characteristics and performance of TFC and TFNC membranes were assessed through the morphological analyses (SEM, AFM), measurement of hydrophilicity and solid–liquid interfacial free energy, water permeability and Mg2+ removal tests. Addition of proper amount of nanoparticles into the polymerization mixture led to the preparation of membranes with more hydrophilic, thinner and smoother active layer as well as higher water permeability compared to TFC control membrane. TFNC membrane prepared with 0.025 g of nanoparticles was the most efficient membrane since it exhibited the highest rejection of MgCl2 and MgSO4 salts. Antifouling capability of membranes, in terms of flux recovery and fouling parameters, demonstrated the high tolerance of TFNC against fouling. Ó 2016 Elsevier Inc. All rights reserved.
1. Introduction
E-mail addresses: [email protected], [email protected] http://dx.doi.org/10.1016/j.jcis.2016.09.035 0021-9797/Ó 2016 Elsevier Inc. All rights reserved.
Hard water is undesirable in the domestic water supply and the industrial applications as a result of its high mineral content which
82
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
Nomenclature
Symbols A AFM CF CP DMAc FR M NF PES PPy PVP PWF Rej Rirf Rrf Rtf RO SEM
& abbreviations active membrane surface area (m2) atomic force microscopy Mg2+ concentration in the feed (mg/l) Mg2+ concentration in the permeate sample (mg/l) N,N-dimethylacetamide flux recovery (%) weight of permeated water through the membrane (kg) nanofiltration polyethersulfone polypyrrole polyvinylpyrrolidone pure water flux (kg/m2 h) rejection (%) irreversible fouling ratio reversible fouling ratio total fouling ratio reverse osmosis scanning electron microscopy
reduces the household cleaning efficiency, makes scaling and corrosion problems, and causes serious difficulties in pipe lines of boilers, heat exchangers and electrical devices. To solve this issue, water softening processes are employed by which calcium and magnesium cations as well as other divalent or multivalent metal ions are eliminated from hard water. Compared with conventional water softening techniques including ion-exchange resins, zeolites or lime-soda ash treatment methods, water softening using membranes have received a specific place in this market due to higher efficiency, lower cost and less chemical and energy consumption [1,2]. Meanwhile reverse osmosis (RO) process as a high performance membrane filtration process serving for removal of solutes and minerals of hard water suffers the problems of high energy consumption and severe scaling and fouling [3,4]. Nanofiltration (NF) membranes have a looser structure rather than RO membranes. Hence, a lower applied pressure is required for NF to achieve an acceptable salt removal and water permeation even higher compared to RO membranes [5–7]. In water softening process for household consumption, it is not needed to completely remove the water hardness, so NF membranes might be a beneficial alternative to RO membranes. Among the currently available NF membranes, thin-film composite (TFC) membranes are particularly applied in water softening processes. These membranes enhance both selectivity and permeability with less energy consumption compared to the typical asymmetric membranes [3,8]. They are usually composed of an ultra-thin active layer responsible for separation and a porous substrate layer providing mechanical resistance. Polyethersulfone (PES) is frequently used for fabrication of substrate in TFC membranes due to the excellent chemical, thermal and mechanical resistance as well as almost a low price. On the other hand, thin-film in the most of TFC membranes is prepared using interfacial polymerization of hydrophilic polymers such as polyamide and polyvinyl alcohol on the substrate. Despite many advantages of a hydrophilic thin-film especially in view of increasing the water permeability, high hydrophilicity is not necessarily the only criterion to choose the material for thin layer. Adhesion between substrate and top layer is really important in the preparation of TFC membranes. It was proved that the adhesion of a hydrophilic layer on a hydrophobic substrate is weaker in comparison with that in the case of a hydrophobic thin-film [3].
SR TFC TFNC TFNC1 TFNC2 TFNC3 WCA WFi WFF WFw DGSL
Dt D
cL
mean surface roughness (nm) thin-film composite thin-film nanocomposite thin-film nanocomposite membrane prepared by 0.005 g of nanoparticles thin-film nanocomposite membrane prepared by 0.025 g of nanoparticles thin-film nanocomposite membrane prepared by 0.05 g of nanoparticles water contact angle (°) initial pure water flux (kg/m2 h) final pure water flux (kg/m2 h) water flux during the filtration of whey solution (kg/ m2 h) surface roughness corrected solid–liquid interfacial free energy (mJ/m2) sampling time (h) relative surface area liquid surface tension (mJ/m2)
Despite the particular characteristics of polypyrrole (PPy) (good thermal and environmental stability as well as ease of preparation [9]), no study has been conducted in which the thin-layer is composed of the hydrophobic PPy. Hence, PPy would be an appropriate choice for fabrication of thin-film on PES substrate; however, it should be noticed that low water permeation and high fouling of TFC membrane with hydrophobic thin-layer cause crucial problems in practice. The key merit for composite membranes is that the characteristics of thin layer and porous substrate layer can be independently optimized to achieve better performance [10]. In this regard, thin-film nanocomposite (TFNC) membranes were developed [3] in which some nanoparticles such as TiO2 [11–16], clay [17], graphene oxide [18–20], iron oxide [21], zeolite [22,23], SiO2 [24,25], and carbon nanotubes [26,27] were embedded within the thin dense layer of TFC membrane mostly with the aim of increasing the hydrophilicity without sacrificing the separation efficiency of membrane [17–19,21]. Boehmite (c-AlOOH) as the other applicable nanoparticles were also employed in the fabrication of nanocomposite membranes due to their specific characteristics particularly high hydrophilicity [28]. Reviewing the literature revealed that addition of boehmite nanoparticles into the membrane matrix resulted in improving the hydrophilicity due to the extra hydroxyl groups of nanoboehmites [29,30]. Moreover, nanoboehmite can be modified using acrylic acid due to the ability of carboxyl groups to be coordinated with aluminum as unidentate and bidentate ligands [28,31,32]. This process leads to the production of acrylate-alumoxane nanoparticles containing C@C bonds with capability of participating in the polymerization reaction [28]. The current study aims to prepare TFNC membrane with improved efficiency in water softening. PES was applied for fabrication of ultrafiltration substrate, and the thin-film was developed through aqueous polymerization of pyrrole on the surface of the substrate. Acrylate-alumoxane nanoparticles were employed as additive for raising the hydrophilicity and improving the quality of PPy thin-film layer. Hence, different concentrations of nanoparticles were introduced as filler into the thin-layer during polymerization reaction. To probe the effects of the employed method on the characteristics and performance of membranes, morphological analyses (scanning electron microscopy (SEM), atomic force
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
microscopy (AFM)), water contact angle (WCA) measurements, water permeability and softening experiments as well as antifouling tests were conducted. 2. Experimental 2.1. Preparation of thin-film composite (TFC) and nanocomposite (TFNC) membranes Ultrafiltration PES membrane as the support layer was fabricated using phase inversion method. A casting solution containing 16 wt.% PES (Ultrason E6020P, MW = 58,000 g/mol, BASF, Germany) as base polymer, 2 wt.% polyvinylpyrrolidone (PVP, MW = 25,000 g/mol, Merck, Germany) as pore former and dimethylacetamide (DMAc, Merck, Germany) as solvent was prepared and mixed by stirring at 400 rpm for half a day until a homogeneous solution was achieved. After ultrasonication for 15 min, solution was casted using a homemade applicator on the non-woven polyester fiber (83 g/m2, USA). The casted film was immersed in the distilled water bath as nonsolvent, and after primary formation of membrane, it was kept in the fresh distilled water for a day in order to guarantee the completion of phase inversion process. Finally, the membrane was kept between two pieces of filter papers to dry for another day. All procedures were done at room temperature. Thin-film of TFC membrane was formed by chemical polymerization of pyrrole on the surface of a pre-cast PES substrate. In this study, distilled water was employed as solvent. For polymerization process, membrane was mounted at the bottom of a container, and 50 ml of 14 wt.% FeCl36H2O (Merck, Germany) aqueous solution (as initiator) was poured into the container and held for 12 h to ensure the penetration of the solution into the pores of the substrate. After draining off the excess amount of FeCl3 solution from the container, 50 ml of 1 wt.% pyrrole (Merck, Germany) aqueous solution was poured into the container. To insert the nanoparticles into the thin film, different quantities of acrylate-alumoxane nanoparticles (0, 0.005, 0.025 and 0.05 g) were added into the pyrrole solution. Acrylate-alumoxane was synthesized based on the method introduced in the previous study [28]. Briefly, bohemite nanoparticles were refluxed in acrylic acid for 72 h and then, washed with n-hexane and dried at 50 °C. Monomer solution containing nanoparticles was sonicated for 30 min to avoid the nanoparticle agglomeration. Then, the sonicated solution was poured onto the surface of the substrate previously exposed to FeCl3 solution and was remained for another 12 h. After draining off the monomer-nanoparticle solution, membrane was thoroughly washed with plenty of distilled water in order to remove unreacted FeCl3 and pyrrole and unattached nanoparticles from the surface of the membrane substrate. For drying the membrane, it was held between two pieces of filter paper at room temperature for 24 h. The thin-film membranes prepared with 0, 0.005, 0.025 and 0.05 g of nanoparticles are denoted as TFC, TFNC1, TFNC2 and TFNC3 in the manuscript. 2.2. Characterization of membranes Membrane structure and surface properties were investigated by use of the typical and highly applicable methods of scanning electron microscopy (SEM) and atomic force microscopy (AFM). Cross-section structure of membranes was monitored using capturing SEM micrographs by a scanning electron microscope (SEM, KYKY-EM3200, China) apparatus. Also, thin-layer thickness of membranes was specified by measuring the thickness in four or five casual points on the top-layer and reporting the average value as the membrane thin-layer thickness. Surface roughness (SR) of membrane as one the characteristics influential on the membrane
83
fouling as well as the mean size of membrane surface pores were determined by scanning the membrane surface using an AFM instrument (ARA-AFM, Ara-Research Co., Iran) in non-contact mode and image analyzer software (version 1.01 Ara-Research). To reach the more precise roughness data, two pieces of each membrane were scanned. The average values of surface roughness were reported as SR. Surface hydrophilicity of membrane was measured by dropping the small droplets of distilled water using a microsyringe on five points on the membrane surface and taking the images by a digital microscope (USB Digital Microscope, 2.0 Mega Pixel Color Video Camera, 500X, China). An Image J freeware was employed to measure the contact angle between the membrane surface and water droplets. It should be mentioned that the reported WCA for each membrane was the average amount of measured values. It is clear that a smaller water contact angle means a higher tendency of water to spread on the surface, i.e., higher hydrophilicity of membrane. Moreover, a modified form of the Young–Dupre equation was used to assess the relative hydrophilicity of membranes by the solid–liquid interfacial free energy [33,34]. It should be noticed that the increase in water contact angle value could not be used to show the increased hydrophobicity of the membrane since the surface roughness can also contribute to the change of contact angle value [33,35]. Therefore, for better representation of hydrophilicity, the surface roughness corrected solid–liquid interfacial free energy, DGSL (mJ/m2) was determined using Eq. (1) [33,34].
cos h DGSL ¼ cL 1 þ D
ð1Þ
where h and cL are the average water contact angle value and the liquid surface tension which is 72.8 mJ/m2 for pure water at 25 °C, respectively. Moreover, D is the relative surface area which was determined using AFM topography images by MATLAB 7.8 software (license num. 161051) [36]. Since the contact angle in a smooth surface (D = 1) is 0°, the maximum solid–liquid interfacial free energy is close to 2cL . Hence, a higher amount of DGSL suggests a higher hydrophilicity of the surface [33,34]. 2.3. Permeability and efficiency of membranes in water softening Membrane performance was evaluated in terms of pure water flux (PWF) and water softening. For measuring the PWF, a defect-free circular piece of membrane was cut and mounted in a dead-end test cell with active membrane surface area of 12.56 cm2. After filling the cell with distilled water, the membrane was compressed under 0.5 MPa pressure for an hour. Then, pressure was reduced to 0.4 MPa and PWF (kg/m2 h) was calculated after an hour using the following equation:
PWF ¼
M At
ð2Þ
where M, A and t are weight of permeated water through the membrane (kg), active membrane surface area (m2) and collecting time of permeated water (h). In order to probe the efficiency of membrane in water softening, magnesium salt solution with high concentration 500 ppm (very hard water) were employed. For investigation of probable effects of counter ions, MgCl2 and MgSO4 solutions were used as feed. To calculate the rejection of magnesium ions (Rej (%)) using fabricated membranes, the following equation was employed:
Rej ð%Þ ¼
CP 100 1 CF
ð3Þ
where C p and C F are concentration (mg/l) of Mg2+ in the permeate and feed solutions, respectively. Magnesium ion concentration
84
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
was measured using chelatometric titration by Na2EDTA (Merck, Germany) solution and in the presence of Eriochrome Black T (Merck, Germany) indicator in the buffer (pH 10.0). It should be mentioned that all filtration experiments were repeated two times using a fresh piece of membrane, and the average amount was reported as the final value.
All tests were carried out twice and each time with a fresh piece of membrane, to increase the accuracy of results.
3. Results and discussion 3.1. Membrane characteristics and water permeability
2.4. Fouling experiments
FR ð%Þ ¼ ðWF F =WF i Þ 100
ð4Þ
where WF F (kg/m2 h) and WF i (kg/m2 h) are the final (aftercleaning) and the initial PWF, respectively. Considering Eq. (4), it is clear that higher amount of FR refers to the higher capability of membrane to be regenerated after cleaning and therefore, higher antifouling property [37,38]. In fact, this parameter indicates what percent of membrane permeability is recovered after fouling and cleaning process. In order to analyze more minutely the antifouling behavior of a membrane, other parameters i.e. the reversible fouling ratio (Rrf) as the fraction of fouling which can be easily removed by simple washing, the irreversible fouling ratio (Rirf) as the fraction of fouling that firmly sticks on the membrane surface and the total fouling ratio (Rtf) as sum of the Rrf and Rirf might be calculated as well. The aforementioned parameters are calculated using the following equations [37,38]:
Rrf ð%Þ ¼
WF F WF w 100 WF i
ð5Þ
Rirf ð%Þ ¼
WF i WF F 100 WF i
ð6Þ
Rtf ð%Þ ¼
WF i WF w 100 WF i
ð7Þ
Results of PWF of prepared membranes are shown in Fig. 1. As expected, membrane permeability decreased from 34 for PES substrate to 9 kg/m2 h for TFC. This decrement is related to the formation of PPy thin-film on the substrate and therefore, enhancement of the membrane thickness and also, the hydrophobicity of PPy. Formation of polymeric layer on the substrate is clearly observable as a result of remarkable change in the color of membrane surface from white to black after polymerization of pyrrole monomers on the membrane substrate (Fig. 2). Also, SEM cross-section image of TFC (Fig. 3) declared the formation of PPy thin-layer with approximately 25 lm thickness on the substrate. By addition of different amounts of nanoparticles into PPy thinfilm, PWF of the prepared membranes increased compared with TFC (see Fig. 1) so that it grew from 9 kg/m2 h for TFC to 15, 28 and 32 kg/m2 h for TFNC1, TFNC2 and TFNC3, respectively. Average values of thin-layer thickness of TFNC membranes measured by digital micrometer are tabulated in Table 1. The data reveals that thickness of thin-layer decreases by addition of nanoparticles. This achievement might be due to the presence of nanoparticles and their beneficial effect on the polymeric thin-film layer [12]. Presence of nanoparticles in the monomer solution decreases the number of monomers which can diffuse to the substrate and contribute
40 35
PWF(kg/m2.h)
One of the reasons of employing acrylate-alumoxane nanoparticles, except betterment of thin-film characteristics, was decreasing the membrane fouling resulted from PPy as a polymer with the hydrophobic nature. Hence, the antifouling properties of membranes with a desirable performance in water softening were investigated. The antifouling capability of membranes was determined using whey solution as a severe foulant. Generally, there are some parameters by which the antifouling characteristic of prepared membranes can be well explored. In order to calculate antifouling parameters, an experimental procedure was thoroughly followed for all membranes. First, pure water flux was determined for each membrane under 0.4 MPa transmembrane pressure during 60 min. This amount was considered as initial water flux and showed by WFi (kg/m2 h). Then, distilled water was replaced by whey solution, and water permeation test was continued under the same pressure during 90 min. Water permeated through the membrane was shown by WFw (kg/m2 h). Cleaning the membrane should be done at this stage of experiments. For this purpose, the membrane was brought out of the cell and immersed in distilled water for 30 min after washing with plenty amount of distilled water. After the cleaning process, membrane was mounted again into the cell, and the first step was repeated. Pure water permeated through the membrane at this stage was represented by WFF (kg/m2 h) in which subscript F is referred to the final stage of filtration tests. It is worthy to mention that in order to reduce the effects of concentration polarization, a constant agitation with speed of 350 rpm was applied during all experiments. Flux recovery (FR (%)) factor calculated using Eq. (4) is one of the most applicable and well-known parameters representing antifouling property of a membrane [37].
34
32 28
30 25 20
15
15 9
10 5 0 Substrate
TFC
TFNC1
TFNC2
TFNC3
Fig. 1. Pure water flux of prepared membranes.
Fig. 2. Surface image of (a) PES substrate and (b) thin-film membrane captured by a digital camera.
85
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
PPy thin-film layer
Fig. 4. Surface image of TFNC3 membrane captured by a digital camera.
Non-woven fibre
Fig. 3. SEM cross-section image of thin-film membrane.
to the polymerization reaction [24,39]. This phenomenon results in a reduction in the thickness of the coated layer (thin-film) in TFNC membranes as observed in TFNC1 and TFNC2 membranes. Since thinner coated layer results in higher permeation through the membrane, the enhancement of PWF of TFNC membranes is reasonable. The appearance of TFNC3 membrane surface (see Fig. 4) as well as the thickness of thin-film in this membrane (see Table 1) prove the aforementioned claim and reveal that the presence of nanoparticles with higher amount (>0.025 g) in the polymerization medium disturbs the process of thin-film formation and leads to a non-uniform deposition of polymer on the substrate. The nonuniform coating of PPy in TFNC3 membrane is the main reason of high flux of this membrane (compare fluxes for substrate and TFNC3 in Fig. 1). Except the thickness of thin-film layer which influences on the water permeability of TFNC membranes, hydrophilicity is another effective factor. Results of water contact angle (WCA), relative surface area ðDÞ and DGSL are tabulated in Table 1. Although WCA of TFC membrane (72.2°) did not change notably in comparison with
that in PES substrate (73°) [28,36] due to the presence of hydrophobic polypyrrole on the substrate [40], TFNC membranes presented higher hydrophilicity (lower WCA values) compared with TFC. Among thin-film nanocomposite membranes, although hydrophilicity of TFNC2 (65.2°) increased in comparison with TFNC1 (69.5°) by enhancing the amount of applied nanoparticle, TFNC3 presented the lowest hydrophilicity (71.7°). Higher hydrophilicity of TFNC membranes compared with TFC is due to the presence of hydrophilic acrylate-alumoxane nanoparticles on the membrane surface. Existence of nanoparticles in the polymerization solution leads to trapping of some of the nanoparticles within the polymer chains growing in the thin-film layer. SEM surface image of TFC and TFNC2 membranes obviously confirm the presence of nanoparticles on the surface of the membrane prepared by addition of nanoparticles (Fig. 5). Hydrophilic acrylate-alumoxane nanoparticles with hydroxyl groups in the thin-film matrix partially improve the surface wettability of membrane. Non-uniform deposition of thin-film, as mentioned earlier, might be responsible for low hydrophilicity of TFNC3 membrane. It should be noticed that the decrement in water contact angle value could not be used as the only criterion to show the increased hydrophilicity of membrane since the surface roughness might also affect the contact angle [33,34]. Hence, DGSL values tabulated in Table 1 were employed as better evidence, instead of WCA, for investigation of surface hydrophilicity. Results of DGSL (Table 1) revealed that the interaction between PPy and acrylatealumoxane nanoparticles changed the chemical surface functionality of membrane and led to an increment in the membrane hydrophilicity for the case of TFNC1 and TFNC2 membranes (higher values of DGSL ). Based on the results, addition of an appropriate nanoparticle into the mixture of polymerization reaction might successfully control the thickness of thin-film and permeability of membrane. Besides, it should be noticed that addition of hydrophilic nanoparticle leads to the improvement of membrane hydrophilicity and thin-film thickness; however, an optimum amount of
Table 1 Characteristics of membranes. Membrane
Membrane thickness (thin-layer thickness)a (lm)
WCA (°)
D ()
DGSL (mJ/m2)
PES substrate TFC TFNC1 TFNC2 TFNC3
160 ± 4 185 ± 5 176 ± 3 170 ± 6 165 ± 5
73.0 ± 2.1 72.2 ± 1.3 69.5 ± 0.9 65.2 ± 1.0 71.7 ± 1.4
– 1.53 1.14 1.10 1.35
– 87.3 95.2 100.5 89.7
(–) (25) (16) (10) (5)
Data are the average values of several measurements. a Thin-layer thickness of each membrane was measured by subtracting the substrate thickness from the total membrane thickness.
86
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
by thin-film membranes might be simply related to the significant reduction in the surface pore size after deposition of dense PPy thin-film on the substrate (see Table 2). On the other hand, based on PPy chemical structure and from the literature at the experimental pH of feed solution (i.e. 5.5) PPy possesses positive charge due to the impossibility of deporotonation of NH groups at this pH [41]. Hence, Donnan repulsion is the dominant mechanism in the rejection of cations. Moreover, there is no critical enhancement in the rejection of MgSO4 compared with MgCl2 implying that Mg2+ ions are mainly rejected ions by the membrane instead of
Fig. 5. SEM surface image of (a) TFC and (b) TFNC2 membranes.
nanoparticle should be consumed to avoid the decomposition of thin-film structure. 3.2. Performance of TFC and TFNC membranes in water softening According to Fig. 6, the water hardness was reduced using thinfilm membranes, particularly thin-film nanocomposite membranes; however, a satisfying result was not achieved using TFNC3 membrane due to the unsuccessful uniform coating of thin-film on the PES substrate. It should be mentioned that Mg2+ removal was also tested by PES substrate, and no rejection of magnesium ions was observed as expected. The reason of enhanced ion rejection
MgCl2
the negative bivalent SO2 4 ions. This further reinforces the concept of positive charge of PPy thin-film. This outcome would be very desirable for water softening because the prepared TFC membranes are able to reduce the water hardness directly by removing the ions which are responsible for water hardening. According to the salt rejection results (Fig. 6), TFNC membranes (except TFNC3 due to non-uniform and inappropriate deposition of PPy thin layer on the substrate) showed a superior salt rejection compared with TFC (composite thin-film without the presence of nanoparticles). This can be explained by existence of crosslinkable acrylate-alumoxane nanoparticles in TFN1 and TFN2 membranes. Based on the acrylate-alumoxane scheme (Fig. 7), there is carbon-carbon double bond on the outer surface of nanoparticles which allows participation of these functional groups in the polymerization reaction as schemed in Fig. 8. This can efficiently enhance the compatibility of hydrophilic nanoparticles with hydrophobic PPy matrix and form a tight and cross-linked structure in the deposited thin-film. Considerable reduction in the mean surface pore size of TFNC1 (13 nm) and TFNC2 (9 nm) membranes in comparison with TFC (38 nm) membrane (Table 2) clearly proves the tighter structure of thin-layer and therefore, higher rejection in TFNC membranes [42]. The similar results were also reported for higher selectivity of membranes with a cross-linked polymer in their structure [24,43,44]. These results illustrated that the presence of acrylatealumoxane nanoparticles efficiently improved the properties of PPy thin-film and created a dense and compact thin layer with
Table 2 Mean surface pore size and roughness of membranes. Membrane
Mean surface pore size (nm)
SR (nm)
PES substrate TFC TFNC1 TFNC2 TFNC3
130 ± 15 38 ± 5 13 ± 1 9±2 81 ± 10
– 22.6 ± 1.5 12.3 ± 0.9 10.7 ± 1.0 –
MgSO4
100 90
83
80
68
Rej (%)
70 60 50
70
59 53 47
40 30 20
10
15
10 0 TFC
TFNC1
TFNC2
Fig. 6. Magnesium rejection by thin-film membranes.
TFNC3 Fig. 7. Schematic of acrylate-alumoxane nanoparticles.
87
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
Fig. 8. Scheme of participation of acrylate-alumoxane nanoparticles in the polymerization and cross-linking.
Wfi
30
WFw
WFF
SR (nm) 80
28
FR(%) 71
69.5 65.2
70
25
Flux (kg/m2.h)
WCA(degree)
72.2
20 20
60
53
50
15 15 11 10
9
8
30
5 5
2
40
33 22.6
20
3
12.3
10.7
10
0 TFC
TFNC1
TFNC2
0 TFC
TFNC1
TFNC2
Fig. 9. Initial, whey solution and after-cleaning water flux. Fig. 10. Roughness, hydrophilicity and flux recovery results of thin-film membranes.
higher rejection of MgCl2 (70%) and MgSO4 (83%) by TFNC2 membrane. Considering the physicochemical properties of TFNC membranes, the chemical structure of PPy thin layer (Donnan repulsion effect) along with the compact thin layer composed of smaller pores resulted in the enhancement in the rejection of TFNC membranes [42,45]. Moreover, the membrane rejections obtained in this study were comparable with other reported results of TFNC nanofiltration membranes [1], and as expected, it was normally lower in comparison with the results obtained by membranes which were employed with higher feed concentrations [19,24]. However, the acrylate-alumoxane nano-enhanced thin-film composite membrane (TFNC2) showed relatively higher performance compared with other published studies [16]. Another important factor which declares the proficiency of prepared TFNC membranes in this study is lower amount of employed nanoparticles compared with other researches since the low consumption of nanofiller is an advantage from environmental point of view and really important in industrial applications. 3.3. Antifouling characteristics of membranes Fig. 9 shows the results of initial, whey solution and recovered water flux for thin-film membranes. It should be mentioned that TFNC3 membrane due to the unacceptable results in Mg2+ removal was not assessed in this stage of experiments. As seen, water flux during filtration of whey solution considerably decreased for all
tested thin-film membranes due to the sever fouling caused by whey solution with high concentration; however, in this case, the membrane efficiency in the flux recovery and fouling mitigation has more significant importance. The obtained FR data (Fig. 10) reveal an increment in the order of TFNC2 > TFNC1 > TFC. Generally, surface roughness and hydrophilicity are two important parameters affecting on the membrane fouling. It has been proved that higher hydrophilicity and lower roughness are in favor of lower fouling [38,46]. In order to probe the surface topography and roughness of membranes, AFM images of membranes are exhibited in Fig. 11. Moreover, the FR, surface roughness (SR) and WCA data are illustrated all together in Fig. 10 to efficiently assess the relationship between theses parameters. Considering the AFM images and SR data show that addition of nanoparticles into the thin-film results in formation of a smoother surface in TFNC membranes compared with that in TFC. According to Fig. 10, the surface roughness and WCA enhancement in the membranes are in the order of TFC > TFNC1 > TFNC2 which proves that both higher hydrophilicity and smoother surface are responsible for the observed trend in FR (TFNC2 > TFNC1 > TFC). Hence, the usage of acrylate-alumoxane nanoparticles was influential on the restoration of membrane permeability after becoming fouled. In order to clarify the antifouling characteristics of membranes, more information is presented in Fig. 12. As it is observed, Rtf parameter for TFNC membranes decreased in comparison with that
88
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
TFC
TFNC1
TFNC2
Fig. 11. Three-dimensional AFM images of thin-film membranes.
Rrf (%)
Rirf (%)
Rtf (%)
90 78
80 67
70
67 61
60 47
50 40
32
30 20
29
20 11
High surface roughness is another factor involved in intensifying fouling of TFC membrane. Results of irreversible fouling of TFNC membranes confirm the effect of surface properties on diminishing the permanent adhesion of foulants on the membrane. As expected, lower surface roughness and higher hydrophilicity of TFNC membranes (Fig. 10) are responsible for superior resistance against fouling. Accordingly, in addition to the elevated salt removal, the applied modification method introduced in this study i.e. embedding functionalized nanoparticles into the thin-film matrix is an effective approach on reducing the fouling problems in prepared membranes.
10
4. Conclusion
0 TFC
TFNC1
TFNC2
Fig. 12. Fouling parameters of thin-film membranes.
for TFC. To better illuminate the membrane resistance against fouling, Rrf and Rirf parameters should be considered. Rrf is the portion of fouling which is removed by a simple washing process. All TFNC membranes offer higher Rrf compared to TFC membrane demonstrating that TFNC membranes are applicable in decreasing the fouling. It is really desired for a membrane employed under severe fouling conditions to be cleaned just by a simple washing process, and to reach a similar initial performance. Resistance of membranes against permanent fouling is examined using irreversible fouling ratio parameter. Rirf decreased considerably from 67% for TFC to 47 and 29% for TFNC1 and TFNC2, respectively. TFC membrane due to the hydrophobic nature of PPy is susceptible to adsorb hydrophobic material (hydrophobic interaction) and therefore, present higher value for Rirf [47–49].
In this study, a novel method was introduced for the first time for fabrication of TFC membrane using polypyrrole as thin-film, and cross-linkable acrylate-alumoxane nanoparticles were employed in the thin-film to achieve a nano-enhanced TFC nanofiltration membrane with promoted water softening and foulingresistant capability. Pyrrole monomers were polymerized using a simple method on the surface of PES ultrafiltration substrate, and different concentrations of cross-linkable acrylate-alumoxane nanoparticles were embedded as the multifunctional modifier into the thin-film during polymerization. From the results of the morphological analyses and performance experiments, the outstanding conclusions are summarized as follows: 1. Addition of cross-linkable nanoparticles into the polymerization solution led to the formation of thinner and denser thinfilm with smaller pore size in TFNC compared with control TFC membrane due to the chemical structure of nanoparticles and participation in the polymerization reaction.
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
2. TFNC membranes prepared by 0.005 and 0.025 g of nanoparticles revealed higher hydrophilicity in comparison with TFC membrane. Higher hydrophilicity and solid–liquid interfacial free energy along with thinner thin-film improved the pure water flux of membranes. 3. Thin-film membrane presented a desirable efficiency in Mg2+ rejection due to the combined effects of dense PPy thin-film on the substrate as well as the electrostatic repulsion between positive charge of membrane surface and Mg2+ ions. 4. Thin-film nanocomposite membrane prepared by 0.025 g of nanoparticles (TFNC2) offered a superior salt rejection (around 80%) compared with TFC membrane due to the formation of a tighter and cross-linked structure with the aid of C@C bonds in acrylate-alumoxane nanoparticles. 5. Flux recovery and antifouling characteristics of membranes increased in the order of TFNC2 > TFNC1 > TFC. Moreover, thinfilm nanocomposite membranes exhibited satisfying resistance against fouling due to higher hydrophilicity and lower surface roughness in comparison with TFC membrane. As an overall analysis, although incorporating the appropriate amount of acrylate-alumoxane nanoparticles into the PPy thinfilm was an effective approach for preparation of TFNC membranes with improved efficiency in water softening, more studies are still required for improving the membrane performance in water softening. For example, functionalization of alumoxane nanoparticles with other functional groups such as carboxyl and amide and studying the effects of nanoparticles on TFNC membrane structure and performance can be conducted as future beneficial researches. Moreover, there are limited studies on the development of multifunctional TFNC membranes with simultaneously superior performance in water softening and antifouling capability. In this regard, further research studies would result in more successful achievements and industrial applications of TFNC membranes in desalination processes. References [1] J. Yin, B. Deng, Polymer-matrix nanocomposite membranes for water treatment, J. Membr. Sci. 479 (2015) 256–275. [2] C. Gabrielli, G. Maurin, H. Francy-Chausson, P. Thery, T.T.M. Tran, M. Tlili, Electrochemical water softening: principle and application, Desalination 201 (1–3) (2006) 150–163. [3] A.F. Ismail, M. Padaki, N. Hilal, T. Matsuura, W.J. Lau, Thin film composite membrane — recent development and future potential, Desalination 356 (2015) 140–148. [4] W. Fang, L. Shi, R. Wang, Interfacially polymerized composite nanofiltration hollow fiber membranes for low-pressure water softening, J. Membr. Sci. 430 (2013) 129–139. [5] J.M. Gohil, P. Ray, Polyvinyl alcohol as the barrier layer in thin film composite nanofiltration membranes: preparation, characterization, and performance evaluation, J. Colloid Interface Sci. 338 (1) (2009) 121–127. [6] R.R. Sharma, S. Chellam, Solute rejection by porous thin film composite nanofiltration membranes at high feed water recoveries, J. Colloid Interface Sci. 328 (2) (2008) 353–366. [7] A.I. Schafer, A.G. Fane, T.D. Waite, Nanofiltration—Principles andApplications, Elsevier, Oxford, UK, 2002. [8] X. Dong, Q. Zhang, S. Zhang, S. Li, Thin film composite nanofiltration membranes fabricated from quaternized poly(ether ether ketone) with crosslinkable moiety using a benign solvent, J. Colloid Interface Sci. 463 (2016) 332–341. [9] S. Hosseini, N.H.M. Ekramul Mahmud, R. Binti Yahya, F. Ibrahim, I. Djordjevic, Polypyrrole conducting polymer and its application in removal of copper ions from aqueous solution, Mater. Lett. 149 (2015) 77–80. [10] R.L. McGinnis, M. Elimelech, Global challenges in energy and water supply: the promise of engineered osmosis, Environ. Sci. Technol. 42 (2008) 8625–8629. [11] M. Peyravi, M. Jahanshahi, A. Rahimpour, A. Javadi, S. Hajavi, Novel thin film nanocomposite membranes incorporated with functionalized TiO2 nanoparticles for organic solvent nanofiltration, Chem. Eng. J. 241 (2014) 155–166. [12] D. Emadzadeh, W.J. Lau, M. Rahbari-Sisakht, A. Daneshfar, M. Ghanbari, A. Mayahi, T. Matsuura, A.F. Ismail, A novel thin film nanocomposite reverse osmosis membrane with superior anti-organic fouling affinity for water desalination, Desalination 368 (2015) 106–113.
89
[13] M. Dinari, H. Ahmadizadegan, Synthesis, structural characterization and properties of novel functional poly(ether imide)/titania nanocomposite thin films, Polymer 55 (24) (2014) 6252–6260. [14] D. Emadzadeh, W.J. Lau, T. Matsuura, M. Rahbari-Sisakht, A.F. Ismail, A novel thin film composite forward osmosis membrane prepared from PSf–TiO2 nanocomposite substrate for water desalination, Chem. Eng. J. 237 (2014) 70– 80. [15] M. Ghanbari, D. Emadzadeh, W.J. Lau, T. Matsuura, M. Davoody, A.F. Ismail, Super hydrophilic TiO2/HNT nanocomposites as a new approach for fabrication of high performance thin film nanocomposite membranes for FO application, Desalination 371 (2015) 104–114. [16] B. Rajaeian, A. Heitz, M.O. Tade, S. Liu, Improved separation and antifouling performance of PVA thin film nanocomposite membranes incorporated with carboxylated TiO2 nanoparticles, J. Membr. Sci. 485 (2015) 48–59. [17] H. Dong, L. Wu, L. Zhang, H. Chen, C. Gao, Clay nanosheets as charged filler materials for high-performance and fouling-resistant thin film nanocomposite membranes, J. Membr. Sci. 494 (2015) 92–103. [18] M. Safarpour, V. Vatanpour, A. Khataee, M. Esmaeili, Development of a novel high flux and fouling-resistant thin film composite nanofiltration membrane by embedding reduced graphene oxide/TiO2, Sep. Purif. Technol. 154 (2015) 96–107. [19] G.S. Lai, W.J. Lau, P.S. Goh, A.F. Ismail, N. Yusof, Y.H. Tan, Graphene oxide incorporated thin film nanocomposite nanofiltration membrane for enhanced salt removal performance, Desalination 387 (2016) 14–24. [20] J. Yin, G. Zhu, B. Deng, Graphene oxide (GO) enhanced polyamide (PA) thinfilm nanocomposite (TFN) membrane for water purification, Desalination 379 (2016) 93–101. [21] A.S. Al-Hobaib, K.M. Al-Sheetan, L. El Mir, Effect of iron oxide nanoparticles on the performance of polyamide membrane for ground water purification, Mater. Sci. Semicond. Process. 42 (Part 1) (2016) 107–110. [22] J. Duan, Y. Pan, F. Pacheco, E. Litwiller, Z. Lai, I. Pinnau, High-performance polyamide thin-film-nanocomposite reverse osmosis membranes containing hydrophobic zeolitic imidazolate framework-8, J. Membr. Sci. 476 (2015) 303– 310. [23] H. Dong, L. Zhao, L. Zhang, H. Chen, C. Gao, W.S. Winston Ho, High-flux reverse osmosis membranes incorporated with NaY zeolite nanoparticles for brackish water desalination, J. Membr. Sci. 476 (2015) 373–383. [24] Q. Li, Y. Wang, J. Song, Y. Guan, H. Yu, X. Pan, F. Wu, M. Zhang, Influence of silica nanospheres on the separation performance of thin film composite poly (piperazine-amide) nanofiltration membranes, Appl. Surf. Sci. 324 (2015) 757– 764. [25] N. Rakhshan, M. Pakizeh, Removal of triazines from water using a novel OA modified SiO2/PA/PSf nanocomposite membrane, Sep. Purif. Technol. 147 (2015) 245–256. [26] G.N.B. Baroña, M. Choi, B. Jung, High permeate flux of PVA/PSf thin film composite nanofiltration membrane with aluminosilicate single-walled nanotubes, J. Colloid Interface Sci. 386 (1) (2012) 189–197. [27] M. Tian, Y.-N. Wang, R. Wang, Synthesis and characterization of novel highperformance thin film nanocomposite (TFN) FO membranes with nanofibrous substrate reinforced by functionalized carbon nanotubes, Desalination 370 (2015) 79–86. [28] P. Daraei, S.S. Madaeni, N. Ghaemi, M.A. Khadivi, L. Rajabi, A.A. Derakhshan, F. Seyedpour, PAA grafting onto new acrylate-alumoxane/PES mixed matrix nano-enhanced membrane: preparation, characterization and performance in dye removal, Chem. Eng. J. 221 (2013) 111–123. [29] X.-M. Wang, X.-Y. Li, K. Shih, In situ embedment and growth of anhydrous and hydrated aluminum oxide particles on polyvinylidene fluoride (PVDF) membranes, J. Membr. Sci. 368 (1–2) (2011) 134–143. [30] V. Vatanpour, S.S. Madaeni, L. Rajabi, S. Zinadini, A.A. Derakhshan, Boehmite nanoparticles as a new nanofiller for preparation of antifouling mixed matrix membranes, J. Membr. Sci. 401–402 (2012) 132–143. [31] A.A. Derakhshan, L. Rajabi, H. Karimnezhad, Morphology and production mechanism of the functionalized carboxylate alumoxane micro and nanostructures, Powder Technol. 225 (2012) 156–166. [32] W. Chen, S. Wu, Y. Lei, Z. Liao, B. Guo, X. Liang, D. Jia, Interfacial structure and performance of rubber/boehmite nanocomposites modified by methacrylic acid, Polymer 52 (19) (2011) 4387–4395. [33] N. Misdan, W.J. Lau, A.F. Ismail, T. Matsuura, Formation of thin film composite nanofiltration membrane: effect of polysulfone substrate characteristics, Desalination 329 (2013) 9–18. [34] A.K. Ghosh, B.-H. Jeong, X. Huang, E.M.V. Hoek, Impacts of reaction and curing conditions on polyamide composite reverse osmosis membrane properties, J. Membr. Sci. 311 (1–2) (2008) 34–45. [35] M. Fathizadeh, A. Aroujalian, A. Raisi, Effect of lag time in interfacial polymerization on polyamide composite membrane with different hydrophilic sub layers, Desalination 284 (2012) 32–41. [36] N. Ghaemi, P. Daraei, Enhancement in copper ion removal by PPy@Al2O3 polymeric nanocomposite membrane, J. Ind. Eng. Chem. 40 (2016) 26–33. [37] H. Rajabi, N. Ghaemi, S.S. Madaeni, P. Daraei, M.A. Khadivi, M. Falsafi, Nanoclay embedded mixed matrix PVDF nanocomposite membrane: preparation, characterization and biofouling resistance, Appl. Surf. Sci. 313 (2014) 207–214. [38] H. Rajabi, N. Ghaemi, S.S. Madaeni, P. Daraei, B. Astinchap, S. Zinadini, S.H. Razavizadeh, Nano-ZnO embedded mixed matrix polyethersulfone (PES) membrane: influence of nanofiller shape on characterization and fouling resistance, Appl. Surf. Sci. 349 (2015) 66–77.
90
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
[39] M.T.M. Pendergast, J.M. Nygaard, A.K. Ghosh, E.M.V. Hoek, Using nanocomposite materials technology to understand and control reverse osmosis membrane compaction, Desalination 261 (3) (2010) 255–263. [40] J.H. Chang, I.W. Hunter, Characterization and control of the wettability of conducting polymer thin films, in: MRS Proceedings, 2009, pp. 1228. [41] Q. Pei, R. Qian, Protonation and deprotonation of polypyrrole chain in aqueous solutions, Synth. Met. 45 (1) (1991) 35–48. [42] Y. Mansourpanah, H. Shahebrahimi, E. Kolvari, PEG-modified GO nanosheets, a desired additive to increase the rejection and antifouling characteristics of polyamide thin layer membranes, Chem. Eng. Res. Des. 104 (2015) 530–540. [43] S. Japip, K.-S. Liao, Y. Xiao, T.-S. Chung, Enhancement of molecular-sieving properties by constructing surface nano-metric layer via vapor cross-linking, J. Membr. Sci. 497 (2016) 248–258. [44] M. Askari, T.-S. Chung, Natural gas purification and olefin/paraffin separation using thermal cross-linkable co-polyimide/ZIF-8 mixed matrix membranes, J. Membr. Sci. 444 (2013) 173–183.
[45] D. Rana, T. Matsuura, Surface modifications for antifouling membranes, Chem. Rev. 110 (4) (2010) 2448–2471. [46] R. Kumar, A.F. Ismail, Fouling control on microfiltration/ultrafiltration membranes: effects of morphology, hydrophilicity, and charge, J. Appl. Polym. Sci. 132 (21) (2015) (n/a-n/a). [47] L.D. Nghiem, A.I. Schäfer, Adsorption and transport of trace contaminant estrone in NF/RO membranes, Environ. Eng. Sci. 19 (6) (2004). [48] N. Ghaemi, S.S. Madaeni, A. Alizadeh, H. Rajabi, P. Daraei, M. Falsafi, Effect of fatty acids on the structure and performance of cellulose acetate nanofiltration membranes in retention of nitroaromatic pesticides, Desalination 301 (2012) 26–41. [49] L.D. Nghiem, A.I. Schäfer, T.D. Waite, Adsorptive interactions between membranes and trace contaminants, Desalination 147 (1–3) (2002) 269–274.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis
Novel antifouling nano-enhanced thin-film composite membrane containing cross-linkable acrylate-alumoxane nanoparticles for water softening Negin Ghaemi Department of Chemical Engineering, Kermanshah University of Technology, 67178 Kermanshah, Iran
g r a p h i c a l a b s t r a c t
Mg2+ rejection tests
a r t i c l e
i n f o
Article history: Received 10 June 2016 Revised 26 August 2016 Accepted 15 September 2016 Available online 15 September 2016 Keywords: Thin-film nanocomposite membrane Cross-linkable nanoparticles Polypyrrole Acrylate-alumoxane Water softening Antifouling
a b s t r a c t A novel thin-film composite (TFC) nanofiltration membrane was prepared using polymerization of pyrrole monomers on the PES ultrafiltration membrane. To improve the characteristics of hydrophobic polypyrrole (PPy) thin-film layer, cross-linkable acrylate-functionalized alumoxane nanoparticles with different concentrations were embedded into the thin-film during polymerization process, and thinfilm nanocomposite (TFNC) membranes were prepared. The characteristics and performance of TFC and TFNC membranes were assessed through the morphological analyses (SEM, AFM), measurement of hydrophilicity and solid–liquid interfacial free energy, water permeability and Mg2+ removal tests. Addition of proper amount of nanoparticles into the polymerization mixture led to the preparation of membranes with more hydrophilic, thinner and smoother active layer as well as higher water permeability compared to TFC control membrane. TFNC membrane prepared with 0.025 g of nanoparticles was the most efficient membrane since it exhibited the highest rejection of MgCl2 and MgSO4 salts. Antifouling capability of membranes, in terms of flux recovery and fouling parameters, demonstrated the high tolerance of TFNC against fouling. Ó 2016 Elsevier Inc. All rights reserved.
1. Introduction
E-mail addresses: [email protected], [email protected] http://dx.doi.org/10.1016/j.jcis.2016.09.035 0021-9797/Ó 2016 Elsevier Inc. All rights reserved.
Hard water is undesirable in the domestic water supply and the industrial applications as a result of its high mineral content which
82
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
Nomenclature
Symbols A AFM CF CP DMAc FR M NF PES PPy PVP PWF Rej Rirf Rrf Rtf RO SEM
& abbreviations active membrane surface area (m2) atomic force microscopy Mg2+ concentration in the feed (mg/l) Mg2+ concentration in the permeate sample (mg/l) N,N-dimethylacetamide flux recovery (%) weight of permeated water through the membrane (kg) nanofiltration polyethersulfone polypyrrole polyvinylpyrrolidone pure water flux (kg/m2 h) rejection (%) irreversible fouling ratio reversible fouling ratio total fouling ratio reverse osmosis scanning electron microscopy
reduces the household cleaning efficiency, makes scaling and corrosion problems, and causes serious difficulties in pipe lines of boilers, heat exchangers and electrical devices. To solve this issue, water softening processes are employed by which calcium and magnesium cations as well as other divalent or multivalent metal ions are eliminated from hard water. Compared with conventional water softening techniques including ion-exchange resins, zeolites or lime-soda ash treatment methods, water softening using membranes have received a specific place in this market due to higher efficiency, lower cost and less chemical and energy consumption [1,2]. Meanwhile reverse osmosis (RO) process as a high performance membrane filtration process serving for removal of solutes and minerals of hard water suffers the problems of high energy consumption and severe scaling and fouling [3,4]. Nanofiltration (NF) membranes have a looser structure rather than RO membranes. Hence, a lower applied pressure is required for NF to achieve an acceptable salt removal and water permeation even higher compared to RO membranes [5–7]. In water softening process for household consumption, it is not needed to completely remove the water hardness, so NF membranes might be a beneficial alternative to RO membranes. Among the currently available NF membranes, thin-film composite (TFC) membranes are particularly applied in water softening processes. These membranes enhance both selectivity and permeability with less energy consumption compared to the typical asymmetric membranes [3,8]. They are usually composed of an ultra-thin active layer responsible for separation and a porous substrate layer providing mechanical resistance. Polyethersulfone (PES) is frequently used for fabrication of substrate in TFC membranes due to the excellent chemical, thermal and mechanical resistance as well as almost a low price. On the other hand, thin-film in the most of TFC membranes is prepared using interfacial polymerization of hydrophilic polymers such as polyamide and polyvinyl alcohol on the substrate. Despite many advantages of a hydrophilic thin-film especially in view of increasing the water permeability, high hydrophilicity is not necessarily the only criterion to choose the material for thin layer. Adhesion between substrate and top layer is really important in the preparation of TFC membranes. It was proved that the adhesion of a hydrophilic layer on a hydrophobic substrate is weaker in comparison with that in the case of a hydrophobic thin-film [3].
SR TFC TFNC TFNC1 TFNC2 TFNC3 WCA WFi WFF WFw DGSL
Dt D
cL
mean surface roughness (nm) thin-film composite thin-film nanocomposite thin-film nanocomposite membrane prepared by 0.005 g of nanoparticles thin-film nanocomposite membrane prepared by 0.025 g of nanoparticles thin-film nanocomposite membrane prepared by 0.05 g of nanoparticles water contact angle (°) initial pure water flux (kg/m2 h) final pure water flux (kg/m2 h) water flux during the filtration of whey solution (kg/ m2 h) surface roughness corrected solid–liquid interfacial free energy (mJ/m2) sampling time (h) relative surface area liquid surface tension (mJ/m2)
Despite the particular characteristics of polypyrrole (PPy) (good thermal and environmental stability as well as ease of preparation [9]), no study has been conducted in which the thin-layer is composed of the hydrophobic PPy. Hence, PPy would be an appropriate choice for fabrication of thin-film on PES substrate; however, it should be noticed that low water permeation and high fouling of TFC membrane with hydrophobic thin-layer cause crucial problems in practice. The key merit for composite membranes is that the characteristics of thin layer and porous substrate layer can be independently optimized to achieve better performance [10]. In this regard, thin-film nanocomposite (TFNC) membranes were developed [3] in which some nanoparticles such as TiO2 [11–16], clay [17], graphene oxide [18–20], iron oxide [21], zeolite [22,23], SiO2 [24,25], and carbon nanotubes [26,27] were embedded within the thin dense layer of TFC membrane mostly with the aim of increasing the hydrophilicity without sacrificing the separation efficiency of membrane [17–19,21]. Boehmite (c-AlOOH) as the other applicable nanoparticles were also employed in the fabrication of nanocomposite membranes due to their specific characteristics particularly high hydrophilicity [28]. Reviewing the literature revealed that addition of boehmite nanoparticles into the membrane matrix resulted in improving the hydrophilicity due to the extra hydroxyl groups of nanoboehmites [29,30]. Moreover, nanoboehmite can be modified using acrylic acid due to the ability of carboxyl groups to be coordinated with aluminum as unidentate and bidentate ligands [28,31,32]. This process leads to the production of acrylate-alumoxane nanoparticles containing C@C bonds with capability of participating in the polymerization reaction [28]. The current study aims to prepare TFNC membrane with improved efficiency in water softening. PES was applied for fabrication of ultrafiltration substrate, and the thin-film was developed through aqueous polymerization of pyrrole on the surface of the substrate. Acrylate-alumoxane nanoparticles were employed as additive for raising the hydrophilicity and improving the quality of PPy thin-film layer. Hence, different concentrations of nanoparticles were introduced as filler into the thin-layer during polymerization reaction. To probe the effects of the employed method on the characteristics and performance of membranes, morphological analyses (scanning electron microscopy (SEM), atomic force
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
microscopy (AFM)), water contact angle (WCA) measurements, water permeability and softening experiments as well as antifouling tests were conducted. 2. Experimental 2.1. Preparation of thin-film composite (TFC) and nanocomposite (TFNC) membranes Ultrafiltration PES membrane as the support layer was fabricated using phase inversion method. A casting solution containing 16 wt.% PES (Ultrason E6020P, MW = 58,000 g/mol, BASF, Germany) as base polymer, 2 wt.% polyvinylpyrrolidone (PVP, MW = 25,000 g/mol, Merck, Germany) as pore former and dimethylacetamide (DMAc, Merck, Germany) as solvent was prepared and mixed by stirring at 400 rpm for half a day until a homogeneous solution was achieved. After ultrasonication for 15 min, solution was casted using a homemade applicator on the non-woven polyester fiber (83 g/m2, USA). The casted film was immersed in the distilled water bath as nonsolvent, and after primary formation of membrane, it was kept in the fresh distilled water for a day in order to guarantee the completion of phase inversion process. Finally, the membrane was kept between two pieces of filter papers to dry for another day. All procedures were done at room temperature. Thin-film of TFC membrane was formed by chemical polymerization of pyrrole on the surface of a pre-cast PES substrate. In this study, distilled water was employed as solvent. For polymerization process, membrane was mounted at the bottom of a container, and 50 ml of 14 wt.% FeCl36H2O (Merck, Germany) aqueous solution (as initiator) was poured into the container and held for 12 h to ensure the penetration of the solution into the pores of the substrate. After draining off the excess amount of FeCl3 solution from the container, 50 ml of 1 wt.% pyrrole (Merck, Germany) aqueous solution was poured into the container. To insert the nanoparticles into the thin film, different quantities of acrylate-alumoxane nanoparticles (0, 0.005, 0.025 and 0.05 g) were added into the pyrrole solution. Acrylate-alumoxane was synthesized based on the method introduced in the previous study [28]. Briefly, bohemite nanoparticles were refluxed in acrylic acid for 72 h and then, washed with n-hexane and dried at 50 °C. Monomer solution containing nanoparticles was sonicated for 30 min to avoid the nanoparticle agglomeration. Then, the sonicated solution was poured onto the surface of the substrate previously exposed to FeCl3 solution and was remained for another 12 h. After draining off the monomer-nanoparticle solution, membrane was thoroughly washed with plenty of distilled water in order to remove unreacted FeCl3 and pyrrole and unattached nanoparticles from the surface of the membrane substrate. For drying the membrane, it was held between two pieces of filter paper at room temperature for 24 h. The thin-film membranes prepared with 0, 0.005, 0.025 and 0.05 g of nanoparticles are denoted as TFC, TFNC1, TFNC2 and TFNC3 in the manuscript. 2.2. Characterization of membranes Membrane structure and surface properties were investigated by use of the typical and highly applicable methods of scanning electron microscopy (SEM) and atomic force microscopy (AFM). Cross-section structure of membranes was monitored using capturing SEM micrographs by a scanning electron microscope (SEM, KYKY-EM3200, China) apparatus. Also, thin-layer thickness of membranes was specified by measuring the thickness in four or five casual points on the top-layer and reporting the average value as the membrane thin-layer thickness. Surface roughness (SR) of membrane as one the characteristics influential on the membrane
83
fouling as well as the mean size of membrane surface pores were determined by scanning the membrane surface using an AFM instrument (ARA-AFM, Ara-Research Co., Iran) in non-contact mode and image analyzer software (version 1.01 Ara-Research). To reach the more precise roughness data, two pieces of each membrane were scanned. The average values of surface roughness were reported as SR. Surface hydrophilicity of membrane was measured by dropping the small droplets of distilled water using a microsyringe on five points on the membrane surface and taking the images by a digital microscope (USB Digital Microscope, 2.0 Mega Pixel Color Video Camera, 500X, China). An Image J freeware was employed to measure the contact angle between the membrane surface and water droplets. It should be mentioned that the reported WCA for each membrane was the average amount of measured values. It is clear that a smaller water contact angle means a higher tendency of water to spread on the surface, i.e., higher hydrophilicity of membrane. Moreover, a modified form of the Young–Dupre equation was used to assess the relative hydrophilicity of membranes by the solid–liquid interfacial free energy [33,34]. It should be noticed that the increase in water contact angle value could not be used to show the increased hydrophobicity of the membrane since the surface roughness can also contribute to the change of contact angle value [33,35]. Therefore, for better representation of hydrophilicity, the surface roughness corrected solid–liquid interfacial free energy, DGSL (mJ/m2) was determined using Eq. (1) [33,34].
cos h DGSL ¼ cL 1 þ D
ð1Þ
where h and cL are the average water contact angle value and the liquid surface tension which is 72.8 mJ/m2 for pure water at 25 °C, respectively. Moreover, D is the relative surface area which was determined using AFM topography images by MATLAB 7.8 software (license num. 161051) [36]. Since the contact angle in a smooth surface (D = 1) is 0°, the maximum solid–liquid interfacial free energy is close to 2cL . Hence, a higher amount of DGSL suggests a higher hydrophilicity of the surface [33,34]. 2.3. Permeability and efficiency of membranes in water softening Membrane performance was evaluated in terms of pure water flux (PWF) and water softening. For measuring the PWF, a defect-free circular piece of membrane was cut and mounted in a dead-end test cell with active membrane surface area of 12.56 cm2. After filling the cell with distilled water, the membrane was compressed under 0.5 MPa pressure for an hour. Then, pressure was reduced to 0.4 MPa and PWF (kg/m2 h) was calculated after an hour using the following equation:
PWF ¼
M At
ð2Þ
where M, A and t are weight of permeated water through the membrane (kg), active membrane surface area (m2) and collecting time of permeated water (h). In order to probe the efficiency of membrane in water softening, magnesium salt solution with high concentration 500 ppm (very hard water) were employed. For investigation of probable effects of counter ions, MgCl2 and MgSO4 solutions were used as feed. To calculate the rejection of magnesium ions (Rej (%)) using fabricated membranes, the following equation was employed:
Rej ð%Þ ¼
CP 100 1 CF
ð3Þ
where C p and C F are concentration (mg/l) of Mg2+ in the permeate and feed solutions, respectively. Magnesium ion concentration
84
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
was measured using chelatometric titration by Na2EDTA (Merck, Germany) solution and in the presence of Eriochrome Black T (Merck, Germany) indicator in the buffer (pH 10.0). It should be mentioned that all filtration experiments were repeated two times using a fresh piece of membrane, and the average amount was reported as the final value.
All tests were carried out twice and each time with a fresh piece of membrane, to increase the accuracy of results.
3. Results and discussion 3.1. Membrane characteristics and water permeability
2.4. Fouling experiments
FR ð%Þ ¼ ðWF F =WF i Þ 100
ð4Þ
where WF F (kg/m2 h) and WF i (kg/m2 h) are the final (aftercleaning) and the initial PWF, respectively. Considering Eq. (4), it is clear that higher amount of FR refers to the higher capability of membrane to be regenerated after cleaning and therefore, higher antifouling property [37,38]. In fact, this parameter indicates what percent of membrane permeability is recovered after fouling and cleaning process. In order to analyze more minutely the antifouling behavior of a membrane, other parameters i.e. the reversible fouling ratio (Rrf) as the fraction of fouling which can be easily removed by simple washing, the irreversible fouling ratio (Rirf) as the fraction of fouling that firmly sticks on the membrane surface and the total fouling ratio (Rtf) as sum of the Rrf and Rirf might be calculated as well. The aforementioned parameters are calculated using the following equations [37,38]:
Rrf ð%Þ ¼
WF F WF w 100 WF i
ð5Þ
Rirf ð%Þ ¼
WF i WF F 100 WF i
ð6Þ
Rtf ð%Þ ¼
WF i WF w 100 WF i
ð7Þ
Results of PWF of prepared membranes are shown in Fig. 1. As expected, membrane permeability decreased from 34 for PES substrate to 9 kg/m2 h for TFC. This decrement is related to the formation of PPy thin-film on the substrate and therefore, enhancement of the membrane thickness and also, the hydrophobicity of PPy. Formation of polymeric layer on the substrate is clearly observable as a result of remarkable change in the color of membrane surface from white to black after polymerization of pyrrole monomers on the membrane substrate (Fig. 2). Also, SEM cross-section image of TFC (Fig. 3) declared the formation of PPy thin-layer with approximately 25 lm thickness on the substrate. By addition of different amounts of nanoparticles into PPy thinfilm, PWF of the prepared membranes increased compared with TFC (see Fig. 1) so that it grew from 9 kg/m2 h for TFC to 15, 28 and 32 kg/m2 h for TFNC1, TFNC2 and TFNC3, respectively. Average values of thin-layer thickness of TFNC membranes measured by digital micrometer are tabulated in Table 1. The data reveals that thickness of thin-layer decreases by addition of nanoparticles. This achievement might be due to the presence of nanoparticles and their beneficial effect on the polymeric thin-film layer [12]. Presence of nanoparticles in the monomer solution decreases the number of monomers which can diffuse to the substrate and contribute
40 35
PWF(kg/m2.h)
One of the reasons of employing acrylate-alumoxane nanoparticles, except betterment of thin-film characteristics, was decreasing the membrane fouling resulted from PPy as a polymer with the hydrophobic nature. Hence, the antifouling properties of membranes with a desirable performance in water softening were investigated. The antifouling capability of membranes was determined using whey solution as a severe foulant. Generally, there are some parameters by which the antifouling characteristic of prepared membranes can be well explored. In order to calculate antifouling parameters, an experimental procedure was thoroughly followed for all membranes. First, pure water flux was determined for each membrane under 0.4 MPa transmembrane pressure during 60 min. This amount was considered as initial water flux and showed by WFi (kg/m2 h). Then, distilled water was replaced by whey solution, and water permeation test was continued under the same pressure during 90 min. Water permeated through the membrane was shown by WFw (kg/m2 h). Cleaning the membrane should be done at this stage of experiments. For this purpose, the membrane was brought out of the cell and immersed in distilled water for 30 min after washing with plenty amount of distilled water. After the cleaning process, membrane was mounted again into the cell, and the first step was repeated. Pure water permeated through the membrane at this stage was represented by WFF (kg/m2 h) in which subscript F is referred to the final stage of filtration tests. It is worthy to mention that in order to reduce the effects of concentration polarization, a constant agitation with speed of 350 rpm was applied during all experiments. Flux recovery (FR (%)) factor calculated using Eq. (4) is one of the most applicable and well-known parameters representing antifouling property of a membrane [37].
34
32 28
30 25 20
15
15 9
10 5 0 Substrate
TFC
TFNC1
TFNC2
TFNC3
Fig. 1. Pure water flux of prepared membranes.
Fig. 2. Surface image of (a) PES substrate and (b) thin-film membrane captured by a digital camera.
85
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
PPy thin-film layer
Fig. 4. Surface image of TFNC3 membrane captured by a digital camera.
Non-woven fibre
Fig. 3. SEM cross-section image of thin-film membrane.
to the polymerization reaction [24,39]. This phenomenon results in a reduction in the thickness of the coated layer (thin-film) in TFNC membranes as observed in TFNC1 and TFNC2 membranes. Since thinner coated layer results in higher permeation through the membrane, the enhancement of PWF of TFNC membranes is reasonable. The appearance of TFNC3 membrane surface (see Fig. 4) as well as the thickness of thin-film in this membrane (see Table 1) prove the aforementioned claim and reveal that the presence of nanoparticles with higher amount (>0.025 g) in the polymerization medium disturbs the process of thin-film formation and leads to a non-uniform deposition of polymer on the substrate. The nonuniform coating of PPy in TFNC3 membrane is the main reason of high flux of this membrane (compare fluxes for substrate and TFNC3 in Fig. 1). Except the thickness of thin-film layer which influences on the water permeability of TFNC membranes, hydrophilicity is another effective factor. Results of water contact angle (WCA), relative surface area ðDÞ and DGSL are tabulated in Table 1. Although WCA of TFC membrane (72.2°) did not change notably in comparison with
that in PES substrate (73°) [28,36] due to the presence of hydrophobic polypyrrole on the substrate [40], TFNC membranes presented higher hydrophilicity (lower WCA values) compared with TFC. Among thin-film nanocomposite membranes, although hydrophilicity of TFNC2 (65.2°) increased in comparison with TFNC1 (69.5°) by enhancing the amount of applied nanoparticle, TFNC3 presented the lowest hydrophilicity (71.7°). Higher hydrophilicity of TFNC membranes compared with TFC is due to the presence of hydrophilic acrylate-alumoxane nanoparticles on the membrane surface. Existence of nanoparticles in the polymerization solution leads to trapping of some of the nanoparticles within the polymer chains growing in the thin-film layer. SEM surface image of TFC and TFNC2 membranes obviously confirm the presence of nanoparticles on the surface of the membrane prepared by addition of nanoparticles (Fig. 5). Hydrophilic acrylate-alumoxane nanoparticles with hydroxyl groups in the thin-film matrix partially improve the surface wettability of membrane. Non-uniform deposition of thin-film, as mentioned earlier, might be responsible for low hydrophilicity of TFNC3 membrane. It should be noticed that the decrement in water contact angle value could not be used as the only criterion to show the increased hydrophilicity of membrane since the surface roughness might also affect the contact angle [33,34]. Hence, DGSL values tabulated in Table 1 were employed as better evidence, instead of WCA, for investigation of surface hydrophilicity. Results of DGSL (Table 1) revealed that the interaction between PPy and acrylatealumoxane nanoparticles changed the chemical surface functionality of membrane and led to an increment in the membrane hydrophilicity for the case of TFNC1 and TFNC2 membranes (higher values of DGSL ). Based on the results, addition of an appropriate nanoparticle into the mixture of polymerization reaction might successfully control the thickness of thin-film and permeability of membrane. Besides, it should be noticed that addition of hydrophilic nanoparticle leads to the improvement of membrane hydrophilicity and thin-film thickness; however, an optimum amount of
Table 1 Characteristics of membranes. Membrane
Membrane thickness (thin-layer thickness)a (lm)
WCA (°)
D ()
DGSL (mJ/m2)
PES substrate TFC TFNC1 TFNC2 TFNC3
160 ± 4 185 ± 5 176 ± 3 170 ± 6 165 ± 5
73.0 ± 2.1 72.2 ± 1.3 69.5 ± 0.9 65.2 ± 1.0 71.7 ± 1.4
– 1.53 1.14 1.10 1.35
– 87.3 95.2 100.5 89.7
(–) (25) (16) (10) (5)
Data are the average values of several measurements. a Thin-layer thickness of each membrane was measured by subtracting the substrate thickness from the total membrane thickness.
86
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
by thin-film membranes might be simply related to the significant reduction in the surface pore size after deposition of dense PPy thin-film on the substrate (see Table 2). On the other hand, based on PPy chemical structure and from the literature at the experimental pH of feed solution (i.e. 5.5) PPy possesses positive charge due to the impossibility of deporotonation of NH groups at this pH [41]. Hence, Donnan repulsion is the dominant mechanism in the rejection of cations. Moreover, there is no critical enhancement in the rejection of MgSO4 compared with MgCl2 implying that Mg2+ ions are mainly rejected ions by the membrane instead of
Fig. 5. SEM surface image of (a) TFC and (b) TFNC2 membranes.
nanoparticle should be consumed to avoid the decomposition of thin-film structure. 3.2. Performance of TFC and TFNC membranes in water softening According to Fig. 6, the water hardness was reduced using thinfilm membranes, particularly thin-film nanocomposite membranes; however, a satisfying result was not achieved using TFNC3 membrane due to the unsuccessful uniform coating of thin-film on the PES substrate. It should be mentioned that Mg2+ removal was also tested by PES substrate, and no rejection of magnesium ions was observed as expected. The reason of enhanced ion rejection
MgCl2
the negative bivalent SO2 4 ions. This further reinforces the concept of positive charge of PPy thin-film. This outcome would be very desirable for water softening because the prepared TFC membranes are able to reduce the water hardness directly by removing the ions which are responsible for water hardening. According to the salt rejection results (Fig. 6), TFNC membranes (except TFNC3 due to non-uniform and inappropriate deposition of PPy thin layer on the substrate) showed a superior salt rejection compared with TFC (composite thin-film without the presence of nanoparticles). This can be explained by existence of crosslinkable acrylate-alumoxane nanoparticles in TFN1 and TFN2 membranes. Based on the acrylate-alumoxane scheme (Fig. 7), there is carbon-carbon double bond on the outer surface of nanoparticles which allows participation of these functional groups in the polymerization reaction as schemed in Fig. 8. This can efficiently enhance the compatibility of hydrophilic nanoparticles with hydrophobic PPy matrix and form a tight and cross-linked structure in the deposited thin-film. Considerable reduction in the mean surface pore size of TFNC1 (13 nm) and TFNC2 (9 nm) membranes in comparison with TFC (38 nm) membrane (Table 2) clearly proves the tighter structure of thin-layer and therefore, higher rejection in TFNC membranes [42]. The similar results were also reported for higher selectivity of membranes with a cross-linked polymer in their structure [24,43,44]. These results illustrated that the presence of acrylatealumoxane nanoparticles efficiently improved the properties of PPy thin-film and created a dense and compact thin layer with
Table 2 Mean surface pore size and roughness of membranes. Membrane
Mean surface pore size (nm)
SR (nm)
PES substrate TFC TFNC1 TFNC2 TFNC3
130 ± 15 38 ± 5 13 ± 1 9±2 81 ± 10
– 22.6 ± 1.5 12.3 ± 0.9 10.7 ± 1.0 –
MgSO4
100 90
83
80
68
Rej (%)
70 60 50
70
59 53 47
40 30 20
10
15
10 0 TFC
TFNC1
TFNC2
Fig. 6. Magnesium rejection by thin-film membranes.
TFNC3 Fig. 7. Schematic of acrylate-alumoxane nanoparticles.
87
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
Fig. 8. Scheme of participation of acrylate-alumoxane nanoparticles in the polymerization and cross-linking.
Wfi
30
WFw
WFF
SR (nm) 80
28
FR(%) 71
69.5 65.2
70
25
Flux (kg/m2.h)
WCA(degree)
72.2
20 20
60
53
50
15 15 11 10
9
8
30
5 5
2
40
33 22.6
20
3
12.3
10.7
10
0 TFC
TFNC1
TFNC2
0 TFC
TFNC1
TFNC2
Fig. 9. Initial, whey solution and after-cleaning water flux. Fig. 10. Roughness, hydrophilicity and flux recovery results of thin-film membranes.
higher rejection of MgCl2 (70%) and MgSO4 (83%) by TFNC2 membrane. Considering the physicochemical properties of TFNC membranes, the chemical structure of PPy thin layer (Donnan repulsion effect) along with the compact thin layer composed of smaller pores resulted in the enhancement in the rejection of TFNC membranes [42,45]. Moreover, the membrane rejections obtained in this study were comparable with other reported results of TFNC nanofiltration membranes [1], and as expected, it was normally lower in comparison with the results obtained by membranes which were employed with higher feed concentrations [19,24]. However, the acrylate-alumoxane nano-enhanced thin-film composite membrane (TFNC2) showed relatively higher performance compared with other published studies [16]. Another important factor which declares the proficiency of prepared TFNC membranes in this study is lower amount of employed nanoparticles compared with other researches since the low consumption of nanofiller is an advantage from environmental point of view and really important in industrial applications. 3.3. Antifouling characteristics of membranes Fig. 9 shows the results of initial, whey solution and recovered water flux for thin-film membranes. It should be mentioned that TFNC3 membrane due to the unacceptable results in Mg2+ removal was not assessed in this stage of experiments. As seen, water flux during filtration of whey solution considerably decreased for all
tested thin-film membranes due to the sever fouling caused by whey solution with high concentration; however, in this case, the membrane efficiency in the flux recovery and fouling mitigation has more significant importance. The obtained FR data (Fig. 10) reveal an increment in the order of TFNC2 > TFNC1 > TFC. Generally, surface roughness and hydrophilicity are two important parameters affecting on the membrane fouling. It has been proved that higher hydrophilicity and lower roughness are in favor of lower fouling [38,46]. In order to probe the surface topography and roughness of membranes, AFM images of membranes are exhibited in Fig. 11. Moreover, the FR, surface roughness (SR) and WCA data are illustrated all together in Fig. 10 to efficiently assess the relationship between theses parameters. Considering the AFM images and SR data show that addition of nanoparticles into the thin-film results in formation of a smoother surface in TFNC membranes compared with that in TFC. According to Fig. 10, the surface roughness and WCA enhancement in the membranes are in the order of TFC > TFNC1 > TFNC2 which proves that both higher hydrophilicity and smoother surface are responsible for the observed trend in FR (TFNC2 > TFNC1 > TFC). Hence, the usage of acrylate-alumoxane nanoparticles was influential on the restoration of membrane permeability after becoming fouled. In order to clarify the antifouling characteristics of membranes, more information is presented in Fig. 12. As it is observed, Rtf parameter for TFNC membranes decreased in comparison with that
88
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
TFC
TFNC1
TFNC2
Fig. 11. Three-dimensional AFM images of thin-film membranes.
Rrf (%)
Rirf (%)
Rtf (%)
90 78
80 67
70
67 61
60 47
50 40
32
30 20
29
20 11
High surface roughness is another factor involved in intensifying fouling of TFC membrane. Results of irreversible fouling of TFNC membranes confirm the effect of surface properties on diminishing the permanent adhesion of foulants on the membrane. As expected, lower surface roughness and higher hydrophilicity of TFNC membranes (Fig. 10) are responsible for superior resistance against fouling. Accordingly, in addition to the elevated salt removal, the applied modification method introduced in this study i.e. embedding functionalized nanoparticles into the thin-film matrix is an effective approach on reducing the fouling problems in prepared membranes.
10
4. Conclusion
0 TFC
TFNC1
TFNC2
Fig. 12. Fouling parameters of thin-film membranes.
for TFC. To better illuminate the membrane resistance against fouling, Rrf and Rirf parameters should be considered. Rrf is the portion of fouling which is removed by a simple washing process. All TFNC membranes offer higher Rrf compared to TFC membrane demonstrating that TFNC membranes are applicable in decreasing the fouling. It is really desired for a membrane employed under severe fouling conditions to be cleaned just by a simple washing process, and to reach a similar initial performance. Resistance of membranes against permanent fouling is examined using irreversible fouling ratio parameter. Rirf decreased considerably from 67% for TFC to 47 and 29% for TFNC1 and TFNC2, respectively. TFC membrane due to the hydrophobic nature of PPy is susceptible to adsorb hydrophobic material (hydrophobic interaction) and therefore, present higher value for Rirf [47–49].
In this study, a novel method was introduced for the first time for fabrication of TFC membrane using polypyrrole as thin-film, and cross-linkable acrylate-alumoxane nanoparticles were employed in the thin-film to achieve a nano-enhanced TFC nanofiltration membrane with promoted water softening and foulingresistant capability. Pyrrole monomers were polymerized using a simple method on the surface of PES ultrafiltration substrate, and different concentrations of cross-linkable acrylate-alumoxane nanoparticles were embedded as the multifunctional modifier into the thin-film during polymerization. From the results of the morphological analyses and performance experiments, the outstanding conclusions are summarized as follows: 1. Addition of cross-linkable nanoparticles into the polymerization solution led to the formation of thinner and denser thinfilm with smaller pore size in TFNC compared with control TFC membrane due to the chemical structure of nanoparticles and participation in the polymerization reaction.
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
2. TFNC membranes prepared by 0.005 and 0.025 g of nanoparticles revealed higher hydrophilicity in comparison with TFC membrane. Higher hydrophilicity and solid–liquid interfacial free energy along with thinner thin-film improved the pure water flux of membranes. 3. Thin-film membrane presented a desirable efficiency in Mg2+ rejection due to the combined effects of dense PPy thin-film on the substrate as well as the electrostatic repulsion between positive charge of membrane surface and Mg2+ ions. 4. Thin-film nanocomposite membrane prepared by 0.025 g of nanoparticles (TFNC2) offered a superior salt rejection (around 80%) compared with TFC membrane due to the formation of a tighter and cross-linked structure with the aid of C@C bonds in acrylate-alumoxane nanoparticles. 5. Flux recovery and antifouling characteristics of membranes increased in the order of TFNC2 > TFNC1 > TFC. Moreover, thinfilm nanocomposite membranes exhibited satisfying resistance against fouling due to higher hydrophilicity and lower surface roughness in comparison with TFC membrane. As an overall analysis, although incorporating the appropriate amount of acrylate-alumoxane nanoparticles into the PPy thinfilm was an effective approach for preparation of TFNC membranes with improved efficiency in water softening, more studies are still required for improving the membrane performance in water softening. For example, functionalization of alumoxane nanoparticles with other functional groups such as carboxyl and amide and studying the effects of nanoparticles on TFNC membrane structure and performance can be conducted as future beneficial researches. Moreover, there are limited studies on the development of multifunctional TFNC membranes with simultaneously superior performance in water softening and antifouling capability. In this regard, further research studies would result in more successful achievements and industrial applications of TFNC membranes in desalination processes. References [1] J. Yin, B. Deng, Polymer-matrix nanocomposite membranes for water treatment, J. Membr. Sci. 479 (2015) 256–275. [2] C. Gabrielli, G. Maurin, H. Francy-Chausson, P. Thery, T.T.M. Tran, M. Tlili, Electrochemical water softening: principle and application, Desalination 201 (1–3) (2006) 150–163. [3] A.F. Ismail, M. Padaki, N. Hilal, T. Matsuura, W.J. Lau, Thin film composite membrane — recent development and future potential, Desalination 356 (2015) 140–148. [4] W. Fang, L. Shi, R. Wang, Interfacially polymerized composite nanofiltration hollow fiber membranes for low-pressure water softening, J. Membr. Sci. 430 (2013) 129–139. [5] J.M. Gohil, P. Ray, Polyvinyl alcohol as the barrier layer in thin film composite nanofiltration membranes: preparation, characterization, and performance evaluation, J. Colloid Interface Sci. 338 (1) (2009) 121–127. [6] R.R. Sharma, S. Chellam, Solute rejection by porous thin film composite nanofiltration membranes at high feed water recoveries, J. Colloid Interface Sci. 328 (2) (2008) 353–366. [7] A.I. Schafer, A.G. Fane, T.D. Waite, Nanofiltration—Principles andApplications, Elsevier, Oxford, UK, 2002. [8] X. Dong, Q. Zhang, S. Zhang, S. Li, Thin film composite nanofiltration membranes fabricated from quaternized poly(ether ether ketone) with crosslinkable moiety using a benign solvent, J. Colloid Interface Sci. 463 (2016) 332–341. [9] S. Hosseini, N.H.M. Ekramul Mahmud, R. Binti Yahya, F. Ibrahim, I. Djordjevic, Polypyrrole conducting polymer and its application in removal of copper ions from aqueous solution, Mater. Lett. 149 (2015) 77–80. [10] R.L. McGinnis, M. Elimelech, Global challenges in energy and water supply: the promise of engineered osmosis, Environ. Sci. Technol. 42 (2008) 8625–8629. [11] M. Peyravi, M. Jahanshahi, A. Rahimpour, A. Javadi, S. Hajavi, Novel thin film nanocomposite membranes incorporated with functionalized TiO2 nanoparticles for organic solvent nanofiltration, Chem. Eng. J. 241 (2014) 155–166. [12] D. Emadzadeh, W.J. Lau, M. Rahbari-Sisakht, A. Daneshfar, M. Ghanbari, A. Mayahi, T. Matsuura, A.F. Ismail, A novel thin film nanocomposite reverse osmosis membrane with superior anti-organic fouling affinity for water desalination, Desalination 368 (2015) 106–113.
89
[13] M. Dinari, H. Ahmadizadegan, Synthesis, structural characterization and properties of novel functional poly(ether imide)/titania nanocomposite thin films, Polymer 55 (24) (2014) 6252–6260. [14] D. Emadzadeh, W.J. Lau, T. Matsuura, M. Rahbari-Sisakht, A.F. Ismail, A novel thin film composite forward osmosis membrane prepared from PSf–TiO2 nanocomposite substrate for water desalination, Chem. Eng. J. 237 (2014) 70– 80. [15] M. Ghanbari, D. Emadzadeh, W.J. Lau, T. Matsuura, M. Davoody, A.F. Ismail, Super hydrophilic TiO2/HNT nanocomposites as a new approach for fabrication of high performance thin film nanocomposite membranes for FO application, Desalination 371 (2015) 104–114. [16] B. Rajaeian, A. Heitz, M.O. Tade, S. Liu, Improved separation and antifouling performance of PVA thin film nanocomposite membranes incorporated with carboxylated TiO2 nanoparticles, J. Membr. Sci. 485 (2015) 48–59. [17] H. Dong, L. Wu, L. Zhang, H. Chen, C. Gao, Clay nanosheets as charged filler materials for high-performance and fouling-resistant thin film nanocomposite membranes, J. Membr. Sci. 494 (2015) 92–103. [18] M. Safarpour, V. Vatanpour, A. Khataee, M. Esmaeili, Development of a novel high flux and fouling-resistant thin film composite nanofiltration membrane by embedding reduced graphene oxide/TiO2, Sep. Purif. Technol. 154 (2015) 96–107. [19] G.S. Lai, W.J. Lau, P.S. Goh, A.F. Ismail, N. Yusof, Y.H. Tan, Graphene oxide incorporated thin film nanocomposite nanofiltration membrane for enhanced salt removal performance, Desalination 387 (2016) 14–24. [20] J. Yin, G. Zhu, B. Deng, Graphene oxide (GO) enhanced polyamide (PA) thinfilm nanocomposite (TFN) membrane for water purification, Desalination 379 (2016) 93–101. [21] A.S. Al-Hobaib, K.M. Al-Sheetan, L. El Mir, Effect of iron oxide nanoparticles on the performance of polyamide membrane for ground water purification, Mater. Sci. Semicond. Process. 42 (Part 1) (2016) 107–110. [22] J. Duan, Y. Pan, F. Pacheco, E. Litwiller, Z. Lai, I. Pinnau, High-performance polyamide thin-film-nanocomposite reverse osmosis membranes containing hydrophobic zeolitic imidazolate framework-8, J. Membr. Sci. 476 (2015) 303– 310. [23] H. Dong, L. Zhao, L. Zhang, H. Chen, C. Gao, W.S. Winston Ho, High-flux reverse osmosis membranes incorporated with NaY zeolite nanoparticles for brackish water desalination, J. Membr. Sci. 476 (2015) 373–383. [24] Q. Li, Y. Wang, J. Song, Y. Guan, H. Yu, X. Pan, F. Wu, M. Zhang, Influence of silica nanospheres on the separation performance of thin film composite poly (piperazine-amide) nanofiltration membranes, Appl. Surf. Sci. 324 (2015) 757– 764. [25] N. Rakhshan, M. Pakizeh, Removal of triazines from water using a novel OA modified SiO2/PA/PSf nanocomposite membrane, Sep. Purif. Technol. 147 (2015) 245–256. [26] G.N.B. Baroña, M. Choi, B. Jung, High permeate flux of PVA/PSf thin film composite nanofiltration membrane with aluminosilicate single-walled nanotubes, J. Colloid Interface Sci. 386 (1) (2012) 189–197. [27] M. Tian, Y.-N. Wang, R. Wang, Synthesis and characterization of novel highperformance thin film nanocomposite (TFN) FO membranes with nanofibrous substrate reinforced by functionalized carbon nanotubes, Desalination 370 (2015) 79–86. [28] P. Daraei, S.S. Madaeni, N. Ghaemi, M.A. Khadivi, L. Rajabi, A.A. Derakhshan, F. Seyedpour, PAA grafting onto new acrylate-alumoxane/PES mixed matrix nano-enhanced membrane: preparation, characterization and performance in dye removal, Chem. Eng. J. 221 (2013) 111–123. [29] X.-M. Wang, X.-Y. Li, K. Shih, In situ embedment and growth of anhydrous and hydrated aluminum oxide particles on polyvinylidene fluoride (PVDF) membranes, J. Membr. Sci. 368 (1–2) (2011) 134–143. [30] V. Vatanpour, S.S. Madaeni, L. Rajabi, S. Zinadini, A.A. Derakhshan, Boehmite nanoparticles as a new nanofiller for preparation of antifouling mixed matrix membranes, J. Membr. Sci. 401–402 (2012) 132–143. [31] A.A. Derakhshan, L. Rajabi, H. Karimnezhad, Morphology and production mechanism of the functionalized carboxylate alumoxane micro and nanostructures, Powder Technol. 225 (2012) 156–166. [32] W. Chen, S. Wu, Y. Lei, Z. Liao, B. Guo, X. Liang, D. Jia, Interfacial structure and performance of rubber/boehmite nanocomposites modified by methacrylic acid, Polymer 52 (19) (2011) 4387–4395. [33] N. Misdan, W.J. Lau, A.F. Ismail, T. Matsuura, Formation of thin film composite nanofiltration membrane: effect of polysulfone substrate characteristics, Desalination 329 (2013) 9–18. [34] A.K. Ghosh, B.-H. Jeong, X. Huang, E.M.V. Hoek, Impacts of reaction and curing conditions on polyamide composite reverse osmosis membrane properties, J. Membr. Sci. 311 (1–2) (2008) 34–45. [35] M. Fathizadeh, A. Aroujalian, A. Raisi, Effect of lag time in interfacial polymerization on polyamide composite membrane with different hydrophilic sub layers, Desalination 284 (2012) 32–41. [36] N. Ghaemi, P. Daraei, Enhancement in copper ion removal by PPy@Al2O3 polymeric nanocomposite membrane, J. Ind. Eng. Chem. 40 (2016) 26–33. [37] H. Rajabi, N. Ghaemi, S.S. Madaeni, P. Daraei, M.A. Khadivi, M. Falsafi, Nanoclay embedded mixed matrix PVDF nanocomposite membrane: preparation, characterization and biofouling resistance, Appl. Surf. Sci. 313 (2014) 207–214. [38] H. Rajabi, N. Ghaemi, S.S. Madaeni, P. Daraei, B. Astinchap, S. Zinadini, S.H. Razavizadeh, Nano-ZnO embedded mixed matrix polyethersulfone (PES) membrane: influence of nanofiller shape on characterization and fouling resistance, Appl. Surf. Sci. 349 (2015) 66–77.
90
N. Ghaemi / Journal of Colloid and Interface Science 485 (2017) 81–90
[39] M.T.M. Pendergast, J.M. Nygaard, A.K. Ghosh, E.M.V. Hoek, Using nanocomposite materials technology to understand and control reverse osmosis membrane compaction, Desalination 261 (3) (2010) 255–263. [40] J.H. Chang, I.W. Hunter, Characterization and control of the wettability of conducting polymer thin films, in: MRS Proceedings, 2009, pp. 1228. [41] Q. Pei, R. Qian, Protonation and deprotonation of polypyrrole chain in aqueous solutions, Synth. Met. 45 (1) (1991) 35–48. [42] Y. Mansourpanah, H. Shahebrahimi, E. Kolvari, PEG-modified GO nanosheets, a desired additive to increase the rejection and antifouling characteristics of polyamide thin layer membranes, Chem. Eng. Res. Des. 104 (2015) 530–540. [43] S. Japip, K.-S. Liao, Y. Xiao, T.-S. Chung, Enhancement of molecular-sieving properties by constructing surface nano-metric layer via vapor cross-linking, J. Membr. Sci. 497 (2016) 248–258. [44] M. Askari, T.-S. Chung, Natural gas purification and olefin/paraffin separation using thermal cross-linkable co-polyimide/ZIF-8 mixed matrix membranes, J. Membr. Sci. 444 (2013) 173–183.
[45] D. Rana, T. Matsuura, Surface modifications for antifouling membranes, Chem. Rev. 110 (4) (2010) 2448–2471. [46] R. Kumar, A.F. Ismail, Fouling control on microfiltration/ultrafiltration membranes: effects of morphology, hydrophilicity, and charge, J. Appl. Polym. Sci. 132 (21) (2015) (n/a-n/a). [47] L.D. Nghiem, A.I. Schäfer, Adsorption and transport of trace contaminant estrone in NF/RO membranes, Environ. Eng. Sci. 19 (6) (2004). [48] N. Ghaemi, S.S. Madaeni, A. Alizadeh, H. Rajabi, P. Daraei, M. Falsafi, Effect of fatty acids on the structure and performance of cellulose acetate nanofiltration membranes in retention of nitroaromatic pesticides, Desalination 301 (2012) 26–41. [49] L.D. Nghiem, A.I. Schäfer, T.D. Waite, Adsorptive interactions between membranes and trace contaminants, Desalination 147 (1–3) (2002) 269–274.