- Email: [email protected]
Nanofiltration thin-film composite polyester polyethersulfone-based membranes prepared by interfacial polymerization
Journal of Membrane Science 348 (2010) 109–116
Contents lists available at ScienceDirect
Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci
Nanofiltration thin-film composite polyester polyethersulfone-based membranes prepared by interfacial polymerization M.N. Abu Seman a , M. Khayet b , N. Hilal a,∗ a b
Centre for Clean Water Technologies, The University of Nottingham, University Park, Nottingham, NG7 2RD, UK Department of Applied Physics I, Faculty of Physics, University Complutense of Madrid, Av. Complutense s/n, 28040, Madrid, Spain
a r t i c l e
i n f o
Article history: Received 19 August 2009 Received in revised form 19 October 2009 Accepted 28 October 2009 Available online 1 November 2009 Keywords: Nanofiltration Polyester Thin-film composite membrane Interfacial polymerization Humic acid
a b s t r a c t Nanofiltration polyester thin-film composite membranes have been prepared by interfacial polymerization using commercial polyethersulfone membrane support. Different monomer bisphenol A (BPA) concentrations in the aqueous solution and various interfacial polymerization times in the organic solution containing trimesoyl chloride (TMC) were studied. The success of the conducted interfacial polymerization procedure was corroborated by FTIR-ATR. Irreversible fouling of both the unmodified polyethersulfone and the modified polyester thin-film composite polyethersulfone membranes have been studied using humic acid model solutions at different pH values. It was observed that polyester thin-film composite membranes exhibited practically no tendency to be irreversibly fouled by humic acid molecules at neutral environment. However, the permeate flux was decreased and the irreversible fouling factor was enhanced with decreasing the pH to a value of 3. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Surface chemistry and morphology of membranes play an important role in the transmembrane transport of components as well as on the efficiency of the membrane process [1]. Besides bulk modification of membrane material, surface modification of previously formed membranes is a promising approach to confer new properties to the existing membranes providing surfaces with tailor-made separation properties, energies and chemical functionalities different from those of the bulk membrane material. Generally, the objective for modification is not only to increase the flux and/or selectivity but also to improve the chemical resistance (i.e. solvent resistance, swelling, or fouling resistance), control of pore size, elimination of defects, etc. [2]. Among the different successful membrane surface modification techniques, interfacial polymerization is of particular interest. This method is a breakthrough in the history of membrane technology since it was developed by Cadotte at the North Star Research Institute for reverse osmosis applications [2]. Thin-film composite membrane prepared by interfacial polymerization was developed in order to overcome the limitation and problems encountered by asymmetric membrane formed by the phase inversion method [3].
∗ Corresponding author at: Centre for Clean Water Technologies, The University of Nottingham, University Park, Nottingham, NG7 2RD, UK. Tel.: +44 0115 951 4168; fax: +44 0115 951 4115. E-mail address: [email protected] (N. Hilal). 0376-7388/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2009.10.047
The advantage of interfacial polymerization is that the reaction is self-inhibiting through passage of a limited supply of reactants through the already formed layer resulting in an extremely thin film of thickness within 50 nm range [4]. The skin or thin layer produced by this technique will determine the overall solute retention, permeate flux and, in general, will control the efficiency of the membrane process. It must be pointed out that this technique also can be applied to obtain modified membranes with pores larger than those of reverse osmosis (RO) applications and useful for UF and possibly for nanofiltration (NF) processes [5]. One of the advantages of interfacial polymerization technique is that the thin layer can be optimized for particular function by varying the monomer concentration in each solution (both aqueous and organic solutions), monomer ratios or reaction time of the interfacial polymerization [6–8]. In general, there are various factors affecting the structural morphology and composition of the formed thin layer such as the concentration of monomers in the corresponding liquid solutions, the partition coefficients of the monomers, the reactivity ratios where blends of monomers are employed, the solubility of nascent polymer in the solvent phase, the overall kinetics and diffusion rates of the monomers, the presence of by products (such as hydrogen chloride in the case of amine with acyl chloride), the hydrolysis or other potentially competitive by-reactions, the crosslinking reaction, the support material and the post-reactions or treatments of the resulting thin layer [9]. However, some problems may be faced during interfacial polymerization decreasing the efficiency of the process such as the highly non-uniform pattern of polymer density and charge across the
110
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
active layer, the interfacial polymerization reaction may take place partly in aqueous phase and the microvoids in the formed active thin layer [10]. Compared to thin-film polyamide membranes prepared by interfacial polymerization, only few researcher studies have been carried out using this technique to prepare other polymeric thin-film polyester and polyesteramide membranes. Polyester RO membrane has been synthesized by Kwak et al. [11] and compared to polyamide membrane. It was claimed that the RO polyester membrane had water permeate flux eleven times higher than that of polyamide membrane and therefore could be used in low RO pressure application maintaining a reasonable rejection. By varying monomer concentration and reaction time, Mohammad et al. [8] produced polyester NF membranes exhibiting rejection factors of NaCl (0.001 M aqueous solution) with values up to 60% and permeate flux of about 1.8 × 10−2 kg/m2 s under operating pressure of 4.5 × 105 Pa. Polyesteramide NF membranes have been prepared by Razdan and Kulkarni [6] and Jayarani and Kulkarni [12] by combining phenol and amine monomers in aqueous solutions during interfacial polymerization. The incorporation of ester linkage increased the oxidation resistance of the membrane and this significantly increased the membrane tolerance on chlorine attack. This indicate that this composite membrane is practically used in desalination process to remove high concentration of salt and in treating waste water from pulp and paper industry which contain very high chlorine content. Polyamide membrane has been explored since over the last 30 years and continued until now with application of variation of amine monomers and acyl chlorides [13,14]. This may be one of the reasons why polyester polymer membrane is not popular and therefore it is still considered new in the field of composite thin-film membrane development. The main objective of the present paper is to prepare polyester thin-film composite membranes with improved NF antifouling properties by means of interfacial polymerization using bisphenol A (BPA) and trimesoyl chloride (TMC) as monomers. A high negatively charged thin layer that improve antifouling characteristics by increasing repulsion force towards humic acid molecules, can be produced by interfacial polymerization technique [8]. Different BPA monomer concentrations and interfacial polymerization time have been considered. Humic acid has been employed as a model organic foulant and different humic acid aqueous solutions of different pH values have been tested. The effects of pH of aqueous solutions on the irreversible fouling of both the modified and unmodified membranes have been studied.
Fig. 1. Chemical structure of the monomers used: bisphenol A (BPA) and trimesoyl chloride (TMC).
2. Experimental 2.1. Materials The monomers bisphenol A (BPA) and trimesoyl chloride (TMC) were purchased from Sigma–Aldrich Co. Their chemical structures are shown in Fig. 1. The solvent hexane (reagent grade) was purchased from Fisher Scientific, UK. Humic acid (Sigma–Aldrich Co.) of molecular weight of 4.1 kDa was chosen as a model organic foulant [15]. To adjust the pH of the feed humic acid solutions to the required values, hydrochloric acid (HCl) and sodium hydroxide (NaOH) were supplied by Sigma–Aldrich Co. and Acros Organics, respectively. The asymmetric commercial membrane NFPES10 purchased from Hoechst Company (Germany) with pore size within 0.2–0.5 nm [16] and 75 m thickness (based on SEM image without backing material) was used as base support for interfacial polymerization. 2.2. Preparation of thin-film composite polyester membrane by interfacial polymerization technique The membrane NFPES10 was first cleaned with distilled water and subsequently left in ultrasonic distilled water bath for 1 min. These two steps were repeated 3 times in order to remove any preservative agent (i.e. glycerine) added by manufacturer during production. The cleaned membrane was immersed in a BPA aqueous solution for 15 min. Since the monomer BPA has very low solubility in water, it was dissolved in an aqueous solution of NaOH at pH of about 11. The BPA monomer concentration in the aqueous solution was varied between 0.1% (w/v) and 2% (w/v). The pre-soaked membrane taken out from the aqueous solution was positioned vertically for 2 min to drain the excess monomer on the membrane surface. Then, this membrane was dipped in the
Fig. 2. Reaction carried out between the monomers BPA and TMC for formation of polyester.
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
111
containing different concentrations of humic acid was determined. The humic acid rejection factor (˛) was calculated as follows:
˛=
Fig. 3. Schematic representation of the cross-flow NF unit (1: 5 L stainless steel tank; 2: stopcock; 3: gear pump; 4: flow meter; 5: valve; 6: pressure gauge; 7: membrane filtration cell; 8: electronic balance).
organic phase prepared by mixing the monomer TMC in hexane (0.15%, w/v) and left in the organic solution for a predetermined time (10 s, 30 s or 60 s) for interfacial polymerization. The reaction of the monomer BPA and TMC occurs at the membrane surface and produces a polyester polymer as shown in Fig. 2. All experiments were performed at ambient temperature. Finally, the membrane was dried in air for 30 min for organic solvent evaporation and then the membrane was stored in de-ionized water overnight before characterization tests. 2.3. Membrane characterization Both the unmodified NFPES10 and the modified membranes have been characterized. Attenuated Total Reflectance-Fourier Transform Infrared Spectroscopy (ATR-FTIR Bruker model Tensor 27) was used to analyse the functional groups on the membrane surface and test if the interfacial polymerization was carried out effectively. In this technique, infrared light enters the device through a set of mirrors. Part of this light is adsorbed by the membrane sample placed above a germanium crystal. The remaining light is reflected back to a detector with the use of mirrors. More details may be found elsewhere [17]. Membrane morphology was observed with a Scanning Electron Microscope, SEM (Model Quanta 600, USA). NF experiments have been carried out using the stainless steel system schematized in Fig. 3. The membrane module has an effective area of 12.6 cm2 and can be operated under the transmembrane pressure range of 1 × 105 to 9 × 105 Pa. The feed solution is circulated through the membrane module by means of a pressure pump (Tuthill Pump Co. of Carlifonia). The transmembrane pressure together with the volumetric flow rate can be adjusted using the concentrate outlet valve and variable speed key of the pump. Both the feed and retentate flow rates were measured at the inlet and outlet of the membrane module by two flowmeters (MPB Industrties, Kent, England). In all NF experiments the feed and retentate flow rates were maintained at 0.4 L/min. The permeate flux (J) of each membrane sample was determined by weighing the obtained permeate during a predetermined time using an electronic balance (Precisa, Model XB3200C) connected to a computer and the following equation. J=
W At
(1)
where W is the weight of the obtained permeate during a predetermined NF operation time t and A is the membrane area. Shimadzu UVmini-1240 was used to determine the concentration of humic acid in the feed, retentate and permeate aqueous solutions. As the spectrophotometer detector reads the concentration in absorbance unit, a calibration curve using different solutions
Cp 1− Cf
× 100
(2)
where Cp and Cf are the humic acid concentration in the permeate and in the feed solutions, respectively. To adjust the pH of the feed solutions, Jenway pH meter model 3540 was used. In this case, the pH probe was calibrated before readings were taken at 20 ◦ C using standard buffer solutions (phthalate, phosphate and borate) of pH values 4, 7 and 10 (±0.01). All these buffer solutions were supplied by Fisher Scientific, UK. Before all NF experiments, each membrane was first immersed in distilled water and then placed in the filtration cell (7 in Fig. 3) and pressurized at 7 × 105 Pa for at least 2 h using distilled water. Subsequently, the pure water experiments were conducted at different transmembrane pressures, P (4 × 105 , 5 × 105 , 6 × 105 and 7 × 105 Pa) in order to determine the pure water permeation flux (Jw ) using Eq. (1). The membrane permeance, Pm , was determined from the slope of the straight line that can be obtained by plotting the permeate flux (Jw ) against P following the following equation. Pm =
Jw P
(3)
Then, humic acid aqueous solutions were used and the product permeate rate (Jt ) as well as the rejection factor (˛) were determined as a function of time applying a transmembrane pressure of 6 × 105 Pa. A stock aqueous solution of 1 g/L humic acid was prepared using distilled water and adjusting the pH to a value slightly higher than 9 in order to make sure that all humic acid powder was completely dissolved. In this study, the humic acid feed solution pH was varied using 0.1 M NaOH or 0.1 M HCl as mentioned previously. For both the unmodified and the modified membranes, after the NF experiments conducted using humic acid solution, the system was washed with distilled water and pure water permeation flux (Jwf ) was measured again in order to evaluate the irreversible fouling in terms of pure water flux reduction, called hereafter irreversible fouling factor (FRW ) and determined as follows [18,19]. FRW =
Jw0 − Jwf Jw0
× 100
(4)
3. Results and discussions As stated earlier both the unmodified and the modified NFPES10 membranes were characterized by FTIR-ATR. Attenuated total reflectance (ATR) is a characterization technique used in conjunction with infrared spectroscopy, which enables samples to be examined directly in the solid or liquid state without further preparation and at the same time to solve the spectral reproducibility encountered by the routine IR analysis [20]. Fig. 4 shows as an example the spectra of the modified membrane with 2% (w/v) BPA at 60 s reaction time compared to the spectra of the original unmodified membrane. The reaction between the monomer BPA in water (pH = 11) and TMC in hexane, acting as organic phase, produces polyester polymer layer on membrane surface as explained in Fig. 3. This was corroborated by FTIR-ATR analysis where it can be seen, besides the typical PES bands of the unmodified membrane, the IR spectra of the modified membrane contained additional peaks at 1750 cm−1 and 1246 cm−1 that correspond to C O and C–O (ester group) bands, respectively [12,21]. Fig. 5 shows the unmodified membrane and membrane modified with 2% (w/v) BPA at 10 s and 60 s respectively. The SEM pictures was taken under 1800× magnification, compared to unmodified membrane (Fig. 5a), one can see that interfacial polymerization generate another layer on top of PES supporting
112
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
Fig. 4. FTIR-ATR analysis of the unmodified NFPES10 membrane and the thin-film composite polyester membrane (2% (w/v) BPA at 60 s reaction time). Fig. 6. Water permeance of the modified membranes with different BPA monomer concentrations versus interfacial polymerization reaction time.
membrane (Fig. 5b and c). Membrane modified with 2% (w/v) BPA at shorter reaction time (10 s) produced a layer with corrugated in shape while a grainy structure generated when the longer reaction time (60 s) was applied. The different of structures become clearer when comparing the membrane at higher magnification (Fig. 5e–f) where nodule and finely dispersed grainy structure was observed for membrane modified with 60 s and 10 s respectively, however the unmodified membrane image remain unchanged. Pure water permeation flux (Jw ) was measured for both the modified and unmodified membranes at different transmembrane
pressures, P, after compaction. Straight lines were obtained between Jw and P and the water permeance, Pm , was determined as stated previously. The effects of the BPA monomer concentration as well as the reaction time on the water permeance of the modified NFPES10 were investigated. Fig. 6 shows the water permeance of the unmodified and modified membranes prepared with different monomer concentrations (0.1, 0.5, 1 and 2%, w/v) as a function of the reaction time. It was found that the water permeance of the
Fig. 5. SEM pictures of surface (a) unmodified membrane (b) modified with 2% (w/v) BPA at 10 s reaction time (c) modified with 2% (w/v) BPA at 60 s reaction time (with 1800× magnification) and (d)–(f) with higher magnification.
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
113
Fig. 8. Permeate flux versus transmembrane pressure of 5 mg/L humic acid solution. Fig. 7. Effect of BPA monomer concentration on water permeance at different interfacial polymerization reaction times.
modified membranes decreased with increasing both the interfacial polymerization reaction time and BPA monomer concentration. At low BPA monomer concentration, 0.1% (w/v), the reaction time has no significant effect on water permeance compared to higher BPA concentrations. This result can be observed clearly in Fig. 7 in which the water permeance is also compared to that of the unmodified NFPES10 membrane. It is worth noting that all modified membranes exhibited smaller permeate fluxes than the unmodified NFPES10 membrane. This is attributed to the formed denser layer of polyester on the NFPES10 membrane surface. Similar results have been observed by Mohammad et al. [8] and Ji and Mehta [22]. It was claimed that the reactant concentration and reaction time have significant effect on the growth of thin film. In our case, as the BPA concentration was increased, the thin-film composite layer was expected to be thicker resulting in membranes with lower permeabilities. Similar conclusion can be made for the effect of the reaction time where longer reaction time may induce thicker polyester layer on the top of the membrane support. Fouling in NF was studied using humic acid aqueous solutions (15 mg/L and different pH values) and tested with both modified and unmodified membranes. The effects of membrane modification by interfacial polymerization on fouling tendency were investigated. In this case, four modified membranes modified with different BPA monomer concentrations (0.1BPA/60 s, 0.5BPA/10 s, 1BPA/10 s and 2BPA/10 s) were considered. The modified membranes were named hereafter 1BPA, 5BPA, 10BPA and 20BPA, respectively. The same NF operation conditions were applied for both the unmodified and the modified membranes. The feed solution was circulated through the NF membrane system at a constant operating pressure of 6 × 105 Pa and a flow rate of 0.4 L/min. It was observed for all membranes that the initial permeate flux when using humic acid solutions were lower than the corresponding pure water permeation flux. This result was also observed when testing low humic acid concentration (5 mg/L) in the feed aqueous solution (Fig. 8). This is attributed to both the osmotic effect and to less extent to the concentration polarization phenomenon. It was observed that the permeation flux of all membranes decreased when the pH value was diminished from 9 to 3. Fig. 9a and b shows as example the obtained permeate flux and the normalized permeate flux of the unmodified membrane. This reduction of the permeate flux may be due to the change of both the humic acid components configuration and to the membrane characteristics. In fact, the pure water permeation flux was also measured at different pH values (Fig. 10) and a reduction of 24% of the permeate flux of the membrane NFPES10 was detected when the pH value was varied from 9 to 3. When using pure water and different pH values in the range 3–9, no change was observed for the water permeability of polyethersulfone MF membrane [23].
Fig. 9. Normalized permeate flux, Jt /J0 versus time (a) and irreversible fouling, FRW (b) of the unmodified membrane for different pH of 15 mg/L humic acid solution.
However, in the present study, it was observed that the water permeance of NF membranes was increased as the pH value was enhanced from 3 to 9. These observations were also reported by Boussu [24] when using NF membrane NFPES10. This phenomenon
Fig. 10. Effect of pH on pure water permeance of the unmodified NFPES10 membrane.
114
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
Fig. 11. Normalized permeate flux of the unmodified and the modified membranes versus time when using 15 mg/L humic acid solutions of different pH values: 7 (a) and 3 (b).
was explained by Barghetta et al. [25] in terms of swollen membrane matrix at different environments. At high pH values, the charged functional groups of the membrane matrix force adjacent polymers apart leading to an increase of the membrane permeance. In contrast, at low pH values, the charge of the membrane matrix is reduced or shielded and membrane polymers come close to each other causing a reduction of the membrane permeance. Moreover, electrostatic charge repulsion force may affect membrane structure [1,26,27]. The membrane matrix having a negative charge would be in a more compacted state under low pH value due to a less intramembrane electrostatic repulsion (i.e. less charged). However, this condition is reversed at high pH value where the membrane matrix is expected to be in a more expanded state (i.e. more swollen) as response to a greater intramembrane electrostatic repulsion (i.e. more charged). As can be seen in Fig. 9, fouling occurred for all studied pH values. However a significant flux decline was observed at a pH value 3 (Fig. 9b), although at this pH the membrane exhibited the lowest initial permeate flux. For example, after 80 min NF operation time, the flux decline is about 10% for pH values 7 and 9 while for pH 3 the reduction reached 30%. For both pH values 7 and 9, the flux decline is almost the same. This means that the pH range neutral and alkaline do not affect too much the membrane characteristics. Furthermore, this result can be related to the change of humic acid configuration at different environment conditions [28,29]. This fact will be explained in details later on. The irreversible fouling factor, FRW , has been evaluated as explained earlier using Eq. (4). The FRW factors of the unmodified membrane at different pH values are shown in Fig. 9. The irreversible fouling increased up to 32% with decreasing the pH value from 9 to 3. At low pH, the permeate flux decreased and the fouling effect is more significant. In Figs. 11 and 12 the normalized permeate flux of both the unmodified and the modified membranes is plot against time for
Fig. 12. Irreversible fouling factor, FRW , of the unmodified and the modified membranes calculated after 250 min NF of 15 mg/L humic acid solutions of different pH values: 7 (a) and 3 (b).
two pH values 7 and 3, respectively. As can be seen in Fig. 10, at pH 7 no permeate flux decline was detected for all modified membranes during 240 min (4 h) NF experiment with humic acid solution. In contrast, under an acidic condition (pH = 3), all the membranes not only showed lower initial permeate fluxes than the ones measured at pH value of 7 but also its decline with time. This may be due to the highly negative charge of the modified membrane that increases repulsion forces between the negative charge of the humic acid molecules at pH 7, reducing the tendency of humic acid molecules to block the pores and/or prevent cake layer formation. It must be pointed out that for pH value of 3, all modified membranes exhibit lower reduction of the permeate flux than the unmodified membrane and for both pH values 7 and 3 the decrease of the permeate flux with time is smaller at higher BPA monomer concentration. The irreversible fouling factor, FRW , has been calculated and the results are shown in Fig. 12. At neutral pH all modified membranes exhibit negative irreversible fouling factor. This means that not only the pure water permeation flux of all modified membranes was fully recovered after cleaning but it was increased after NF of humic acid solution. Interestingly, very few modified membranes showed higher water flux than its initial flux after humic acid filtration. This phenomenon was also experienced by Mänttäri et al. [18] and Schäfer et al. [30]. The membrane 20BPA exhibits an enhancement of water permeation flux of about 13%. This may be due to certain amount of adsorbed humic acid molecules on the membrane surface leading to an increase of the hydrophilicity of the membrane and the permeate flux. According to Mänttäri et al. [18], humic acid contains both hydrophobic and hydrophilic units. During NF, the hydrophobic parts of humic acid bind on the hydrophobic ones of the membrane surface while the hydrophilic parts of humic acid are directed towards the feed solution. Therefore, the membrane surface becomes more hydrophilic and more negatively charged.
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
Mänttäri et al. [18] studied fouling effects of humic acid in NF membranes. The results showed that during 90 min humic acid filtration at pH values 7–8, the water permeation increased and this enhancement was obvious at higher humic acid concentrations. Schäfer et al. [30] found that after humic acid filtration the permeate water flux increased 13% of its initial flux and claimed that the humic acid made the membrane more hydrophilic as it was explained previously. In fact, based on Nuclear Magnetic Resonance (NMR) spectra it was shown that humic acid supplied by Sigma–Aldrich Co. contains larger hydrophilic fraction than hydrophobic one [31]. At low pH value, as can be seen in Fig. 12b, only the modified membrane with the highest BPA monomer concentration exhibits the lowest irreversible fouling factor. The other modified membranes are more susceptible to be fouled by humic acid under acidic environment than the unmodified membrane. Under acidic environment, carboxylic acid groups of humic acid lost their charge. At low pH (<4), the macromolecules of humic acid have smaller macromolecular configuration due to the increased hydrophobicity and reduced inter-chain electrostatic repulsion [18,28,29,32]. In fact, the isoelectric point of humic acid is approximately at pH 3.5 [18]. That is the reason explaining why at pH value 3 (Fig. 12b), all membranes were fouled by humic acid molecules since both humic acid and membranes are not charged (no electrostatic repulsion force can occur). Furthermore, molecules of humic acid become coiled, spherical in shape and compact leading to high permeate flux decline compared to the obtained results at neutral environment (pH = 7) [18,28]. As can be seen in Fig. 12b, when comparing membrane 20BPA with other membrane which modified with lower concentration like membrane 10BPA, membrane 20BPA has higher antifouling characteristics. This could be related to non-uniformity of new layer formed on the membrane support as described by Song et al. [10]. It could be the higher content of BPA monomer produced denser and more uniform layer than lower concentration. At lower concentration of BPA, some part of PES support which not covered by the polyester layer adsorbed very high hydrophobic humic acid at pH 3, thereby increasing degree of irreversible fouling. Regarding humic acid removal, not much difference was observed between the unmodified and the modified membranes under acidic environment. All membranes show rejection factors (˛) higher than 98%. Some of the modified membranes exhibit slightly higher rejection factors. At pH 7, removal of humic acid molecules may be attributed to charge and sieving effects while at lower pH (pH3) sieving effect is more predominant because at acidic environment charge effect is not significant (humic acid molecules have low charge) [28,32]. Schäfer et al. [33] reported that the ratio of the membrane pore size to foulant size is one of the factors determining the mechanisms involved in rejection and fouling. The humic acid molecule size may range between 1.7 nm and 3.5 nm, depending on their molecular weight 1 kDa and 10 kDa, respectively [30]. In our case, the molecular weight of the used humic acid is 4.1 kDa [15], therefore the molecule size can be between 1.7 nm and 3.5 nm. The reported pore size of the unmodified NFPES10 membrane used is less than 0.3 nm [16]. This indicates the high rejection factor observed for this membrane and that the followed interfacial polymerization method does not damage it. In term of fouling, as mentioned by Schäfer et al. [33] which foulants larger than the pore size will only cause surface fouling, this seems that all the membranes only experienced with surface fouling since the higher ratio of foulants/pore size where the foulants cannot easily penetrate into the pores. 4. Conclusions Thin-film composite polyester nanofiltration membranes were successfully prepared by interfacial polymerization technique as indicated by ATR-FTIR and SEM images.
115
At neutral environment, the modified membranes exhibited practically no irreversible fouling compared to the unmodified membrane. For some modified membranes, the irreversible fouling factor was found to be negative indicating that an increase of the water permeate flux after humic acid experiment. This result was attributed to the hydrophilic fraction of humic acid itself. However, most of the membranes were found to be more susceptible to irreversible fouling by humic acid components at more acidic environment (pH = 3) and under this condition the membrane modified with high BPA monomer concentration (2%, w/v) during 10 s interfacial polymerization time exhibited the lowest fouling factor. Acknowledgments We would like to thank the Ministry of Higher Education, Malaysia for supporting Mazrul Abu Seman’s study. We also like to thank the University Complutense of Madrid (UCM) for financially supporting M. Khayet during his stay in the UK. References [1] T. Matsuura, Synthetic Membranes and Membrane Separation, CRC Press, Inc., Boca Raton, Fla., 1994, pp. 467. [2] I. Pinnau, B.D. Freeman, Membrane formation and modification, in: ACS Sym. Series 744, American Chemical Society, Washington, DC, 2000. [3] A.P. Rao, N.V. Desai, R. Rangarajan, Interfacially synthesized thin film composite RO membranes for seawater desalination, J. Membr. Sci. 124 (1997) 263– 272. [4] M. Mulder, Basic Principles of Membrane Technology, second ed., Kluwer Academic Publishers, Dordrecht, The Netherlands, 2000. [5] A.P. Korikov, P.B. Kosaraju, K.K. Sirkar, Interfacially polymerized hydrophilic microporous thin film composite membranes on porous polypropylene hollow fibers and flat films, J. Membr. Sci. 279 (2006) 588–600. [6] U. Razdan, S.S. Kulkarni, Nanofiltration thin-film-composite polyesteramide membranes based on bulky diols, Desalination 161 (2004) 25–32. [7] A.L. Ahmad, B.S. Ooi, J.P. Choudhury, Preparation and characterization of co-polyamide thin film composite membrane from piperazine and 3,5diaminobenzoic acid, Desalination 158 (2003) 101–108. [8] A.W. Mohammad, N. Hilal, M.N. Abu Seman, A study on producing composite nanofiltration membranes with optimized properties, Desalination 158 (2003) 73–78. [9] R.J. Petersen, Composite reverse osmosis and nanofiltration membranes, J. Membr. Sci. 83 (1993) 81–150. [10] Y. Song, P. Sun, L.L. Henry, B. Sun, Mechanisms of structure and performance controlled thin film composite membrane formation via interfacial polymerization process, J. Membr. Sci. 251 (2005) 67–79. [11] S.Y. Kwak, M.O. Yeom, I.J. Roh, D.Y. Kim, J.J. Kim, Correlations of chemical structure, atomic force microscopy (AFM) morphology, and reverse osmosis (RO) characteristics in aromatic polyester high-flux RO membranes, J. Membr. Sci. 132 (1997) 183–191. [12] M.M. Jayarani, S.S. Kulkarni, Thin-film composite poly(esteramide)-based membranes, Desalination 130 (2000) 17–30. [13] V. Freger, J. Gilron, S. Belfer, TFC polyamide membranes modified by grafting of hydrophilic polymers: an FT-IR/AFM/TEM study, J. Membr. Sci. 209 (2002) 283–292. [14] J.M.M. Peeters, Characterization of nanofiltration membranes, Ph.D. Thesis, Universiteit Twente, 1997. [15] Y.P. Chin, G. Aiken, E. O’Loughlin, Molecular weight, polydispersity and spectroscopic properties of aquatic humic substances, Environ. Sci. Technol. 28 (1994) 1853–1858. [16] H. Al Zoubi, Development of novel approach to the prediction of nanofiltration membrane performance using advanced atomic force microscopy, Ph.D. Dissertation, University of Nottingham, UK, 2006. [17] http://en.wikipedia.org/wiki/Attenuated total reflectance. [18] M. Mänttäri, L. Puro, J. Nuortila-Jokinen, M. Nyström, Fouling effects of polysaccharides and humic acid in nanofiltration, J. Membr. Sci. 165 (2000) 1–17. [19] M. Mänttäri, M. Nystrom, Critical flux in NF of high molar mass polysaccharides and effluents from the paper industry, J. Membr. Sci. 170 (2000) 257– 273. [20] FT-IR Spectroscopy Attenuated Total Reflectance (ATR) Perkin Elmer Life and Analytical Sciences, Technical Note, 2005. [21] D. He, Surface-selective and controllable photo-grafting for synthesis of tailored macroporous membrane adsorbers, Ph.D. Thesis, Universitat Duisburg-Essen, Germany. [22] J. Ji, M. Mehta, Mathematical model for the formation of thin-film composite hollow fiber and tubular membranes by interfacial polymerization, J. Membr. Sci. 192 (2001) 41–54. [23] L. Al-Khatib, Development of (bio)fouling resistant membranes for water treatment applications. Ph.D. Dissertation, University of Nottingham, UK, 2006.
116
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
[24] K. Boussu, Influence of membrane characteristics on flux decline and retention in nanofiltration, Ph.D. Dissertation, Katholieke Universiteit Leuven, Belgium, 2007. [25] A. Barghetta, F.A. DiGiano, W.P. Ball, Nanofiltration of natural organic matter: pH and ionic strength effects, J. Environ. Eng. 123 (7) (1997) 628–641. [26] M. Tasaka, S. Tamura, N. Takemura, Concentration dependence of electroosmosis and streaming potential across charged membrane, J. Membr. Sci. 12 (1982) 169–182. [27] H.P. Gregor, in: E. Selegny (Ed.), Fixed-Charged Ultrafiltration Membranes. Charged Gels and Membranes. Part I, Reidel, Dordrecht, The Netherlands, 1976, pp. 57–69. [28] S. Hong, M. Elimelech, Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes, J. Membr. Sci. 132 (1997) 159–181.
[29] A.E. Childress, M. Elimelech, Effect of solution chemistry on surface charge of polymeric reverse osmosis and nanofiltration membranes, J. Membr. Sci. 119 (1996) 253–268. [30] A.I. Schäfer, A.G. Fane, T.D. Waite, Nanofiltration of natural organic matter: removal, fouling and the influence of multivalent ions, Desalination 118 (1998) 109–122. [31] H Susanto, M. Ulbricht, High-performance thin-layer hydrogel composite membranes for ultrafiltration of natural organic matter, Water Res. 42 (2008) 2827–2835. [32] K. Ghosh, M. Schnitzer, Macromolecular structures of humic substances, Soil Sci. 129 (5) (1980) 266–276. [33] A.I. Schäfer, A.G. Fane, T.D. Waite, Fouling effects on rejection in the membrane filtration of natural waters, Desalination 131 (2000) 215–224.
Contents lists available at ScienceDirect
Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci
Nanofiltration thin-film composite polyester polyethersulfone-based membranes prepared by interfacial polymerization M.N. Abu Seman a , M. Khayet b , N. Hilal a,∗ a b
Centre for Clean Water Technologies, The University of Nottingham, University Park, Nottingham, NG7 2RD, UK Department of Applied Physics I, Faculty of Physics, University Complutense of Madrid, Av. Complutense s/n, 28040, Madrid, Spain
a r t i c l e
i n f o
Article history: Received 19 August 2009 Received in revised form 19 October 2009 Accepted 28 October 2009 Available online 1 November 2009 Keywords: Nanofiltration Polyester Thin-film composite membrane Interfacial polymerization Humic acid
a b s t r a c t Nanofiltration polyester thin-film composite membranes have been prepared by interfacial polymerization using commercial polyethersulfone membrane support. Different monomer bisphenol A (BPA) concentrations in the aqueous solution and various interfacial polymerization times in the organic solution containing trimesoyl chloride (TMC) were studied. The success of the conducted interfacial polymerization procedure was corroborated by FTIR-ATR. Irreversible fouling of both the unmodified polyethersulfone and the modified polyester thin-film composite polyethersulfone membranes have been studied using humic acid model solutions at different pH values. It was observed that polyester thin-film composite membranes exhibited practically no tendency to be irreversibly fouled by humic acid molecules at neutral environment. However, the permeate flux was decreased and the irreversible fouling factor was enhanced with decreasing the pH to a value of 3. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Surface chemistry and morphology of membranes play an important role in the transmembrane transport of components as well as on the efficiency of the membrane process [1]. Besides bulk modification of membrane material, surface modification of previously formed membranes is a promising approach to confer new properties to the existing membranes providing surfaces with tailor-made separation properties, energies and chemical functionalities different from those of the bulk membrane material. Generally, the objective for modification is not only to increase the flux and/or selectivity but also to improve the chemical resistance (i.e. solvent resistance, swelling, or fouling resistance), control of pore size, elimination of defects, etc. [2]. Among the different successful membrane surface modification techniques, interfacial polymerization is of particular interest. This method is a breakthrough in the history of membrane technology since it was developed by Cadotte at the North Star Research Institute for reverse osmosis applications [2]. Thin-film composite membrane prepared by interfacial polymerization was developed in order to overcome the limitation and problems encountered by asymmetric membrane formed by the phase inversion method [3].
∗ Corresponding author at: Centre for Clean Water Technologies, The University of Nottingham, University Park, Nottingham, NG7 2RD, UK. Tel.: +44 0115 951 4168; fax: +44 0115 951 4115. E-mail address: [email protected] (N. Hilal). 0376-7388/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2009.10.047
The advantage of interfacial polymerization is that the reaction is self-inhibiting through passage of a limited supply of reactants through the already formed layer resulting in an extremely thin film of thickness within 50 nm range [4]. The skin or thin layer produced by this technique will determine the overall solute retention, permeate flux and, in general, will control the efficiency of the membrane process. It must be pointed out that this technique also can be applied to obtain modified membranes with pores larger than those of reverse osmosis (RO) applications and useful for UF and possibly for nanofiltration (NF) processes [5]. One of the advantages of interfacial polymerization technique is that the thin layer can be optimized for particular function by varying the monomer concentration in each solution (both aqueous and organic solutions), monomer ratios or reaction time of the interfacial polymerization [6–8]. In general, there are various factors affecting the structural morphology and composition of the formed thin layer such as the concentration of monomers in the corresponding liquid solutions, the partition coefficients of the monomers, the reactivity ratios where blends of monomers are employed, the solubility of nascent polymer in the solvent phase, the overall kinetics and diffusion rates of the monomers, the presence of by products (such as hydrogen chloride in the case of amine with acyl chloride), the hydrolysis or other potentially competitive by-reactions, the crosslinking reaction, the support material and the post-reactions or treatments of the resulting thin layer [9]. However, some problems may be faced during interfacial polymerization decreasing the efficiency of the process such as the highly non-uniform pattern of polymer density and charge across the
110
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
active layer, the interfacial polymerization reaction may take place partly in aqueous phase and the microvoids in the formed active thin layer [10]. Compared to thin-film polyamide membranes prepared by interfacial polymerization, only few researcher studies have been carried out using this technique to prepare other polymeric thin-film polyester and polyesteramide membranes. Polyester RO membrane has been synthesized by Kwak et al. [11] and compared to polyamide membrane. It was claimed that the RO polyester membrane had water permeate flux eleven times higher than that of polyamide membrane and therefore could be used in low RO pressure application maintaining a reasonable rejection. By varying monomer concentration and reaction time, Mohammad et al. [8] produced polyester NF membranes exhibiting rejection factors of NaCl (0.001 M aqueous solution) with values up to 60% and permeate flux of about 1.8 × 10−2 kg/m2 s under operating pressure of 4.5 × 105 Pa. Polyesteramide NF membranes have been prepared by Razdan and Kulkarni [6] and Jayarani and Kulkarni [12] by combining phenol and amine monomers in aqueous solutions during interfacial polymerization. The incorporation of ester linkage increased the oxidation resistance of the membrane and this significantly increased the membrane tolerance on chlorine attack. This indicate that this composite membrane is practically used in desalination process to remove high concentration of salt and in treating waste water from pulp and paper industry which contain very high chlorine content. Polyamide membrane has been explored since over the last 30 years and continued until now with application of variation of amine monomers and acyl chlorides [13,14]. This may be one of the reasons why polyester polymer membrane is not popular and therefore it is still considered new in the field of composite thin-film membrane development. The main objective of the present paper is to prepare polyester thin-film composite membranes with improved NF antifouling properties by means of interfacial polymerization using bisphenol A (BPA) and trimesoyl chloride (TMC) as monomers. A high negatively charged thin layer that improve antifouling characteristics by increasing repulsion force towards humic acid molecules, can be produced by interfacial polymerization technique [8]. Different BPA monomer concentrations and interfacial polymerization time have been considered. Humic acid has been employed as a model organic foulant and different humic acid aqueous solutions of different pH values have been tested. The effects of pH of aqueous solutions on the irreversible fouling of both the modified and unmodified membranes have been studied.
Fig. 1. Chemical structure of the monomers used: bisphenol A (BPA) and trimesoyl chloride (TMC).
2. Experimental 2.1. Materials The monomers bisphenol A (BPA) and trimesoyl chloride (TMC) were purchased from Sigma–Aldrich Co. Their chemical structures are shown in Fig. 1. The solvent hexane (reagent grade) was purchased from Fisher Scientific, UK. Humic acid (Sigma–Aldrich Co.) of molecular weight of 4.1 kDa was chosen as a model organic foulant [15]. To adjust the pH of the feed humic acid solutions to the required values, hydrochloric acid (HCl) and sodium hydroxide (NaOH) were supplied by Sigma–Aldrich Co. and Acros Organics, respectively. The asymmetric commercial membrane NFPES10 purchased from Hoechst Company (Germany) with pore size within 0.2–0.5 nm [16] and 75 m thickness (based on SEM image without backing material) was used as base support for interfacial polymerization. 2.2. Preparation of thin-film composite polyester membrane by interfacial polymerization technique The membrane NFPES10 was first cleaned with distilled water and subsequently left in ultrasonic distilled water bath for 1 min. These two steps were repeated 3 times in order to remove any preservative agent (i.e. glycerine) added by manufacturer during production. The cleaned membrane was immersed in a BPA aqueous solution for 15 min. Since the monomer BPA has very low solubility in water, it was dissolved in an aqueous solution of NaOH at pH of about 11. The BPA monomer concentration in the aqueous solution was varied between 0.1% (w/v) and 2% (w/v). The pre-soaked membrane taken out from the aqueous solution was positioned vertically for 2 min to drain the excess monomer on the membrane surface. Then, this membrane was dipped in the
Fig. 2. Reaction carried out between the monomers BPA and TMC for formation of polyester.
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
111
containing different concentrations of humic acid was determined. The humic acid rejection factor (˛) was calculated as follows:
˛=
Fig. 3. Schematic representation of the cross-flow NF unit (1: 5 L stainless steel tank; 2: stopcock; 3: gear pump; 4: flow meter; 5: valve; 6: pressure gauge; 7: membrane filtration cell; 8: electronic balance).
organic phase prepared by mixing the monomer TMC in hexane (0.15%, w/v) and left in the organic solution for a predetermined time (10 s, 30 s or 60 s) for interfacial polymerization. The reaction of the monomer BPA and TMC occurs at the membrane surface and produces a polyester polymer as shown in Fig. 2. All experiments were performed at ambient temperature. Finally, the membrane was dried in air for 30 min for organic solvent evaporation and then the membrane was stored in de-ionized water overnight before characterization tests. 2.3. Membrane characterization Both the unmodified NFPES10 and the modified membranes have been characterized. Attenuated Total Reflectance-Fourier Transform Infrared Spectroscopy (ATR-FTIR Bruker model Tensor 27) was used to analyse the functional groups on the membrane surface and test if the interfacial polymerization was carried out effectively. In this technique, infrared light enters the device through a set of mirrors. Part of this light is adsorbed by the membrane sample placed above a germanium crystal. The remaining light is reflected back to a detector with the use of mirrors. More details may be found elsewhere [17]. Membrane morphology was observed with a Scanning Electron Microscope, SEM (Model Quanta 600, USA). NF experiments have been carried out using the stainless steel system schematized in Fig. 3. The membrane module has an effective area of 12.6 cm2 and can be operated under the transmembrane pressure range of 1 × 105 to 9 × 105 Pa. The feed solution is circulated through the membrane module by means of a pressure pump (Tuthill Pump Co. of Carlifonia). The transmembrane pressure together with the volumetric flow rate can be adjusted using the concentrate outlet valve and variable speed key of the pump. Both the feed and retentate flow rates were measured at the inlet and outlet of the membrane module by two flowmeters (MPB Industrties, Kent, England). In all NF experiments the feed and retentate flow rates were maintained at 0.4 L/min. The permeate flux (J) of each membrane sample was determined by weighing the obtained permeate during a predetermined time using an electronic balance (Precisa, Model XB3200C) connected to a computer and the following equation. J=
W At
(1)
where W is the weight of the obtained permeate during a predetermined NF operation time t and A is the membrane area. Shimadzu UVmini-1240 was used to determine the concentration of humic acid in the feed, retentate and permeate aqueous solutions. As the spectrophotometer detector reads the concentration in absorbance unit, a calibration curve using different solutions
Cp 1− Cf
× 100
(2)
where Cp and Cf are the humic acid concentration in the permeate and in the feed solutions, respectively. To adjust the pH of the feed solutions, Jenway pH meter model 3540 was used. In this case, the pH probe was calibrated before readings were taken at 20 ◦ C using standard buffer solutions (phthalate, phosphate and borate) of pH values 4, 7 and 10 (±0.01). All these buffer solutions were supplied by Fisher Scientific, UK. Before all NF experiments, each membrane was first immersed in distilled water and then placed in the filtration cell (7 in Fig. 3) and pressurized at 7 × 105 Pa for at least 2 h using distilled water. Subsequently, the pure water experiments were conducted at different transmembrane pressures, P (4 × 105 , 5 × 105 , 6 × 105 and 7 × 105 Pa) in order to determine the pure water permeation flux (Jw ) using Eq. (1). The membrane permeance, Pm , was determined from the slope of the straight line that can be obtained by plotting the permeate flux (Jw ) against P following the following equation. Pm =
Jw P
(3)
Then, humic acid aqueous solutions were used and the product permeate rate (Jt ) as well as the rejection factor (˛) were determined as a function of time applying a transmembrane pressure of 6 × 105 Pa. A stock aqueous solution of 1 g/L humic acid was prepared using distilled water and adjusting the pH to a value slightly higher than 9 in order to make sure that all humic acid powder was completely dissolved. In this study, the humic acid feed solution pH was varied using 0.1 M NaOH or 0.1 M HCl as mentioned previously. For both the unmodified and the modified membranes, after the NF experiments conducted using humic acid solution, the system was washed with distilled water and pure water permeation flux (Jwf ) was measured again in order to evaluate the irreversible fouling in terms of pure water flux reduction, called hereafter irreversible fouling factor (FRW ) and determined as follows [18,19]. FRW =
Jw0 − Jwf Jw0
× 100
(4)
3. Results and discussions As stated earlier both the unmodified and the modified NFPES10 membranes were characterized by FTIR-ATR. Attenuated total reflectance (ATR) is a characterization technique used in conjunction with infrared spectroscopy, which enables samples to be examined directly in the solid or liquid state without further preparation and at the same time to solve the spectral reproducibility encountered by the routine IR analysis [20]. Fig. 4 shows as an example the spectra of the modified membrane with 2% (w/v) BPA at 60 s reaction time compared to the spectra of the original unmodified membrane. The reaction between the monomer BPA in water (pH = 11) and TMC in hexane, acting as organic phase, produces polyester polymer layer on membrane surface as explained in Fig. 3. This was corroborated by FTIR-ATR analysis where it can be seen, besides the typical PES bands of the unmodified membrane, the IR spectra of the modified membrane contained additional peaks at 1750 cm−1 and 1246 cm−1 that correspond to C O and C–O (ester group) bands, respectively [12,21]. Fig. 5 shows the unmodified membrane and membrane modified with 2% (w/v) BPA at 10 s and 60 s respectively. The SEM pictures was taken under 1800× magnification, compared to unmodified membrane (Fig. 5a), one can see that interfacial polymerization generate another layer on top of PES supporting
112
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
Fig. 4. FTIR-ATR analysis of the unmodified NFPES10 membrane and the thin-film composite polyester membrane (2% (w/v) BPA at 60 s reaction time). Fig. 6. Water permeance of the modified membranes with different BPA monomer concentrations versus interfacial polymerization reaction time.
membrane (Fig. 5b and c). Membrane modified with 2% (w/v) BPA at shorter reaction time (10 s) produced a layer with corrugated in shape while a grainy structure generated when the longer reaction time (60 s) was applied. The different of structures become clearer when comparing the membrane at higher magnification (Fig. 5e–f) where nodule and finely dispersed grainy structure was observed for membrane modified with 60 s and 10 s respectively, however the unmodified membrane image remain unchanged. Pure water permeation flux (Jw ) was measured for both the modified and unmodified membranes at different transmembrane
pressures, P, after compaction. Straight lines were obtained between Jw and P and the water permeance, Pm , was determined as stated previously. The effects of the BPA monomer concentration as well as the reaction time on the water permeance of the modified NFPES10 were investigated. Fig. 6 shows the water permeance of the unmodified and modified membranes prepared with different monomer concentrations (0.1, 0.5, 1 and 2%, w/v) as a function of the reaction time. It was found that the water permeance of the
Fig. 5. SEM pictures of surface (a) unmodified membrane (b) modified with 2% (w/v) BPA at 10 s reaction time (c) modified with 2% (w/v) BPA at 60 s reaction time (with 1800× magnification) and (d)–(f) with higher magnification.
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
113
Fig. 8. Permeate flux versus transmembrane pressure of 5 mg/L humic acid solution. Fig. 7. Effect of BPA monomer concentration on water permeance at different interfacial polymerization reaction times.
modified membranes decreased with increasing both the interfacial polymerization reaction time and BPA monomer concentration. At low BPA monomer concentration, 0.1% (w/v), the reaction time has no significant effect on water permeance compared to higher BPA concentrations. This result can be observed clearly in Fig. 7 in which the water permeance is also compared to that of the unmodified NFPES10 membrane. It is worth noting that all modified membranes exhibited smaller permeate fluxes than the unmodified NFPES10 membrane. This is attributed to the formed denser layer of polyester on the NFPES10 membrane surface. Similar results have been observed by Mohammad et al. [8] and Ji and Mehta [22]. It was claimed that the reactant concentration and reaction time have significant effect on the growth of thin film. In our case, as the BPA concentration was increased, the thin-film composite layer was expected to be thicker resulting in membranes with lower permeabilities. Similar conclusion can be made for the effect of the reaction time where longer reaction time may induce thicker polyester layer on the top of the membrane support. Fouling in NF was studied using humic acid aqueous solutions (15 mg/L and different pH values) and tested with both modified and unmodified membranes. The effects of membrane modification by interfacial polymerization on fouling tendency were investigated. In this case, four modified membranes modified with different BPA monomer concentrations (0.1BPA/60 s, 0.5BPA/10 s, 1BPA/10 s and 2BPA/10 s) were considered. The modified membranes were named hereafter 1BPA, 5BPA, 10BPA and 20BPA, respectively. The same NF operation conditions were applied for both the unmodified and the modified membranes. The feed solution was circulated through the NF membrane system at a constant operating pressure of 6 × 105 Pa and a flow rate of 0.4 L/min. It was observed for all membranes that the initial permeate flux when using humic acid solutions were lower than the corresponding pure water permeation flux. This result was also observed when testing low humic acid concentration (5 mg/L) in the feed aqueous solution (Fig. 8). This is attributed to both the osmotic effect and to less extent to the concentration polarization phenomenon. It was observed that the permeation flux of all membranes decreased when the pH value was diminished from 9 to 3. Fig. 9a and b shows as example the obtained permeate flux and the normalized permeate flux of the unmodified membrane. This reduction of the permeate flux may be due to the change of both the humic acid components configuration and to the membrane characteristics. In fact, the pure water permeation flux was also measured at different pH values (Fig. 10) and a reduction of 24% of the permeate flux of the membrane NFPES10 was detected when the pH value was varied from 9 to 3. When using pure water and different pH values in the range 3–9, no change was observed for the water permeability of polyethersulfone MF membrane [23].
Fig. 9. Normalized permeate flux, Jt /J0 versus time (a) and irreversible fouling, FRW (b) of the unmodified membrane for different pH of 15 mg/L humic acid solution.
However, in the present study, it was observed that the water permeance of NF membranes was increased as the pH value was enhanced from 3 to 9. These observations were also reported by Boussu [24] when using NF membrane NFPES10. This phenomenon
Fig. 10. Effect of pH on pure water permeance of the unmodified NFPES10 membrane.
114
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
Fig. 11. Normalized permeate flux of the unmodified and the modified membranes versus time when using 15 mg/L humic acid solutions of different pH values: 7 (a) and 3 (b).
was explained by Barghetta et al. [25] in terms of swollen membrane matrix at different environments. At high pH values, the charged functional groups of the membrane matrix force adjacent polymers apart leading to an increase of the membrane permeance. In contrast, at low pH values, the charge of the membrane matrix is reduced or shielded and membrane polymers come close to each other causing a reduction of the membrane permeance. Moreover, electrostatic charge repulsion force may affect membrane structure [1,26,27]. The membrane matrix having a negative charge would be in a more compacted state under low pH value due to a less intramembrane electrostatic repulsion (i.e. less charged). However, this condition is reversed at high pH value where the membrane matrix is expected to be in a more expanded state (i.e. more swollen) as response to a greater intramembrane electrostatic repulsion (i.e. more charged). As can be seen in Fig. 9, fouling occurred for all studied pH values. However a significant flux decline was observed at a pH value 3 (Fig. 9b), although at this pH the membrane exhibited the lowest initial permeate flux. For example, after 80 min NF operation time, the flux decline is about 10% for pH values 7 and 9 while for pH 3 the reduction reached 30%. For both pH values 7 and 9, the flux decline is almost the same. This means that the pH range neutral and alkaline do not affect too much the membrane characteristics. Furthermore, this result can be related to the change of humic acid configuration at different environment conditions [28,29]. This fact will be explained in details later on. The irreversible fouling factor, FRW , has been evaluated as explained earlier using Eq. (4). The FRW factors of the unmodified membrane at different pH values are shown in Fig. 9. The irreversible fouling increased up to 32% with decreasing the pH value from 9 to 3. At low pH, the permeate flux decreased and the fouling effect is more significant. In Figs. 11 and 12 the normalized permeate flux of both the unmodified and the modified membranes is plot against time for
Fig. 12. Irreversible fouling factor, FRW , of the unmodified and the modified membranes calculated after 250 min NF of 15 mg/L humic acid solutions of different pH values: 7 (a) and 3 (b).
two pH values 7 and 3, respectively. As can be seen in Fig. 10, at pH 7 no permeate flux decline was detected for all modified membranes during 240 min (4 h) NF experiment with humic acid solution. In contrast, under an acidic condition (pH = 3), all the membranes not only showed lower initial permeate fluxes than the ones measured at pH value of 7 but also its decline with time. This may be due to the highly negative charge of the modified membrane that increases repulsion forces between the negative charge of the humic acid molecules at pH 7, reducing the tendency of humic acid molecules to block the pores and/or prevent cake layer formation. It must be pointed out that for pH value of 3, all modified membranes exhibit lower reduction of the permeate flux than the unmodified membrane and for both pH values 7 and 3 the decrease of the permeate flux with time is smaller at higher BPA monomer concentration. The irreversible fouling factor, FRW , has been calculated and the results are shown in Fig. 12. At neutral pH all modified membranes exhibit negative irreversible fouling factor. This means that not only the pure water permeation flux of all modified membranes was fully recovered after cleaning but it was increased after NF of humic acid solution. Interestingly, very few modified membranes showed higher water flux than its initial flux after humic acid filtration. This phenomenon was also experienced by Mänttäri et al. [18] and Schäfer et al. [30]. The membrane 20BPA exhibits an enhancement of water permeation flux of about 13%. This may be due to certain amount of adsorbed humic acid molecules on the membrane surface leading to an increase of the hydrophilicity of the membrane and the permeate flux. According to Mänttäri et al. [18], humic acid contains both hydrophobic and hydrophilic units. During NF, the hydrophobic parts of humic acid bind on the hydrophobic ones of the membrane surface while the hydrophilic parts of humic acid are directed towards the feed solution. Therefore, the membrane surface becomes more hydrophilic and more negatively charged.
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
Mänttäri et al. [18] studied fouling effects of humic acid in NF membranes. The results showed that during 90 min humic acid filtration at pH values 7–8, the water permeation increased and this enhancement was obvious at higher humic acid concentrations. Schäfer et al. [30] found that after humic acid filtration the permeate water flux increased 13% of its initial flux and claimed that the humic acid made the membrane more hydrophilic as it was explained previously. In fact, based on Nuclear Magnetic Resonance (NMR) spectra it was shown that humic acid supplied by Sigma–Aldrich Co. contains larger hydrophilic fraction than hydrophobic one [31]. At low pH value, as can be seen in Fig. 12b, only the modified membrane with the highest BPA monomer concentration exhibits the lowest irreversible fouling factor. The other modified membranes are more susceptible to be fouled by humic acid under acidic environment than the unmodified membrane. Under acidic environment, carboxylic acid groups of humic acid lost their charge. At low pH (<4), the macromolecules of humic acid have smaller macromolecular configuration due to the increased hydrophobicity and reduced inter-chain electrostatic repulsion [18,28,29,32]. In fact, the isoelectric point of humic acid is approximately at pH 3.5 [18]. That is the reason explaining why at pH value 3 (Fig. 12b), all membranes were fouled by humic acid molecules since both humic acid and membranes are not charged (no electrostatic repulsion force can occur). Furthermore, molecules of humic acid become coiled, spherical in shape and compact leading to high permeate flux decline compared to the obtained results at neutral environment (pH = 7) [18,28]. As can be seen in Fig. 12b, when comparing membrane 20BPA with other membrane which modified with lower concentration like membrane 10BPA, membrane 20BPA has higher antifouling characteristics. This could be related to non-uniformity of new layer formed on the membrane support as described by Song et al. [10]. It could be the higher content of BPA monomer produced denser and more uniform layer than lower concentration. At lower concentration of BPA, some part of PES support which not covered by the polyester layer adsorbed very high hydrophobic humic acid at pH 3, thereby increasing degree of irreversible fouling. Regarding humic acid removal, not much difference was observed between the unmodified and the modified membranes under acidic environment. All membranes show rejection factors (˛) higher than 98%. Some of the modified membranes exhibit slightly higher rejection factors. At pH 7, removal of humic acid molecules may be attributed to charge and sieving effects while at lower pH (pH3) sieving effect is more predominant because at acidic environment charge effect is not significant (humic acid molecules have low charge) [28,32]. Schäfer et al. [33] reported that the ratio of the membrane pore size to foulant size is one of the factors determining the mechanisms involved in rejection and fouling. The humic acid molecule size may range between 1.7 nm and 3.5 nm, depending on their molecular weight 1 kDa and 10 kDa, respectively [30]. In our case, the molecular weight of the used humic acid is 4.1 kDa [15], therefore the molecule size can be between 1.7 nm and 3.5 nm. The reported pore size of the unmodified NFPES10 membrane used is less than 0.3 nm [16]. This indicates the high rejection factor observed for this membrane and that the followed interfacial polymerization method does not damage it. In term of fouling, as mentioned by Schäfer et al. [33] which foulants larger than the pore size will only cause surface fouling, this seems that all the membranes only experienced with surface fouling since the higher ratio of foulants/pore size where the foulants cannot easily penetrate into the pores. 4. Conclusions Thin-film composite polyester nanofiltration membranes were successfully prepared by interfacial polymerization technique as indicated by ATR-FTIR and SEM images.
115
At neutral environment, the modified membranes exhibited practically no irreversible fouling compared to the unmodified membrane. For some modified membranes, the irreversible fouling factor was found to be negative indicating that an increase of the water permeate flux after humic acid experiment. This result was attributed to the hydrophilic fraction of humic acid itself. However, most of the membranes were found to be more susceptible to irreversible fouling by humic acid components at more acidic environment (pH = 3) and under this condition the membrane modified with high BPA monomer concentration (2%, w/v) during 10 s interfacial polymerization time exhibited the lowest fouling factor. Acknowledgments We would like to thank the Ministry of Higher Education, Malaysia for supporting Mazrul Abu Seman’s study. We also like to thank the University Complutense of Madrid (UCM) for financially supporting M. Khayet during his stay in the UK. References [1] T. Matsuura, Synthetic Membranes and Membrane Separation, CRC Press, Inc., Boca Raton, Fla., 1994, pp. 467. [2] I. Pinnau, B.D. Freeman, Membrane formation and modification, in: ACS Sym. Series 744, American Chemical Society, Washington, DC, 2000. [3] A.P. Rao, N.V. Desai, R. Rangarajan, Interfacially synthesized thin film composite RO membranes for seawater desalination, J. Membr. Sci. 124 (1997) 263– 272. [4] M. Mulder, Basic Principles of Membrane Technology, second ed., Kluwer Academic Publishers, Dordrecht, The Netherlands, 2000. [5] A.P. Korikov, P.B. Kosaraju, K.K. Sirkar, Interfacially polymerized hydrophilic microporous thin film composite membranes on porous polypropylene hollow fibers and flat films, J. Membr. Sci. 279 (2006) 588–600. [6] U. Razdan, S.S. Kulkarni, Nanofiltration thin-film-composite polyesteramide membranes based on bulky diols, Desalination 161 (2004) 25–32. [7] A.L. Ahmad, B.S. Ooi, J.P. Choudhury, Preparation and characterization of co-polyamide thin film composite membrane from piperazine and 3,5diaminobenzoic acid, Desalination 158 (2003) 101–108. [8] A.W. Mohammad, N. Hilal, M.N. Abu Seman, A study on producing composite nanofiltration membranes with optimized properties, Desalination 158 (2003) 73–78. [9] R.J. Petersen, Composite reverse osmosis and nanofiltration membranes, J. Membr. Sci. 83 (1993) 81–150. [10] Y. Song, P. Sun, L.L. Henry, B. Sun, Mechanisms of structure and performance controlled thin film composite membrane formation via interfacial polymerization process, J. Membr. Sci. 251 (2005) 67–79. [11] S.Y. Kwak, M.O. Yeom, I.J. Roh, D.Y. Kim, J.J. Kim, Correlations of chemical structure, atomic force microscopy (AFM) morphology, and reverse osmosis (RO) characteristics in aromatic polyester high-flux RO membranes, J. Membr. Sci. 132 (1997) 183–191. [12] M.M. Jayarani, S.S. Kulkarni, Thin-film composite poly(esteramide)-based membranes, Desalination 130 (2000) 17–30. [13] V. Freger, J. Gilron, S. Belfer, TFC polyamide membranes modified by grafting of hydrophilic polymers: an FT-IR/AFM/TEM study, J. Membr. Sci. 209 (2002) 283–292. [14] J.M.M. Peeters, Characterization of nanofiltration membranes, Ph.D. Thesis, Universiteit Twente, 1997. [15] Y.P. Chin, G. Aiken, E. O’Loughlin, Molecular weight, polydispersity and spectroscopic properties of aquatic humic substances, Environ. Sci. Technol. 28 (1994) 1853–1858. [16] H. Al Zoubi, Development of novel approach to the prediction of nanofiltration membrane performance using advanced atomic force microscopy, Ph.D. Dissertation, University of Nottingham, UK, 2006. [17] http://en.wikipedia.org/wiki/Attenuated total reflectance. [18] M. Mänttäri, L. Puro, J. Nuortila-Jokinen, M. Nyström, Fouling effects of polysaccharides and humic acid in nanofiltration, J. Membr. Sci. 165 (2000) 1–17. [19] M. Mänttäri, M. Nystrom, Critical flux in NF of high molar mass polysaccharides and effluents from the paper industry, J. Membr. Sci. 170 (2000) 257– 273. [20] FT-IR Spectroscopy Attenuated Total Reflectance (ATR) Perkin Elmer Life and Analytical Sciences, Technical Note, 2005. [21] D. He, Surface-selective and controllable photo-grafting for synthesis of tailored macroporous membrane adsorbers, Ph.D. Thesis, Universitat Duisburg-Essen, Germany. [22] J. Ji, M. Mehta, Mathematical model for the formation of thin-film composite hollow fiber and tubular membranes by interfacial polymerization, J. Membr. Sci. 192 (2001) 41–54. [23] L. Al-Khatib, Development of (bio)fouling resistant membranes for water treatment applications. Ph.D. Dissertation, University of Nottingham, UK, 2006.
116
M.N.A. Seman et al. / Journal of Membrane Science 348 (2010) 109–116
[24] K. Boussu, Influence of membrane characteristics on flux decline and retention in nanofiltration, Ph.D. Dissertation, Katholieke Universiteit Leuven, Belgium, 2007. [25] A. Barghetta, F.A. DiGiano, W.P. Ball, Nanofiltration of natural organic matter: pH and ionic strength effects, J. Environ. Eng. 123 (7) (1997) 628–641. [26] M. Tasaka, S. Tamura, N. Takemura, Concentration dependence of electroosmosis and streaming potential across charged membrane, J. Membr. Sci. 12 (1982) 169–182. [27] H.P. Gregor, in: E. Selegny (Ed.), Fixed-Charged Ultrafiltration Membranes. Charged Gels and Membranes. Part I, Reidel, Dordrecht, The Netherlands, 1976, pp. 57–69. [28] S. Hong, M. Elimelech, Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes, J. Membr. Sci. 132 (1997) 159–181.
[29] A.E. Childress, M. Elimelech, Effect of solution chemistry on surface charge of polymeric reverse osmosis and nanofiltration membranes, J. Membr. Sci. 119 (1996) 253–268. [30] A.I. Schäfer, A.G. Fane, T.D. Waite, Nanofiltration of natural organic matter: removal, fouling and the influence of multivalent ions, Desalination 118 (1998) 109–122. [31] H Susanto, M. Ulbricht, High-performance thin-layer hydrogel composite membranes for ultrafiltration of natural organic matter, Water Res. 42 (2008) 2827–2835. [32] K. Ghosh, M. Schnitzer, Macromolecular structures of humic substances, Soil Sci. 129 (5) (1980) 266–276. [33] A.I. Schäfer, A.G. Fane, T.D. Waite, Fouling effects on rejection in the membrane filtration of natural waters, Desalination 131 (2000) 215–224.