This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright

com/locate/memsci A new outlook on membrane enhancement with nanoparticles: The alternative of ZnO Stefan Balta a. Romania c Department of Chemical and Energetic Technology. In and scaling may also occur. gold (Au). A the membranes properties.V. Inha University. Two methods are an economic or even technical point of view. It depends on use. University Dunarea de Jos. 1. but also to the synthesis procedure. and the overall cost of the TiO2 enhanced membranes. ∗ Corresponding author. such as the simultaneous filtration and photocatalytic activity that is aimed at. Republic of Korea a r t i c l e i n f o a b s t r a c t Article history: Although several studies explored the use of nanoparticles as additives in membrane structures. (Al2 O3 ). having unique physico-chemical properties that application of membranes is threatened by the critical formation differ from bulk materials.1016/j. Spain d Department of Environmental Engineering. Membrane fouling is the main problem that using nanoparticles. Leuven. Patricia Luis a .035–4 wt%). Arcadio Sotto a. contact angle measurement. compounds [8–10] and permeation properties [11–13] caused by TiO2 is a functional nanomaterial. many types of nanoparticles have been studied to improve ing is a dramatic decrease of the flow through the membrane. fouling and the ability to produce desired structures as well as Strong efforts are being done in several membrane applications functionalities [1–5]. Incheon.kr (J. addition.V. Belgium b Department of Environmental and Material Engineering. silica (SiO2 ). All rights reserved. The results showed an overall improvement compared to the neat membranes in terms of permeability as well as dye rejection and fouling resistance by adding ZnO nanoparticles even in small and ultralow concentrations.memsci. TiO2 nanoparticles [5]. alumina diminished membrane performance occurs together with exces. Polyethersulfone (PES) membranes manufactured by diffusion induced phase inver- Membranes Nanofiltration sion in N-methyl-pyrrolidone (NMP) using a range of procedures were blended with ZnO nanoparticles ZnO in a wide range of concentrations from ultralow to high (0. as most studies focus on. The Nanoparticles influence of the ZnO nanoparticles on the characteristics of PES/ZnO membranes was investigated with microscopic observations. carbon nanotubes. E-mail address: jeonghwankim@inha. Tel.elsevier. The main effect of membrane foul. 0376-7388/$ – see front matter © 2011 Elsevier B. advancing towards a good performance and limits the use of membranes in a wide range of applications from trying to understand the fouling phenomenon. All rights reserved. filtration experiments.U.ac. K. Rey Juan Carlos University. zerovalent iron (Fe0 ). Accepted 21 October 2011 Although ZnO has attractive features that potentially could fill the objectives of mixed matrix membranes Available online 25 October 2011 with lower cost and better performance. but also has disadvantages in the the higher hydrophobicity of the membrane surface. fouling resistance deter- mination and observation of the rejection of selected dyes.2011. membrane via phase inversion [19–25] and a second by dipping the brane surface. Author's personal copy Journal of Membrane Science 389 (2012) 155–161 Contents lists available at SciVerse ScienceDirect Journal of Membrane Science journal homepage: www. which is not only related to the intrinsic cost of the nanoparticles. It was shown that the new mem- Membrane synthesis brane materials embedded with ZnO nanoparticles have significantly improved membrane features. are of high interest in the manufacturing of fouling and the reality shows that their use in industrial applica- of membranes to achieve a high degree of control over membrane tions is extremely restricted if this problem is not solved. This paper investigates the synthesis of ZnO enhanced membranes and evaluates the performance of mixed matrix membranes with Keywords: ZnO nanoparticles. sive operating costs [6. mixed Received 16 April 2011 matrix membranes still suffer from difficulties in synthesis and applications. doi:10.27]. Bart Van der Bruggen a . Introduction the membrane characteristics [14–17] and on the filtration mode (cross-flow or dead-end filtration) [18] but it can be said that the Nanoparticles. although other types of fouling such as biofouling prepared membrane in a suspension with nanoparticles [26. Lidia Benea b . Kim). a new outlook Received in revised form 18 October 2011 on enhancement of membranes with nanoparticles is proposed by using ZnO as an alternative to TiO2 . the dispersion in the bulk of the polymer or in the top layer.7] and a decrease in rejection of the target palladium (Pd) and.025 . silver (Ag).∗ a Department of Chemical Engineering.c . The fouling phenomenon is persing the nanoparticles in the casting solution and prepare the often attributed to adsorption of organic compounds on the mem.10. Madrid. In this paper.: +82 32 860 7502. one by dis- offer a key to resolve this problem. challenges in development remain. Jeonghwan Kim d. Galati.b . Nanoparticles may reported in general to prepare composite membranes. zirconia (ZrO2 ). © 2011 Elsevier B.

Elcometer) scale make this alternative a potential system that can meet the in an atmosphere with controlled relative humidity on nonwoven demand of an efficient and lower-cost device. membrane sheet for eight membranes of each type were tested. The thoroughly washed membrane chased from Acros Organics (Belgium) were selected in order to was cut into the desired shape and fitted in the dead end device. Germany). Balta et al. The eventual membrane structure was also stud- ied in this work. 0. Compar- ison of the fouling-resistant ability of the manufactured neat and 2. in the pores of the support layer. 30 and 32 wt%) was used as the Jw PWP = (1) polymer matrix. 0.21 Da). The PWP was calculated were prepared using phase inversion induced by immersion by the following equation: precipitation. To visualize membrane surface characteristics. humic acid (HA) were purchased from Sigma–Aldrich (St. the support layer Furthermore. type Radel) supplied by Solvay (Belgium) a wide range of membranes.750. 0. membrane was measured until no further change was observed. Experimental rejection studies using dead end filtration experiments. Already in 2007 Franklin et al. After formation of [28]. excellent optical. Subsequently. 2 cation in photo-catalysis and anti-bacterial materials. the films were cast with 250 ␮m thickness volume ratio obtained when ZnO is used as particles in a nano-sized using a filmograph (K4340 Automatic Film Applicator. temperature of 25 ◦ C in dead-end mode with a Sterlitech HP4750 tion products from plant and animal residues [35]. [36]. 1. Cross-sections were prepared by The novelty of this paper is in the synthesis of ZnO enhanced fracturing the membranes in liquid nitrogen. SEM for the development of low-cost and fouling-prevention membrane images were made with a Philips XL30 FEG instrument with an technology.250. methylene blue flux was measured 30 s after the pressurization. were used: 0. A nitrogen cylinder coupled with the pressure regulator Six different dyes were used to explore the size interaction was connected to the top of vessel to pressurize the cell. Characterization of the membrane surface and ZnCl2 and observed that toxicity is attributable solely to dis- solved zinc. a new outlook on the use of nanoparticles in mem. pure water permeability (membrane hydraulic resistance) and dye 2.67 Da). through the use of zinc oxide (ZnO) volume of NMP for 3 h by mechanical stirring at 200 rpm and room nanoparticles.085. The bioavailability could be significantly greater than that of larger membrane was afterwards repeatedly washed with distilled water particles. rejection and 2. ZnO nanoparticle and 8% for PES and PES/ZnO membranes. 27.85 Da). polyester as support layer. a homogenous solution. 0. The support layer (Viledon Therefore. membranes in view to significantly enhance the performance of nanofiltration technology in terms of permeation. the slope of the linear regression of the water flux as a function of transmembrane pres- Neat PES membranes and ZnO-entrapped PES membranes sure was determined as the permeability. a series of 8 measurements for each of the different membrane Thus. 0. ZnO is one of the most important multifunctional temperature. Eleven different concentrations of ZnO nanoparticles semiconductor materials and exceptionally important for appli.07.500. the use of ZnO in the nano-size scale incorporated in mem. dissolving different amounts of nanoparticles in the corresponding brane structures is explored. [34] compared the toxicity in algae of nanoparticulate ZnO.54 Da). electrical. Prior to the casting. Also. The lower cost of ZnO and the increase of the surface-to. the obtained results are the average of 32 experimental FO2471) used for the PES membrane manufacturing was obtained values. Four membrane coupons of the same was employed as the base polymer. membrane area was 14. Permeate was col- (319. Author's personal copy 156 S. a stable system can be developed.77 Da). respectively. The ZnO-entrapped membranes were prepared by P . 99. For each polymer obtained [29–33]. 0375. Louis. / Journal of Membrane Science 389 (2012) 155–161 In this work. another issue that can limit the application of was wetted with NMP to prevent the polymer solution of intruding nanoparticles is their toxicity since it is thought that nanoparti. For ZnO nanoparticles. The initial water methyl red (269.6 cm2 . area are not related to toxicity. Filtration experiments fouling resistance. HA is known as a foulant in natural waters being fraction of iment at a transmembrane pressure of 10 bar and a constant humic substances obtained from chemical and biological degrada. 0. The selected dyes were volume of the appropriate solution was 250 mL.125.4.1.2. and stored wet. surfaces. five identical membrane sheets were made duce an increase in toxicity since the size distribution and surface and tested to obtain an average value of flux and solute rejection. keeping their physical properties associated to their size and the The average contact angle for distilled water was determined in chemical activity related to their availability in the membrane. PES cast from four different concentrations in N- methyl-pyrrolidone (NMP) (25. Sudan black (456. The pure water flux was determined from a compaction exper- MO). Stirred Cell. the pure water permeability (PWP) was determined by measuring the pure water flux (Jw ) at different trans- 2. due to its and 4 wt%. A water droplet was placed on a dry flat homogeneous of bulk ZnO. Materials blended membranes was explored by cross-flow experiments.035. The maximum experimental errors were less than 5% and from Freudenberg (Weinheim. Pure water permeability and dye rejection were determined for Polyethersulfone (PES.5%) was used as the polymer solvent. 1-methyl-2-pyrrolidone (NMP. to simple solubility of the compounds since A contact angle measuring system DSA 10 Mk2 (Krüss. The prepared membranes were characterized for water flux. since ZnO nanoparticles will be embedded in a membrane surface and the contact angle between the water and solid matrix (the membrane).10 Da). solution composition. Organic compounds pur. Victoria blue (506. The cover a large range of molecular mass. positive conclusions have been to remove the remaining solvent. neutral red (288. bulk ZnO 2. The use of nanoparticulate ZnO does not pro. Germany) ZnO nanoparticles aggregate in freshwater systems forming flocs was used to measure the water contact angle of the synthesized of even several microns with a saturation solubility similar to that membranes. accelerating voltage of 20 keV. lected in a graduated cylinder for a time interval until steady state Congo red (696.3. The prepared films were immersed cles may persist as small particles in aquatic systems and that their in a non-solvent bath (distilled water at 20 ◦ C) for precipitation. Preparation of membrane membrane pressures (P) from 2 to 14 bar. this is. scanning branes is a promising and novel system that may be the solution electron microscopy (SEM) measurements were performed. mechanical and chemical properties which was stirred for 24 h at 500 rpm and 40 ◦ C. the polymer was added to the solution. After compaction. The active in the interface solute-membrane pore.

To compare flux decline between different membranes. Balta et al. In order to study the effect of membrane fouling. and where Cp and Cf are the permeate and feed concentrations of dyes. 1. (B) 0. Regression factors sized PES membrane embedded with TiO2 nanoparticles from our (R2 ) obtained for calibrations within the experimental concentra. Four samples of permeate. Membrane characterization with SEM observations membrane. previous study [38]. SEM images of the surface of the synthesized membranes at different concentrations of ZnO (PES 27%) (A) Neat PES membrane.000. the membranes Membrane surfaces synthesized were observed with scanning were tested in a cross-flow filtration set-up [37] fed with 5 ppm of electron microscopy (SEM). mer concentrations.7 wt%. the MWCOs between control PES membranes tion range were above 0.7 wt% promoted the formation of clusters or aggregates of Jw ZnO on the membrane surface significantly. clusions cannot be drawn about the relative pore sizes between the control PES membrane and the PES/ZnO membrane from the 2. which is situated far in the turbulent region. Assuming that the pores of UF or NF membranes are in the same range as A Shimadzu UV-1601 double beam spectrophotometer was molecular weight cut-off (MWCO) measurements using synthe- used to determine the concentration of dyes. The influence of the addition of ZnO HA. Analytical methods SEM images. (C) 0.1. further increase of added nanoparti- RF = (2) cles to 0. were taken.   Cross-sectional observations of synthesized membranes were Cp R(%) = 1− × 100 (3) also made with SEM. The rejection R of the dissolved dyes and PES/ZnO membranes were compared using model dye com- was calculated as follows: pounds (results are discussed later). Author's personal copy S. Results and discussion 10 bar.5 m s−1 .5 wt%. 2 shows that the number of Cf macrovoids decreases with increasing polymer concentration. The fluxes were defined as the relation of the permeate flux to the pure SEM images of the PES membrane surface before addition of ZnO water flux of the respective membrane as follows: nanoparticles was similar to these of the same membranes after Jv embedding ZnO.99. However. This feed velocity corresponds to a Reynolds was improved by the addition of ZnO nanoparticles. 5 mL each. / Journal of Membrane Science 389 (2012) 155–161 157 Rejections were measured at a transmembrane pressure of 3.125 wt%. the sponge-like structure is more pronounced at higher poly- respectively. However. 1. . which velocity of 4. con- number of 30. relative nanoparticles to the membrane surface is shown in Fig. Concentration polarization at the membrane surface is min- imized by driven a Teflon coated magnetic stirring bar on top of the 3. and (D) 0. Fig. The membranes with ZnO nanoparticles in Fig. and more quantitative analysis is needed.5. The SEM images show Concentration polarization is minimized by using a cross-flow that pore-like structures on the membrane surface appeared.

40 ticles to water. 60 3. Similar results were observed at 70 higher concentrations of nanoparticles than proposed in this work Contact angle (º) (0. membranes (0. the addition of ZnO can increase the hydrophilicity 25 27 30 32 of membrane. the casting solutions showed that the separation layer (upper layer) became thinner than that of the control PES membrane. thus promot. As shown in 50 Fig. and the contact angle should be lower because a larger fraction of water Fig. indicating that these membranes are more hydrophobic.125% ZnO) [41]. . Hydrophilicity of membranes Figs. (C) 32% PES.2. an increasing polymer concentration increases the contact angle of the membrane surface. / Journal of Membrane Science 389 (2012) 155–161 Fig. 3.% ing macrovoid formation [39.125 wt% of ZnO) at different PES concentrations (A) 27% PES.40]. 2. Contact angles measured for neat membranes (no ZnO used) and PES/ZnO diffuses through the membrane structure. Balta et al. This fact can be explained by the presence of PES concentraon (%) hydrophilic ZnO nanoparticles in membrane structures. Cross-sectional SEM images of synthesized PES (without ZnO) and PES/ZnO (0.125 wt.125 wt% ZnO) at different PES concentrations. As a result of the high affinity of nanopar. (B) 30% PES. (E) 30% PES/ZnO. 3 and 4 show contact angle measurements for both dif- ferent polymer and ZnO nanoparticle concentrations. Furthermore. macrovoid formation of PES/ZnO membranes was more improved for higher polymer concentrations by adding 80 ZnO nanoparticles. 3. and (F) 32% PES/ZnO. (D) 27% PES/ZnO. Author's personal copy 158 S. PES/ZnO 0. The exchange between solvent (NMP) and non-solvent (water) is slower because of a hindrance effect of PES nanoparticles during the phase-inversion process.

4. Fig. 3).5 wt% ZnO). Increasing the nanoparticles concentration to 0. As shown in Fig. As shown in Fig. the observation of higher permeabilities for the higher ZnO concentra- tion can be explained by the more hydrophilic characteristic of the PES membrane due to the presence of ZnO nanoparticles. 4.035 wt%). The polymer con. 5. Comparison of permeability between PES and PES/ZnO at lower polymer concentrations. the adsorption of organic pollutants within membrane structure can be reduced.5 0. 3.4%.125 0.75 1 4 Fig. A concentration of 5 mg/L of humic acid was selected in this test. 5 shows the combined effect of the polymer and ZnO and 0. the maximum per- meability was found in the concentration interval between 0. suggesting that an optimum concentration of ZnO concentration on the membrane permeability.07 0.07 0. 4 is that even ilar increase of water fluxes (almost double) has been observed at ultra-low concentration (0.4. i.75 1 2 4 50 ZnO content (wt.375 0. The permeability increased inter-diffusion rate can be decreased. no further increase in the flux was observed. however.3 Fig.035 0.125 wt% increased the relative flux. an enhancement of 25%.3. / Journal of Membrane Science 389 (2012) 155–161 159 75 70 27 wt.% 50 2 -1 Contact angle (º) 65 40 30 60 20 55 10 0 0. Using ZnO nanoparticles in membrane structures. Balta et al. the water perme- ability increases and this effectiveness becomes more pronounced 3. 2) and also with the increase of hydrophilicity observed by contact angle measurements (Fig. In addition. In the literature [40]. the permeate flux. This result ZnO concentrations between 0. 7. In general.3 and 0. 6 also shows a sudden decrease of permeability at centration has a negative effect on the water permeate flux as a nanoparticle concentrations of 0. The results also indicate again that an optimum con- centration of ZnO nanoparticle exists. the permeabil- concentration as observed in Fig. was reported for a .e.035 0. Author's personal copy S.. is in agreement with the improvement of macrovoids formation described by SEM images (Fig.0850. the contact angle of a PES/ZnO membrane becomes smaller as the ZnO concentration is higher than 1 wt%. The remarkable conclusion from Fig. if the polymer fraction becomes higher. Reduction of membrane fouling In order to explore the fouling resistance of PES/ZnO mem- branes. Due to the increase of the hydrophilic- ity of the membrane by the addition of ZnO nanoparticles. embedding ZnO with addition of TiO2 nanoparticles.5 0. Permeability of the newly synthesized membranes (L h−1 m−2 bar) in the ZnO Concentraon (wt. but the TiO2 concentration nanoparticles into PES membrane can drop contact angle signifi- was much higher (about 1 wt%) than the proposed in this work. as the ZnO concentration was higher than 1 wt% addition of ZnO nanoparticles has a positive effect on the water and it was recovered to the permeability values obtained under the permeability for all of polymer concentrations tested. the membrane porosity which is evidenced by the increase of contact angle measurement could be smaller during phase-inversion because the solvent observed in these concentration values. the systematic again.75 and 1%. Thus. however. hydrophilicity did not change ity of the membranes was two times higher than the initial value significantly. However. In contrast to the increase of the contact angle with polymer For ZnO nanoparticle concentrations below 0. A similar effect on tion of ZnO concentration and PES concentration. as the ZnO concentration increases.. membrane At nanoparticle concentrations below 1 wt%. the addition of ZnO nanoparticles improves the fouling resistance considerably. in spite of the increasing content of ZnO nanoparti- in the absence of ZnO nanoparticles. Permeability of the newly synthesized membranes (L h−1 m−2 bar) as a func- in permeate flux was observed (0. showing the relative fluxes at different nanoparticle concen- trations. ciated with the reduction of the dispersion rate of nanoparticles.25 0. cantly from 70◦ (control PES membrane) to about 56◦ . should exist.% PES 60 Permeability ( L h m bar ) -1 70 PES concentraon 27 wt.25 0. 4).125 0. 6. a 23% increase Fig.%) 0 0. Fig. a sim- cles (Fig. 3.375 0. but at higher concentrations. Contact angles measured for PES/ZnO membranes at different nanoparticle concentrations. filtration experiments were performed using humic acid as a model organic foulant in aqueous solution.5 wt% interval eventually.5 wt%. This fact could be asso- result of the increasing hydrophobicity of membrane.%) ultralow to high concentration range for ZnO at 27 wt% of PES.

the rejection of selected dye compound with PES/ZnO membrane is higher than the one with control PES 4. This suggests that the charge between 250 and 400 Da. the extent of flux improvement was even lower than observed with PES/ZnO membranes here [43].9 80 Δ P = 10 bar T = 25 ºC Relave flux 0.8 Da) was measured dye solute may diffuse through the PES/ZnO membrane at a slower at different PES concentrations. B 100 PES/TiO2 composite membrane [42]. which is shown in Fig. Author's personal copy 160 S. dye compound and ZnO nanoparticles should be involved under lar mass of dye was higher than 400 Da. electrostatic repulsion between the between these two membranes was observed when the molecu.% PES/ZnO 0. an excellent competitor to TiO2 as anti-fouling material. 0.125 ZnO wt. the increasing hydrophilicity of the membrane with In order to better understand the rejection of organic dye addition of ZnO can also decrease solute adsorption in the mem- compounds smaller than 400 Da with PES/ZnO membranes. It is interesting to note that the rejection of PES/ZnO membranes was higher when the molecular weight of dye compounds was smaller than 400 Da while no difference in the rejection pattern nanoparticles is at pH of ca. increasing hydrophilicity and charge effect between methylene blue and ZnO nanoparticles is an impor- effects by the addition of ZnO nanoparticles may play an important tant factor that determines the rejection potential of the organic role in the rejection of dye compounds in the PES/ZnO membrane compound with the PES/ZnO membrane. PES/ZnO 0.125 wt% of ZnO). The rejection of 5 mg/L methylene blue (MW = 319. For the rate than the control PES membrane and this leads to lower con- control PES membrane.8 Rejecon (%) 60 23 % 0.0 Humic acids 5mgL -1 A 100 0. After 1 h contact years due to the attractive properties of ZnO. although the sieving mechanism becomes more dominant for the Considering the hydrophobic character of the selected dye rejection potential as the size of dye compound increases. in the interval the neutral pH tested in this study.6 PES/ZnO 0.125 ZnO wt.035. and 0. Since methylene ter permeability compared to neat polymeric membrane due to blue is a cationic dye and the zero point of charge (ZPC) of ZnO a higher hydrophilicity of the ZnO membranes.% 0 0. 0. 2.5 ZnO wt. which tion batch tests were performed with the same ZnO content as may lead to new applications of membranes in the forthcoming tested (0.8.125. solute adsorption into the membrane can be promoted Increased rejection of the methylene blue with the control PES with hydrophobic membranes due to hydrophobic–hydrophobic membrane is assumed to be caused by the macrovoids decrease in interactions [13]. Perhaps. centrations in the membrane permeate. the 0 24 26 28 30 32 difference of MWCO between the PES/ZnO membrane and the con- trol PES membrane was not significant. the concentration 80 Δ P = 10 bar of TiO2 was about five times higher than proposed in this study.125 wt% size of PES membranes. suggesting that the addition PES content (wt. To con- firm if the ZnO nanoparticles embedded in PES membrane have any The results from this study show that ZnO nanoparticles are adsorption capacity of methylene blue compound or not.125 ZnO wt. 8. we observed that the adsorption effect of ZnO nanoparticles ZnO blended membrane showed lower flux decline and a bet- for the methylene blue was almost negligible. Usually for hydrophobic cant effect on the rejection potential of the selected dye compound. 8B. improve the rejection of hydrophobic organic compounds. the polymer concentration has a signifi. / Journal of Membrane Science 389 (2012) 155–161 1.5 0 5 10 15 20 25 300 400 500 600 700 Time (h) Molecular weight (Da) Fig.035 ZnO wt. compound. adsorp. and therefore increase the rejection potential.%) of ZnO nanoparticles did not yield a large improvement in the pore Fig. It was shown that time. However. 8A shows a typical tendency for the PES rejection of organic compounds by a nanofiltration (NF) membrane. This has not been .% which can be quantified as about 500 Da of MWCO.% PES/ZnO 0. In T = 25 ºC some cases.5. As shown in Fig. ZnO (A) and rejection performance at different polymer concentrations (B). 7.% 20 PES PES/ZnO 0. Avoiding hydrophobic interaction in the inter- the membrane because higher polymer content can decrease the face solute-membrane surface is thought to be an effective way to inter-diffusion rate during phase inversion as discussed with Fig.7 40 PES 0. Balta et al. MWCO measurement and rejection performance 40 -1 Methylene blue 5 mgL In order to estimate the MWCO of prepared PES/ZnO membranes (0.125 wt%) and 5 mg/L of methylene blue. However. Relative flux of PES membranes with four concentrations of ZnO nanoparti- cles (0. compounds. 8B. Rejecon (%) 60 3. 9.5 wt%) in transient regime. MWCO measurement of control PES and PES/ZnO membranes with 0. Fig. the brane structure. Conclusions membrane under all polymer contents tested in this study. different commercial dyes with increasing 20 molecular mass were used.

J. Li. C. Barron. B.M.B.H. Vandecasteele. Li.W. Zha. 193 (2001) casteele. Ramaekers. 63 (2008) 251–263. Tak. S. retention properties and economic evaluation. Y. Sci. L.L. Preparation and characterizations of a new PS/TiO2 hybrid (2010-0024286).H. C. Coleman. G. The rejection of on the surface morphology and performance of microporous PES membrane. S.F. Membr. W. J. Am.B. M.J. methylene blue increased from 47. tration ceramic membranes derived from alumoxane nanoparticles.D. Y.C. This was observed even at unusually low ultrafiltration membrane modified by nano-sized alumina (Al2 O3 ) and its concentration (0. Batley.R. Yang. Comparative 35 (1980) 59–103. Hong. 313 (2008) 158–169.N. port provided by the European Union.X. Res. de Pinho. Nghiem.E.035 wt%) of ZnO nanoparticles. Romanian Government and [25] C. morphology and performance of microporous PES membrane. Hydrophilic modification of poly(ether sul- fone) ultrafiltration membrane surface by self-assembly of TiO2 nanoparticles. K. R. ultrafiltration membranes. 1–8.C. 205 (2002) 33–43.W. T. Toxicol. Preparation and characterization of fouling-resistant provided by the Regional Government of Madrid through project TiO2 self-assembled nanocomposite membranes. Correlation between retention and polyethersulfone nanofiltration membranes: antifouling effect observed at adsorption of phenolic compounds in nanofiltration membranes. Membr. Surf. 11 (2009) 25–39. 29 (2002) 23–31. Wu. performance of crosslinked polyimide organic solvent nanofiltration (OSN) Exposure to ZnO nanoparticles induces oxidative stress and cytotoxicity in membranes. fabrication of iron oxide ultrafiltration membranes. J. Luo. C. Arancia. Environ. [40] M. Nanosized zinc Sci. C. ron. R. [37] B. Bottero. Narbaitz. C. You. L. P. Envi- membranes. Xiang. C. Chem. Science and Technology (2006) 9–16. [24] Y. M. Surf. Vandecasteele. ˜ A. [4] I. Yang. 6/1. C. Y. Role of the dissolved zinc membrane structures: review of manufacturing procedures and performance ion and reactive oxygen species in cytotoxicity of ZnO nanoparticles. B 170 (2010) 88–92. Membr. [9] B. Gan. Boromand. 275 (2006) REMTAVARES (S2009/AMB-1588). J. Y. [38] A. through the project POSDRU. fone) ultrafiltration membrane surface by self-assembly of TiO2 nanoparticle. Sci. In addition. 276 (2006) 162. C. Membrane fouling test: filtration membranes for activated sludge filtration. Yan. Wang. Shi. M. Sci. Determination. Kim. [15] M. Huang.and micro-sized ZnO particles in human lung epithelial cells. Balta et al. Sci. Membr. Huang. S.R. K. Sep. Acknowledgements [22] L. Hu. [19] Y. Maes. J. Y. Zhang. 249 (2005) apparatus evaluation. antifouling research. toxicity of nanoparticulate ZnO. J. Sci. Desalination [34] N. Y. Zhu.E. M. Fundamental studies of a new hybrid Dunarea de Jos University of Galati. Berardis. Vande. Balta. Condello.M. [29] W.Q. C.H. Effect of TiO2 nanoparticles able improvement in the dye rejection potential. Yahui. Sci. Hagstrom.R. Effect of TiO2 nanoparticles on fouling mitigation of ultra- [18] D. Appl. C. Effect of TiO2 nanoparticle size on the perfor- mance of PVDF membrane. Civitelli. Taheri. J. W.H. 249 (2005) 76–84. J. 252 (2005) 195–203. Cortalezzi. Rose. Wang.S. Van der Bruggen. Cho. Zhao. B. Xu. H. J. M. I. Technol. Appl. Zhang. Mater. Surf. W. Q. [2] M. [23] P. [5] J. 216 (2003) 269–278. 199 (2010) 389–397. Yan. Xia. Van der Bruggen. 249 (2005) 76–84.X. F. Appl. Membr. J. S. 253 (2006) 2003–2010. Chen. Membr. References Appl. 7. X. Desalination 36 (1981) 235–264. Zhang. Zhang. S. Drawbacks of applying nanofil. J. Barron.L. – ASCE 130 (2004) 90–98. Nystrom. Mechanisms underlying the effects of Polymer 47 (10) (2006) 3464–3476. G. Van der Bruggen.3% [20] L.H. Study of the characteristics due to adsorption of organic compounds. Chen. Preparation of polysulfone–Fe3 O4 composite gram through the National Research Foundation of Korea (NRF) ultrafiltration membrane and its behavior in magnetic field. Techol. Li. G.Q. Impact of TiO2 nanoparticles on morphology and [32] B. Influence of hydrophobicity on retention in nanofiltration of aqueous 239–248. Ahlert. 21 (2011) 250 (2010) 829–832. Sci. Livingston. Mater. Luo. Van der Bruggen. Kim.S. Z. Xu. Stefan Balta would like to acknowledge the sup- membranes by sol–gel process. H. M. Linlin.J. J. 81 (1989) 61–67. H. Bulk ZnO. Arsuaga.Y. Toxicol. Van der Bruggen. Eng. B. [35] A. Sci. Vandecasteele. M. L. Author's personal copy S. Wiesner. Apte. Yuan. Membr. Li. A.Y. X. J. Membr. G. Membr. Mallevialle. X. [43] T. Pu. Surf. 343 (2009) 189–198. Jian.5/S/15. J.C. J. Xu. C. Everaert. human colon carcinoma cells. Cortalezzi. H. Deng. J. Yang. López-Munoz. Sci. Boussu. Xu. J.H. Pharm. D. Zheng.D. Hazard. Shun. 254 (2008) 7080–7086. Concentration polarization and fouling. L.M. J.R. P.F. Sci. Membr. filtration membranes.S. A. Elimelech. Kim. Modeling of the [39] A. Costa. [14] J. Sci. 41 (2007) 8484–8490. (inorganic–organic) positively charged membrane: membrane preparation and characterizations. Espendiller. improvement for water treatment. Mater. Z. [27] M. J.H. X. S. J.B. Polymer 46 (2005) 7701–7707. S. Chemphyschem 6 (2005) 1606–1612. and ZnCl2 to a freshwater microalga [7] D. Potts. Song. This research was supported by Basic Science Research Pro. R. H. Jiao.P. Tang. Desalination ultralow concentrations of TiO2 nanoparticles. Li. B. Sci. P.M. D. [13] L. Chou. Li. Surf.H. B. Effects of ultrafiltration mem. Mansourpanah. B. 97 (1994) 37–52.L. Colloids 207–217. Jucker. M. Technol. Application of nation 250 (2010) 682–687. Cao. B 76 (2010) 145–150. Childress. Tang. 284 funded by the Ministry of Education. Role of membrane surface Appl. Clark. Tak. Q. Boussu. G. Facile synthesis of ZnO nanoparticles based [1] J. membrane fouling on the nanofiltration of trace organic contaminants. P. Y. M. . Wiesner. Braeken.A. R. J. C.M. Matthiasson. J. [10] L. Madaeni. J.Y. Membr. Effect of TiO2 nanoparticles on the surface on bacterial cellulose. Yang. Song. Rahimpour. Polymer 47 (2006) 2683–2688. Membr. G.H. Sun. Franklin.S. Braeken. Bae. L. Cui. Y. Wu. J. oxide particles induce neural stem cell apoptosis. Bae. H. N. J. Z. T. W. [42] G. W. Lett. 158 (2010) 2335–2349. T. Sci. 1–5. ZnO nanoparticles provide a remark. L. antibacterial and polyamide reverse osmosis membranes. Liu.B. J. Van der Bruggen. Purif. K. nanofiltration for the removal of pesticides and hardness from ground water: [11] L. A. Shiyan. Flux decline in nanofiltration [36] K.J. Pu. Sotto. J. Damodar. 40 (1999) 131–139. Coupling TiO2 adsorption of organic compounds on polymeric nanofiltration membranes in nanoparticles with UV irradiation for modification of polyethersulfone ultra- solutions containing two compound. J. 127 (1997) 101–109. Mosqueda-Jimenez. Y. Effect of nano-sized Al2 O3 -particle addi- tion on PVDF ultrafiltration membrane performance. Doping of [12] J. Characteristics of ultrafil.5% for neat membranes to 82. Z. Sep. Pozzi. Wang. 227 (2003) characterization and cytotoxicity on HELF cells of ZnO nanoparticles. [28] W.M. Wang. Preparation of poly(vinylidene fluoride) (PVDF) for blended membranes. G. Environ. Zhao. / Journal of Membrane Science 389 (2012) 155–161 161 evidenced before. Vandecasteele. Wu. M.E. and the performance of self-made nanoporous polyethersulfone membranes. Appl. B. Y. Ma. Wells. J. rejection and fouling. Soroko. Eng. 115101. M. Chen. Yu. Xiang. [6] E. Study the self cleaning. A. Sci. Membr. Manttari. J. morphology in colloidal fouling of cellulose acetate and composite aromatic [41] R. Li. Meschini. T. Laine. Ding.M.S. [8] B.S.M. Sci.N. [30] X. Arcadio Sotto would like to acknowledge the support [26] T. M. [16] C. Interactions between natural organic matter and membranes: posite membranes. The use of nanoparticles in polymeric and ceramic [33] W. J.Q. Nanotechnology 20 (2009) [3] M. nano. Sci. Water Works Assoc. 281 (2006) 716–725. Van der Bruggen.N. Surf. Sivik. Wang. Adsorption of aquatic humic substances on hydrophobic 172 (2009) 1321–1328. Mechanisms of colloidal natural organic tration and how to avoid them: a review. [21] X. Gadd. solutions containing organic compounds. P.Z. Water Sci. Pollut.S. A. Wilms. Membr.E. Ma. Guo. 246 (2010) 116–127. Matsuura. S. Zhou. H.C. 10311–10320. Lin. J.C. M. Lista. Clark. Luan. Elimelech. Van der Bruggen. A critical review of fouling of reverse osmosis (Pseudokirchneriella subcapitata): the importance of particle solubility.R. Rogers. A. Ceramic membranes derived from ferroxane nanoparticles: a new route for the [31] J. Sci. Ren. matter fouling in ultrafiltration.J. Casey. Nanopart.R. Membr.J. Purif. Shannon. Sotto. Preparation and characterization of PES/TiO2 com- [17] J. Braeken. Zhang. Amy. C. Sci. Sci.M. Yang. Toxicity of 255 (2009) 4725–4732. J. photocatalytic properties of TiO2 entrapped PVDF membranes. Sci. Desali. Sci. J. 288 (2007) 231–238. Rose. Techol. Hydrophilic modification of poly(ether sul- brane composition.