We present a new method for impregnation of silver nanoparticles (Ag NPs) at high loading on polyethersulfone (PES) membrane's external surface, simultaneously retaining native membrane's porosity – to achieve a high water permeate flux without biofouling. This was possible by PES membrane's surface modification with acrylic acid (AA), finally leading to AA-Ag-PES membrane. AA-Ag-PES had a high (9.04%) Ag-NP loading selectively on membrane surface, as discrete, smaller (mean size: 20 nm) nanoparticles (NPs). In nonfunctionalized Ag-PES, aggregated (mean size: 70 nm) NPs, with lower Ag loading (0.73 wt.%) was obtained, with NP being present both on membrane surface and inside pores. Consequently, AA-Ag-PES could maintain similar water permeability and porosity (10,153.05 Lm−2 h−1bar−1 and 69.98%, respectively), as in native PES (11,368.74 Lm−2 h−1bar−1 and 68.86%, respectively); whereas both parameters dropped significantly for Ag-PES (4,869.66 Lm−2 h−1bar−1 and 49.02%, respectively). AA-Ag-PES also showed least flux reduction (7.7%) due to its anti-biofouling property and high flux recovery after usage and cleaning, compared to native PES and Ag-PES membrane's much higher flux reduction (54.29% and 36.7%, respectively). Hence, discrete NP impregnation, avoiding pore blockage, is key for achieving high water flux and anti-biofouling properties (in AA-Ag-PES), compared to non-functionalized Ag-PES, due to aggregated Ag-NPs inside its pores.

  • Surface modification of PES help silver nanoparticle impregnation on the surface.

  • Higher silver loading (9.4 wt%) achieved through surface modification.

  • Water permeability decreased for surface unmodified silver loaded membrane.

  • Surface modified silver impregnated membrane shows better biofouling prevention.

  • Silver leaching is within limit by surface modification than without modification.

Graphical Abstract

Graphical Abstract
Graphical Abstract

Membrane technology has been playing a vital role in water purification for decades due to its high selectivity, compact footprint, feasibility of easy scale-up, as well as its environment friendly properties (Liu et al. 2015a). Polyethersulfone (PES), a synthetic polymer, is used widely for the preparation of ultrafiltration and microfiltration membranes, for the purpose of water treatment, because of its high thermal stability, excellent chemical resistance and good mechanical strength (McKeen 2010; Mukherjee & Bandyopadhyaya 2020). However, biofouling of membrane is an eventuality, resulting in a decrease in membrane permeate flux and hence membrane lifetime, also increase in operational cost in terms of cleaning as well as replacement (Haider et al. 2016).

Biofouling is initiated by the adhesion of either a single or a cluster of microorganisms on the membrane surface (Khan et al. 2016). After attaching on the membrane surface, bacteria starts to colonize rapidly, with the help of soluble nutrients present in the water stream or absorbed proteins or humic materials on the membrane surface, leading to the formation of a thick biofilm layer over time (Nguyen et al. 2012; Ng et al. 2013; Zhu et al. 2014). Moreover, microorganisms secrete extracellular polymeric substances (EPS), which help them to anchor strongly over the membrane surface, to propagate further colonization, resulting in pore blockage (Marshall 1985). The matrix of EPS also offers protection from the removal of the biofilm by a simple physical or chemical process and other disinfectants (Binahmed et al. 2018). Hence, prevention of membrane fouling is of serious consideration and an active research focus in membrane technology.

Conventional fouling prevention processes like hydraulic cleaning (Hilal et al. 2005), acid and alkali wash (Liu et al. 2001; Zondervan & Roffel 2007), use of disinfecting products including chlorine, hydrogen peroxide, etc., can remove 99.9% of microorganisms. Still, they cannot ensure the prevention of bacterial regrowth by the remaining 0.1% population, which can rapidly multiply and regenerate a biofilm (Liu et al. 2010). Besides, use of these chemical treatments damage the membrane's inherent structural properties (pore size, porosity, mechanical property) and its integrity (Liu et al. 2015b; Biswas & Bandyopadhyaya 2017). Therefore, an alternative strategy to reduce biofouling is by modifying the membrane surface with antibacterial agents to provide the anti-biofouling effect for a longer duration. Another major concern is that the PES membrane being hydrophobic, it is more susceptible to biofouling. Literature has shown that, converting hydrophobic PES membrane to a hydrophilic one, by adding polar functional groups, reduces biofouling to some extent (Kouwonou et al. 2008; Zhao et al. 2013). Most bacteria being hydrophobic, tend to adhere more on a hydrophobic surface of a filtering membrane (Zhang et al. 2014a). However, it is important to note that the mere addition of a hydrophilic coating on a membrane does not fully eliminate the process of bacterial fouling (Binahmed et al. 2018). Hence, adding another biocide on the hydrophilic membrane surface in order to prevent biofouling has become a major research focus.

Antimicrobial properties of nanoparticles create an opportunity to impregnate it on the existing polymeric membrane, to prevent bacterial growth during the disinfection process. Nanomaterials like silver (Ag-NP) (Ben-Sasson et al. 2014), copper (Cu-NP) (Ben-Sasson et al. 2014, 2016), TiO2 nanoparticles (Mansourpanah & Habili 2015), graphene oxide nanosheets (GO) (Perreault et al. 2016), etc. have been used for antimicrobial properties. In this regard, several methods of embedding nanoparticles on a polymer matrix have been described. Mixing of nanoparticles in the polymer solution during the casting of membranes (Zodrow et al. 2009; Zhang et al. 2012, 2014b) is one method for nanoparticle embedding. Another strategy is to add the nanoparticles onto a hydrophilic polymer solution, followed by coating or grafting of the solution onto the pre-fabricated membrane surface (Mauter et al. 2011; Nisola et al. 2012; De Faria et al. 2015). The third strategy is to make the nanoparticles selectively bind on the membrane surface through some functional groups like sulfonic acid (–SO3H) (Cao et al. 2010), amine (–NH2) (Haider et al. 2016), carboxylic acid (–COOH) (Díez et al. 2017) etc., by modifying the polymer, followed by preparation of the membrane. In all three methods, the antimicrobial effect of nanoparticles was shown to be potentiated by increase in hydrophilicity of the membrane, whether with or without surface modification (Zhao et al. 2013; Zhang et al. 2014a; Qi et al. 2019). However, it is important to emphasize that neither the mere presence of nanoparticles, nor the increase in membrane hydrophilicity, fully contributes to the complete prevention of biofouling. Rather, a more important concern here is the generation of nanoparticles with the appropriate size, and having a selective uniform distribution, being present only on the external surface of a hydrophilic membrane. Additionally, the modification should not affect the inherent property of the native membrane or block its pores, thereby sacrificing permeate water flux. Leaching of nanoparticles during the filtration process should also remain within the permissible limit of metal concentration in the treated water for this to be successful.

It has already been reported in literature that, acrylic acid increases the hydrophilicity of the membrane by adding polar functional groups (–COOH) (Kouwonou et al. 2008). In addition, it is well known that Ag-NP has a high affinity to polar functional groups (like –C=O, –COOH, –OH, –NH2) (Srinivasan et al. 2013). We have shown that alteration of surface of the pre-fabricated PES membrane with acrylic acid and selective binding of Ag-NP on the membrane external surface prevents membrane biofouling to a significant extent (Mukherjee & Bandyopadhyaya 2020). The effect of increasing Ag-NP loading on nanoparticle distribution and subsequent filtration performance was also illustrated. However, the role of surface modification to embed the Ag-NP selectively and uniformly on the membrane surface and its subsequent biofouling prevention was not emphasized. To investigate this, here in this study, we have conducted experiments, where commercially available hydrophilic PES membrane was impregnated with Ag-NP, either with or without surface modification (acrylic acid grafting) and was thereafter compared in terms of spatial location and Ag-NP distribution in both cases. Finally, these membranes were tested for performance quality in water disinfection.

Hydrophilic polyethersulfone membrane (Supor 200; Pall India) was used as a base membrane for the modification of external surface and later used for embedding of silver. Acrylic acid (procured from SRL) and UV lamp (Phillips, 250 watt) were used for functionalization of the membrane. Silver nanoparticles were synthesized from silver nitrate and trisodium citrate (purchased from Qualigens, India). Escherichia coli wild type strain (MG1655) collected from the Microbial Type Culture Collection and Gene Bank (MTCC, Chandigarh, India), was used as a model in all experiments on water disinfection and biofouling. E. coli cell culture was done in Lauria broth, obtained from Himedia, India. Sodium chloride (NaCl), potassium chloride (KCl), disodium hydrogen phosphate (Na2HPO4) and potassium dihydrogen phosphate (KH2PO4) were acquired from Merk, India, were used to make phosphate buffer saline (PBS). MiliQ water was used for all experiments.

UV assisted graft polymerization of PES membrane by acrylic acid

The surface of pre-fabricated, commercially available PES membrane was modified by graft polymerization method using acrylic acid in presence of UV radiation. The membrane was functionalized by following the previously reported protocol (Mukherjee & Bandyopadhyaya 2020). In this method, PES flat sheet membrane was immersed in a Petri dish containing 300 mM acrylic acid solution and exposed to UV radiation for 15 min at an intensity of 2.06 mW cm−2. The membrane was then removed, and washing was carried out with Milli-Q water for 3–4 h in a rotary shaker for removing acrylic acid which was in excess. After washing with water, the membrane was soaked for 30 min in 0.1 M NaOH solution followed by washing with ultrapure water for removing the unreacted NaOH. The membrane was then kept for air drying for further use. This modified PES membrane was termed as AA-PES membrane. Grafting density of AA-PES was calculated by gravimetric method (details of the procedure and data mentioned in supplementary material).

Embedding of silver nanoparticle on AA-PES and native PES membranes surface

Silver nanoparticles were synthesized on the PES membrane exterior surface using the UV radiation. It is reported that acrylic acid modification adds carboxyl acid functional groups on the membrane surface (Kouwonou et al. 2008), which would form carboxylate anion after activating it with NaOH solution. On subsequent addition of silver nitrate solution, Ag+ will bind with the –COO group present on the surface. Deposited Ag+ ions were further reduced to form silver nanoparticles. Here in this study, the silver ions were reduced by using trisodium citrate in the presence of UV radiation (UV lamp 365 nm wavelength and 8 W power) by the following reaction.
formula
formula

Both AA-PES and unmodified PES membranes were soaked in silver nitrate solution (with 10 mM concentration) in the dark for 1 h to adsorb the silver ions on the membrane surface. After that, equimolar concentration of trisodium citrate was put in to the solution and exposed to UV radiation in a UV chamber for 12 h. Membranes were then rinsed with ultrapurewater to wash out the unattached nanoparticles and finally kept in room temperature for drying. Silver loaded unmodified membrane and acrylic acid modified PES membrane were named Ag-PES and AA-Ag-PES, respectively.

Characterisation of native PES, AA-Ag-PES and Ag-PES membranes

Functional groups of native PES and AA-PES were detected by FTIR-ATR spectroscopy in our previous work (Mukherjee & Bandyopadhyaya 2020) to prove the addition of –COOH groups on the external side of the membrane, after functionalization using acrylic acid. Further, X-ray photoelectron spectroscopy (XPS) (Kratos Analytical (UK)) was used to verify the appearance of silver nanoparticles and its oxidation state. X-ray diffraction (XRD) (PANalytical, The Netherlands), FEG SEM (JEOL, JSM-7600F) of PES, Ag-PES, and AA-Ag-PES were done to check the Ag-Nps distribution on the external part of membrane. Membrane roughness was checked by atomic force microscopy (AFM) (Asylum Research, USA. MFP-3D BIO). Hardness of PES and AA-Ag-PES surfaces were determined using nanoindentation (TI- 900, Hysitron Inc., USA). Ten indents were made for both samples to measure sample hardness. Loading of silver on Ag-PES and AA-Ag-PES membranes were quantified by ICP-AES (SPECTRO Analytical Instruments, Germany).

Evaluation of anti-biofouling performance of native PES, AA-Ag-PES, and Ag-PES in a laboratory-designed membrane module with cell attachment experiments

Laboratory-designed membrane module setup (Mukherjee & Bandyopadhyaya 2020) was used to check the biofouling prevention performance of native PES, Ag-PES, and AA-Ag-PES. The membranes were compacted by passing Milli-Q water at 70 psi pressure to make sure that no structural change occurred in the membrane while doing the filtration experiments. E. coli cells were used for checking all anti-biofouling experiments and 104 CFU mL−1 bacterial cell concentration was considered as initial bacterial load for preparing the synthetic contaminated water. E. coli cells were dispersed in PBS buffer to maintain the cellular osmotic balance. E. coli containing water was passed over the membrane with three selected inlet flow rates (0.5, 1, 1.5 LPM) for 3 h. The permeate flux was checked at a definite time interval of 10 min. The E. coli count was checked at the permeate water by the plate count method. Bacterial concentration in permeate water was also counted. The larger dimension of the bacteria (0.8–1 μM in width and 1.5–2.5 μM in length), than the pore diameter of PES membrane (225 nm) (Mukherjee & Bandyopadhyaya 2020), for any membrane, presence of viable bacterial cell cells in the filtrate water should be zero unless there is leakage in the module.

After performing the filtration experiments with E. coli containing water, the membranes were backwashed with Milli-Q water for 15 min. The ultrapure water was run through the membrane after the hydraulic cleaning and the permeate flux of the ultrapure pure water was checked to calculate the flux recovery of the membrane after biofouling.

For checking the cell attachment and killing attached bacteria on the membranes, cryo-SEM and live dead assay were done. Contaminated water was filtered through the membrane for 3 h, following which, each membrane was dipped in sterile PBS buffer and sonicated for 5 min so that the attached cells can enter the solution. The cells were finally incubated with BacLight live dead assay kit (containing SYTO-9 and propidium iodide dye) and visualized under confocal microscope (Carl Zeiss, LSM 780). Here, live cells and dead cells were stained by SYTO-9 (green fluorescence in 488 nm laser) and PI (red fluorescence in 561 nm laser), respectively. To detect the biofilm layer, the membranes (PES and AA-Ag-Pes) were dipped in E. coli containing water for 48 h and then visualized under FEG-SEM using cryo mode (JEOL, Japan).

Silver release in the filtrate water

The leaching of silver from Ag-PES and AA-Ag-PES in permeate water were determined during the decontamination process with a cross-flow module. The quantity of silver was obtained by ICP-AES analysis and was plotted with time.

Characterisation of native PES, AA-Ag-PES and Ag-PES

Silver nanoparticle impregnation on membrane surface

XPS was done to detect the presence of silver nanoparticles and its electronic state on AA-Ag-PES surface. The spectra obtained from XPS analysis of the AA-Ag-PES membrane surface showed the existence of silver peak at its corresponding binding energy (Figure 1(a)). High-resolution XPS spectra of Ag (3d) doublet showed an asymmetrically shaped peak with two spin-orbit components at a binding energy of 368.6 eV (Ag3d5/2) and 374.6 eV (Ag3d3/2), which were separated by 6 eV. These two characteristics clearly indicated the existence of metallic Ag-NP on the modified membrane (Uznanski et al. 2017). In addition, deconvolution of the spectra showed other two-component peaks at 367.56 eV (Ag3d5/2) and 373.5 e V (Ag3d3/2), indicating the presence of Ag+ state (Liu et al. 2017).

Figure 1

Nature of silver nanoparticles: (a) XPS spectra of acrylic acid modified silver-impregnated PES (AA-Ag-PES), (b) XRD of AA-Ag-PES.

Figure 1

Nature of silver nanoparticles: (a) XPS spectra of acrylic acid modified silver-impregnated PES (AA-Ag-PES), (b) XRD of AA-Ag-PES.

Figure 4(b) represents the characteristic XRD pattern of AA-Ag-PES. Four peaks were obtained for the 2θ range of 20–80 degrees, which indicates the crystalline nature of Ag-NP. The peaks at 2θ position of 38.1, 44.2, 64.4 and 77.4 corresponds to (111), (200), (220) and (311) planes of Ag (JCPDS No. 04–0783 ‘Joint Committee on Powder Diffraction Standards’) (Haider et al. 2016; Ahmad et al. 2019), respectively. Hence AA-PES composite is composed of Ag-NP.

Location and morphology of Ag-NP in AA-Ag-PES and Ag-PES

The appearance of Ag-NP on the external membrane surface of AA-Ag-PES was captured by using FEG-SEM (Figure 2(a)). Well dispersed spherical Ag-NPs were observed on the exterior part of AA-PES membrane. Further, size distribution of these particles was done by measuring the size of 200 particles from different images by the ‘Image J’ software. A 20 nm mean particle diameter with narrow size distribution was observed, which also indicates the monodisperse Ag-NPs exists on the exterior part of AA-PES membrane (Figure 2(b)).

Figure 2

Location and size distribution of silver nanoparticles: FEG-SEM image of (a) acrylic acid modified silver nanoparticle impregnated PES membrane (AA-Ag-PES) showing discrete nanoparticles selectively on the membrane surface and (b) its corresponding size distribution (200 particles were measured for size distribution), (c) unmodified silver nanoparticle impregnated PES membrane (Ag-PES) showing agglomerated nanoparticles both on the surface and pores of membrane (d) size distribution of silver nanoparticles on Ag-PES membrane.

Figure 2

Location and size distribution of silver nanoparticles: FEG-SEM image of (a) acrylic acid modified silver nanoparticle impregnated PES membrane (AA-Ag-PES) showing discrete nanoparticles selectively on the membrane surface and (b) its corresponding size distribution (200 particles were measured for size distribution), (c) unmodified silver nanoparticle impregnated PES membrane (Ag-PES) showing agglomerated nanoparticles both on the surface and pores of membrane (d) size distribution of silver nanoparticles on Ag-PES membrane.

However, in contrast, Ag-PES membrane was covered by aggregated Ag-NPs, which were found to be located, both on the surface and inside the PES membrane pores (Figure 2(c)). A broad size distribution with a mean particle diameter of 70 nm was obtained from FEG-SEM images by the above-mentioned method, indicating the presence of polydisperse particles (Figure 2(d)). This observation suggests that alteration of PES membrane surface improved the selective impregnation and formation of monodisperse Ag-NP on PES membrane surface, compared to Ag-PES.

To describe the location of silver nanoparticles, the cross-section images and EDX mapping of both AA-Ag-PES and Ag-PES are given in the supplementary information (Fig. S3).

Loading of Ag-NP on AA-Ag-PES and Ag-PES

Quantity of silver loaded on A-Ag-PES and Ag-PES was estimated by ICP-AES. Loading of both the samples is listed in Table 1. It can be concluded from the value that, functionalization with acrylic acid improves the loading of silver in AA-Ag-PES by an order of magnitude, compared to the silver loading in unmodified PES membrane (Ag-PES).

Table 1

Loading of silver on AA-Ag-PES and Ag-PES

SamplesSilver loading (wt%)
AA-Ag-PES 9.04 
Ag-PES 0.73 
SamplesSilver loading (wt%)
AA-Ag-PES 9.04 
Ag-PES 0.73 

Mechanical property of AA-Ag-PES and Ag-PES

Stress vs strain curve was obtained from the UTM analysis of AA-Ag-PES and Ag-PES and subsequently modulus of elasticity and tensile strength was extracted from the plot and compared with PES as shown in Figure 3(a)3(c). After addition of Ag-NP, there was not much variation in Young's modulus for either PES or AA-Ag-PES, but a slight decrease was observed for Ag-PES (Figure 3(b)). Similarly, a slight decrease in tensile strength was observed for Ag-PES, in comparison with AA-Ag-PES and the original PES membrane. This result signifies that mechanical properties were not compromised after Ag-NP addition, followed by surface modification with acrylic acid. However, a slight drop of mechanical properties in case of Ag-PES may be attributed to the agglomeration and nonuniform distribution of Ag-NP, both on the surface and inside the pores.

Figure 3

Mechanical properties of membrane before and after Ag-NP addition: (a) stress-strain curve of the membranes for the analysis of the mechanical property, (b), (c) Young's modulus and tensile strength of native PES, acrylic acid modified silver loaded membrane (AA-Ag-PES) and surface unmodified silver loaded membrane (Ag-PES) respectively (d) hardness measurements by nanoindentation.

Figure 3

Mechanical properties of membrane before and after Ag-NP addition: (a) stress-strain curve of the membranes for the analysis of the mechanical property, (b), (c) Young's modulus and tensile strength of native PES, acrylic acid modified silver loaded membrane (AA-Ag-PES) and surface unmodified silver loaded membrane (Ag-PES) respectively (d) hardness measurements by nanoindentation.

The hardness of the membrane is also an important parameter to be considered, which denotes the ability to withstand transmembrane pressure during membrane operation. Results from Figure 3(d) show that the hardness of AA-Ag-PES was higher (14.06 Mpa), than that of the native PES membrane (6.87 Mpa). This data indicated that incorporation of Ag-NP on exterior surface of the membrane made the membrane mechanically more stable in terms of sustaining higher transmembrane pressure during the filtration process.

Effect of Ag-NP addition on membrane morphology

Roughness of membrane is an important factor that needs to be considered in the context of bacterial adhesion on the membrane surface. Three-dimensional surface morphology of PES, AA-Ag-PES, and Ag-PES, obtained from AFM, as shown in Figure 4(a)4(c), respectively. Table 2 represents quantitative analysis of nano-scale surface roughness in terms of root mean square roughness (Rq) and average roughness (Ra).

Figure 4

Membrane surface characteristic before and after Ag-NP addition: two- and three-dimensional AFM images of membrane surface (scanning area 5 μm × 5 μm). (a) Native PES membrane, (b) acrylic acid modified silver loaded PES membrane (AA-Ag-PES), (c) surface unmodified silver loaded PES membrane (Ag-PES).

Figure 4

Membrane surface characteristic before and after Ag-NP addition: two- and three-dimensional AFM images of membrane surface (scanning area 5 μm × 5 μm). (a) Native PES membrane, (b) acrylic acid modified silver loaded PES membrane (AA-Ag-PES), (c) surface unmodified silver loaded PES membrane (Ag-PES).

Table 2

Surface roughness parameters of native PES, AA-Ag-PES and Ag-PES obtained from AFM

MembranesRoot mean square roughness (nm)Average roughness (nm)
Native PES 130 103.51 
AA-Ag-PES 120 88.16 
Ag-PES 248 200.80 
MembranesRoot mean square roughness (nm)Average roughness (nm)
Native PES 130 103.51 
AA-Ag-PES 120 88.16 
Ag-PES 248 200.80 

It was observed that the surface property was influenced after incorporation of Ag-NP in both cases of AA-Ag-PES and Ag-PES. The surface roughness (Rq: 120 nm) of AA-Ag-PES was lesser than the native PES (Rq: 130 nm). However, for Ag-PES, the roughness of the membrane (Rq: 248 nm) increased, compared to PES and AA-Ag-PES. Appearance of the smoother surface of AA-Ag-PES was probably because of the smaller size Ag-NP, which had a higher affinity to bind with the polymer matrix (Zhang et al. 2014a). In contrast, it may be speculated that, for Ag-PES, bigger sized agglomerated nanoparticles increased the membrane's surface roughness. It has been found in the literature that there is a positive correlation between the roughness of membrane surface and membrane fouling (Rahimpour et al. 2011; Zhu et al. 2011). Thus, decreased surface roughness in AA-Ag-PES compared to native PES and Ag-PES might have a favourable impact on its anti-biofouling performance.

Permeability and porosity measurements of native PES, AA-Ag-PES and Ag-PES

Permeability is an inherent quality of a membrane and should not be affected after embedding of Ag-NP. The impact of Ag-NP addition on both surface-modified PES and unmodified PES was evaluated in the membrane module set-up and compared with the native PES and AA-PES membrane. Figure 5(a) illustrates that pure water permeability of native PES, AA-PES and AA-Ag-PES membranes are comparable, which indicates that permeability of the membrane is not compromised after functionalization with acrylic acid. However, in case of Ag-PES, permeability dropped significantly. To support this fact, porosity was checked for all three membranes by gravimetry (detailed experimental method in supplementary material). This was observed that the porosity of PES and AA-Ag-PES were comparable, whereas the same was significantly lesser for Ag-PES membrane. Drop in porosity and permeability values of Ag-PES is mostly due to the agglomeration of silver nanoparticles, in absence of membrane modification, which increased the possibility of pore blockage.

Figure 5

Effect of membrane's intrinsic properties after Ag-NP addition: (a) permeability of ultrapure water of native PES, AA-PES, AA-Ag-PES and Ag-PES, (b) porosity of PES, AA-Ag-PES, and Ag-PES.

Figure 5

Effect of membrane's intrinsic properties after Ag-NP addition: (a) permeability of ultrapure water of native PES, AA-PES, AA-Ag-PES and Ag-PES, (b) porosity of PES, AA-Ag-PES, and Ag-PES.

Comparison of permeate flux and bacterial cell adhesion between AA-Ag-PES and Ag-PES

Disinfection performance of native PES, AA-Ag-PES and Ag-PES were tested in cross-flow membrane module set up. As stated previously, E. coli cell diameter being larger than the pores of PES membranes, the cells flow mainly on the surface of the membrane. However, in the case of native PES membrane, cells flowing on the membrane can cause biofouling with time, since they are not killed. In the case of membranes having silver nanoparticles, there is a possibility of bacteria getting killed as well, while they were in contact with the silver nanoparticle existing on the external membrane surface. The effect of biofouling would be less prominent if a higher number of bacteria gets killed, because of effective exposure with Ag-NP.

The effect of bio-fouling was demonstrated with a drop in permeate flux after achieving steady state with respect to initial flux, during the passage of bacterial contaminated water across the membrane. It was found that initial permeate flux of native PES and AA-Ag-PES were almost identical, but after passing of contaminated water (with cell conc. 104 CFU mL−1), there was a significant drop in flux at steady state in case of native PES (54.1%), signifying considerable bio-fouling due to bacterial cell attachment (Figure 6(a)). Whereas, for AA-Ag-PES, due to the antibacterial property of Ag-NP, bacterial cells died and could not adhere to the membrane surface in reasonable amount, reducing the magnitude of bio-fouling and thus maintaining the steady state permeate flux almost identical to initial flux (least flux reduction of 7.7%) (Figure 6(a) and 6(b)). For Ag-PES membrane, the initial permeate flux was lesser due to pore blockage (as reflected in permeability and porosity data, Figure 5(a) and 5(b)), when compared to native PES and AA-Ag-PES membrane. The decrease in permeate flux of PES, AA-PES, AA-Ag-PES, and Ag-PES due to bacterial cell attachment was analysed in detail with respect to time (data shown in Fig. S1 in supplementary information).

Figure 6

Permeate flux analysis during disinfection of E. coli containing water: (a) initial permeate flux, steady state permeate flux after passing E. coli containing water and its corresponding permeate flux recovered after hydraulic cleaning for native PES, AA-Ag-PES, and Ag-PES membrane, (b) rate of permeate flux reduction and its flux recovery after biofouling followed by hydraulic cleaning, (c), (d) permeate flux at three inlet flow rates (0.5, 1, 1.5 LPM) with PES, and AA-Ag-PES, respectively.

Figure 6

Permeate flux analysis during disinfection of E. coli containing water: (a) initial permeate flux, steady state permeate flux after passing E. coli containing water and its corresponding permeate flux recovered after hydraulic cleaning for native PES, AA-Ag-PES, and Ag-PES membrane, (b) rate of permeate flux reduction and its flux recovery after biofouling followed by hydraulic cleaning, (c), (d) permeate flux at three inlet flow rates (0.5, 1, 1.5 LPM) with PES, and AA-Ag-PES, respectively.

However, percentage flux reduction during filtration process, was lesser for Ag-PES (36.8%) compared to native PES membrane (54.1%), due to some antibacterial effect of silver nanoparticles (Figure 6(b)). Nonetheless, the extremely low initial flux made this membrane unsuitable for the process of water decontamination. It was important to note that, bacterial concentration in permeate water was zero in case of all three membranes, as explained in the experimental section (results of cell concentration in permeate water depicted in supplementary section; Fig S2).

Subsequently flux recovery was checked for all three membranes. The flux recovery was lowest (85.63%) in case of PES membrane. Comparatively, there was an improvement of flux recovery up to 89.7% for Ag-PES, which may be because of antibiofouling activity of Ag-Np (Figure 6(b)). However, the permeate flux recovery was highest (98.9%) with AA-Ag-PES, which might be attributed to its strong antibiofouling activity, where pores were not clogged by the bacterial cells. As biofouling is a surface phenomenon, a reversible or irreversible layer of cells is formed on the membrane's active surface during the decontamination process, depending on the strength of the bacterial attachments (Ahmad et al. 2019). Formation of reversible biofouling can be removed easily by simple hydraulic washing. In AA-Ag-PES, a well-established exposure between the Ag-NP and bacteria (existence of of Ag-NP on the surface part of membrane) ensured no live bacterial attachments on the surface and least decline of permeate flux as well as highest flux recovery with simple backwashing.

Lastly, the result of variation of inlet flow rates was tested on flux reduction. Here decontamination performance was examined at three inlet flow rates (0.5, 1, and 1.5 LPM) for both PES and AA-Ag-PES membranes (Table 3; Figure 6(c) and 6(d)). It was found that flux reduction was less prominent at lower flow rates and increased with increasing inlet flow rate. A similar trend was found in case of both the membranes. This phenomenon could be explained by the fact that, at a higher inlet flow rate, the total bacterial load also increased, which led to a higher amount of fouling. The effect of different flow rates was not tested for Ag-PES membrane as the initial permeate flux (at inlet flow rate 0.5 LPM as shown in Figure 6(a)) and permeability (Figure 5(a)) of Ag-PES was significantly lower compared to PES and AA-Ag-PES, which made it unsuitable for membrane operations. Moreover, due to agglomeration of Ag-NP (shown in Figure 2(c)) and pore blockage (shown in change in porosity data Figure 5(b)), surface transport resistance increased (McCloskey et al. 2010) during water flow and pressure drop also increased at higher flow rates in case of Ag-PES.

Table 3

Changes in permeate flux with variation of inlet flow for native PES, and AA-Ag-PES

Sample nameInlet flow (LPM)
0.5
1
1.5
Initial flux Lm−2 h−1Steady state flux Lm−2 h−1Flux reduction (%)Initial flux Lm−2 h−1Steady state flux Lm−2 h−1Flux reduction (%)Initial flux Lm−2 h−1Steady state flux Lm−2 h−1Flux reduction (%)
PES 884 404 54.29 1,266 528 58.29 2,606 973.2 62.66 
AA-Ag-PES 872 804 7.79 1,338 1,188 11.21 2,190 1,192 12.69 
Sample nameInlet flow (LPM)
0.5
1
1.5
Initial flux Lm−2 h−1Steady state flux Lm−2 h−1Flux reduction (%)Initial flux Lm−2 h−1Steady state flux Lm−2 h−1Flux reduction (%)Initial flux Lm−2 h−1Steady state flux Lm−2 h−1Flux reduction (%)
PES 884 404 54.29 1,266 528 58.29 2,606 973.2 62.66 
AA-Ag-PES 872 804 7.79 1,338 1,188 11.21 2,190 1,192 12.69 

Effect of silver impregnation on antibacterial activity

Confocal laser scanning microscopy (CLSM) was performed to check the E. coli cell attachment on native and AA-Ag-PES membrane surfaces shown in Figure 7. BacLight cell viability kit was used to differentiate the viability of the bacteria, which is adhered on the membrane surface. BacLight is a mixture of SYTO9 (shows green fluorescence) and propidium iodide (shows red fluorescence) dyes. SYTO 9 labels those cells, which are having both intact and damaged cell membranes, but PI can only penetrate into cells having damaged cell membranes (Molecular Probes 2004). Likewise, this cell viability kit differentiates between live and dead cells. After 48 h of incubation, intense green fluorescence and negligible red fluorescence (Figure 7(a) and 7(b)) were observed, indicating the presence of all live adhered cells on the native PES membrane. In contrast, a strong red signal from PI and no signal from SYTO 9 (Figure 7(c) and 7(d)) signify that all cells adhered to the AA-Ag-PES were dead, after similar incubation time. These images demonstrate that the incorporation of silver nanoparticles on the PES membrane surface kills the bacteria and prevents biofilm formation.

Figure 7

Cell viability and biofilm formation with native PES and AA-Ag-PES: CLSM image of (a), (b) native PES membrane, and (c), (d) AA-Ag-PES membrane. Cryo-SEM images of PES membrane (e) surface, (f) cross-section and AA-Ag-PES, (g) surface, (h) cross-section.

Figure 7

Cell viability and biofilm formation with native PES and AA-Ag-PES: CLSM image of (a), (b) native PES membrane, and (c), (d) AA-Ag-PES membrane. Cryo-SEM images of PES membrane (e) surface, (f) cross-section and AA-Ag-PES, (g) surface, (h) cross-section.

Cryo-SEM images reconfirm the formation of biofilm by the adhered E. coli bacteria on the native PES membrane surface after 48 h of incubation. Additionally, cross-sectional view also showed the layer of biofilm on the upper membrane surface (Figure 7(e) and 7(f)), which clogged the membrane pores and reduced the permeate flux over time. To understand the biofilm formation more clearly, Cryo SEM of PES and AA-Ag-PES and its cross-section were done before biofouling (at time t = 0) and is shown in supplementary information (Fig. S4). In comparison, the attached cells were found to be dead due to the presence of Ag-NP over AA-Ag-PES membrane. However, cross-sectional views of the same sample did not show any biolayer formed by E. coli (Figure 7(g) and 7(h)). Hence, this demonstrates that, silver nanoparticles on the external side of the membrane effectively prevents biofilm formation and maintains a high permeate flux in contrast to the native membrane, for a longer duration.

Release of silver in the permeate water

The concentration of silver in permeate water was checked during the decontamination process (in membrane module set-up), shown in Figure 8. Results demonstrate that, silver release is higher in case of Ag-PES, over the permissible limit (100 ppb) set by World Health Organization (WHO) and USEPA (World Health Organization 2011; U.S. EPA 2020). However, in AA-Ag-PES, the silver concentration in the filtrate remained within the safety limit, throughout the experiment, and reached up to 29.8 ppb at steady state, after passing of 120 L of water in 240 min, with an inlet flow rate of 1.5 LPM.

Figure 8

Release of silver in permeate stream: comparison of silver release between acrylic acid modified silver nanoparticles impregnated PES membrane (AA-Ag-PES) and surface unmodified silver loaded PES membrane (Ag-PES).

Figure 8

Release of silver in permeate stream: comparison of silver release between acrylic acid modified silver nanoparticles impregnated PES membrane (AA-Ag-PES) and surface unmodified silver loaded PES membrane (Ag-PES).

It has been found in literature that carbonyl groups bound strongly with silver nanoparticles (AbdelRahim et al. 2017). Hence, impregnation of Ag-NP via carbonyl groups attached on the membrane surface through acrylic acid treatment helped binding of silver strongly with the membrane surface, maintaining concentration of silver in the filtrate within the permissible limit. This possibility was further confirmed by the higher silver release in Ag-PES, in which the binding of Ag-NP to polymer matrix might not be strong enough to sustain the Ag-NP in the membrane at a high inlet flow. Therefore, alteration of the surface of the membrane helped to reduce the silver release, even when there was high silver loading.

Table 4 shows the comparison of biofouling prevention performance between our present study and different nanocomposite membranes reported in the existing literature. Permeability and permeate flux are the two key parameters for successful membrane operation. Permeability of the membrane should not be compromised after any modification and higher permeate flux at steady state indicates effective biofouling prevention. Therefore, these two parameters are considered here, so as to compare our work with the previously reported work. In our earlier work (Mukherjee & Bandyopadhyaya 2020), we had shown higher permeability and permeate flux at steady state, than that of earlier literature reports (Zhang et al. 2012; Dolina et al. 2015) mentioned in Table 1. The lower permeability and permeate in their work was due to the presence of Ag-NP inside the pores of their membrane. Placement of Ag-NP inside the membrane pores after silver addition, blocked the membrane pores, resulting in lower permeability and lower permeate flux, thereby having least biofouling prevention efficiency.

Table 4

Comparison of anti-biofouling performances of different Ag-NP impregnated PES membrane with the present study

Nanocomposite membraneSurface modificationParticle size (nm)Nanoparticle loading (wt%)Permeability (L m−2 h−1 bar−1)Permeate flux at steady state after biofoulingReferences
PEI-Ag-PES No – 2.9 648 – Dolina et al. (2015)  
Bio-Ag-PES No wet phase inversion 11.2 – 275 Zhang et al. (2012)  
AA-PES Acrylic acid No Ag-NP – – 11 Kouwonou et al. (2008)  
AA-Ag-PES AA modification 20 9.04 10,153 1,913 Our previous work and present work 
Ag-PES No surface modification 70 0.78 4,869 192 Our present work 
Nanocomposite membraneSurface modificationParticle size (nm)Nanoparticle loading (wt%)Permeability (L m−2 h−1 bar−1)Permeate flux at steady state after biofoulingReferences
PEI-Ag-PES No – 2.9 648 – Dolina et al. (2015)  
Bio-Ag-PES No wet phase inversion 11.2 – 275 Zhang et al. (2012)  
AA-PES Acrylic acid No Ag-NP – – 11 Kouwonou et al. (2008)  
AA-Ag-PES AA modification 20 9.04 10,153 1,913 Our previous work and present work 
Ag-PES No surface modification 70 0.78 4,869 192 Our present work 

In our earlier work (Mukherjee & Bandyopadhyaya 2020), we had modified the surface of PES membrane with acrylic acid and added silver nanoparticles on its external surface to make AA-Ag-PES, which showed highest permeability and antibiofouling performance among all earlier reported studies. Further, in our present work, in addition to that, silver is now added to PES membrane without any surface modification (Ag-PES), to compare the antibiofouling performance with AA-Ag-PES. In case of Ag-PES, lower permeability and permeate flux is observed than AA-Ag-PES, which is because of pore blockage and agglomerated particles in PES membrane. This result proves the hypothesis that, selective binding of Ag-NP is important to maintain higher permeability as well as establish bacteria to nanoparticle contact, resulting in higher antibiofouling.

Membrane biofouling is a major hindrance in operation, as it decreases the overall membrane lifetime. To overcome this, controlled Ag-NP impregnation was achieved to form well-separated, smaller sized NPs with high loading, which can bind selectively on the external part of the membrane surface to mitigate the effect of biofouling. Ag-NP was put on hydrophilic PES membrane via carbonyl groups through acrylic acid functionalization in AA-Ag-PES membrane. Due to individual, discrete NPs being present on its surface (compared to aggregated NPs inside membrane pores, as in Ag-PES membrane), the former provides a better contact between bacteria and NPs, resulting in improved antibiofouling performance. This hypothesis was further confirmed by comparing the performance of biofouling prevention with and without surface engineered, Ag-NP loaded, PES membranes, namely AA-Ag-PES and Ag-PES, respectively. In Ag-PES, Ag-NP formed both on the external surface of membrane and inside the membrane pores in an aggregated manner, with significantly lower Ag-NP loading (0.73 wt%), compared to that in AA-Ag-PES. Aggregated Ag-NPs inside membrane pores of Ag-PES resulted in blockage of pores and decrease in nanoparticle to bacteria contact, leading to both decreased water permeability than the native PES membrane and also increased permeate flux reduction (indicating less efficient bacterial killing due to lower Ag-NP loading) compared to AA-Ag-PES. In comparison, in the engineered AA-Ag-PES membrane, increased silver loading (9.04 wt%) with a uniform distribution of nanoparticles all over the membrane surface without compromising the porosity and water permeability of the native PES membrane was achieved. As a result, with AA-Ag-PES, we could achieve higher biofouling prevention in terms of lesser permeate flux reduction (signifying enhanced bacterial killing) and higher recovery of permeate flux over Ag-PES and PES membranes. The concentration of silver in the filtrate stream was also maintained well-within a safe limit (29.8 ppb at steady state) which was much higher in Ag-PES (230 ppm at steady state).

Hence, the present study indicates the significance of uniform, selective embedding of individual NPs on the exterior part of membrane surface, via surface modification (despite having a hydrophilic membrane surface), to establish better bacteria to nanoparticle contact. It also demonstrates better anti-biofouling performance, preserving both mechanical strength of membrane and water permeability, simultaneously producing E. coli free permeate water, restricting silver concentration within permissible limit in the treated water.

We are thankful to the Metallurgical Engineering & Materials Science Department, IIT Bombay for their support in UTM analysis. We express our gratitude to the Sophisticated Analytical Instrument Facility (SAIF), and Industrial Research & Consultancy Centre (IRCC), IIT Bombay for supplying the facilities of sophisticated analytical instruments used for the experiments. RB acknowledges funding agency for their financial supports to carry out the research, acquired from Water Innovation Centre: Technology, Research and Education (WICTRE), IIT Bombay, set-up by Department of Science and Technology India, via award no. DST/TM/WTI/WIC/2K17/100(c), dated 5/12/2018.

All relevant data are included in the paper or its Supplementary Information.

AbdelRahim
K.
Mahmoud
S. Y.
Ali
A. M.
Almaary
K. S.
Mustafa
A. E. Z. M. A.
Husseiny
S. M.
2017
Extracellular biosynthesis of silver nanoparticles using Rhizopus stolonifer
.
Saudi Journal of Biological Sciences
24
(
1
),
208
216
.
Ben-Sasson
M.
Zodrow
K. R.
Genggeng
Q.
Kang
Y.
Giannelis
E. P.
Elimelech
M.
2014
Surface functionalization of thin-film composite membranes with copper nanoparticles for antimicrobial surface properties
.
Environmental Science and Technology
48
(
1
),
384
393
.
Ben-Sasson
M.
Lu
X.
Nejati
S.
Jaramillo
H.
Elimelech
M.
2016
In situ surface functionalization of reverse osmosis membranes with biocidal copper nanoparticles
.
Desalination
388
,
1
8
.
Binahmed
S.
Hasane
A.
Wang
Z.
Mansurov
A.
Romero-Vargas Castrillón
S.
2018
Bacterial adhesion to ultrafiltration membranes: role of hydrophilicity, natural organic matter, and cell-surface macromolecules
.
Environmental Science and Technology
52
(
1
),
162
172
.
Biswas
P.
Bandyopadhyaya
R.
2017
Biofouling prevention using silver nanoparticle impregnated polyethersulfone (PES) membrane: E. coli cell-killing in a continuous cross-flow membrane module
.
Journal of Colloid and Interface Science
491
,
13
26
.
http://dx.doi.org/10.1016/j.jcis.2016.11.060
.
Cao
X.
Tang
M.
Liu
F.
Nie
Y.
Zhao
C.
2010
Immobilization of silver nanoparticles onto sulfonated polyethersulfone membranes as antibacterial materials
.
Colloids and Surfaces B: Biointerfaces
81
(
2
),
555
562
.
De Faria
A. F.
Perreault
F.
Shaulsky
E.
Arias Chavez
L. H.
Elimelech
M.
2015
Antimicrobial electrospun biopolymer nanofiber mats functionalized with graphene oxide-silver nanocomposites
.
ACS Applied Materials and Interfaces
7
(
23
),
12751
12759
.
Available from: http://www.ncbi.nlm.nih.gov/pubmed/25980639 (accessed 16 March 2020)
.
Díez
B.
Roldán
N.
Martín
A.
Sotto
A.
Perdigón-Melón
J. A.
Arsuaga
J.
Rosal
R.
2017
Fouling and biofouling resistance of metal-doped mesostructured silica/polyethersulfone ultrafiltration membranes
.
Journal of Membrane Science
526
,
252
263
.
Available from: https://www.sciencedirect.com/science/article/pii/S0376738816315046 (accessed 5 September 2019)
.
Dolina
J.
Dlask
O.
Lederer
T.
Dvořák
L.
2015
Mitigation of membrane biofouling through surface modification with different forms of nanosilver
.
Chemical Engineering Journal
275
,
125
133
.
Haider
M. S.
Shao
G. N.
Imran
S. M.
Park
S. S.
Abbas
N.
Tahir
M. S.
Hussain
M.
Bae
W.
Kim
H. T.
2016
Aminated polyethersulfone-silver nanoparticles (AgNPs-APES) composite membranes with controlled silver ion release for antibacterial and water treatment applications
.
Materials Science and Engineering C
62
,
732
745
.
http://dx.doi.org/10.1016/j.msec.2016.02.025
.
Hilal
N.
Ogunbiyi
O. O.
Miles
N. J.
Nigmatullin
R.
2005
Methods employed for control of fouling in MF and UF membranes: a comprehensive review
.
Separation Science and Technology
40
(
10
),
1957
2005
.
Khan
S. J.
Hankins
N. P.
Shen
L. C.
2016
Submerged and attached growth membrane bioreactors and forward osmosis membrane bioreactors for wastewater treatment
. In:
Emerging Membrane Technology for Sustainable Water Treatment
(N. P. Hankins & R. Singh, eds).
Elsevier Inc.
,
Oxford, UK
, pp.
277
296
.
Kouwonou
Y.
Malaisamy
R.
Jones
K. L.
2008
Modification of PES membrane: reduction of biofouling and improved flux recovery
.
Separation Science and Technology
43
(
16
),
4099
4112
.
Liu
C.
Caothien
S.
Hayes
J.
Caothuy
T.
Otoyo
T.
Ogawa
T.
2001
Membrane Chemical Cleaning: From Art to Science
.
American Water Works Association
,
Port Washington, NY, USA
, p.
25
.
Liu
C. X.
Zhang
D. R.
He
Y.
Zhao
X. S.
Bai
R.
2010
Modification of membrane surface for anti-biofouling performance: effect of anti-adhesion and anti-bacteria approaches
.
Journal of Membrane Science
346
(
1
),
121
130
.
Liu
L.
Di
D. Y. W.
Park
H.
Son
M.
Hur
H. G.
Choi
H.
2015a
Improved antifouling performance of polyethersulfone (PES) membrane via surface modification by CNTs bound polyelectrolyte multilayers
.
RSC Advances
5
(
10
),
7340
7348
.
http://dx.doi.org/10.1039/C4RA14113 J
.
Liu
S.
Fang
F.
Wu
J.
Zhang
K.
2015b
The anti-biofouling properties of thin-film composite nanofiltration membranes grafted with biogenic silver nanoparticles
.
Desalination
375
(
November 2017
),
121
128
.
http://dx.doi.org/10.1016/j.desal.2015.08.007
.
Liu
S.
Yang
Q.
Zhu
J.
Zhu
D.
Wang
X.
2017
Preparation and characterization of Ag@AgCl-doped fly ash cenospheres for photocatalytic applications
.
Research on Chemical Intermediates
43
(
2
),
703
716
.
Mansourpanah
Y.
Habili
E. M.
2015
Investigation and characterization of TiO2-TFC nanocomposite membranes; membrane preparation and UV studies
.
Journal of Membrane Science and Research
1
(
1
),
26
33
.
Marshall
K. C.
1985
Mechanisms of bacterial adhesion at solid-water interfaces
. In:
Bacterial Adhesion
(D. C. Savage & M. Fletcher, eds).
Springer
,
Boston, MA, USA
, pp.
133
161
.
Mauter
M. S.
Wang
Y.
Okemgbo
K. C.
Osuji
C. O.
Giannelis
E. P.
Elimelech
M.
2011
Antifouling ultrafiltration membranes via post-fabrication grafting of biocidal nanomaterials
.
ACS Applied Materials and Interfaces
3
(
8
),
2861
2868
.
McKeen
L. W.
2010
Fatigue and tribological properties of plastics and elastomers
. In:
Fatigue and Tribological Properties of Plastics and Elastomers
(L. McKeen, ed.). 2nd edn.
William Andrew Publishing
,
Norwich, NY, USA
, pp.
265
286
.
Molecular Probes
I.
2004
LIVE/DEAD ® bacLight TM bacterial viability kits
.
Product Information
,
1
5
. .
Perreault
F.
Jaramillo
H.
Xie
M.
Ude
M.
Nghiem
L. D.
Elimelech
M.
2016
Biofouling mitigation in forward osmosis using graphene oxide functionalized thin-film composite membranes
.
Environmental Science and Technology
50
(
11
),
5840
5848
.
Available from: http://www.ncbi.nlm.nih.gov/pubmed/27160324 (accessed 16 March 2020)
.
Rahimpour
A.
Jahanshahi
M.
Rajaeian
B.
Rahimnejad
M.
2011
TiO2 entrapped nano-composite PVDF/SPES membranes: preparation, characterization, antifouling and antibacterial properties
.
Desalination
278
(
1–3
),
343
353
.
U.S. EPA
2020
Secondary Drinking Water Standards: Guidance for Nuisance Chemicals
.
Safe Drinking Water Act (SDWA), U.S. EPA
. .
Uznanski
P.
Zakrzewska
J.
Favier
F.
Kazmierski
S.
Bryszewska
E.
2017
Synthesis and characterization of silver nanoparticles from (bis)alkylamine silver carboxylate precursors
.
Journal of Nanoparticle Research
19
(
3
),
1
20
.
World Health Organization
2011
Guidelines for Drinking-Water Quality
, 4th edn. .
Zhang
M.
Zhang
K.
De Gusseme
B.
Verstraete
W.
2012
Biogenic silver nanoparticles (bio-Ag0) decrease biofouling of bio-Ag0/PES nanocomposite membranes
.
Water Research
46
(
7
),
2077
2087
.
Zhang
M.
Field
R. W.
Zhang
K.
2014a
Biogenic silver nanocomposite polyethersulfone UF membranes with antifouling properties
.
Journal of Membrane Science
471
,
274
284
.
http://dx.doi.org/10.1016/j.memsci.2014.08.021
.
Zhang
M.
Field
R. W.
Zhang
K.
2014b
Biogenic silver nanocomposite polyethersulfone UF membranes with antifouling properties
.
Journal of Membrane Science
471
,
274
284
.
Zhao
C.
Xue
J.
Ran
F.
Sun
S.
2013
Modification of polyethersulfone membranes – a review of methods
.
Progress in Materials Science
58
(
1
),
76
150
.
http://dx.doi.org/10.1016/j.pmatsci.2012.07.002 (accessed 16 March 2020)
.
Zhu
X.
Tang
L.
Wee
K. H.
Zhao
Y. H.
Bai
R.
2011
Immobilization of silver in polypropylene membrane for anti-biofouling performance
.
Biofouling
27
(
7
),
773
786
.
Available from: http://www.tandfonline.com/doi/abs/10.1080/08927014.2011.603830 (accessed 9 March 2020)
.
Zodrow
K.
Brunet
L.
Mahendra
S.
Li
D.
Zhang
A.
Li
Q.
Alvarez
P. J. J.
2009
Polysulfone ultrafiltration membranes impregnated with silver nanoparticles show improved biofouling resistance and virus removal
.
Water Research
43
(
3
),
715
723
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Supplementary data