- Email: [email protected]
PSF composite membranes for treatment of wastewater containing oil
Journal of Membrane Science 361 (2010) 113–119
Contents lists available at ScienceDirect
Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci
Study on phosphorylated Zr-doped hybrid silicas/PSF composite membranes for treatment of wastewater containing oil Yuqing Zhang a,b,∗ , Zhenhua Jin a , Yueling Wang a , Ping Cui a a b
School of Chemical Engineering and Technology, Tianjin University, Tianjin, 300072, PR China ARC Centre of Excellence for Functional Nanomaterials, AIBN and School of Engineering, The University of Queensland, Brisbane, 4072, Australia
a r t i c l e
i n f o
Article history: Received 7 March 2010 Received in revised form 29 April 2010 Accepted 1 June 2010 Available online 10 June 2010 Keywords: Phosphorylated Hybrid silica Zr-doped Composite membrane Wastewater containing oil
a b s t r a c t Polysulfone (PSF) membranes are broadly applied in the field of the treatment for wastewater containing oil owing to their good physicochemical stability, resistance to oxidation and chlorine. But they are easy to be contaminated by oil for its hydrophobic property, which limits their application in large scale. To enhance the capability of PSF membrane, such as hydrophilic property, anti-fouling ability and tensile strength, phosphorylated Zr-doped hybrid silica particles (SZP particles) were added to the porous matrix of PSF and a novel composite membrane (SZP/PSF) was prepared through a sol–gel process under optimum preparation conditions. The results of tensile strength and contact angle measurements show that the mechanical strength and hydrophilic property of composite membrane have been enhanced to a large extent respectively. SEM micrographs indicate that composite membrane with the asymmetry structure has both layers of compact layer and porous layer with SZP particles uniformly dispersed in PSF. The result shows that the oil concentration of 0.84 mg/L in permeation meets the standard for wastewater reuse (less than 10 mg/L). It can be concluded that the anti-fouling ability and hydrophilic property of composite membrane are significantly enhanced, and therefore, the novel composite membrane is desirable in the treatment of wastewater containing oil. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Wastewater containing oil produced from industry pollutes the environment and is difficult to be treated. Many conventional methods including gravity settling, dewatering and incineration perform cannot efficiently treat emulsified and soluble oil in wastewater. Membrane technology is currently utilized in treating wastewater containing oil; this is because wastewater containing oil may be effectively treated using membrane technology [1]. However, membranes are easy to be contaminated by oil, which forms an oil-layer on membrane surface [2]. Hence, the enhancement of the hydrophilic and anti-fouling property of PSF membranes has become a focus of many researchers. The method of doping inorganic oxide particles to polymer to prepare organic–inorganic composite membranes is attractive, owing to its simple operating process and preparation technology. Bottino et al. [3] improved membrane’s thermodynamic property, mechanical strength, tenacity and so on by adding nanosilica to poly(vinyl alcohol) membrane. Nunes et al. [4] prepared composite
∗ Corresponding author at: School of Chemical Engineering and Technology, Tianjin University, Tianjin, 300072, PR China. Tel.: +86 22 2789 0470; fax: +86 22 2740 3389. E-mail address: [email protected] (Y. Zhang). 0376-7388/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2010.06.002
membrane with SiO2 nano-dispersed in polyetherimides (PEI) and the resistance of the membrane to compactability has increased to some extent. Zhang and Ding [5] doped A12 O3 to PSF membrane to enhance the hydrophilic property and anti-fouling ability of the membrane. Though the capability of polymer membranes can be enhanced by adding inorganic oxide particles, further enhancement is limited because there are few Lewis acid sites and hydroxide radicals on the surface of stoichiometric monocomponent inorganic oxide particles [6]. Nonstoichiometric inorganic oxide nanoparticles have many point defects inside and lots of exposed hydroxide radicals on the surface, so these nanoparticles show stronger activity in the course of chemical bonding than stoichiometric monocomponent inorganic oxide particles [7,8]; especially when nonstoichiometric inorganic oxide nanoparticles are filled in polymer membranes, the capability of membranes is evidently improved. Recently, Zhang et al. doped small-sized particles such as Ce-doped nonstoichiometric nanosilica, Y-doped nonstoichiometric zirconnia to PSF membrane to prepare composite membranes for enhancing its hydrophilic property and anti-fouling ability [9,10].In this paper, a novel composite membrane (SZP/PSF) was prepared by adding SZP particles to the porous matrix of PSF to enhance the capability of PSF membrane, such as hydrophilic property, anti-fouling ability and tensile strength. The properties of membranes were evaluated by permeability, tensile strength, hydrophilic property and so on. Further
114
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
suring instrument supplied by Micromeritics Corporation, America. The accuracy of measurements is ±0.1 ◦ C.
2. Experimental
2.2.5. Measurement of porosity and pore diameter Round membrane piece with its diameter of 6 cm was weighed accurately with vacuum drying for 8 h at 50 ◦ C. After being immersed in water for 48 h, the same piece was wiped with filter paper and weighed again. Porosity (Pr) was calculated as a function of the membrane weight, which can be described by the following equation:
2.1. Materials and reagents
Pr =
Fig. 1. Shape of specimen for tensile test.
more, these composite membranes could be used to treat wastewater containing oil.
PSF was purchased from Dalian Polysulfone Co., Ltd. and its MW and polydispersity were 84,400 Da and 1.37, respectively. SZP particles with the diameter from 1 to 3 m were prepared in our laboratory. Bovine serum albumin (BSA), bought from Beijing Aobo Star Biotechnology GmbH. Polyethylene glycol with average MW 400 Da (PEG400) was supplied by Tianjin Jinyu Fine Chemical Factory. N,N-dimethylacetamide (DMAC) was obtained from Tianjin Damao Service of Chemical Instruments. The last two kinds of reagents were analytical grade and used as received. 2.2. Preparation and characterization of membranes 2.2.1. Preparation of SZP particles Preparation process of SZP particles is similar to literature [11]. 2.2.2. Preparation of membranes with different casting solutions Preparation process of SZP/PSF membranes is similar to literature; the details are as follows [12]. Firstly, DMAC in a 500 mL flask was heated to 40–50 ◦ C in a water bath. PSF was then added and dissolved with stirring. And then, PEG 400 with a mass ratio of 10% to PSF was added as porogen to promote the yield of pores in the gelation process. After that, SZP particles with a mass ratio of 10% to PSF were added into the mixture. They were mixed with vigorous stirring under the condition of ultrasounds until a homogenous solution was obtained. And then the solution was stilly kept for 24–48 h. Then the solution was poured onto a dense glass plate and cast to form thin films (thickness ca. 0.3 mm) that after a 10 s exposure period in air (20 ◦ C and 60% relatively humidity) were immersed into a water bath at 20 ◦ C. Membranes were leached under running water for at least 2 d prior to being soaked in 30 wt.% glycerin aqueous solution. Finally membranes were stored in de-ionized water containing 1 wt.% formaldehyde to avoid bacteria growth. PSF membranes and nanosilica/PSF membranes were prepared by using the same procedure mentioned above, except for inorganic oxide nanoparticles were doped to casting solutions. Nothing was added to the former casting solution and nanosilica particles instead of SZP particles were added to the latter casting solution. 2.2.3. Measurement of tensile strength Specimen was cut out from a membrane sample which should be level led off. Dogbone-shaped specimen is shown in Fig. 1. The specimen was tested with a M350 almighty material extensometer supplied by Testometric Corporation, England. Operating parameters were: the length of the specimen in gauge section, 25 mm; the width, 4 mm; the thickness, 0.30 mm; applied stroke speed, 15 mm/min; measuring range, 0–3 MPa; test temperature, 21 ◦ C. 2.2.4. Measurement of hydrophilic property Contact angles of membrane samples cast from different casting solutions were measured with a 1501 dynamic contact angle mea-
A=
Ww − Wd × 100% Al 1 2 d 4
(1) (2)
where Ww represents weights of membrane containing water at equilibrium swelling; Wd the weights of membrane at dry state; A the area of the membrane; l the thickness of the membrane; d the average diameter of the membrane; the density of water. Membrane pore diameter was measured by thermoporometry. Thermal effect of water’s liquid–solid transition in membrane was measured by DSC. Assumes that the pore channels in membrane are cylindrical, pore diameter can be calculated by the following formula: rp = 0.68 −
32.33 T
w = −0.155T 2 − 11.39T − 332
(3) (4)
where rp is the pole radius (nm); T the subcooled temperature (◦ C); w is the thermal effects of liquid–solid transformation (J/g). 2.2.6. SEM studies of membrane The composite membrane pieces were thoroughly rinsed by de-ionized water, immersed in 30% glycerin aqueous solution and then dried in air for the SEM analysis. Cross-section samples were obtained by being freeze-fractured in liquid nitrogen to obtain a tidy cross-section and then sputtered with gold. Membrane surfaces and cross-section were observed under a JEOL JSM-6400F scanning electron microscope (SEM). 2.3. Oil–water separation studies The model waste water containing oil (oil-in-water emulsion) was created by machine oil and de-ionized water with vigorous stirring at 3000 rpm speed over 30 min until a homogenous solution was obtained, whereupon the emulsion was prepared with an oil concentration of 80 mg/L. The stability of the emulsion was observed visually over 24 h period and the mixture maintained cloudy, turbid, indicating that oil was in emulsified and soluble condition. Diameter of the used membrane sample was 50 mm, hole diameter of the used membrane sample between 0.1 and 0.2 m, thickness of the used membrane sample 0.2 mm. Membrane evaluation device was made in our laboratory. The conditions for the long term run of the separation experiments was intermittent mode. 2.4. Cleaning and operation studies of membrane In general, the permeation flux of membrane would decline owing to fouling and block, even under optimum operating condition. So it is necessary to clean membrane to remove pollutant and resume the permeate flux as much as possible. To regenerate membranes and resume the permeation flux of membranes, combination of backflushing and chemical cleaning method are adopted in this paper. PSF membrane and SZP/PSF composite membrane (the amount of adding SZP particles 10 wt.%) are compared, and
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119 Table 3 Effect of SZP particles on porosity and pore diameter of membrane.
Table 1 Tensile strength of membranes with different casting solutions. Content of particles (wt.% PSF)
Tensile strength (MPa)
0 10a 5b 10b 15b 20b
1.837 2.095 2.764 3.310 2.987 2.896
a b
115
Remarks: adding silica particles. Adding SZP particles.
Membrane code number
Content of SZP particles (wt.% PSF)
Porosity (%)
Mean skin pore diameter (m)
P0 P1 P2 P3 P4
0 5 10 15 20
30 42 60 58 56
0.11 0.12 0.16 0.14 0.13
Remarks: P0, pure PSF membrane; P1–P4, SZP/PSF membrane; PSF, 12 wt.%; PEG400, 10 wt.%.
the used membranes were backflushed once each hour, 40–140 s period once; and chemical cleaning once every month, 30 min period once. The chemical cleaning process condition is as follows: (1) Clean with 3% caustic soda solution, and then rinse with pure water; (2) Clean with 3% hydrogen chloride solution, then rinse with pure water to gain neutral. The membrane experiment carried out with wastewater of oil concentration of 80 mg/L, at the condition of room temperature and operating pressure of 0.20 MPa. 3. Results and discussion 3.1. Membrane characterization 3.1.1. Analysis of tensile strength As indicated in Table 1, the tensile strength of the membrane increases as the addition amount of the SZP particles increases. It can be explained that the SZP particles with small size and large curvature can effectively decentralize the stress and impact from outside and make the composite membranes have the good capability of anti-impact and the tensile strength. The maximum tensile strength of SZP/PSF composite membrane prepared by the solution containing 10 wt.% SZP particles reaches 3.310 MPa. By contrast, the tensile strength of SZP/PSF composite membrane decreases as the amount of adding SZP particles increases again (the amount of adding SZP particles is more than 10 wt.%), which is caused by particles combination. 3.1.2. Analysis of hydrophilic property As is known when the contact angle between water droplet and the surface of SZP/PSF composite membrane is smaller, the hydrophilic property of SZP/PSF composite membrane is better [13]. The contact angles between water droplet and SZP/PSF composite membranes with different amount of adding SZP particles were tested, and the results are shown in Table 2. It can be easily indicated that the hydrophilic property of composite membrane containing SZP particles increases with increasing of the amount of adding SZP particles. And its hydrophilic property is also bet-
ter than that of composite membrane filled by SiO2 particles. This is because SZP particles with nonstoichiometric structure show stronger activity in the process of chemical bonding as a result of many point defects inside and lots of exposed hydroxide radicals on their surface, especially when SZP particles were filled in PSF membrane, hydrophilic property of the membrane was evidently improved. In Table 2, it can be observed that the contact angle between water droplet and SZP/PSF composite membrane reaches 42.6◦ (minimum) when the amount of adding SZP particles is 10 wt.%. However, the hydrophilic property of SZP/PSF composite membrane decreases when the amount of adding SZP particles is more than 10 wt.%. This is because the amount of hydroxide radicals and exposed hydroxide radicals on the SZP particles’ surface decreases owing to combination between particles. 3.1.3. Membrane porosity and pore diameter analysis Membrane porosity and pore diameter is listed in Table 3. The datum present that, as the amount of SZP particles added into the composite membrane increases, the membrane porosity increases, and bigger than that of the PSF membrane. It reached highest porosity and biggest pore diameter at the amount of adding SZP particles was 10 wt.%. The phenomenon can be explained that, the addition of SZP particles increases crystallinity of PSF, leading to increase of porosity. However, excess addition of SZP particles caused decrease of crystallinity, owing to conglobation of small particles. 3.1.4. Morphological analysis of SZP/PSF composite membrane Figs. 2–5 show SEM micrographs of the cross-section and surface of the SZP/PSF composite membrane (the amount of adding SZP particles 10 wt.%). In Fig. 2, it is observed that SZP/PSF composite membrane has the asymmetric structure with a compact skin layer and porous support substrate. So the skin layer holds back
Table 2 Effect of particles to membrane hydrophilicity. Membrane code number
Content of particles (wt.% PSF)
Contact angle (◦ )
P0 PS P1 P2 P3 P4
0 10 5 10 15 20
80.1 67.5 49.8 42.6 43.8 45.7
Remarks: P0, pure PSF membrane; PS, silica/PSF membrane; P1–P4, SZP/PSF membrane.
Fig. 2. SEM of the cross-section of the composite membrane.
116
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
Fig. 3. SEM of the upper surface of the composite membrane.
Fig. 4. SEM of the bottom surface of the composite membrane.
macromolecular substance while the finger-like substrate layer functions as mechanical supporting layer. Figs. 3 and 4 are the pictures of upper and bottom surface respectively. The white spots in Fig. 3 show that particles are uniformly dispersed inside membrane, and they make the surface of membrane crude. The SZP particles, which have abundant hydrophilic radicals on the surface and strong chemical bonding ability with PSF, effectively improve the intensity, tenacity and hydrophilic property of the composite membrane. A hydrophilic layer formed on the surface of the membrane plays an important role in removing the gel layer, because the SZP particles added in composite membrane have the intense hydrophilic property. Even though gel layer was formed on the surface of the membrane, it could be easily eliminated under the feed rising. Therefore, the membrane fouling is significantly resisted. It can be seen from Fig. 4, the bottom surface of net-like structure indicates that the membrane has mechanical support function. Fig. 5
shows SEM of the cross-section of the composite membrane with different SZP contents: (a) pure PSF membrane; (b) SZP/PSF composite membrane (the amount of adding SZP particles 5 wt.%); (c) SZP/PSF composite membrane (the amount of adding SZP particles 10 wt.%); (d) SZP/PSF composite membrane (the amount of adding SZP particles 20 wt.%). Fig. 5(b) and (c) shows that particles are uniformly dispersed inside membrane, but in Fig. 5(d), conglobation between particles can be observed. 3.2. Analysis for oil–water separation 3.2.1. Effect of operating pressure As the process of membrane is drived by pressure, so a suitable operating pressure has an important influence on the process of membrane. The composite membrane added by SZP particles (the amount of adding SZP particles 10 wt.%) was used to treat wastewa-
Fig. 5. SEM of the cross-section of the composite membrane with different SZP contents.
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
Fig. 6. Relation of permeation flux and operation pressure.
ter containing oil (oil concentration 80 mg/L) at different pressures, and the temperature was fixed at room temperature. As shown in Fig. 6, permeation flux increases with operating pressure increasing. Before reaching pressure of 0.20 MPa, the increase of operating pressure has a positive effect on the increase of permeation flux, but its permeation flux increases little when pressure is higher than 0.20 MPa. If operating pressure is much lower than pressure of 0.20 MPa, it is far more difficult to compensate for the resistance loss along the system flow channel, which reduces the efficiency of the entire membrane system. By contrast, when operating pressure exceeds 0.20 MPa, pore blocking by oil droplets induces the declination of permeation flux for the compressible property of oil droplets. And compaction of membrane reduces permeation flux at the same time. Therefore, the suitable operating pressure can be fixed at 0.20 MPa, when experiment results and cost are taken into consideration. 3.2.2. Effect of operating time Operating time also has significant effect on permeation flux. The composite membrane added by SZP particles (the amount of adding SZP particles 10 wt.%) was used to treat different concentration of wastewater containing oil, under pressure of 0.20 MPa and room temperature. As is shown in Fig. 7, it can be observed that the flux declined rapidly at initial stage, and then it gradually declines and reaches steady state after about 7 h of operation. One of the possible reasons for flux declination is concentration polarization due
117
Fig. 8. Relation of permeation flux and feed temperature.
to the increase of retentive concentration. With carrying on membrane process, retentive concentration on the membrane surface is higher than that of the feed concentration to cause concentration polarization which will form a gel layer on the membrane surface. Another possible reason for flux declination is pore blocking due to the existence of membrane pores impervious for large oil droplets. And pore blocking formed by oil droplets could be the major factor for initial flux declination. 3.2.3. Effect of operating temperature To study the effect of operating temperature to membrane separation process, experiment was carried out with SZP/PSF composite membrane (the amount of adding SZP particles 10 wt.%), treated pure water and wastewater containing oil, under operating pressure of 0.20 MPa. Fig. 8 presents that the flux of pure water increases with operating temperature increasing, while wastewater containing oil flux is not apparently changing as the temperature grows. The viscosity of water declines when heating water, which causes permeation flux increase [14]. On the contrary, high temperature has negative effect on the particle size of oil molecular, owing to interferential forces between oil and surface of membrane increases adsorption pollution. Therefore, suitable operating temperature is room temperature. 3.2.4. Rejection of oil The effect of membrane separation can be learned by the rejection of oil. The membrane experiment was carried out with SZP/PSF composite membrane (with different amount of adding SZP particles), treated wastewater containing oil at the condition of room temperature and operating pressure of 0.20 MPa. Table 4 shows that the percent rejection of oil for membranes increases with the amount of adding SZP particles increasing. The optimum membrane is fixed at 10 wt.% SZP/PSF composite membrane after analysis of hydrophilic property and anti-fouling ability are taken into account. The quality of permeate with oil contamination of Table 4 Analysis results of water samples.
Fig. 7. Relation of permeation flux and operation time at different concentration of wastewater containing oil.
Membrane
Oil concentration in permeation (mg/L)
Oil concentration in feed PSF 5 wt.% SZY/PSF 10 wt.% SZY/PSF 15 wt.% SZY/PSF 20 wt.% SZY/PSF
80.00 5.10 1.25 0.84 0.83 0.82
Oil retention rate (%)
93.63 98.70 98.95 98.96 98.98
118
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
Fig. 9. Relation of permeation flux and backwashing.
Fig. 10. Relation of flux recovery ratio and backwashing time.
0.84 mg/L meets the environment standard for wastewater discharge (<10 mg/L). The results indicate that SZP/PSF membrane has a potential application in treatment of wastewater containing oil. 3.3. Membrane regeneration 3.3.1. Effect of backflushing Backflushing uses gas or liquid as medium to remove the particles and cake produced in holes and on the surface of the membrane for regenerating membrane [15]. It can effectively remove gel layer formed on the surface and channels of the membrane while preventing membrane from being polluted. In this work, wastewater containing oil was backflushed once each hour. After backflushed relation of flux recovery ratio and backwashing time of PSF membrane and SZP/PSF composite membrane were tested and researched. 3.3.1.1. Compare of PSF membrane and SZP/PSF composite membrane. After running several days, permeate flux of membranes tended to decline. After backflushing SZP/PSF composite membrane still keeps a higher permeate flux (about 210 L/m2 h), however, PSF membrane is lower (the flux is about 150 L/m2 h). The results are caused by the hydrophilicity of membrane filled by SZP particles. The results are shown as Fig. 9. 3.3.1.2. Effect of backflushing time for flux recovery. Effect of backflushing time for flux recovery is given in Fig. 10. As can be seen the permeate flux increases as the backflashing time increases. After backflushing for 120 s, the permeation flux of membranes trends to a steady state, and flux recovery ratio can reach almost 99%, so the suitable backflushing time is 120 s. Thereby, all membranes backflushed 120 s each time. 3.3.2. Effect of chemical washing Chemical methods are widely used procedures in membrane cleaning [15]. The effect of chemical washing is related with the type of various cleaning agents, temperature, pH and so on. In this work, the process condition of the paper is used to do chemical cleaning to membranes. Fig. 11 presents that chemical cleaning has a positive effect on resuming permeation flux of membranes and keeping higher flux. After doing chemical cleaning, the permeation flux of the SZP/PSF composite membrane (about 220 L/m2 h) is higher than that of the PSF membrane (the flux is about 150 L/m2 h). This is because the SZP/PSF composite membrane has a good hydrophilicity and antifouling capability.
Fig. 11. Relation of permeation flux and chemical washing.
Table 5 Oil retention rate and content of oil after regeneration. Membrane
Oil concentration in permeation (mg/L)
Oil retention rate (%)
PSF 10 wt.% SZY/PSF
5.21 0.85
91.75 97.96
The used PSF membrane and SZP/PSF composite membrane were backflushed once each hour, 120 s period once; and chemical cleaning once every month, 30 min period once. After cleaning, waste water containing oil (concentration of oil, 80 mg/L) was filtered, and oil retention rate and the content of oil are presented in Table 5. 4. Conclusion SZP/PSF composite membrane with high activity and hydrophilic property can be synthesized by adulterating SZP to PSF membrane. Tensile strength, hydrophilic property, porosity and pore diameter of the composite membrane are improved when SZP particles are adulterated to PSF membrane. SEM micrographs indicate that SZP/PSF composite membrane has the asymmetry structure with compact skin layer and porous substrate. SZP particles are uniformly dispersed in PSF membrane. The optimum operating conditions for oil–water separation are determined: the
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
amount of adding SZP particles to SZP/PSF composite membrane, 10 wt.%; operating pressure, 0.20 MPa; operating temperature, room temperature; operating time 7 h. Backflushing and chemical washing have been used to regenerate membrane. SZP/PSF composite membrane with strong hydrophilic property and antifouling capability has practical application potential in treating wastewater containing oil. Acknowledgments This project is supported by the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry (No. 2009-1341), by the Basic Research of Tianjin Municipal Science and Technology Commission (No. 07JCYBJC00700), by Petro China Innovation Fund (No. 060511-6-3), and by the Program of Introducing Talents of Discipline to Universities (No. B06006). References [1] M. Ebrahimi, D. Willershausen, et al., Investigations on the use of different ceramic membranes for efficient oil-field produced water treatment, Desalination 250 (2009) 991–996. [2] K.K. Malgorzata, M.N. Katarzyna, W. Tjomasz, Analyse of membrane fouling in the treatment of water solutions containing humicacids and mineral salts, Desalination 126 (1999) 179–189.
119
[3] A. Bottino, G. Capannelli, A. Comite, Preparation and characterization of novel porous PVDF-ZrO2 composite membranes, Desalination 146 (2002) 35–40. [4] S.P. Nunes, K.V. Peinemann, K. Ohlrogge, et al., Membranes of poly(ether imide) and nanodispersed silica, J. Membr. Sci. 157 (1999) 219–226. [5] Y.Q. Zhang, J. Ding, Effect of the addition of alumina on performance of polysulfone based membrane, Chin. J. Chem. Eng. 28 (2000) 42–44. [6] G.Y. Hong, Surface chemistry of solid, in: G.Y. Hong (Ed.), Inorganic Solid State Chemistry, Science Press, Beijing, 2002, pp. 226–255. [7] V.P. Zlomanov, A.J. Zavrazhnov, A.V. Davydov, Nonstoichiometry and P–T–x diagrams of binary systems, Intermetallics 11 (2003) 1287–1291. [8] G.G. Libowitz, J.B. Lightstone, Characterization of point defects in nonstoichiometric compounds from thermodynamic considerations, J. Phys. Chem. Solids 28 (1967) 1145–1154. [9] Y.Q. Zhang, L.B. Shan, Z.Y. Tu, Y.H. Zhang, Preparation and characterization of novel Ce-doped nonstoichiometric nanosilica/polysulfone composite membranes, Sep. Purif. Technol. 63 (2008) 207–212. [10] Y.Q. Zhang, P. Cui, et al., Development of a sulfated Y-doped nonstoichiometric zirconia/polysulfone composite membrane for treatment of wastewater containing oil, Sep. Purif. Technol. 70 (2009) 153–159. [11] Y.Q. Zhang, T.D. Du, et al., Study on Ce-doped nonstoichiometric nanosilica for promoting properties of polysulfone membrane, Chem. Eng. Technol. (2010), doi:10.1002/ceat.200900529. [12] Y.Q. Zhang, X.Q. Gao, Y.L. Wang, et al., Study on the build of channels in accurate separation membrane and its selective mechanism, J. Membr. Sci. 339 (2009) 100–108. [13] G. Xu, J.M. Yang, P. Chen, Determination of contact angle of hydrophobic composite membrane, J. Xi’an Inst. Technol. 24 (2004) 65–69. [14] R.V. López, S. Elmaleh, N. Ghaffor, Cross-flow ultrafiltration of hydrocarbon emulsions, J. Membr. Sci. 102 (1995) 55–64. [15] H. Peng, A.Y. Tremblay, Membrane regeneration and filtration modeling in treating oily wastewaters, J. Membr. Sci. 324 (2008) 59–66.
Contents lists available at ScienceDirect
Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci
Study on phosphorylated Zr-doped hybrid silicas/PSF composite membranes for treatment of wastewater containing oil Yuqing Zhang a,b,∗ , Zhenhua Jin a , Yueling Wang a , Ping Cui a a b
School of Chemical Engineering and Technology, Tianjin University, Tianjin, 300072, PR China ARC Centre of Excellence for Functional Nanomaterials, AIBN and School of Engineering, The University of Queensland, Brisbane, 4072, Australia
a r t i c l e
i n f o
Article history: Received 7 March 2010 Received in revised form 29 April 2010 Accepted 1 June 2010 Available online 10 June 2010 Keywords: Phosphorylated Hybrid silica Zr-doped Composite membrane Wastewater containing oil
a b s t r a c t Polysulfone (PSF) membranes are broadly applied in the field of the treatment for wastewater containing oil owing to their good physicochemical stability, resistance to oxidation and chlorine. But they are easy to be contaminated by oil for its hydrophobic property, which limits their application in large scale. To enhance the capability of PSF membrane, such as hydrophilic property, anti-fouling ability and tensile strength, phosphorylated Zr-doped hybrid silica particles (SZP particles) were added to the porous matrix of PSF and a novel composite membrane (SZP/PSF) was prepared through a sol–gel process under optimum preparation conditions. The results of tensile strength and contact angle measurements show that the mechanical strength and hydrophilic property of composite membrane have been enhanced to a large extent respectively. SEM micrographs indicate that composite membrane with the asymmetry structure has both layers of compact layer and porous layer with SZP particles uniformly dispersed in PSF. The result shows that the oil concentration of 0.84 mg/L in permeation meets the standard for wastewater reuse (less than 10 mg/L). It can be concluded that the anti-fouling ability and hydrophilic property of composite membrane are significantly enhanced, and therefore, the novel composite membrane is desirable in the treatment of wastewater containing oil. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Wastewater containing oil produced from industry pollutes the environment and is difficult to be treated. Many conventional methods including gravity settling, dewatering and incineration perform cannot efficiently treat emulsified and soluble oil in wastewater. Membrane technology is currently utilized in treating wastewater containing oil; this is because wastewater containing oil may be effectively treated using membrane technology [1]. However, membranes are easy to be contaminated by oil, which forms an oil-layer on membrane surface [2]. Hence, the enhancement of the hydrophilic and anti-fouling property of PSF membranes has become a focus of many researchers. The method of doping inorganic oxide particles to polymer to prepare organic–inorganic composite membranes is attractive, owing to its simple operating process and preparation technology. Bottino et al. [3] improved membrane’s thermodynamic property, mechanical strength, tenacity and so on by adding nanosilica to poly(vinyl alcohol) membrane. Nunes et al. [4] prepared composite
∗ Corresponding author at: School of Chemical Engineering and Technology, Tianjin University, Tianjin, 300072, PR China. Tel.: +86 22 2789 0470; fax: +86 22 2740 3389. E-mail address: [email protected] (Y. Zhang). 0376-7388/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2010.06.002
membrane with SiO2 nano-dispersed in polyetherimides (PEI) and the resistance of the membrane to compactability has increased to some extent. Zhang and Ding [5] doped A12 O3 to PSF membrane to enhance the hydrophilic property and anti-fouling ability of the membrane. Though the capability of polymer membranes can be enhanced by adding inorganic oxide particles, further enhancement is limited because there are few Lewis acid sites and hydroxide radicals on the surface of stoichiometric monocomponent inorganic oxide particles [6]. Nonstoichiometric inorganic oxide nanoparticles have many point defects inside and lots of exposed hydroxide radicals on the surface, so these nanoparticles show stronger activity in the course of chemical bonding than stoichiometric monocomponent inorganic oxide particles [7,8]; especially when nonstoichiometric inorganic oxide nanoparticles are filled in polymer membranes, the capability of membranes is evidently improved. Recently, Zhang et al. doped small-sized particles such as Ce-doped nonstoichiometric nanosilica, Y-doped nonstoichiometric zirconnia to PSF membrane to prepare composite membranes for enhancing its hydrophilic property and anti-fouling ability [9,10].In this paper, a novel composite membrane (SZP/PSF) was prepared by adding SZP particles to the porous matrix of PSF to enhance the capability of PSF membrane, such as hydrophilic property, anti-fouling ability and tensile strength. The properties of membranes were evaluated by permeability, tensile strength, hydrophilic property and so on. Further
114
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
suring instrument supplied by Micromeritics Corporation, America. The accuracy of measurements is ±0.1 ◦ C.
2. Experimental
2.2.5. Measurement of porosity and pore diameter Round membrane piece with its diameter of 6 cm was weighed accurately with vacuum drying for 8 h at 50 ◦ C. After being immersed in water for 48 h, the same piece was wiped with filter paper and weighed again. Porosity (Pr) was calculated as a function of the membrane weight, which can be described by the following equation:
2.1. Materials and reagents
Pr =
Fig. 1. Shape of specimen for tensile test.
more, these composite membranes could be used to treat wastewater containing oil.
PSF was purchased from Dalian Polysulfone Co., Ltd. and its MW and polydispersity were 84,400 Da and 1.37, respectively. SZP particles with the diameter from 1 to 3 m were prepared in our laboratory. Bovine serum albumin (BSA), bought from Beijing Aobo Star Biotechnology GmbH. Polyethylene glycol with average MW 400 Da (PEG400) was supplied by Tianjin Jinyu Fine Chemical Factory. N,N-dimethylacetamide (DMAC) was obtained from Tianjin Damao Service of Chemical Instruments. The last two kinds of reagents were analytical grade and used as received. 2.2. Preparation and characterization of membranes 2.2.1. Preparation of SZP particles Preparation process of SZP particles is similar to literature [11]. 2.2.2. Preparation of membranes with different casting solutions Preparation process of SZP/PSF membranes is similar to literature; the details are as follows [12]. Firstly, DMAC in a 500 mL flask was heated to 40–50 ◦ C in a water bath. PSF was then added and dissolved with stirring. And then, PEG 400 with a mass ratio of 10% to PSF was added as porogen to promote the yield of pores in the gelation process. After that, SZP particles with a mass ratio of 10% to PSF were added into the mixture. They were mixed with vigorous stirring under the condition of ultrasounds until a homogenous solution was obtained. And then the solution was stilly kept for 24–48 h. Then the solution was poured onto a dense glass plate and cast to form thin films (thickness ca. 0.3 mm) that after a 10 s exposure period in air (20 ◦ C and 60% relatively humidity) were immersed into a water bath at 20 ◦ C. Membranes were leached under running water for at least 2 d prior to being soaked in 30 wt.% glycerin aqueous solution. Finally membranes were stored in de-ionized water containing 1 wt.% formaldehyde to avoid bacteria growth. PSF membranes and nanosilica/PSF membranes were prepared by using the same procedure mentioned above, except for inorganic oxide nanoparticles were doped to casting solutions. Nothing was added to the former casting solution and nanosilica particles instead of SZP particles were added to the latter casting solution. 2.2.3. Measurement of tensile strength Specimen was cut out from a membrane sample which should be level led off. Dogbone-shaped specimen is shown in Fig. 1. The specimen was tested with a M350 almighty material extensometer supplied by Testometric Corporation, England. Operating parameters were: the length of the specimen in gauge section, 25 mm; the width, 4 mm; the thickness, 0.30 mm; applied stroke speed, 15 mm/min; measuring range, 0–3 MPa; test temperature, 21 ◦ C. 2.2.4. Measurement of hydrophilic property Contact angles of membrane samples cast from different casting solutions were measured with a 1501 dynamic contact angle mea-
A=
Ww − Wd × 100% Al 1 2 d 4
(1) (2)
where Ww represents weights of membrane containing water at equilibrium swelling; Wd the weights of membrane at dry state; A the area of the membrane; l the thickness of the membrane; d the average diameter of the membrane; the density of water. Membrane pore diameter was measured by thermoporometry. Thermal effect of water’s liquid–solid transition in membrane was measured by DSC. Assumes that the pore channels in membrane are cylindrical, pore diameter can be calculated by the following formula: rp = 0.68 −
32.33 T
w = −0.155T 2 − 11.39T − 332
(3) (4)
where rp is the pole radius (nm); T the subcooled temperature (◦ C); w is the thermal effects of liquid–solid transformation (J/g). 2.2.6. SEM studies of membrane The composite membrane pieces were thoroughly rinsed by de-ionized water, immersed in 30% glycerin aqueous solution and then dried in air for the SEM analysis. Cross-section samples were obtained by being freeze-fractured in liquid nitrogen to obtain a tidy cross-section and then sputtered with gold. Membrane surfaces and cross-section were observed under a JEOL JSM-6400F scanning electron microscope (SEM). 2.3. Oil–water separation studies The model waste water containing oil (oil-in-water emulsion) was created by machine oil and de-ionized water with vigorous stirring at 3000 rpm speed over 30 min until a homogenous solution was obtained, whereupon the emulsion was prepared with an oil concentration of 80 mg/L. The stability of the emulsion was observed visually over 24 h period and the mixture maintained cloudy, turbid, indicating that oil was in emulsified and soluble condition. Diameter of the used membrane sample was 50 mm, hole diameter of the used membrane sample between 0.1 and 0.2 m, thickness of the used membrane sample 0.2 mm. Membrane evaluation device was made in our laboratory. The conditions for the long term run of the separation experiments was intermittent mode. 2.4. Cleaning and operation studies of membrane In general, the permeation flux of membrane would decline owing to fouling and block, even under optimum operating condition. So it is necessary to clean membrane to remove pollutant and resume the permeate flux as much as possible. To regenerate membranes and resume the permeation flux of membranes, combination of backflushing and chemical cleaning method are adopted in this paper. PSF membrane and SZP/PSF composite membrane (the amount of adding SZP particles 10 wt.%) are compared, and
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119 Table 3 Effect of SZP particles on porosity and pore diameter of membrane.
Table 1 Tensile strength of membranes with different casting solutions. Content of particles (wt.% PSF)
Tensile strength (MPa)
0 10a 5b 10b 15b 20b
1.837 2.095 2.764 3.310 2.987 2.896
a b
115
Remarks: adding silica particles. Adding SZP particles.
Membrane code number
Content of SZP particles (wt.% PSF)
Porosity (%)
Mean skin pore diameter (m)
P0 P1 P2 P3 P4
0 5 10 15 20
30 42 60 58 56
0.11 0.12 0.16 0.14 0.13
Remarks: P0, pure PSF membrane; P1–P4, SZP/PSF membrane; PSF, 12 wt.%; PEG400, 10 wt.%.
the used membranes were backflushed once each hour, 40–140 s period once; and chemical cleaning once every month, 30 min period once. The chemical cleaning process condition is as follows: (1) Clean with 3% caustic soda solution, and then rinse with pure water; (2) Clean with 3% hydrogen chloride solution, then rinse with pure water to gain neutral. The membrane experiment carried out with wastewater of oil concentration of 80 mg/L, at the condition of room temperature and operating pressure of 0.20 MPa. 3. Results and discussion 3.1. Membrane characterization 3.1.1. Analysis of tensile strength As indicated in Table 1, the tensile strength of the membrane increases as the addition amount of the SZP particles increases. It can be explained that the SZP particles with small size and large curvature can effectively decentralize the stress and impact from outside and make the composite membranes have the good capability of anti-impact and the tensile strength. The maximum tensile strength of SZP/PSF composite membrane prepared by the solution containing 10 wt.% SZP particles reaches 3.310 MPa. By contrast, the tensile strength of SZP/PSF composite membrane decreases as the amount of adding SZP particles increases again (the amount of adding SZP particles is more than 10 wt.%), which is caused by particles combination. 3.1.2. Analysis of hydrophilic property As is known when the contact angle between water droplet and the surface of SZP/PSF composite membrane is smaller, the hydrophilic property of SZP/PSF composite membrane is better [13]. The contact angles between water droplet and SZP/PSF composite membranes with different amount of adding SZP particles were tested, and the results are shown in Table 2. It can be easily indicated that the hydrophilic property of composite membrane containing SZP particles increases with increasing of the amount of adding SZP particles. And its hydrophilic property is also bet-
ter than that of composite membrane filled by SiO2 particles. This is because SZP particles with nonstoichiometric structure show stronger activity in the process of chemical bonding as a result of many point defects inside and lots of exposed hydroxide radicals on their surface, especially when SZP particles were filled in PSF membrane, hydrophilic property of the membrane was evidently improved. In Table 2, it can be observed that the contact angle between water droplet and SZP/PSF composite membrane reaches 42.6◦ (minimum) when the amount of adding SZP particles is 10 wt.%. However, the hydrophilic property of SZP/PSF composite membrane decreases when the amount of adding SZP particles is more than 10 wt.%. This is because the amount of hydroxide radicals and exposed hydroxide radicals on the SZP particles’ surface decreases owing to combination between particles. 3.1.3. Membrane porosity and pore diameter analysis Membrane porosity and pore diameter is listed in Table 3. The datum present that, as the amount of SZP particles added into the composite membrane increases, the membrane porosity increases, and bigger than that of the PSF membrane. It reached highest porosity and biggest pore diameter at the amount of adding SZP particles was 10 wt.%. The phenomenon can be explained that, the addition of SZP particles increases crystallinity of PSF, leading to increase of porosity. However, excess addition of SZP particles caused decrease of crystallinity, owing to conglobation of small particles. 3.1.4. Morphological analysis of SZP/PSF composite membrane Figs. 2–5 show SEM micrographs of the cross-section and surface of the SZP/PSF composite membrane (the amount of adding SZP particles 10 wt.%). In Fig. 2, it is observed that SZP/PSF composite membrane has the asymmetric structure with a compact skin layer and porous support substrate. So the skin layer holds back
Table 2 Effect of particles to membrane hydrophilicity. Membrane code number
Content of particles (wt.% PSF)
Contact angle (◦ )
P0 PS P1 P2 P3 P4
0 10 5 10 15 20
80.1 67.5 49.8 42.6 43.8 45.7
Remarks: P0, pure PSF membrane; PS, silica/PSF membrane; P1–P4, SZP/PSF membrane.
Fig. 2. SEM of the cross-section of the composite membrane.
116
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
Fig. 3. SEM of the upper surface of the composite membrane.
Fig. 4. SEM of the bottom surface of the composite membrane.
macromolecular substance while the finger-like substrate layer functions as mechanical supporting layer. Figs. 3 and 4 are the pictures of upper and bottom surface respectively. The white spots in Fig. 3 show that particles are uniformly dispersed inside membrane, and they make the surface of membrane crude. The SZP particles, which have abundant hydrophilic radicals on the surface and strong chemical bonding ability with PSF, effectively improve the intensity, tenacity and hydrophilic property of the composite membrane. A hydrophilic layer formed on the surface of the membrane plays an important role in removing the gel layer, because the SZP particles added in composite membrane have the intense hydrophilic property. Even though gel layer was formed on the surface of the membrane, it could be easily eliminated under the feed rising. Therefore, the membrane fouling is significantly resisted. It can be seen from Fig. 4, the bottom surface of net-like structure indicates that the membrane has mechanical support function. Fig. 5
shows SEM of the cross-section of the composite membrane with different SZP contents: (a) pure PSF membrane; (b) SZP/PSF composite membrane (the amount of adding SZP particles 5 wt.%); (c) SZP/PSF composite membrane (the amount of adding SZP particles 10 wt.%); (d) SZP/PSF composite membrane (the amount of adding SZP particles 20 wt.%). Fig. 5(b) and (c) shows that particles are uniformly dispersed inside membrane, but in Fig. 5(d), conglobation between particles can be observed. 3.2. Analysis for oil–water separation 3.2.1. Effect of operating pressure As the process of membrane is drived by pressure, so a suitable operating pressure has an important influence on the process of membrane. The composite membrane added by SZP particles (the amount of adding SZP particles 10 wt.%) was used to treat wastewa-
Fig. 5. SEM of the cross-section of the composite membrane with different SZP contents.
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
Fig. 6. Relation of permeation flux and operation pressure.
ter containing oil (oil concentration 80 mg/L) at different pressures, and the temperature was fixed at room temperature. As shown in Fig. 6, permeation flux increases with operating pressure increasing. Before reaching pressure of 0.20 MPa, the increase of operating pressure has a positive effect on the increase of permeation flux, but its permeation flux increases little when pressure is higher than 0.20 MPa. If operating pressure is much lower than pressure of 0.20 MPa, it is far more difficult to compensate for the resistance loss along the system flow channel, which reduces the efficiency of the entire membrane system. By contrast, when operating pressure exceeds 0.20 MPa, pore blocking by oil droplets induces the declination of permeation flux for the compressible property of oil droplets. And compaction of membrane reduces permeation flux at the same time. Therefore, the suitable operating pressure can be fixed at 0.20 MPa, when experiment results and cost are taken into consideration. 3.2.2. Effect of operating time Operating time also has significant effect on permeation flux. The composite membrane added by SZP particles (the amount of adding SZP particles 10 wt.%) was used to treat different concentration of wastewater containing oil, under pressure of 0.20 MPa and room temperature. As is shown in Fig. 7, it can be observed that the flux declined rapidly at initial stage, and then it gradually declines and reaches steady state after about 7 h of operation. One of the possible reasons for flux declination is concentration polarization due
117
Fig. 8. Relation of permeation flux and feed temperature.
to the increase of retentive concentration. With carrying on membrane process, retentive concentration on the membrane surface is higher than that of the feed concentration to cause concentration polarization which will form a gel layer on the membrane surface. Another possible reason for flux declination is pore blocking due to the existence of membrane pores impervious for large oil droplets. And pore blocking formed by oil droplets could be the major factor for initial flux declination. 3.2.3. Effect of operating temperature To study the effect of operating temperature to membrane separation process, experiment was carried out with SZP/PSF composite membrane (the amount of adding SZP particles 10 wt.%), treated pure water and wastewater containing oil, under operating pressure of 0.20 MPa. Fig. 8 presents that the flux of pure water increases with operating temperature increasing, while wastewater containing oil flux is not apparently changing as the temperature grows. The viscosity of water declines when heating water, which causes permeation flux increase [14]. On the contrary, high temperature has negative effect on the particle size of oil molecular, owing to interferential forces between oil and surface of membrane increases adsorption pollution. Therefore, suitable operating temperature is room temperature. 3.2.4. Rejection of oil The effect of membrane separation can be learned by the rejection of oil. The membrane experiment was carried out with SZP/PSF composite membrane (with different amount of adding SZP particles), treated wastewater containing oil at the condition of room temperature and operating pressure of 0.20 MPa. Table 4 shows that the percent rejection of oil for membranes increases with the amount of adding SZP particles increasing. The optimum membrane is fixed at 10 wt.% SZP/PSF composite membrane after analysis of hydrophilic property and anti-fouling ability are taken into account. The quality of permeate with oil contamination of Table 4 Analysis results of water samples.
Fig. 7. Relation of permeation flux and operation time at different concentration of wastewater containing oil.
Membrane
Oil concentration in permeation (mg/L)
Oil concentration in feed PSF 5 wt.% SZY/PSF 10 wt.% SZY/PSF 15 wt.% SZY/PSF 20 wt.% SZY/PSF
80.00 5.10 1.25 0.84 0.83 0.82
Oil retention rate (%)
93.63 98.70 98.95 98.96 98.98
118
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
Fig. 9. Relation of permeation flux and backwashing.
Fig. 10. Relation of flux recovery ratio and backwashing time.
0.84 mg/L meets the environment standard for wastewater discharge (<10 mg/L). The results indicate that SZP/PSF membrane has a potential application in treatment of wastewater containing oil. 3.3. Membrane regeneration 3.3.1. Effect of backflushing Backflushing uses gas or liquid as medium to remove the particles and cake produced in holes and on the surface of the membrane for regenerating membrane [15]. It can effectively remove gel layer formed on the surface and channels of the membrane while preventing membrane from being polluted. In this work, wastewater containing oil was backflushed once each hour. After backflushed relation of flux recovery ratio and backwashing time of PSF membrane and SZP/PSF composite membrane were tested and researched. 3.3.1.1. Compare of PSF membrane and SZP/PSF composite membrane. After running several days, permeate flux of membranes tended to decline. After backflushing SZP/PSF composite membrane still keeps a higher permeate flux (about 210 L/m2 h), however, PSF membrane is lower (the flux is about 150 L/m2 h). The results are caused by the hydrophilicity of membrane filled by SZP particles. The results are shown as Fig. 9. 3.3.1.2. Effect of backflushing time for flux recovery. Effect of backflushing time for flux recovery is given in Fig. 10. As can be seen the permeate flux increases as the backflashing time increases. After backflushing for 120 s, the permeation flux of membranes trends to a steady state, and flux recovery ratio can reach almost 99%, so the suitable backflushing time is 120 s. Thereby, all membranes backflushed 120 s each time. 3.3.2. Effect of chemical washing Chemical methods are widely used procedures in membrane cleaning [15]. The effect of chemical washing is related with the type of various cleaning agents, temperature, pH and so on. In this work, the process condition of the paper is used to do chemical cleaning to membranes. Fig. 11 presents that chemical cleaning has a positive effect on resuming permeation flux of membranes and keeping higher flux. After doing chemical cleaning, the permeation flux of the SZP/PSF composite membrane (about 220 L/m2 h) is higher than that of the PSF membrane (the flux is about 150 L/m2 h). This is because the SZP/PSF composite membrane has a good hydrophilicity and antifouling capability.
Fig. 11. Relation of permeation flux and chemical washing.
Table 5 Oil retention rate and content of oil after regeneration. Membrane
Oil concentration in permeation (mg/L)
Oil retention rate (%)
PSF 10 wt.% SZY/PSF
5.21 0.85
91.75 97.96
The used PSF membrane and SZP/PSF composite membrane were backflushed once each hour, 120 s period once; and chemical cleaning once every month, 30 min period once. After cleaning, waste water containing oil (concentration of oil, 80 mg/L) was filtered, and oil retention rate and the content of oil are presented in Table 5. 4. Conclusion SZP/PSF composite membrane with high activity and hydrophilic property can be synthesized by adulterating SZP to PSF membrane. Tensile strength, hydrophilic property, porosity and pore diameter of the composite membrane are improved when SZP particles are adulterated to PSF membrane. SEM micrographs indicate that SZP/PSF composite membrane has the asymmetry structure with compact skin layer and porous substrate. SZP particles are uniformly dispersed in PSF membrane. The optimum operating conditions for oil–water separation are determined: the
Y. Zhang et al. / Journal of Membrane Science 361 (2010) 113–119
amount of adding SZP particles to SZP/PSF composite membrane, 10 wt.%; operating pressure, 0.20 MPa; operating temperature, room temperature; operating time 7 h. Backflushing and chemical washing have been used to regenerate membrane. SZP/PSF composite membrane with strong hydrophilic property and antifouling capability has practical application potential in treating wastewater containing oil. Acknowledgments This project is supported by the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry (No. 2009-1341), by the Basic Research of Tianjin Municipal Science and Technology Commission (No. 07JCYBJC00700), by Petro China Innovation Fund (No. 060511-6-3), and by the Program of Introducing Talents of Discipline to Universities (No. B06006). References [1] M. Ebrahimi, D. Willershausen, et al., Investigations on the use of different ceramic membranes for efficient oil-field produced water treatment, Desalination 250 (2009) 991–996. [2] K.K. Malgorzata, M.N. Katarzyna, W. Tjomasz, Analyse of membrane fouling in the treatment of water solutions containing humicacids and mineral salts, Desalination 126 (1999) 179–189.
119
[3] A. Bottino, G. Capannelli, A. Comite, Preparation and characterization of novel porous PVDF-ZrO2 composite membranes, Desalination 146 (2002) 35–40. [4] S.P. Nunes, K.V. Peinemann, K. Ohlrogge, et al., Membranes of poly(ether imide) and nanodispersed silica, J. Membr. Sci. 157 (1999) 219–226. [5] Y.Q. Zhang, J. Ding, Effect of the addition of alumina on performance of polysulfone based membrane, Chin. J. Chem. Eng. 28 (2000) 42–44. [6] G.Y. Hong, Surface chemistry of solid, in: G.Y. Hong (Ed.), Inorganic Solid State Chemistry, Science Press, Beijing, 2002, pp. 226–255. [7] V.P. Zlomanov, A.J. Zavrazhnov, A.V. Davydov, Nonstoichiometry and P–T–x diagrams of binary systems, Intermetallics 11 (2003) 1287–1291. [8] G.G. Libowitz, J.B. Lightstone, Characterization of point defects in nonstoichiometric compounds from thermodynamic considerations, J. Phys. Chem. Solids 28 (1967) 1145–1154. [9] Y.Q. Zhang, L.B. Shan, Z.Y. Tu, Y.H. Zhang, Preparation and characterization of novel Ce-doped nonstoichiometric nanosilica/polysulfone composite membranes, Sep. Purif. Technol. 63 (2008) 207–212. [10] Y.Q. Zhang, P. Cui, et al., Development of a sulfated Y-doped nonstoichiometric zirconia/polysulfone composite membrane for treatment of wastewater containing oil, Sep. Purif. Technol. 70 (2009) 153–159. [11] Y.Q. Zhang, T.D. Du, et al., Study on Ce-doped nonstoichiometric nanosilica for promoting properties of polysulfone membrane, Chem. Eng. Technol. (2010), doi:10.1002/ceat.200900529. [12] Y.Q. Zhang, X.Q. Gao, Y.L. Wang, et al., Study on the build of channels in accurate separation membrane and its selective mechanism, J. Membr. Sci. 339 (2009) 100–108. [13] G. Xu, J.M. Yang, P. Chen, Determination of contact angle of hydrophobic composite membrane, J. Xi’an Inst. Technol. 24 (2004) 65–69. [14] R.V. López, S. Elmaleh, N. Ghaffor, Cross-flow ultrafiltration of hydrocarbon emulsions, J. Membr. Sci. 102 (1995) 55–64. [15] H. Peng, A.Y. Tremblay, Membrane regeneration and filtration modeling in treating oily wastewaters, J. Membr. Sci. 324 (2008) 59–66.