Biogenic silver nanocomposite polyethersulfone UF membranes with antifouling properties

Journal of Membrane Science 471 (2014) 274–284

Contents lists available at ScienceDirect

Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci

Biogenic silver nanocomposite polyethersulfone UF membranes with antifouling properties Manying Zhang a, Robert W. Field b, Kaisong Zhang a,n a b

Key Laboratory of Urban Pollutant Conversion,Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China Department of Engineering Science, University of Oxford, Oxford OX1 3PJ, UK

art ic l e i nf o

a b s t r a c t

Article history: Received 18 June 2014 Received in revised form 9 August 2014 Accepted 11 August 2014 Available online 22 August 2014

Biofouling remains one of the most challenging issues of membrane application in water industry. One of the practical strategies to control fouling is anti-biofouling membrane. High concentration and high stable biogenic nanoparticle silver with the averaged diameter of only 6 nm (Bio-Ag0-6) was firstly extracted from the supernatant of Lactobacillus fermentum. The biogenic nanocomposite polyethersulfone (Bio-Ag0-6/PES) membranes were prepared by adding different amounts of biogenic nanoparticle silver into the dope solution. The nanocomposite membranes were systematically tested for physical properties with pure water permeability, MWCO (molecular weight cut-off), contact angle, scanning electron microscopy (SEM) and atomic force microscopy (AFM). The results demonstrated that the Bio-Ag0-6 nanoparticles well dispersed into PES matrix without aggregation. Bio-Ag0-6 slightly increased the hydrophilicity of the PES membranes, improved the water permeability and did not sacrifice the selectivity of BSA. The protein adsorption on the membrane surface decreased significantly due to the increased hydrophilicity and the improved smoothness of membrane surfaces. The evaluation of silver release from the composite membranes indicated the good stability of immobilized Bio-Ag0-6 in the membranes. The results of disk diffusion test revealed the excellent antibacterial activity of the nanocomposite membranes. In addition, the sludge immersion and the bacterial suspension filtration experiments showed that the Bio-Ag0-6/PES composite membranes not only prevent the bacteria attachment on the membrane surface but also inhibit the reproduction and development of biofims. The Bio-Ag0-6/PES composite membranes were considered to be an effective strategy to decrease the biofouling in the membrane process. & 2014 Elsevier B.V. All rights reserved.

Keywords: Biogenic silver nanoparticles Ultrafiltration membrane Antibacterial Biofouling Flux recovery

1. Introduction The ultrafiltration (UF) process is an attractive technology due to low capital cost, small footprint, low operation pressure and no phase change during the filtration process [1,2]. Membrane fouling, especially biofouling, remains one of the most challenging issues in membrane separation processes which hinder wider applications of UF in wastewater treatment system [3,4]. Biofouling takes place through a series of steps including the attachment of microorganisms on the membrane surface, accumulation of assimilable organics, multiplication, colony formation and finally biofilm maturation [5]. As the most complicated fouling, biofouling usually causes various negative effects on membrane performances such as flux decline, increment of operation or maintenance costs and membrane degradation [6,7]. Therefore, many efforts have been made to develop anti-fouling strategies.

n

Corresponding author. Tel.: þ 86 592 6190782; fax: þ 86 592 6190534. E-mail address: [email protected] (K. Zhang).

http://dx.doi.org/10.1016/j.memsci.2014.08.021 0376-7388/& 2014 Elsevier B.V. All rights reserved.

To reduce biofouling, functional membranes containing biocides or antibacterial materials have attracted tremendous interest. Silver is one of the most widely studied biocides because of its excellent biocidal properties [8,9]. Silver nanoparticles have been successfully introduced into various membrane materials such as polysulfone [2,10,11], polyethersulfone [12–15], polyvinylidene fluoride [16,17], polyamide [18,19] and chitosan [20–22]. The addition of silver nanoparticles into the polymer membranes improved the membrane performance in terms of their flux and fouling resistance, attributing to an increase of hydrophilicity or change in membrane morphology. However, the chemically produced silver nanoparticles often have problems with particle stability. The beneficial effects of added particles are often limited by aggregation and poor compatibility with the polymeric matrix [6,7]. Besides the improved durability of silver-containing membrane and the simultaneously reduced potential risks of released silver ions at high load to the environment and filtration process are still challenges for excellent membrane performance. Recently, biosynthesis of silver nanoparticles as an environmentfriendly method has attracted increasing attention. In our previous

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

study, biogenic silver nanoparticles with the mean size of 11 nm (denoted as Bio-Ag0-11) was introduced into PES membranes (Bio-Ag0-11/PES). The attachment of the nanoparticles with the micro scale surface of the bacterium on which they were formed prevents them from aggregating [6,7]. However the large amount of existed biomass not only significantly decreased the pure silver content in the biogenic silver (150 mgAg g  1) but also decreased the compatibility with PES matrix as shown that many silvercontaining microstructures dispersed on the membrane surfaces. Besides the reduced compatibility of nanoparticles enhanced the silver leaching in the filtration test which may increase the potential risks to the environment. Herein, we improved a novel biogenic silver nanoparticles with an average size of only 6 nm (Bio-Ag0-6), extracted from the supernatant of Lactobacillus fermentum and no attachment to any whole bacteria cells. The fabrication method, the size and the morphology of nanoparticles were quite different from our previous report [23]. The modified synthesis largely improved the silver content from 150 mgAg g  1 to 450–500 mgAg g  1. The main objective of this study is to fabricate and characterize properties of the Bio-Ag0-6/PES mixed matrix membranes. The silver release from the nanocomposite membranes in both static immersion and dead-end filtration tests was assessed. Furthermore, we studied the effect of silver release on the filtration performance of the nanocomposite membranes. The antibacterial and antibiofouling performances were evaluated by the disk diffusion method, activated sludge immersion test as well as bacterial suspension filtration experiment.

2. Materials and methods 2.1. Materials Polyethersulfone (PES, E6020P, BASF Co., Germany) was dried at 120 1C overnight in a vacuum oven prior to dope preparation. N, N-Dimethylacetamide (DMAC) was obtained from Shanghai Jinshan Jingwei Chemical Co., Ltd. Silver standard solution (1000 mg L  1) was supplied by Sinopharm Chemical Reagent Co., Ltd. The standard constituents of the Luria-Bertani medium and agar plates used for bacteria incubation were purchased from Oxoid Co., Ltd. Trypsin (24 KDa), pepsin (35 KDa), egg albumin (45 KDa) and Bovine serum albumin (BSA, 69 KDa) was obtained from Solarbio Science & Technology Co., Ltd. The ultrapure water used in all experiments was supplied by a Milli-Q system (Millipore Corp., USA). 2.2. Synthesis and characterizeation of biogenic silver (Bio-Ag0-6) Biogenic silver (Bio-Ag0-6) was synthesized with Lactobacillus fermentum LMG 8900, provided by Laboratory of Microbial Ecology and Technology (LabMET), Ghent University in a modified method. The detailed preparation process and characterization of Bio-Ag0-6 have been reported in the previous paper [24]. 2.3. Preparation of composite membranes Different amounts of dry Bio-Ag0-6 were added to DMAC and ultrasonicated for 30 min for well dispersion. Then PES was dissolved in the solvent while stirring for 24 h at 60 1C until a homogeneous solution was obtained. The casting solution was degassed at 60 1C overnight without stirring to completely remove air bubbles. The solution was cast on a non-woven support with a casting knife gap setting of 200 mm. The fresh membrane was immediately immersed in the deionized water bath at room temperature to induce phase separation. After coagulation, the membranes were transferred to fresh distilled water to remove all

275

Table 1 The composition and the basic parameters of nanocomposite membranes. Membrane no.

Silver content (wt%)

CA (1)

porosity (%)

MWCO (KDa)

Water uptake (%)

M0 M1 M2 M3 M4

0 0.1 0.3 0.5 1.0

68.9 60.8 57.5 54.7 54.6

74.9 76.3 76.8 77.3 75.7

45 45 45 45 45

59.5 7 2.1 62.6 7 1.5 62.9 7 1.3 65.2 7 2.4 65.3 7 1.7

the residual solvent before characterization. Table 1 shows the composition of all the membranes prepared in the study. 2.4. Membrane characterization Morphological structures of the prepared membranes were examined using a scanning electron microscopy (FESEM) (HITACHI S-4800). For the cross-section observation, the membrane samples were frozen and fractured in liquid nitrogen. All the samples were sputtered by gold for observation. Presence of silver nanoparticles was confirmed by energy dispersive X-ray spectra (EDX) and elemental mapping (see Supporting information). The surfaces of membranes were scanned in a scan size of 5 μm  5 μm by atomic force microscopy (AFM) to determine the roughness of membranes. The surface roughness parameters which are expressed in terms of the mean roughness (Sa) and the root mean square roughness (Sq) were calculated from AFM images. The hydrophilicity of all the membrane surfaces was characterized by contact angle goniometer (DSA100, German KRUSS). At least five contact angles at different locations were recorded to get a reliable value. The membrane porosity ε(%) was defined as the ratio between the volume of the pores and the total volume of the membrane. For porous membranes, it could be determined by a gravimetric method, determining the weight of liquid contained in the membrane pores [27]:

εð%Þ ¼

ðW w W d Þ=Dw  100% ðW w  W d Þ=Dw þW d =Dp

ð1Þ

where ε is the porosity of membrane (%), W w is the wet sample weight (g), W d is the dry sample weight (g), Dw (0.998 g cm  3) and Dp (0.37 g cm  3)are the density of the water and polymer, respectively. Water uptake tests were conducted to evaluate the adsorption capability of water to membranes with biogenic silver (Bio-Ag0-6). Pieces of different membrane samples were immersed in Milli-Q water at room temperature for 24 h and the weight of wetted membrane W w was measured after dabbing it with a filter paper. The dry weight W d was determined after 24 h drying at 60 1C. The water uptake ratio was calculated with Eq. (2):   Ww Wd  100% ð2Þ U¼ Ww 2.5. Filtration performance of composite membranes A dead-end filtration cell (Model 8010, Millipore Corp. USA) was used to evaluate the filtration performance of membranes. The effective area of the membrane was 4.1 cm2. All the experiments were performed at room temperature (257 1 1C). To measure the pure water permeation, the membranes were initially compacted at 0.2 MPa for 30 min, and then the cumulative permeate weight was measured by an electronic balance (Sartorius BS224S, Germany) with Wedge software. The permeate

276

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

flux ðJÞ was calculated with Eq. (3) J¼

V A Δt

2.8. Antibiofouling experiment ð3Þ

where J (L m  2 h  1) is the membrane flux, V(L) was the volume of permeated water, A(m2) was the membrane area and Δt (h) was the permeation time. Molecular weight cut-off (MWCO) of the membrane was determined using the proteins with different molecular weights such as trypsin (24 kDa), pepsin (35 kDa), egg albumin (45 kDa) and BSA (69 kDa) [28]. The protein solutions were prepared in phosphate buffer saline (PBS) solution (0.1 M, 1 g L  1, pH 7.0). The rejection of protein (R) was calculated by the following equation:   Cp  100% ð4Þ R ¼ 1 Cf where C p and C f (g L  1) are the permeate concentration and the feed concentration, respectively. Proteins concentrations of both feed and permeate solutions were measured by a UV–vis spectrophotometer (DR5000 HACH) at 280 nm. The smallest molecular weight that is rejected by 90% is taken as the MWCO of the membranes.

2.6. Protein adsorption experiment For static protein adsorption experiment, a piece of membrane (d ¼35 mm) was immersed in 10 mL PBS solution containing BSA (1 g L  1) which was placed in a vial. After incubated for 12 h at room temperature, the membrane was rinsed slightly with PBS solution and then immersed in a washing solution (2% sodium dodecyl sulfate (SDS), 0.05 M NaOH) at 37 1C for 2 h with shaking to remove the adsorbed protein. The protein concentration of the washing solution was determined by using the Micro BCA™ Protein Assay Reagent Kit [11].

2.7. Silver release To study the static kinetics of silver release from the membranes, the composite membrane was cut into a rectangle shape with 1.0 cm  3.0 cm and immersed in 10 mL Milli-Q water at 37 1C. After a determined time, the membranes were transferred to 10 mL fresh Milli-Q water. The concentration of silver in the water was analyzed by inductively coupled plasma mass spectrometry (ICP-MS, Agilent, model 7500CX)with argon as carrier gas. To measure the silver content incorporated in the composite membrane, the membrane was digested by sonication in concentrated HNO3 for 3 days. After digestion, the suspension was filtered to remove large particles and analyzed by ICP-MS for total silver content [23]. Further, in order to observe the depletion behavior of silver from the membrane, filtration experiment was performed. Milli-Q water was filtrated with the composite membranes at 0.1 MPa for 18 h. The silver concentrations in the permeate samples were analyzed by ICP-MS. Besides, the effect of silver depletion on the change of two kinds of composite membrane (Bio-Ag0-6/PES and Bio-Ag0-11/PES) filtration performance was also studied. First, the pure water fluxes and the initial BSA rejection of the composite membranes were tested. Then the membranes were immersed in 2% HNO3 at 37 1C with shaking for 4 weeks to maximize the release of silver. After that, the pure water flux and BSA rejection were measured again. Based on the pure water flux before and after immersion, the rate of flux change can be calculated.

2.8.1. Disk diffusion method The antibacterial activities of the composite membrane were investigated by agar diffusion method as we reported earlier [29,30]. Gram-negative bacteria, Escherichia coli (ATCC15597) and Pseudomonas aeruginosa (ATCC27853) were used as test microorganisms. The tested bacterial were grown overnight and 100 mL of the suspension was spread onto LB agar plate. After sterilized by ultraviolet radiation for 30 min, membrane with a diameter of 25 mm was placed on the LB agar. After 24 h incubation at 37 1C, the inhibition zone around the membrane was observed as the indicator for the antibacterial activity.

2.8.2. Activated sludge immersion experiment The adhesion and the formation of biofilm on the membrane surface were evaluated by immersing the tested membranes in an activated sludge tank for up to 9 weeks. The activated sludge suspension used in this study was cultivated in a pilot scale membrane bioreactor (MBR) operated in our lab. A piece of membrane sample was cut from the examined membranes for SEM observation at different time intervals. The samples were fixed in 2.5% glutaraldehyde at 4 1C for 5 h, step dehydrations in ethanol for 15 min and finally dried in a vacuum oven at 80 1C over 24 h before SEM observation.

2.8.3. Bacterial suspension filtration Bacterial suspension filtration was conducted to further evaluate the antibiofouling performance of composite membranes. The filtration experiment was carried out using the dead-end filtration cell as described below. E. coli (ATCC15597) was grown overnight in LB broth at 37 1C and shaken at 180 rpm. The overnight culture was re-suspended in PBS (0.1 M, 1 g L  1, pH 7.0) and diluted to a concentration of 105 CFU mL  1. Pure water ðJ w1 Þ was firstly filtrated for 1 h and then E. coli cells were inoculated onto the membrane surface by filtration of 5 mL bacterial suspension. Following filtration of E. coli suspension, each membrane was incubated on LB agar plate for 24 h at 37 1C to maximize bacterial growth. Then the pure water was filtered through the cell to determine the membrane flux ðJ e Þ after 24 h biofilm had been formed on the composite membranes. The membrane was cleaned with sodium hypochlorite (NaOCl, 200 mg L  1) solution by shaking for 30 min. After chemical cleaning the pure water flux ðJ w2 Þ was measured [27]. In order to evaluate the fouling resistance of the membrane, the relative flux reduction (RFR) and the flux recovery ratio (FRR) were calculated as [10,13,14,23]: RFR ¼

  J w1  J e  100% J w1

ð5Þ

FRR ¼

  J w2  100% J w1

ð6Þ

The antibiofouling performance of composite membranes was confirmed by SEM after 24 h incubation. The viability of bacteria on the membrane surface after the biofouling experiment was determined using the LIVE/DEAD BacLight Bacterial Viability Kits (L13152; Molecular Probes, USA) stain and confocal laser scanning microscopy (CLSM).

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

3. Results and discussion 3.1. Characterization of Bio-Ag0-6/PES composite membranes 3.1.1. Morphologies of membranes The membrane surface and cross-section morphologies were observed by SEM. As shown in Fig. 1a, all of the pure PES membrane and the Bio-Ag0-6/PES composite membranes have similar flat smooth surfaces without obvious aggregation on membrane surfaces. The presence of silver nanoparticles can be confirmed by the EDS spectra and elemental mapping. From Fig. 1b we can see that the EDS peak that corresponds to silver became more significant with the increase of Bio-Ag0-6 content in the casting solutions. The silver mapping (Fig. S1) on the surface of composite membranes showed that Ag (the bright spots) was evenly distributed in the membrane surface, indicating that the Bio-Ag0-6 nanoparticles can be successfully introduced into the PES membrane by directly blending. This is in contrast with our previous report and other's reports, in which biogenic silver–containing microstructures or chemically produced silver nanoparticles seemed to aggregate easily on the membrane surface [15]. Fig. 1a shows that all the membranes exhibit the typical asymmetric structure, with a dense skin layer, a finger-like porous sub layer and sponge like bottom layer. Moreover, the pure PES membrane has a narrow finger-like structure with an obvious sponge structure at the bottom. By introducing Bio-Ag0-6 nanoparticles, the finger-like micro voids seemed to elongate across the whole thickness, and become wider close to the back side of membrane. Besides, the width of micro voids in the bottom layer became bigger and also there were more circular voids close to the back side of the membrane, which would enhance the permeation flux of the membrane. Generally, membrane structure is determined by the driving force and relative diffusion rate between solvent and non-solvent [31]. It is also believed that if the nanoparticles had higher affinity to non-solvent, elongated macrovoids and finger-like structure could appear due to the higher solvent/non-solvent exchange rate [32]. In this study, the hydrophilic functional groups around the Bio-Ag0-6 nanoparticles may facilitate the exchange rate of solvent and non-solvent. Meanwhile, the incorporation of Bio-Ag0-6 nanoparticles into casting solution decreased the interaction between polymer and solvent molecules, which led to an acceleration of solvent and non-solvent exchange during the demixing process. Hence, the macrovoids formed at the bottom of the composite membranes (M4). The cross-section morphologies changes were all caused by the accelerated demixing process and were in well agreement with other reports [2,33]. Moreover, no obvious aggregation of Bio-Ag0-6 nanoparticles was observed in the membrane cross-section micrographs, which indicated that the Bio-Ag0-6 was well dispersed within the membrane matrix. AFM analysis was done to determine the roughness of bare and composite membrane surfaces. AFM images and summary of the surface roughness values of membranes are presented in Fig. 2. The Bio-Ag0-6/PES composite membranes had relatively lower values in terms of the mean roughness (Sa) and root mean square roughness (Sq) than the pure PES membrane. These data indicated that the Bio-Ag0-6/PES composite membranes had a smoother surface than the PES membrane. As the concentration of Bio-Ag0-6 was increased, smoother surfaces appeared which was probably due to the changes in casting solution viscosity [33]. It was also known that the smaller particles had higher ability to fit into the matrix of polymer and could apparently make polymeric membranes surface smoother [10,34,35]. This behavior was also reported elsewhere in which smoother surface was formed by the introduction of inorganic nanoparticles into polymeric membranes

277

[36]. It had been known that surface roughness, which can affect the affinity between foulants and membrane surfaces, is an important factor in the extent of bacterial adhesion [21]. Thus, the improved smoothness of membrane surfaces may be expected to have a positive effect on the antibiofouling performance of the composite membranes. 3.1.2. Hydrophilicity, porosity and pure water uptake Water contact angle is usually used to characterize the membrane surface hydrophilicity. The surface water contact angles were measured as shown in Table 1. The pure PES membrane had the highest contact angle of 68.9 71.31. The contact angle of composite membrane decreases with the increase of Bio-Ag0-6 content in the mixed matrix membranes, which indicates the hydrophilicity is improved. This may attribute to the well dispersed Bio-Ag0-6 nanoparticles which contain a great deal of hydrophilic groups such as hydroxyl groups and amino groups, responsible for the hydrophilicity increase. It has been demonstrated that most bacteria are hydrophobic and hence tend to attach to hydrophobic surface [28]. Thus, the higher hydrophilicity of the composite membrane surface may provide another positive contribution to the anti-adhesion and anti-biofouling performance of the membranes. The effect of Bio-Ag0-6 addition on porosity of the prepared membranes is presented in Table 1. The existence of Bio-Ag0-6 nanoparticles causes a slightly increased porosity of the composite membranes. The reason may be the increase of more open structure with wider internal macro voids due to the addition of Bio-Ag0-6 nanoparticles, as observed from the cross-section images of the membranes [25,27,33]. The addition of nanoparticles does not have any significant effect on the surface pore size of the membrane and all the membranes have the same MWCO of 45 KDa. Incorporation of the Bio-Ag0-6 nanoparticles can change the water uptake ability of membranes. As shown in Table 1, the water uptake ratio increased with the increase of Bio-Ag0-6 loading. The increased surface hydropilicity and the higher porosity may account for the difference [21]. 3.2. Ultrafiltration performance of membranes Ultrafiltration experiments were conducted to study the permeability of PES membrane with different contents of Bio-Ag0-6 nanoparticles as shown in Fig. 3. The presence of Bio-Ag0-6 nanoparticles in the composite membranes effectively enhanced the pure water flux which could be explained by the increased hydrophilicity and water uptake ratio. The flux decreased dramatically when the solution was changed to BSA due to the membrane fouling. It was caused by the deposition and the adsorption of protein molecules onto the membrane surface or in the membrane pores [23]. The BSA solution flux of PES membrane is 54 L m  2 h  1. With the incorporation of Bio-Ag0-6 nanoparticles, the protein flux slightly increased to 62.5, 68.5, 75 and 85 L m  2 h  1 for M1, M2, M3 and M4, respectively. The BSA rejection of PES membrane was 95.7%, and 98.3, 98.6, 97.7, 97.9 for M1, M2, M3, M4, respectively. Therefore, it can be concluded that the addition of Bio-Ag0-6 nanoparticles can enhance the permeability of composite membranes and does not sacrifice the rejection capability of composite membranes. 3.3. Protein static adsorption The effect of Bio-Ag0-6 nanoparticles on membrane antifouling performance was also investigated by the BSA static adsorption test. Fig. 4 shows that the Bio-Ag0-6/PES composite membranes

278

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

0.7

S 0.6

0.5

M0 M2 M4

C

KCnt

0.4

O

Pt

0.3

0.2

Ag 0.1

0.0 0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Energy-Kev Fig. 1. The surface, cross-sectional SEM morphologies (a) and the energy dispersive X-ray spectra (EDX) (b) of M0, M2 and M4.

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

279

Membrane Sa (nm)

No.

Sq (nm)

Sz (nm)

M0

5.96

7.49

69.3

M2

5.27

6.75

69.8

M4

5.35

6.74

56.4

Fig. 2. The surface AFM images (a, b and c) and the surface roughness parameters (d) of M0, M2 and M4. Sa: The mean roughness; Sq: The root mean square roughness Sz: The mean difference between the five highest peaks and the lowest valleys.

100

400

60

350

150 80

-2

200

50

ug cm

90

250

-2

-1

Flux (L m h )

Purewater BSA

BSA Rejection (%)

55 300

45

40

100 50

35 70

0

M0

M1

M2

M3

M4

Fig. 3. The pure water flux, BSA (in 0.1 M, pH 7.0 PBS, 1 g L  1) flux and the BSA rejection of PES membranes with different Bio-Ag0-6 loadings.

30

M0

M1

M2

M4

M3 0

Fig. 4. The BSA adsorption of membranes with different Bio-Ag -6 contents.

3.4. Analysis of silver leaching exhibit more resistance against the protein adsorption. The adsorbed BSA amount to the PES membrane (M0) was about 57.3 μg cm  2, while those were lower for the blended membranes, especially for the M4 samples (46.6 μg cm  2). These results suggest that the incorporation of Bio-Ag0-6 nanoparticles benefits a lot for the anti-adsorption of BSA. It is believed that hydrophilic modification of membranes may be helpful to hinder protein adsorption because large amount of water could bind the surface and the inner of membranes and thus mitigate the protein adsorption [11].

Silver static release during 80 days test was analyzed by ICP-MS and the results are presented in Fig. 5a. Silver released gradually from all the composite membranes and with the increase of BioAg0-6 content in the membrane, the total released silver increased. The silver leaching mainly happened during the first 15 days with a relatively high released rate. After 20 days immersion, the release rate nearly decreased to zero for M1, M2 and M3. The ratio of remained silver to the total one in the composite membrane was 75%, 83.7%, 85.6% and 69% for M1, M2, M3 and

280

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

In addition, the released silver only occupied 2.4%, 3.5%, 2.3% and 3.7% of the total silver for M1, M2, M3 and M4, respectively. This suggested the application of Bio-Ag0-6/PES composite membranes is much safer than our previous reported Bio-Ag0-11/PES composite membranes and the release of silver as well as the antibacterial properties of the nanocomposite membranes may last for longer time. To investigate the influence of silver leaching to the membrane filtration property, pure water flux and selectivity of BSA before and after treatment by 2% HNO3 were measured. As shown in Fig. 6a, the water flux of all the composite membranes increased and for the Bio-Ag0-11/PES composite membranes, the influence was much more significant than the Bio-Ag0-6/PES composite membranes. With the increase of water flux, the BSA rejection of all composite membranes decreased. From Fig. 6b we can see that the initial BSA rejection of Bio-Ag0-11/PES and Bio-Ag0-6/PES composite membranes were around 93% and 98%, respectively. After treatment by 2% HNO3, the BSA rejection of Bio-Ag0-11/PES decreased to 85%. But the BSA rejection of Bio-Ag0-6/PES still remained as high as around 92%. The change tendency is consistent between water flux and BSA rejection. The release of silver from the membrane may form cavities in or on the membrane surfaces, change the pore structure and increase the water flux. The results suggested that the Bio-Ag0-6 nanoparticles may have a stronger binding strength with PES matrix and could be held tightly in the membrane which is consistent with the results of silver leaching during 18 h filtration. So it is expected that a more

2.5

M1 M2 M3 M4

1.5

-2

ug cm day

-1

2.0

1.0

0.5

0.0

0

20

40

60

80

Time (day)

M1 M2 M3 M4

20

15

10

+

Ag concentration (ppb)

25

5 6

Bio-Ag 11 Bio-Ag

60 0 2h

4h

6h

8h

10h

12h

14h

16h

18h

M4 respectively after 80 days tests, indicating that most of the silver remained inside the membranes. It can be concluded that stable and extended anti-biofouling performance of the composite Bio-Ag0-6/PES membranes were expected, due to the stable immobilized silver [37]. Fig. 5b shows the silver leaching from the Bio-Ag0-6/PES composite membranes during 18 h filtration. The silver concentration in the filtrate decreased as more water was filtrated. For all the test membranes, the leaching of silver has a positive relationship with the embedded content of Bio-Ag0-6 in composite membranes. For M4, the silver concentration in the permeate dropped from 24 ppb to 5 ppb after 18 h filtration which was far less than we reported earlier [11,35]. Thus it appears that the BioAg0-6 nanoparticles have a strong interaction with the PES matrix. It was believed that the loss of silver during filtration was mostly from the surface. The evidence from the SEM images (Fig. 1) revealed that the Bio-Ag0-6 nanoparticles are well dispersed in the composite membrane and there was no observable aggregation on the membrane surface even for M4. This is a great advantage over our previous report in which the Bio-Ag0-11 containing microstructures were dispersed on the membrane surface and also contributed to the much less silver leaching. According to the world health organization (WHO) guideline and the regulation of U.S. Environmental Protection Agency (USEPA), the highest silver concentration in drinking water is 100 ppb, which is 4 times higher than the highest silver concentration released from M4.

40

20

0 M1

M2

M3

M4

100

95

BSA Rejection (%)

Fig. 5. (a) The releasing rate of silver during the 80 days immersion test; (b) The concentration of released silver in the permeate filtered by the nanocomposite membranes.

(J1-J0) /J0

Time (h)

90

85

6

Bio-Ag -initial 11 Bio-Ag -initial

80

6

Bio-Ag -4weeks 11 Bio-Ag -4weeks

75 M1

M2

M3

M4

Fig. 6. The effect of silver depletion on the change of pure water flux (a), BSA selectivity (b).

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

stable and lasting filtration performance could be achieved by the Bio-Ag0-6/PES composite membranes. 3.5. Anti-biofouling performances of the Bio-Ag0-6/PES composite membranes 3.5.1. Inhibition zone method Fig. 7a and b shows the antibacterial results for the growth of P. aeruginosa and E.coli from the disk diffusion test. The pure PES membrane did not show any inhibition effect against the growth of P. aeruginosa or E.coli. All of the composite membranes showed clear inhibition zone surrounding the samples and with the increase of

281

Bio-Ag0-6 content, the inhibition zone became wider and clearer. This is possibly due to more biotoxic silver ions were released from the membranes to the LB medium. The results indicated that the Bio-Ag0-6/PES composite membranes showed great effectiveness in controlling the growth of P. aeruginosa and E.coli. The good antibacterial effect of the composite membranes is the major premise on which the anti-biofouling performance is based. 3.5.2. Activated sludge immersion test The anti-biofouling performances of the composite membranes were investigated by immersing the membrane samples in an activated sludge tank for up to 9 weeks. The membranes samples

Fig. 7. The antibacterial effect of nanocomposite membranes on (a) Pseudomonas aeruginosa (b) Escherichia coli observed in the disk diffusion test and the biofilm formation on the M0, M2, M4 membranes after immersed in an activated sludge tank for up to 9 weeks (c).

282

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

were taken at different times and observed by SEM as shown in Fig. 7c. The foulants in the activated sludge tank are complicated mixtures composed of biological floc, extracellular polymeric substances (EPS), organic or inorganic substances. The PES membrane (M0) was found to have attached a whole layer of foulants after 1 week immersion. Quite different from the M0, the Bio-Ag06/PES composite membranes (M2 and M4) had very few foulants on the surfaces. After 9 weeks immersion, very thick biofilms were found to cover the pure PES membrane surface and the surfaces of composite membranes (M2 and M4) were still relatively clean and free of biofilm attachment. 80 Flux after E.Coli growth Flux after chemical clean Flux decline

90

60

80 70

40

60 50

Flux decline, %

Normalized permeate flux, J/J0, %

100

20 40 30 0 M0

M1

M2

M3

M4

Fig. 8. The normalized permeate flux of membranes with different Bio-Ag0-6 contents after the biofouling was maximized on the LB agar plate and after chemical cleaning (bars) also the flux decline of all the membranes (points).

From the 9 weeks immersion test, it is clear that by introduction of Bio-Ag0-6 into the PES matrix, the anti-adhesion and antibiofilm formation properties of the composite membranes could be improved a lot. This behavior is consistent with the membrane hydrophilicity and roughness changes as shown in Table 1 and Fig. 2d. As discussed above, the membrane surface hydrophilicity and the surface smoothness were slightly enhanced by the incorporation of Bio-Ag0-6. Both of them may make it more difficult for bacteria to adhere to the membrane surfaces, also the silver nanoparticles in the composite membranes may kill the attached bacteria and prevent the reproduction of bacteria. 3.5.3. Biofouling experiment with E.coli and flux decline The biofouling experiments with the inoculation of E.coli were carried out using fabricated membranes. Fig. 8 shows the normalized water flux of all the samples at different stages. After 24 h incubation on LB agar plate, the pure water flux ðJ e Þ of all the membranes decreased. The flux of pure PES membrane declined almost to 33% of its original pure water flux. The flux decline was only 43%, 34%, 14% and 8% for the Bio-Ag0-6 modified M1, M2, M3 and M4, which was considerably lower than the unmodified PES membrane (66% flux decline). This can be attributed to the bacterial growth and biofilm formation on the membrane surface. From the SEM observation in Fig. 9, it is clear that only a few bacteria existed on the M2 surface, while the bacteria formed very thick biofilms and totally covered the PES membrane surface. The corresponding CLSM images in Fig. 9 demonstrate that most of the bacteria on the modified membrane (M2) are dead (appear in red). On the other hand, there were very thick biofilms consisted of clusters of live cells (appear in green) on the pure PES membranes.

Fig. 9. The SEM and CLSM images of the unmodified membrane (M0) and the modified membrane (M2) after the E. coli cells were inoculated onto the membrane surface and incubated on LB agar plate for 24 h in the biofouling experiments. Green and red colors indicate living and dead cells. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.).

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

Though almost the same numbers of bacteria cells were inoculated onto the membrane surfaces, the excellent bactericidal properties of Bio-Ag0-6 nanoparticles immobilized in the composite membranes can kill the attached cells and effectively prevent or slower the formation of biofilms. After the test of J e through the fouled membranes, the membranes were cleaned with NaOCl to remove any reversible fouling and then the pure water flux was measured again (denoted as J w2 ). Based on J w1 and J w2 , the flux recovery ratio (FRR) was calculated. It was obvious that the flux recovery ratio of Bio-Ag0-6 modified membranes were much higher than the bare PES membrane. The FRR value of M0 was only 56%, indicating that chemical cleaning was inefficient and that fouling was irreversible. While it reached to about 91% for M2, which means that membrane fouling was greatly alleviated by the addition of 0.3% of Bio-Ag0-6 nanoparticles. The higher the FRR value, the better the antifouling property of the membranes. From the results above, the much less flux decrease, the much higher the flux recovery. Also the SEM and CLSM images indicate that the surfaces of composite membranes are more biofouling resistant, that is, a better antibiofouling performance of the modified membranes.

4. Conclusions Bio-Ag0-6 nanoparticles were incorporated into the PES membranes by the phase inversion method. The SEM results indicated that they were well dispersed in the membrane and there was no visual aggregation on the surfaces. The addition of Bio-Ag0-6 nanoparticles elongated the finger-like micro voids in the sublayer and improved the interconnectivity of pores between the sub-layer and bottom layer. The membrane hydrophilicity, porosity, water uptake ratio and the smoothness were enhanced due to the introduction of Bio-Ag0-6 nanoparticles. Filtration performance results revealed that the modified membranes had much higher pure water flux and did not sacrifice the BSA rejection ability. The released silver from the Bio-Ag0-6/PES composite membranes were in the safe range and had less influence on the filtration performance compared to the Bio-Ag0-11/PES composite membranes, indicating the good stability of immobilized silver in the membranes. All the composite membranes showed excellent antibacterial and anti-biofouling performances, demonstrating that the Bio-Ag0-6 nanoparticles in the PES membranes could be an effective approach to reduce membrane biofouling.

Acknowledgments KSZ thanks Royal Academy of Engineering, UK for the Research Exchange with China/India scheme, RWF thanks Chinese Academy of Sciences for the Distinguished Visiting Professor Fellowship. The authors gratefully acknowledge the financial support by Xiamen Municipal Bureau of Science and Technology(3502Z20131159).The authors are also grateful to the reviewers for their helpful and insightful comments.

Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.memsci.2014.08.021. References [1] Y. Mansourpanah, E. Momeni Habili, Preparation and modification of thin film PA membranes with improved antifouling property using acrylic acid and UV irradiation, J. Membr. Sci. 430 (2013) 158–166.

283

[2] A. Mollahosseini, A. Rahimpour, M. Jahamshahi, M. Peyravi, M. Khavarpour, The effect of silver nanoparticle size on performance and antibacteriality of polysulfone ultrafiltration membrane, Desalination 306 (2012) 41–50. [3] Y. Liu, H.N. Liu, L. Cui, K.S. Zhang, The ratio of food-to-microorganism (F/M) on membrane fouling of anaerobic membrane bioreactors treating low-strength wastewater, Desalination 297 (2012) 97–103. [4] M. Yao, B. Ladewig, K. Zhang, Identification of the change of soluble microbial products on membrane fouling in membrane bioreactor (MBR), Desalination 278 (2011) 126–131. [5] Y. Liu, E. Rosenfield, M. Hu, B. Mi, Direct observation of bacterial deposition on and detachment from nanocomposite membranes embedded with silver nanoparticles, Water Res. 47 (2013) 2949–2958. [6] H. Wu, B. Tang, P. Wu, Preparation and characterization of anti-fouling β-cyclodextrin/polyester thin film nanofiltration composite membrane, J. Membr. Sci. 428 (2013) 301–308. [7] H. Wu, J. Mansouri, V. Chen, Silica nanoparticles as carriers of antifouling ligands for PVDF ultrafiltration membranes, J. Membr. Sci. 433 (2013) 135–151. [8] G.A. Sotiriou, S.E. Pratsinis, Antibacterial activity of nanosilver ions and particles, Environ. Sci. Technol. 44 (2010) 5649–5654. [9] C. Levard, E.M. Hotze, G.V. Lowry, G.E. Brown, Environmental transformations of silver nanoparticles: impact on stability and toxicity, Environ. Sci. Technol. 46 (2012) 6900–6914. [10] D.Y. Koseoglu-Imer, B. Kose, M. Altinbas, I. Koyuncu, The production of polysulfone (PS) membrane with silver nanoparticles (AgNP): physical properties, filtration performances, and biofouling resistances of membranes, J. Membr. Sci. 428 (2013) 620–628. [11] K. Zodrow, L.a. 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. [12] I. Sawada, R. Fachrul, T. Ito, Y. Ohmukai, T. Maruyama, H. Matsuyama, Development of a hydrophilic polymer membrane containing silver nanoparticles with both organic antifouling and antibacterial properties, J. Membr. Sci. 387-388 (2012) 1–6. [13] H. Basri, A.F. Ismail, M. Aziz, Polyethersulfone (PES)-silver composite UF membrane: effect of silver loading and PVP molecular weight on membrane morphology and antibacterial activity, Desalination 273 (2011) 72–80. [14] 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. [15] 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. [16] J.H. Li, X.S. Shao, Q. Zhou, M.Z. Li, Q.Q. Zhang, The double effects of silver nanoparticles on the PVDF membrane: surface hydrophilicity and antifouling performance, Appl. Surf. Sci. 265 (2013) 663–670. [17] B. De Gusseme, T. Hennebel, E. Christiaens, H. Saveyn, K. Verbeken, J.P. Fitts, N. Boon, W. Verstraete, Virus disinfection in water by biogenic silver immobilized in polyvinylidene fluoride membranes, Water Res. 45 (2011) 1856–1864. [18] J. Yin, Y. Yang, Z. Hu, B. Deng, Attachment of silver nanoparticles (AgNPs) onto thin-film composite (TFC) membranes through covalent bonding to reduce membrane biofouling, J. Membr. Sci. 441 (2013) 73–82. [19] S.Y. Lee, H.J. Kim, R. Patel, S.J. Im, J.H. Kim, B.R. Min, Silver nanoparticles immobilized on thin film composite polyamide membrane: characterization, nanofiltration, antifouling properties, Polym. Adv. Technol. 18 (2007) 562–568. [20] A. Regiel, S. Irusta, A. Kyziol, M. Arruebo, J. Santamaria, Preparation and characterization of chitosan-silver nanocomposite films and their antibacterial activity against Staphylococcus aureus, Nanotechnology 24 (2013). [21] E.-S. Kim, G. Hwang, M. Gamal El-Din, Y. Liu, Development of nanosilver and multi-walled carbon nanotubes thin-film nanocomposite membrane for enhanced water treatment, J. Membr. Sci. 394–395 (2012) 37–48. [22] X.Y. Zhu, R.B. Bai, K.H. Wee, C.K. Liu, S.L. Tang, Membrane surfaces immobilized with ionic or reduced silver and their anti-biofouling performances, J. Membr. Sci. 363 (2010) 278–286. [23] M. Zhang, K. Zhang, B. De Gusseme, W. Verstraete, Biogenic silver nanoparticles (bio-Ag0) decrease biofouling of bio-Ag0/PES nanocomposite membranes, Water Res. 46 (2012) 2077–2087. [24] M. Zhang, K. Zhang, B. De Gusseme, W. Verstraete, R. Field, The antibacterial and anti-biofouling performance of biogenic silver nanoparticles by Lactobacillus fermentum, Biofouling (2014) 1–11. [25] C. Dong, G. He, H. Li, R. Zhao, Y. Han, Y. Deng, Antifouling enhancement of poly (vinylidene fluoride) microfiltration membrane by adding Mg(OH)2 nanoparticles, J. Membr. Sci. 387-388 (2012) 40–47. [27] F. Zhang, W. Zhang, Y. Yu, B. Deng, J. Li, J. Jin, Sol–gel preparation of PAA-gPVDF/TiO2 nanocomposite hollow fiber membranes with extremely high water flux and improved antifouling property, J. Membr. Sci. 432 (2013) 25–32. [28] 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. Mater. Sci. 369 (2011) 258–266. [29] R. Malaisamy, D. Berry, D. Holder, L. Raskin, L. Lepak, K.L. Jones, Development of reactive thin film polymer brush membranes to prevent biofouling, J. Membr. Sci. 350 (2010) 361–370.

284

M. Zhang et al. / Journal of Membrane Science 471 (2014) 274–284

[30] Y. Zhang, X. Yu, S. Gong, C. Ye, Z. Fan, H. Lin, Antibiofilm activity of Bacillus pumilus SW9 against initial biofouling on microfiltration membranes, Appl. Microbiol. Biotechnol. (2013) 1–12. [31] W. Zhao, Y. Su, C. Li, Q. Shi, X. Ning, Z. Jiang, Fabrication of antifouling polyethersulfone ultrafiltration membranes using Pluronic F127 as both surface modifier and pore-forming agent, J. Membr. Sci. 318 (2008) 405–412. [32] V. Vatanpour, S.S. Madaeni, A.R. Khataee, E. Salehi, S. Zinadini, H.A. Monfared, TiO2 embedded mixed matrix PES nanocomposite membranes: influence of different sizes and types of nanoparticles on antifouling and performance, Desalination 292 (2012) 19–29. [33] 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.

[34] A. Rahimpour, M. Jahanshahi, B. Rajaeian, M. Rahimnejad, TiO2 entrapped nano-composite PVDF/SPES membranes: preparation, characterization, antifouling and antibacterial properties, Desalination 278 (2011) 343–353. [35] X. Zhu, L. Tang, K.-H. Wee, Y.-H. Zhao, R. Bai, Immobilization of silver in polypropylene membrane for anti-biofouling performance, Biofouling 27 (2011) 773–786. [36] J. María Arsuaga, A. Sotto, G. del Rosario, A. Martínez, S. Molina, S.B. Teli, J. de Abajo, Influence of the type, size, and distribution of metal oxide particles on the properties of nanocomposite ultrafiltration membranes, J. Membr. Sci. 428 (2013) 131–141. [37] X. Liu, S. Qi, Y. Li, L. Yang, B. Cao, C.Y. Tang, Synthesis and characterization of novel antibacterial silver nanocomposite nanofiltration and forward osmosis membranes based on layer-by-layer assembly, Water Res. 47 (2013) 3081–3092.