Preparation and antibacterial property of polyethersulfone ultrafiltration hybrid membrane containing halloysite nanotubes loaded with copper ions

Chemical Engineering Journal 210 (2012) 298–308

Contents lists available at SciVerse ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Preparation and antibacterial property of polyethersulfone ultrafiltration hybrid membrane containing halloysite nanotubes loaded with copper ions Yifeng Chen a, Yatao Zhang a,⇑, Jindun Liu a, Haoqin Zhang a,⇑, Kaijuan Wang b a b

School of Chemical Engineering and Energy, Zhengzhou University, Zhengzhou 450001, PR China College of Public Health, Zhengzhou University, Zhengzhou 450001, PR China

h i g h l i g h t s " Halloysite nanotubes act as a carrier of antibacterial agent for loading copper ions. " PES UF membranes containing HNTs loaded with copper ions were prepared. " The membranes showed good antibacterial performance against Escherichia coli and S. aureus.

a r t i c l e

i n f o

Article history: Received 29 April 2012 Received in revised form 28 August 2012 Accepted 29 August 2012 Available online 10 September 2012 Keywords: Halloysite nanotubes (HNTs) Copper ions Polyethersulfone ultrafiltration membrane Antibacterial activity

a b s t r a c t Polyethersulfone ultrafiltration membrane containing halloysite nanotubes loaded with copper ions (Cu2+-HNTs/PES) were prepared via phase inversion method using polyethersulfone (PES) as membrane material and Cu2+-HNTs as an antibacterial agent, which were synthesized by chemical modification of HNTs with silane coupling agent, and then mixed with copper dichloride for complexing copper ions. The morphology and performance of the membranes were characterized by SEM, AFM, TEM, contact angle, and mechanical measurements. The hybrid membranes were shown to be more hydrophilic, with a higher pure water flux. Mechanical test revealed that the mechanical strength of hybrid membranes increased as the addition of Cu2+-HNTs. It was also found that Cu2+-HNTs were dispersed uniformly in the membrane. The antibacterial test indicated that the hybrid membranes showed good antibacterial performance against Gram-negative bacteria (Escherichia coli) and Gram-positive bacteria (Staphylococcus aureus). Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction Polyethersulfone (PES) membranes have been widely used in water treatment, food processing and biomedical fields because of their excellent chemical resistance, good thermal stability, oxidation resistance and mechanical properties [1–5]. However, PES membrane is liable to be attached by bacteria in water treatment, which leads to membrane biofouling. Membrane biofouling results in the loss of water throughput increased the operational pressure and system downtime for cleaning [6,7]. Furthermore, biofouling cannot be reduced by pretreatment, because of its self-replicating nature [8]. One of the current trends in preventing membrane biofouling is to improve the antibacterial property of membranes by chemical modification and physical blending. Qiu et al. grafted Poly (4-vinylpyridine) (P4VP) onto polysulfone (PSF) membranes by surface⇑ Corresponding authors. Tel.: +86 371 67781724; fax: +86 371 67781734. E-mail addresses: [email protected] (Y. Zhang), [email protected] (H. Zhang). 1385-8947/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cej.2012.08.100

initiated atom transfer radical polymerization (SI-ATRP) and then immobilized copper (II) ions on the modified membrane. Copperloaded membranes exhibited excellent antibacterial properties [9]. Yao et al. prepared novel antibacterial polyurethane (PU) fibrous membranes by electro-spinning the polymer followed by plasma pretreatment, UV-induced graft copolymerization and quaternization reaction. The antibacterial activities of the modified PU fibrous membranes were assessed against S. aureus and Escherichia coli [10]. Taurozzi et al. prepared polysulfone–silver nanocomposite membranes via the wet phase inversion process by physical blending PSF casting solution with silver. The polysulfone–silver nanocomposite membrane was shown to effectively prevent biofouling [11]. Compared with chemical modification, the physical blending with antibacterial agent was a convenient and effective method for practical utilization. Recently, various heavy metals as antibacterial agent has been widely used or studied in the ultrafiltration (UF) process including silver, copper, zinc, and so on. Among them, copper represents a more promising metal for antibacterial applications due to its excellent antimicrobial properties, low toxicity, and low cost [9,12]. The

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

literature reported that the theory of antibacterial mode of copper ions may interact with cell wall membrane, leading to the breakage of cell wall and solidify structure of proteins structure [13,14]. However, it has been known that the copper ions by physical blending are easily washed away from membranes, which will lead to a shorter antibacterial time [2,15]. To slow the release of copper ions, several methods have been used to load copper ions onto various inorganic carriers [12,16,17]. Tan et al. immobilized copper on the porous carbonaceous materials by direct immersion of ordered mesoporous carbon in CuCl2 aqueous solution and it showed that the release time of copper ions can be delayed for a long time [12]. Özdemir et al. loaded copper ions onto montmorillonite by ion-exchange and it showed that the copper ions released from the surface very slowly [16]. Halloysite (Al2Si2O5 (OH)42H2O) is a kind of naturally occurring aluminosilicate clay with hollow nanotubular structure, which has been used as catalyst support, nanoreactors and adsorbents [18– 21]. In addition, as reported by Abdullayev et al., silver nanorods for antibacterial composite coating have been synthesized using HNTs as templates [22]. Furthermore, modified HNTs loaded with silver nanoparticles as an antibacterial agent was synthesized via a series of reactions including in-situ polymerization in our previous work [23]. The studies on HNTs loading copper ions as antibacterial agent are rarely reported. In this paper, HNTs loading copper ions (Cu2+-HNTs) were prepared and used as a novel antibacterial agent. In contrast with other inorganic carriers, the adequate hydroxyl groups and tubular structure of HNTs make them easily dispersed in the polymer matrix [24]. Zhang et al. also reported that the modified HNTs were dispersed uniformly and individually in the membrane and there were few large clusters [23]. More importantly, the characteristics of HNTs resemble that of multi-walled carbon nanotube (MWCNT), such as the tubular structure and the hydrophilic radical on the surface. Therefore, HNTs have the same advantage with MWCNT because of their similar characteristics. Celik et al. have shown that MWCNT blended PES ultrafiltration membranes appeared to be more hydrophilic, with a higher pure water flux than PES membranes. Moreover, this study reported that the carbon nanotube content of the MWCNT/PES membranes was shown to alleviate the membrane fouling caused by natural water [25]. Furthermore, compared with MWCNT, HNTs are cheap, rich in supply, and more easily modified. In order to make HNTs have the capability of loading copper ions, silane coupling agent was chosen to modify HNTs, which has lone pairs of electrons that can bind copper ion through an electron pair sharing to form a complex [26,27]. Silane is considered as a double purpose agent. Primarily, it will provide improved compatibility between the polymer matrix and the inorganic phase [26]. Secondly, Cu2+-HNTs were prepared via two steps, silanemodification and complexation reaction. This method is simple, effective, and applied to practical production. In the present study, PES ultrafiltration membranes were modified by blending with Cu2+-loaded HNTs, which were synthesized by chemical modification with silane coupling agent, and then the silane-modified HNTs were mixed with copper dichloride for complexing copper ions. The effects of Cu2+-HNTs content on the structure, characteristic, and separation performance of the membranes were discussed in detail. Finally, the antibacterial activities of the hybrid membranes were assessed against both Gram-positive (S. aureus) and Gram-negative (E. coli). 2. Experimental 2.1. Materials Halloysite nanotubes (HNTs) were refined from clay minerals in Henan province, China. N-b-(aminoethyl)-c-aminopropyltrimeth-

299

oxy silane (AEAPTMS) was used for the chemical modification of HNTs and was purchased from Merck Company. Polyethersulfone (PES) was supplied by BASF Company. The test strains, E. coli (8099) and S. aureus (ATCC6538) used for this study were provided by College of Public Health of Zhengzhou University. Other reagents were all of analytical grade and used without further purification. The used water is de-ionized water. 2.2. Copper ions loaded on HNTs Chemical modification of HNTs by AEAPTMS was carried out by applying this procedure as follows: AEAPTMS (15 g) was dissolved in toluene (150 ml) by shaking and then HNTs (10 g) was poured into the solution. The resulting mixture was then refluxed at 125 °C for 24 h under rigorous stirring. Then, the surface modified HNTs were washed with isopropanol and collected by centrifugation. The product was dried in a vacuum oven at 50 °C. The silane-modified halloysite nanotubes (7 g) were mixed with 100 mL 0.5 mol/L CuCl2 solutions and stirred continuously at 60 °C for 24 h. After the exchange, the sample was washed with a large amount of de-ionized water. Then, the product was collected by centrifugation and dried in a vacuum oven at 60 °C. 2.3. Preparation of membranes Pure PES membrane and HNTs/PES hybrid membranes were prepared by the phase-inversion method. Casting solution of the PES dissolved in N,N-dimethylacetamide (DMAc) was prepared using polyvinylpyrrolidone (PVP) as pore former by stirring at room temperature. After formation of homogeneous solution, HNTs was added into the casting solution under stirring. In order to obtain optimal dispersions of particles in the polymer solutions, agitation was required for at least 12 h. The casting solutions were then kept in the dark for at least 24 h to remove air bubbles. Different HNTs concentrations (0–3%, by weight of PES) of polymer dopes consisting of PES (20%, by weight of the solution), DMAc (71.2%, by weight of the solution), PVP (8%, by weight of the solution), and acetone (0.8%, by weight of the solution) were prepared. The solutions were cast uniformly onto a glass substrate by means of a hand-casting knife with the thickness of 0.3 mm and then immersed in a bath filled with de-ionized water. The formed membranes were stored in water and this allows the water soluble components in the membrane to be leached out. 2.4. Characterization of HNTs 2.4.1. Fourier transformed infra red spectroscopy (FTIR) FTIR spectra of the HNTs were performed at 2 cm1 resolution with Thermo Nicolet IR 200 spectroscope (Thermo Nicolet Corporation, USA). Typically, 64 scans were signal-averaged to reduce spectral noise. The spectra were recorded in the 400–4000 cm1 range using KBr pellets. 2.4.2. Thermogravimetric analysis (TGA) TGA measurements were carried out using a TG-DTA, DT-40 system (Shimadzu, Japan). Samples (3 mg) were heated from 0 to 800 °C at 10 °C/min under flowing nitrogen. 2.4.3. Specific surface area and pore size analyzer All specific surface area and pore size measurements were performed with a Micromeritics NOVA 4200e automatic physisorption analyzer (Quantachrome, USA) with multi-gas option. Moisture was removed from the samples prior to the surface area measurement by placing them in a vacuum oven for 24 h at 70 °C at a pressure of 500 mbar.

300

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

2.4.4. Transmission electron microscopy (TEM) A FEI model TECNAI G2 transmission electron microscope (200 KV acceleration voltages) was used to study the nanotubular shapes of the HNTs. The samples for analysis were ground in ethanol and agitated in a glass vial to disperse the particles within the solvent. The suspended particles were transferred to and allowed to dry on a copper grid (400 meshes) coated with a strong carbon film. 2.4.5. Wavelength dispersive X-ray fluorescence (WDXRF) spectrometer To know the elemental composition of the halloysite nanotubes, HNTs was determined by wavelength dispersive X-ray fluorescence (WDXRF) spectrometer using S4 PIONEER (BRUKER AXS GMBH). 2.5. Characterization of membranes 2.5.1. Scanning electron microscopy (SEM) Samples of the membranes were frozen in liquid nitrogen and then fractured. Cross section of the membranes was sputtered with gold, which were viewed with the microscope at 10 kV. The structure of the membranes was inspected by SEM using a JEOL Model JSM-6700F scanning electron microscope (Tokyo, Japan). 2.5.2. Atomic force microscope (AFM) For analyzing the surface morphology and roughness of the membranes, atomic force microscopy was employed using the AFM apparatus (DI Nanoscope IIIa, Veeco, USA). Small squares of the prepared membranes (approximately 1 cm2) were cut and glued on glass substrate. The membrane surfaces were examined in a scan size of 10 lm  10 lm. 2.5.3. Transmission electron microscopy (TEM) Cu2+-HNTs were observed with a FEI model TECNAI G2 transmission electron microscope operated at 200 kV. The membranes were embedded in epoxy resin and cross sections with a thickness of 50 nm were obtained by sectioning with a Leica Ultracut UCT ultramicrotome. Then these thin sections were mounted on the carbon-coated TEM copper grids. 2.5.4. X-ray photoelectron spectroscopy (XPS) The presence of copper on membrane surfaces was analyzed by the XPS (AXIS Ultra, Kratos, England). For the XPS analysis, the base pressure of the analyzer chamber was about 5  107 Pa. The survey spectra (from 0 to 1400 eV) were recorded. During the widescan, peak for C 1s was observed at binding energy 284.7 eV. All readings were calibrated with the corresponding C 1s as the standard for the correction of charging effects. 2.5.5. Contact angle Water contact angle (h) was measured at 25 °C and 50% RH on a contact angle system (OCA20, Dataphysics Instruments, Germany) for the evaluation of the membrane hydrophilicity. 1 lL of de-ionized water was carefully dropped on the top surface and the contact angle between the water and membrane was measured until no further change was observed. To minimize the experimental error, the contact angle was measured at five random locations for each sample and then the average was reported. 2.5.6. Porosity and pore size The porosity (e) was determined by gravimetric method, as defined in the following equation [28,29]:

e¼

m1  m2

qw  A  l

ð1Þ

where m1 is the weight of the wet membrane; m2 the weight of the dry membrane; qw the water density (0.998 g cm3); A the effective area of the membrane (m2), and l is the membrane thickness (m). Guerout–Elford–Ferry equation (Eq. (2)) was utilized to determine membrane mean pore radius (rm) on the basis of the pure water flux and porosity data [28,29].

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð2:9  1:75eÞ  8glQ rm ¼ e  A  DP

ð2Þ

where g is the water viscosity (8.9  104 Pa s), Q the volume of the permeate pure water per unit time (m3 s1), and DP is the operation pressure (0.1 MPa). 2.5.7. Separation performance of membranes A cross flow system (feed stream flowing tangentially to the membrane surface) was used to characterize the performance of the prepared membrane. The details of the description of the UF setup are given elsewhere [30,31]. The flat sheet membrane pieces with an effective area of 22.2 cm2. Each membrane was first prepressured using pure water at 200 kPa for 30 min to get a steady flux, and then the pure water flux was recorded at 100 kPa and a system temperature of 25 ± 2 °C. After that, the PEG (0.5 g/L) solution was forced to permeate through the membrane at the same pressure. The permeation flux (J) and rejection (R) were calculated using the following equation:

J¼

V A  Dt 

%R ¼

1

ð3Þ  Cp  100% Cf

ð4Þ

where V is the volume of permeate pure water (L), A the effective area of the membrane (m2), and Dt the permeation time (h), Cp the permeate concentration and Cf is the feed concentration. The concentrations of PEG was obtained by UV spectrophotometer. All of the filtration processes were repeated three times on different membrane sheets, the average value were calculated and reported. Molecular weight cut-off (MWCO) of the membrane was determined by measuring the rejection of PEG (10 kDa and 20 kDa). The smallest molecular weight that has a rejection of 90% is taken as the MWCO of the membrane. 2.5.8. Mechanical properties The mechanical properties of the prepared membranes were measured according to the standard method (ASTM D 882) using Shimazu AG-10-TB tensile test machine under ambient conditions. The tested samples were prepared by cutting membranes into 25 mm  50 mm pieces. Measurements were carried out with the rate of pull at 10 mm/min. For each test, five samples were used and the average values are reported for tensile strength and elongation. 2.6. Antibacterial activity tests The antibacterial activity of HNTs and the membranes were tested by an inhibition zone method. To measure antibacterial activity, all samples were sterilized by autoclaving for 30 min. Then the samples were placed on E. coli and S. aureus bacteria agar plate at an inoculum concentration of 105 colonies forming units per ml (cfu/ml), respectively, and then incubated at 37 °C for 24 h. All samples were tested for two groups. The diameters of inhibitory zones surrounding the samples disks were observed.

301

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

The plates were photographed and the average inhibition zone diameters were measured. Moreover, the antibacterial duration of these membranes were also investigated from examining the bacteriostasis rate by the viable cell counting technique. E. coli and S. aureus were inoculated in 5 ml of LB liquid nutrient medium respectively, and shaken for 12 h at 37 °C. The actual number of cells used for a given experiment was determined by the standard serial dilution method. 0.03 g various membranes including the membrane after prepared and the membrane which were kept in distilled water for 12 months, were cut and sterilized by autoclaving for 20 min, respectively. To test the antibacterial activity, the membranes were added into the 5 ml solution inoculated by about 106 cfu/ml E. coli and S. aureus, respectively, and then which were incubated at room-temperature. After 24 h, membranes were retrieved from cultures and washed by normal saline. The wash solutions were collected and diluted it with de-ionized water till its concentration becomes to 103 of the original value. 0.2 ml of dilution solution was spread onto LB culture medium and all plates were incubated at 37 °C for 24 h. The numbers of colonies on the plates were determined by the plate count method.

3. Results and discussion

Fig. 2. Thermogravimetric analysis (TGA) curves of (a) raw halloysite nanotubes, and (b) silane-modified halloysite nanotubes.

Table 1 Specific surface area, total pore volume and apparent average pore radius of halloysite nanotubes and silane-modified halloysite nanotubes.

3.1. Characterization of HNTs Fig. 1 shows the FTIR spectra of raw halloysite nanotubes (a) and silane-modified halloysite nanotubes (b). Compared to the unmodified HNTs, silane-modified HNTs exhibit some new peaks, such as the stretching CH2 vibration at 2954 cm1, and the stretching NH2 vibration at 3452 cm1 [23,32]. The peak at 3452 cm1 and 2954 cm1 were due to surface modification by the silane coupling agent. All of these peaks showed the presence of the AEAPTMS moieties in the modified halloysite nanotubes. Fig. 2 illustrates the thermogravimetric analysis (TGA) curves of raw nanotubes (a) and silane-modified nanotubes (b). In the curves of raw nanotubes (a), one major mass loss was resolved in the temperature range of 450–600 °C. This mass loss was assigned to the dehydroxylation of structural AlOH groups of halloysite [18,33]. Another mass loss in the range 0–150 °C, corresponding to the loss of adsorbed water (surface and interlayer) [18]. For AEAPTMS modified sample, three mass losses were resolved. The first mass loss in

Fig. 1. FTIR spectra of (a) raw halloysite nanotubes and (b) silane-modified halloysite nanotubes.

Specific surface area (m2/g) Total pore volume (cm3/g) Apparent average pore radius (nm)

HNTs

Modified HNTs

59.6 0.24 80

21.7 0.1 95.6

the range 0–150 °C was due to physically adsorbed water or AEAPTMS. The second mass loss occurring from approximately 250–475 °C was due to decomposition of the grafted silane [26,32]. The last mass loss over the range 475–575 °C corresponds to dehydroxylation of the residual structural AlOH groups. From these TGA analyses one may make two important conclusions. First, the silylation took place successfully, as also evidenced by the FTIR analysis. Second, the content of the AEAPTMS grafted onto HNTs was 0.105 g (KH-792)/g (HNTs). As shown in Table 1, the silane-modified halloysite show lower value of specific area, total pore volume and higher value of apparent average pore radius in comparison with raw halloysite. It is because that after the modification by the silane, the micro-pores of the HNTs were partly plugged, resulting from the decrease of the total pore volume and increase of the apparent average pore radius. Morphology of raw nanotubes, silane-modified nanotubes and copper–ion loaded nanotubes were observed by TEM and representative images were shown in Fig. 3. Fig. 3a shows that the halloysite nanotubes used in the experiment have a cylindrical shape, hollow and open-ended. The large and smooth pores can provide sufficient space for flowing of the water, which can improve the water flux of the membrane. Compared with Fig. 3a–c show that the TEM structure of silane-modified nanotubes and copper–ion loaded nanotubes were not much different from raw nanotubes, which suggest that modification had little influence on the skeleton construction of nanotubes. To further confirm Cu2+ load and silylation, the surface elemental composition was determined by WDXRF and the results shown in Table 2. Depending on the molecular formula of the raw halloysite nanotubes, the proportion of Si content to Al content was about 1:1. And according to the table, the ratio of Si content to Al content was 3:2 for copper–ion loaded nanotubes. The Si

302

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

Fig. 4. Static water contact angle of PES membranes with different concentrations of Cu2+-HNTs.

3.2. Characterization of membranes 3.2.1. Hydrophilicity of membranes The water contact angle measurement is one of the methods for characterization the hydrophilic property of the membrane surface. As shown in Fig. 4, the pure PES membrane has the contact angle of 84.9° due to the hydrophobic nature. The water contact angle decreased with increasing the Cu2+-HNTs concentration, which indicates that the membrane surface became more hydrophilic after adding Cu2+-HNTs. The lowest contact angle was decreased to 69.8° for 3 wt.% of Cu2+-HNTs. It can be explained that Cu2+-HNTs had favorable hydrophilicity due to the presence of a great deal of hydroxyl groups on the HNTs. In the phase inversion preparation of hybrid membranes, hydrophilic HNTs migrated spontaneously to the membrane/water interface to reduce the interface energy [25,28]. The result was also in agreement when membranes with other hydrophilic substance were used [28,29, 34].

Fig. 3. TEM images of (a) raw halloysite nanotubes, (b) silane-modified halloysite nanotubes, and (c) copper–ion loaded halloysite nanotubes.

Table 2 Surface elemental composition of copper-ion loaded halloysite nanotubes by WDXRF. Element

Mg

Al

Si

Cl

S

Cu

Others

Content (wt.%)

0.14

28

41.6

9.89

3.18

15.74

1.45

content increased from HNTs to Cu2+-HNTs, demonstrating successful silylation of HNTs. This result is in agreement with the conclusion drawn from FTIR analysis. The complexing amount of copper ions was about 0.117 g Cu2+/g (HNTs) through the calculation.

3.2.2. Cross-section morphology Fig. 5 shows the morphologies of cross-section of the tested membranes. All the membranes had a dense skin layer and a support layer with finger-like structure, which is the typical structure of asymmetric ultrafiltration membrane. These findings indicate that the addition of Cu2+-HNTs did not affect the structures of the cross-section. Therefore, the mechanism of PES membranestructure formation was not altered by the addition of Cu2+-HNTs [34,35]. However, compared with pure PES membrane, the finger-like pore size of the hybrid membrane seemed to become wider, which would enhance the permeation flux of the membrane. The porosity and pore size information of the tested membranes are listed in Table 3. As shown, the porosity of all membranes is nearly similar. But the mean pore size increased with increasing the Cu2+-HNTs content, this is because the Cu2+HNTs may increase the thermodynamic instability of the cast film, which promotes a rapid phase demixing, results in large pore formation at the membrane skin layer [28,36]. 3.2.3. Permeation properties The effect of Cu2+-HNTs concentration on the flux and rejection of PEG were shown in Fig. 6. The experimental results showed that the pure water flux of Cu2+-HNTs/PES hybrid membranes increased

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

303

Fig. 5. SEM images of the cross-section morphology of membranes. (a) Pure PES membrane, and (b) PES membrane with 3% of Cu2+-HNTs.

Table 3 Effect of copper-ion loaded halloysite nanotubes content on the porosity, pore size, and mechanical properties of the membranes. Cu2+-HNTs content (wt.%)

e

rm/nm

Tensile strength (MPa)

Elongation ratio (%)

0 1 2 3

0.656 0.653 0.657 0.652

27.8 30.3 32.8 35.8

3.52 3.65 3.83 3.97

14.75 14.62 14.53 14.37

Fig. 6. Separation performance of PES membranes with different concentrations of copper–ion loaded halloysite nanotubes.

with the addition of Cu2+-HNTs, the pure water flux was 73.2, 85.9, 102.3, and 120.1 L/(m2 h) wherein the content of Cu2+-HNTs (by weight of PES) showed at 0 wt.%, 1 wt.%, 2 wt.%, and 3 wt.%, respectively. Apparently, the pure water fluxes of membrane (with 3 wt.% of Cu2+-HNTs) were 64% higher than that without Cu2+-HNTs. Moreover, the results also showed that all the membranes have MWCO of 20 kDa. However, the rejection of PEG10000 decreased with the increase of the Cu2+-HNTs content. This implies that adding Cu2+-HNTs to PES polymer could enlarge its pore size. And the results shown in Table 3 also clearly showed that the pore size increased with increasing the Cu2+-HNTs content. The increase in pure water flux of the hybrid membranes might be explained as follows: Firstly, the hydrophilicity of the membranes can be improved significantly by the addition of Cu2+-HNTs,

which could attract water molecules inside the membrane matrix and promote them to pass through the membrane and enhance permeability accordingly [35]. Secondly, as shown in Fig. 3, HNTs possess hollow nanotubular structure in the submicrometer range, water molecules can permeate to them [34]. Finally, the addition of the hydrophilic HNTs enhanced the phase separation, which resulted in a bigger pore size and a higher pure water flux [25,35]. 3.2.4. Atomic force microscopy Fig. 7 displays three-dimensional AFM images of the membrane surfaces. In these images, the brightest area presents the highest point of the membrane surface and the dark regions indicate valleys or membrane pores. It can be seen that abundant nodular structure was formed in the top surface of pure PES membrane, whereas the membrane (with 3 wt.% of Cu2+-HNTs) was smooth. This finding is consistent with the other researchers who reported that the surface roughness of the pure membrane was higher than that of the hybrid membrane [28,37]. The influence of the addition of HNTs on the roughness of membrane can be explained as the followings. In the phase inversion, hydrophilic HNTs migrated spontaneously to the membrane/water interface, which resulted in the valleys of the membrane surfaces were filled with HNTs. It is well established that impurities were likely to be absorbed in the valleys of membrane with coarser surfaces, resulting in clogging of the valleys. Furthermore, membrane with lower roughness and surface energy has stronger anti-fouling abilities [28,37,38]. The above results show that addition of Cu2+-HNTs does not have a negative effect on membrane performance; rather, it effectively improves the permeating flux and anti-fouling properties. 3.2.5. Cu2+-HNTs distribution in the membrane Fig. 8 shows TEM images of the Cu2+-HNTs/PES hybrid membranes at different magnifications. Cu2+-HNTs were observed as black spots in the image (indicated by arrows in Fig. 8a), and they can be readily distinguished against the PES matrix. Fig. 8a shows that most Cu2+-HNTs were dispersed uniformly in the membrane, with the exception of a few large clusters that might have resulted from particles overlapping or from particles coalescing in the membrane. The same result was also obtained when PVDF membranes were modified by inorganic nano-sized Al2O3 particles [34]. The individual Cu2+-HNTs was clearly observed in Fig. 8b. Thus, at high magnifications it can be seen that Cu2+-HNTs have tubular structure in the PES matrix. These hybrid membranes were prepared by dispersing organ-modified HNTs into the polymer matrix, which is based on the physical adsorption theory. The interfacial bonding between HNTs and the polymer matrix is on mechanical crosslinking and secondary bonds. Organ-modified

304

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

Fig. 7. AFM surface images of the membranes. (a) pure PES membrane, and (b) PES membrane with 3% of Cu2+-HNTs.

Fig. 8. TEM images of hybrid membranes at different magnifications.

HNTs was dispersed well in the polymer matrix, which could reduce interfacial stress and improve the compatibility of HNTs and polymer matrix.

3.2.6. Mechanical properties of membranes The tensile strength and elongation ratio results are presented in Table 3. With the concentrations of copper-ion loaded halloysite nanotubes increasing from 0 wt.% to 3 wt.%, breaking strength increased from 3.52 MPa to 3.97 MPa while the elongation ratio decreased from 14.75% to 14.37%. The results indicate that the mechanical strength of the membrane enhanced with the increase of the Cu2+-HNTs concentration. This result could be attributed to the interactions between Cu2+-HNTs and PES. Cu2+-HNTs could

act as a crosslinking point in composite membrane to link the polymer chain and increase the rigidity of polymer chain, indicating that more energy is needed to break down the bond between Cu2+-HNTs and PES [39]. 3.3. Antibacterial effects of membranes with Cu2+-HNTs The antibacterial activity of the Cu2+-HNTs/PES hybrid membranes (with 3 wt.% of Cu2+-HNTs) was tested by the method of inhibition zone. It is clearly observed that the membranes with Cu2+-HNTs showed antibacterial activity against both E. coli (Fig. 9a) and S. aureus (Fig. 9b). Cao et al. explained the antibacterial activity of the hybrid membranes was mainly caused by the

Fig. 9. Antibacterial action of the hybrid membranes.

305

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

Fig. 10. Antibacterial actions of (a) and (b) raw halloysite nanotubes, and (c) and (d) copper-ion loaded halloysite nanotubes.

5

3.0x10

285 532

5

Counts / s

2.5x10

5

2.0x10

933

5

1.5x10

5

1.0x10

400

Fig. 11. Antibacterial sketch map of the hybrid membrane. 4

5.0x10

antibacterial agent, not from the PES [2]. Fig. 10 showed that HNTs had no antibacterial activities against the two bacteria, so the antibacterial effect of the hybrid membranes should be due to the Cu2+. And the antibacterial mechanism of the Cu2+-HNTs/PES hybrid membranes, as shown in Fig. 11, could be interpreted as follows: The loaded copper ions on membrane surfaces could absorb onto the surface of bacteria cells and further damage the cell membrane and solidify structure of proteins structure [13,14,16]. To confirm the presence of copper on membrane surfaces, XPS analysis was used to characterize the hybrid membranes surface. As shown in Fig. 12, the emission peak at 933 eV attributed to the binding energy (BE) of Cu 2p [40,41]. The results confirmed copper ions were firmly existed on membrane surfaces. So the antibacterial activity of the hybrid membranes was caused by the loaded copper ions on membrane surfaces. As shown in Fig. 9, the diameter of the hybrid membranes placed at the bottom of Petri dish was 10 mm and 9 mm,

0.0

1200

1000

800

600

400

200

0

Binding Energy (ev) Fig. 12. XPS spectra of the hybrid membrane surface.

respectively, and the average size of its inhibition zone was 2.5 mm and 1.5 mm, respectively. The results indicated that the hybrid membranes had significant inhibition capacity toward both Gram-negative and the Gram-positive bacteria. It was also revealed that the hybrid membranes had better antibacterial activity to Gram-negative bacteria, E. coli. Basri et al. attributed the phenomenon to the difference of structure of the cell walls. Cell walls of Gram-positive species contain 3–20 times more peptidoglycan than Gram-negative bacteria. Since peptidoglycans are

306

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

Fig. 13. Measurement of antibacterial property by the bacteriostasis rate. (a) and (d) pure PES membrane, (b) and (e) PES membrane with 3% of Cu2+-HNTs, and (c) and (f) the hybrid membrane kept in distilled water for 12 months.

Table 4 Antibacterial activities of the tested membranes. E. coli

PES membrane PES membrane with 3% of Cu2+-HNTs Hybrid membrane kept in distilled water for 12 months

S. aureus

The numbers of bacterial colonies (cfu)

Bacteriostasis rate (%)

The numbers of bacterial colonies (cfu)

Bacteriostasis rate (%)

150 0 76

– 100 49.3

365 0 201

– 100 44.9

Table 5 Comparison of antibacterial activity between this work and the membranes with other antibacterial agents. Membranes with different antibacterial agent

Antibacterial efficiency

References

Blending Cu2+-HNTs Modification with quaternary ammonium Polysulfone–silver nanocomposite membranes Blending TiO2 particles 6-O-carboxymethylchitosan/WPU composite membrane

100% (bacteriostasis rate) 99.9% (bacteriostasis rate) Much less bacterial growth on the surface Removal almost complete of E. coli within 1 min of UV light exposure The test bacteria formed on membranes were decreased with the increase of the chitosan derivative

This study [10] [11] [43] [44]

negatively charged, they probably bind to some portion of copper ion in the bacteria culture [42]. Moreover, the antibacterial activity of the membranes including the membrane after prepared and the membrane which were kept in distilled water for 12 months, were tested again by bacteriostasis rate. The viable cell numbers of E. coli and S. aureus after coming into contact with various membranes were evaluated, respectively. The antibacterial effect was shown in Fig. 13 and the result was shown in Table 4. As can be seen, the bacteriostasis rates of the new hybrid membranes against E. coli and S. aureus are 100% and 100%, respectively (the content of Cu2+-HNTs is about 3% by weight of

PES). The hybrid membranes had a high antibacterial efficacy for E. coli and S. aureus. In comparison, the bacteriostasis rates of the hybrid membranes against E. coli and S. aureus are 49.3% and 44.9% after 12 months, respectively. In other words, the bacteriostasis rate of the hybrid membranes against E. coli is higher than that of S. aureus, and corresponds to the result of inhibition zone. It also indicated that the hybrid membranes had a good antibacterial duration. The comparison of antibacterial activity between this work and the membranes with other antibacterial agents showed that Cu2+HNTs/PES hybrid membranes also had high antibacterial property (Table 5). In addition, Cu2+-HNTs/PES hybrid membranes had a

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

good antibacterial duration. Thus, the hybrid membranes will have a potential application to reduce bacterial fouling in membrane treatment of water. 4. Conclusion Cu2+-HNTs/PES hybrid ultrafiltration membranes were prepared via phase inversion by dispersing Cu2+-HNTs in the PES casting solution. The SEM, AFM, and permeation property analyses showed that addition of Cu2+-HNTs did not affect the microstructure of the PES membranes. Although the membrane surface morphology was altered by decreased of surface roughness, it had no negative effects on membrane permeability or anti-fouling performance. The hydrophilicity and mechanical strength of membrane were enhanced after adding Cu2+-HNTs. Furthermore, the pure water fluxes of membrane (with 3% of Cu2+-HNTs) was 64% higher than that without Cu2+-HNTs. TEM indicated that most Cu2+-HNTs were dispersed uniformly throughout the hybrid membrane. Moreover, the antibacterial test indicated that the hybrid membranes were effective against E. coli and S. aureus, which would have a potential application to reduce bacterial fouling in membrane treatment of waste-water. Acknowledgments We gratefully acknowledge the support from National Natural Science Foundation of China (No. 21106137), Research Fund for the Doctoral Program of Higher Education of China (No. 20114101120001) and Innovation Scientists and Technicians Troop Construction Projects of Zhengzhou City. References [1] Q. Wei, J. Li, B. Qian, B. Fang, C. Zhao, Preparation, characterization and application of functional polyethersulfone membranes blended with poly (acrylic acid) gels, J. Membr. Sci. 337 (2009) 266–273. [2] X. Cao, M. Tang, F. Liu, Y. Nie, C. Zhao, Immobilization of silver nanoparticles onto sulfonated polyethersulfone membranes as antibacterial materials, Colloids Surf. B 81 (2010) 555–562. [3] J. Shen, H. Ruan, L. Wu, C. Gao, Preparation and characterization of PES–SiO2 organic–inorganic composite ultrafiltration membrane for raw water pretreatment, Chem. Eng. J. 168 (2011) 1272–1278. [4] A. Rahimpour, S.S. Madaeni, M. Jahanshahi, Y. Mansourpanah, N. Mortazavian, Development of high performance nano-porous polyethersulfone ultrafiltration membranes with hydrophilic surface and superior antifouling properties, Appl. Surf. Sci. 255 (2009) 9166–9173. [5] J.J. Qin, M.H. Oo, Y. Li, Development of high flux polyethersulfone hollow fiber ultrafiltration membranes from a low critical solution temperature dope via hypochlorite treatment, J. Membr. Sci. 247 (2005) 137–142. [6] S. Ciston, R.M. Lueptow, K.A. Gray, Bacterial attachment on reactive ceramic ultrafiltration membranes, J. Membr. Sci. 320 (2008) 101–107. [7] W. Zhao, J. Huang, B. Fang, S. Nie, N. Yi, B. Su, H. Li, C. Zhao, Modification of polyethersulfone membrane by blending semi-interpenetrating network polymeric nanoparticles, J. Membr. Sci. 369 (2011) 258–266. [8] A.K. Singh, P. Singh, S. Mishra, V.K. Shahi, Anti-biofouling organic-inorganic hybrid membrane for water treatment, J. Mater. Chem. 22 (2012) 1834–1844. [9] J. Qiu, Y. Zhang, Y. Zhang, H. Zhang, J. Liu, Synthesis and antibacterial activity of copper-immobilized membrane comprising grafted poly(4-vinylpyridine) chains, J. Colloid Interface Sci. 354 (2011) 152–159. [10] C. Yao, X. Li, K.G. Neoh, Z. Shi, E.T. Kang, Surface modification and antibacterial activity of electrospun polyurethane fibrous membranes with quaternary ammonium moieties, J. Membr. Sci. 320 (2008) 258–267. [11] J.S. Taurozzi, H. Arul, V.Z. Bosak, A.F. Burban, T.C. Voice, M.L. Bruening, V.V. Tarabara, Effect of filler incorporation route on the properties of polysulfonesilver nanocomposite membranes of different porosities, J. Membr. Sci. 325 (2008) 58–68. [12] S. Tan, W. Zou, F. Jiang, S. Tan, Y. Liu, D. Yuan, Facile fabrication of coppersupported ordered mesoporous carbon for antibacterial behavior, Mater. Lett. 64 (2010) 2163–2166. [13] P.C. Liu, J.H. Hsieh, C. Li, Y.K. Chang, C.C. Yang, Dissolution of Cu nanoparticles and antibacterial behaviors of TaN–Cu nanocomposite thin films, Thin Solid Films 517 (2009) 4956–4960. [14] Z.G. Dan, H.W. Ni, B.F. Xu, J. Xiong, P.Y. Xiong, Microstructure and antibacterial properties of AISI 420 stainless steel implanted by copper ions, Thin Solid Films 492 (2005) 93–100.

307

[15] K. Zodrow, L. Brunet, S. Mahendra, D. Li, A. Zhang, Q. Li, P.J.J. Alvarez, Polysulfone ultrafiltration membranes impregnated with silver nanoparticles show improved biofouling resistance and virus removal, Water Res. 43 (2009) 715–723. [16] G. Özdemir, M.H. Limoncu, S. Yapar, The antibacterial effect of heavy metal and cetylpridinium-exchanged montmorillonites, Appl. Clay Sci. 48 (2010) 319– 323. [17] A. Top, S. Ülkü, Silver, zinc, and copper exchange in a Na-clinoptilolite and resulting effect on antibacterial activity, Appl. Clay Sci. 27 (2004) 13–19. [18] S. Barrientos-Ramírez, E.V. Ramos-Fernández, J. Silvestre-Albero, A. SepúlvedaEscribano, M.M. Pastor-Blas, A. González-Montiel, Use of nanotubes of natural halloysite as catalyst support in the atom transfer radical polymerization of methyl methacrylate, Microporous Mesoporous Mater. 120 (2009) 132–140. [19] D.G. Shchukin, G.B. Sukhorukov, R.R. Price, Y.M. Lvov, Halloysite nanotubes as biomimetic nanoreactors, Small 5 (2005) 510–513. [20] P. Luo, Y. Zhao, B. Zhang, J. Liu, Y. Yang, J. Liu, Study on the adsorption of neutral red from aqueous solution onto halloysite nanotubes, Water Res. 44 (2010) 1489–1497. [21] M. Du, B. Guo, D. Jia, Newly emerging applications of halloysite nanotubes: a review, Polym. Int. 59 (2010) 574–582. [22] E. Abdullayev, K. Sakakibara, K. Okamoto, W. Wei, K. Ariga, Y. Lvov, Natural tubule clay template synthesis of silver nanorods for antibacterial composite coating, Appl. Mater. Interfaces 3 (2011) 4040–4046. [23] J. Zhang, Y. Zhang, Y. Chen, L. Du, B. Zhang, H. Zhang, J. Liu, K. Wang, Preparation and characterization of novel polyethersulfone hybrid ultrafiltration membranes bending with modified halloysite nanotubes loaded with silver nanoparticles, Ind. Eng. Chem. Res. 51 (2012) 3081–3090. [24] H. Ismail, P. Pasbakhsh, M.N.A. Fauzi, A.A. Bakar, Morphological, thermal and tensile properties of halloysite nanotubes filled ethylene propylene diene monomer (EPDM) nanocomposites, Polym. Test. 27 (2008) 841–850. [25] E. Celik, H. Park, H. Choi, H. Choi, Carbon nanotube blended polyethersulfone membranes for fouling control in water treatment, Water Res. 45 (2011) 274– 282. [26] A.F. Ismail, S.A. Hashemifard, T. Matsuura, Facilitated transport effect of Ag+ ion exchanged halloysite nanotubes on the performance of polyetherimide mixed matrix membrane for gas separation, J. Membr. Sci. 379 (2011) 378– 385. [27] A. Ghaee, M. Shariaty-Niassar, J. Barzin, T. Matsuura, Effects of chitosan membrane morphology on copper ion adsorption, Chem. Eng. J. 165 (2010) 46–55. [28] 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. [29] J. Li, Z. Xu, H. Yang, L. Yu, M. Liu, Effect of TiO2 nanoparticles on the surface morphology and performance of microporous PES membrane, Appl. Surf. Sci. 255 (2009) 4725–4732. [30] A. Rahimpour, S.S. Madaeni, Y. Mansourpanah, The effect of anionic, non-ionic and cationic surfactants on morphology and performance of polyethersulfone ultrafiltration membranes for milk concentration, J. Membr. Sci. 296 (2007) 110–121. [31] S.S. Madaeni, A. Rahimpour, Effect of type of solvent and non-solvents on morphology and performance of polysulfone and polyethersulfone ultrafiltration membranes for milk concentration, Polym. Adv. Technol. 16 (2005) 717–724. [32] S.A. Hashemifard, A.F. Ismail, T. Matsuura, Mixed matrix membrane incorporated with large pore size halloysite nanotubes (HNT) as filler for gas separation: experimental, J. Colloid Interface Sci. 359 (2011) 359–370. [33] P. Yuan, P.D. Southon, Z. Liu, M.E.R. Green, J.M. Hook, S.J. Antill, C.J. Kepert, Functionalization of halloysite clay nanotubes by grafting with caminopropyltriethoxysilane, J. Phys. Chem. C 112 (2008) 15742–15751. [34] Y. Lu, S.L. Yu, B.X. Chai, X. Shun, Effect of nano-sized Al2O3-particle addition on PVDF ultrafiltration membrane performance, J. Membr. Sci. 276 (2006) 162– 167. [35] C. Liao, P. Yu, J. Zhao, L. Wang, Y. Luo, Preparation and characterization of NaY/ PVDF hybrid ultrafiltration membranes containing silver ions as antibacterial materials, Desalination 272 (2011) 59–65. [36] J. Huang, G. Arthanareeswaran, K. Zhang, Effect of silver loaded sodium zirconium phosphate (nanoAgZ) nanoparticles incorporation on PES membrane performance, Desalination 285 (2012) 100–107. [37] A. Razmjou, J. Mansouri, V. Chen, The effects of mechanical and chemical modification of TiO2 nanoparticles on the surface chemistry, structure and fouling performance of PES ultrafiltration membranes, J. Membr. Sci. 378 (2011) 73–84. [38] Y. Wei, H.Q. Chu, B.Z. Dong, X. Li, S.J. Xia, Z.M. Qiang, Effect of TiO2 nanowire addition on PVDF ultrafiltration membrane performance, Desalination 272 (2011) 90–97. [39] L. Yu, Z. Xu, H. Shen, H. Yang, Preparation and characterization of PVDF–SiO2 composite hollow fiber UF membrane by sol–gel method, J. Membr. Sci. 337 (2009) 257–265. [40] M.C. Rodríguez-Argüelles, R. Villalonga, C. Serra, R. Cao, M.A. Sanromán, M.A. Longo, A copper(II) thiosemicarbazone complex built on gold for the immobilization of lipase and laccase, J. Colloid Interface Sci. 348 (2010) 96– 100. [41] R.S. Vieira, M.L.M. Oliveira, E. Guibal, E. Rodríguez-Castellón, M.M. Beppu, Copper, mercury and chromium adsorption on natural and crosslinked

308

Y. Chen et al. / Chemical Engineering Journal 210 (2012) 298–308

chitosan films: an XPS investigation of mechanism, Colloids Surf, A 374 (2011) 108–114. [42] H. Basri, A.F. Ismail, M. Aziz, K. Nagai, T. Matsuura, M.S. Abdullah, B.C. Ng, Silver-filled polyethersulfone membranes for antibacterial applications – effect of PVP and TAP addition on silver dispersion, Desalination 261 (2010) 264–271.

[43] R.A. Damodar, S.J. You, H.H. Chou, Study the self cleaning, antibacterial and photocatalytic properties of TiO2 entrapped PVDF membranes, J. Hazard. Mater. 172 (2009) 1321–1328. [44] S.H. Yu, F.L. Mi, S.S. Shyu, C.H. Tsai, C.K. Peng, J.Y. Lai, Miscibility, mechanical characteristic and platelet adhesion of 6-O-carboxymethylchitosan/ polyurethane semi-IPN membranes, J. Membr. Sci. 276 (2006) 68–80.