inorganic additives

inorganic additives

Applied Surface Science 351 (2015) 715–724 Contents lists available at ScienceDirect Applied Surface Science journal homepage: www.elsevier.com/loca...

3MB Sizes 0 Downloads 11 Views

Applied Surface Science 351 (2015) 715–724

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Preparation and arsenic adsorption assessment of PPESK ultrafiltration membranes with organic/inorganic additives Jingwen Ji a , Yanbin Yun a,∗ , Zhu Zeng a , Ruochen Wang a , Xiaoyan Zheng a , Lihong Deng b,∗ , Chunli Li c a

School of Environmental Science and Engineering, Beijing Forestry University, Beijing 100083, China College of Materials Science and Technology, Beijing Forestry University, Beijing 100083, China c New Technique Centre, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China b

a r t i c l e

i n f o

Article history: Received 13 December 2014 Received in revised form 25 March 2015 Accepted 31 May 2015 Available online 9 June 2015 Keywords: Arsenic adsorption PPESK membrane Nano-TiO2 PEO–PPO–PEO Oxalic acid

a b s t r a c t In this paper, the effects of PPESK concentration, and additives of PEO–PPO–PEO, oxalic acid/PEO–PPO–PEO, nano-TiO2 /oxalic acid/PEO–PPO–PEO on membrane performances and arsenic adsorption properties were investigated. Compared with single additive of PEO–PPO–PEO, oxalic acid/PEO–PPO–PEO or nano-TiO2 /oxalic acid/PEO–PPO–PEO had better pore-forming ability. It was found that PPESK membranes with additives of PEO–PPO–PEO or oxalic acid/PEO–PPO–PEO had no arsenic adsorption capability. For optimized PPESK membrane (14.0 wt.% PPESK/1.2 wt.% nano-TiO2 /3.0 wt.% oxalic acid/3.0 wt.% PEO–PPO–PEO), with the contact time increasing, the arsenic adsorption efficiency decreased from 100 to 73.7%; with the increase of operation pressure, the efficiency of arsenic adsorption led a light decrease which ranged from 86.6 to 70.8%; when the arsenic (V) concentration increased from 0 ppb to 500 ppb, the arsenic adsorption efficiency reduced from 100 to 8.3%; the effect of feed temperature on arsenic adsorption was indistinctively; as the pH increasing from 3 to 11, the efficiency of arsenic adsorption first kept stable at 85.0%, and then decreased to 63.4%. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Because arsenic is harmful to the human body, arsenic pollution, as a global concern, has been paid more and more attention. Adsorption technique is the main approach to removing arsenic from water. Some materials, such as TiO2 , magnetic Fe3 O4 , etc. have shown high adsorption capacity for arsenic. However, poisonous by-products will be produced during adsorption process [1]. The reverse osmosis (RO) technology has shown a satisfying arsenic removal performance. But, RO membranes require high operating pressure from 0.30 to 6.90 MPa, the operation fee of RO is rather high [2,3]. Unlike RO, ultrafiltration (UF) removes components through physical sieving at lower pressures. And the water recovery and the flux for UF are higher than those for RO. The combination of polyethersulfone (PES) UF membranes and cationic surfactant cetylpyridinium chloride (CPC) had significant effects on arsenic removal while PES membranes without the surfactant micelles

∗ Corresponding authors. Tel.: +86 10 62336615. E-mail address: [email protected] (Y. Yun). http://dx.doi.org/10.1016/j.apsusc.2015.05.183 0169-4332/© 2015 Elsevier B.V. All rights reserved.

were ineffective for it. Highest arsenic removal, 100%, was achieved for the feed water arsenic concentration of 22 ppb using the PES membrane (MW, 5000 Da) with CPC at pH of 5.5 [4]. However, little information has thrown light on arsenic removal using UF membrane combined with adsorption technology. In this study, UF combined with adsorption technique was put forward as a method to remove arsenic (V). That is to say, adding the arsenic adsorption material (such as, nano-TiO2 ) into the UF membrane could make which had the adsorption ability to arsenic. Due to superior mechanical strength, thermo-stability, chemical resistance and very high glass transition temperature [5], poly (phthalazinone ether sulfone ketone) (PPESK) was chosen as membrane material. The effects of polymer concentration and conventional additives (polyethylene oxide–poly propylene oxide–polyethylene oxide triblock copolymer (PEO–PPO–PEO) and oxalic acid) on membrane performances and arsenic adsorption properties were studied and measured respectively, and the optimized PPESK membrane prescription was confirmed, furthermore, the arsenic removal PPESK UF membrane combined with inorganic additive (nano-TiO2 ) was prepared and evaluated. PEO–PPO–PEO, displayed in Fig. 1, is an amphiphilic organic macromolecule which acts as an attractive modifier to increase the hydrophilicity of membranes. It has reported that PEO–PPO–PEO

716

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

CH3 CH3 O

CH 100

CH3 O

CH3 CH3 O 66

100

CH3 Fig. 1. Molecular structure of PEO–PPO–PEO.

as additive could improve the hydrophilicity of the polysulfone (PSf) and PES UF membranes because of the segregation of PEO segments on the membrane surface [6,7]. Also, as a kind of poreforming agent, PEO–PPO–PEO could increase the water flux. For example, PEO–PPO–PEO could enhance permeation of PES and poly (vinylidene fluoride) UF membranes [7,8]. As a small molecular organic additive, oxalic acid shows the strong pore-forming ability [7–10]. It was found that with the oxalic acid concentration increasing in the casting solution, the water flux of membranes increased and the membrane structure became looser which because the oxalic acid could enhance the porosity of the membrane [7,8]. Also Zhao [9] have reported that when oxalic acid concentration increased, the speed of phase separation increased, and the pore size in the skin layer became larger. Nano-TiO2 shows good adsorption ability to arsenic (V) [11–14]. For example, it was found that arsenic (V) could be adsorbed quantitatively on immobilized nano-TiO2 within a pH range of 5.0–7.5, and the adsorption capacity of immobilized nano-TiO2 was found to be 4.2 mg/g [11]. However, Pena found the arsenic (V) adsorption capacity of nano-TiO2 was more than 0.5 mmol/g [12]. Wang reported that with the nano-TiO2 dosage increasing, the arsenic (V) adsorption increased when arsenic (V) concentration was 500 ppb [13]. In addition, when the feed pH was less than 8, arsenic (V) could be effectively adsorbed by nano-TiO2 [14]. Meanwhile, nano-TiO2 , as pore-forming agent, could improve porosity and hydrophilicity of the membranes. Yang found that nano-TiO2 could result in an increase of pore density and porosity of the PSf UF membrane’s skin layer. As the concentration of nano-TiO2 increased, the permeability and the rejection of the membrane were both enhanced [15]. Also, it was found that with the nano-TiO2 concentration increasing from 0 to 3.0 wt.% in the PPESK UF membranes, the contact angle of membranes decreased from 81.0 to 15.6◦ which shown the remarkable improvement in hydrophilic [16]. In this study, the effects of PPESK concentration, and additives of PEO–PPO–PEO, oxalic acid/PEO–PPO–PEO, nano-TiO2 /oxalic acid/PEO–PPO–PEO on membrane performances and arsenic adsorption properties were studied and measured. For optimized PPESK UF membrane, the effects of contact time, operation pressure, arsenic (V) concentration, feed temperature, pH on efficiency of arsenic adsorption were studied in detail.

2. Experimental 2.1. Materials PPESK (sulfone/ketone, 1/1) was supported by Dalian Polymer New Material, Co., Ltd., Liaoning, China. N-methyl-2pyrrolidone (NMP) as solvent was pure analytical grade and obtained from Guangdong Jinhua Co., Ltd., Guangdong, China. PEO–PPO–PEO, oxalic acid and nano-TiO2 were selected as additives. PEO–PPO–PEO (MW 12,600 Da) was obtained from German. Nano-TiO2 particles (diameter, 5 nm) were bought from Xuancheng Jingrui new material CO., Ltd., Anhui, China. Oxalic acid and bull serum albumin (BSA, MW 67,000 Da) were supported from Beijing Chemistry Co. Ltd., Beijing, China. The arsenic (Na2 HAsO4 ) standard solution (100 ppm) was bought from Yixiu Bogu Biotechnology Co. Ltd., Beijing, China.

Fig. 2. Schematic of the cross-flow filtration setup.

2.2. Membrane preparation PPESK membrane was prepared by steps as follows: PPESK and the additives were dissolved in NMP solvent, and vigorous stirred at 40 ◦ C for 24 h until a homogenous casting solution was formed. After being degassed, the solution would be cast on glass plates and precipitated by being immersed into gelation bath tap-water until membrane formed. To remove all residual solvent, the membrane was moved into a brand-new tap-water bath at room temperature for 24 h. 2.3. Evaluation of membrane performance Under operation pressure of 0.10 MPa and feed temperature of 25 ◦ C, the water flux and BSA rejection of the membranes were performed in a cross-flow manner by the permeation test instrument (Fig. 2). De-ionized water was used to measure the water flux of the membrane (F) which was expressed by F = V/(A × t), where V is the total permeate volume (L); A represents the effective membrane area (m2 ); t is the filtration time (h). 0.05 wt.% BSA were used to test the rejection (R) which was calculated with R = (1 − Cp /Cf ) × 100%, where Cp and Cf are BSA concentration of permeate and feed, respectively. The BSA concentration was determined by a UVspectrophotometer (UNICO-UV2102, China) at the wavelength of 280 nm. The viscosity of the casting solution was measured by the rotary viscosimeter (NDJ-1, Shanghai Changji Co., Ltd., Shanghai, China). The contact angle of membrane was measured by the contact angle meter (Shanghai Zhongchen Digital Technology Apparatus Co., Ltd., Shanghai, China). The cross-sectional micrographs of membranes were observed by the scanning electron microscopy (SEM, Quanta200, FEI Co., Ltd.). Each experiment data had been done with three parallel experiments, the numerical value was taken as the average values. The X-ray diffractometer (XRD7000, Beijing Tianlin Hengtai Science and Technology Co., Ltd., Beijing, China) was used to qualitative analysis the membranes. The experimental conditions were as follows: metallic target using copper, acceleration voltage of 40 kV, electric current of 30 mA, sweep speed of 2◦ /min and the range of scanning angle of 20–70◦ . The gelation velocity of casting solution was determined using an online optical microscope–camera system. This system contains an OPTEC BDS200-PH optical inverted microscope, a 1/2 CMOS color image sensor, a computer and two specially designed microscope slides. The optical system allows capturing 30 frames s−1 and the magnification ranges from 40 to 1200 times. The temperature of casting solution and nonsolvent were controlled at the room temperature (20 ◦ C). Two specially designed microscope slides were used to observe the gelation process. After the microscope was adjusted, a drop of the casting solution was placed between two glass slides, and drops of the de-ionized water (DI water) as the precipitant was introduced from the holes and extra care to ensure the

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

717

flattened casting solution droplet was surrounded by the DI water, then the precipitation process was initiated. The whole precipitation process was recorded by the online optical microscope–camera system. The professional image processing software was used for processing the resulting images, and the images were analyzed to determine the changes of gelation front movement distance (X) over time. Finally, the curve of gelation distance and time (X − t) was draw, which was called gelation kinetic curves. 2.4. Evaluation of adsorbability on arsenic For nano-TiO2 particles, the arsenic (V) adsorption kinetic and adsorption isotherm were tested. In the adsorption kinetic Part, the arsenic (V) concentration of initial feed was 100 ppb, the temperature was 25 ◦ C and the adsorption time was ranged from 0 to 60 min. In the adsorption isotherm Part, the adsorption time was 60 min, the temperature was 25 ◦ C and the arsenic (V) concentration was ranged from 0 to 5000 ppb. For PPESK membranes, under the operation pressure of 0.10 MPa, temperature of 25 ◦ C, contact time of 60 min and initial arsenic (V) concentration of 50 ppb, the adsorbability on arsenic y was measured using the permeation test instrument (Fig. 2). For optimized PPESK membrane, the effects of contact time, operation pressure, arsenic (V) concentration, temperature, and pH on adsorbability on arsenic (V) were investigated respectively using experimental method of single factor. The efficiency of arsenic adsorption was calculated with Efficiency = (1 − C1 /C0 ) × 100%, where C1 and C0 were the arsenic (V) concentration of membrane permeate (or feed for adsorbed with t min) and feed (or initial feed), respectively. The arsenic (V) concentration was determined by the atomic fluorescence analyzer (AF-610D2, Beijing Ruili Co., Ltd., Beijing, China). Each data point was tested by three parallel experiments.

Fig. 3. Effects of PPESK concentrations on membrane performance and casting solution viscosity.

3. Results and discussion 3.1. Effects of PPESK concentration on membrane properties Fig. 3 shows that when the PPESK concentration increased from 12.0 to 20.0 wt.%, the rejection of BSA increased from 66.6 to 95.8%, and the water flux decreased from 666 to 186 L/m2 h. With the increasing concentration of PPESK, the viscosity of the casting solution increased (Fig. 3) and the gelation rate decreased (Fig. 4), the intermolecular force between PPESK macromolecules was stronger than that between NMP and PPESK [17], the number of the nucleation formed by the polymer increased; the membrane structure shaped by this nucleation became more compact [16], the pore size changed smaller and fluid resistance got bigger. Thus, the water flux decreased and the rejection increased. Through the tests of adsorbability on arsenic, the membranes of different PPESK concentrations showed no adsorbing ability to arsenic (V). By considering the membrane water flux and rejection, 14.0 wt.% of PPESK concentration was selected as the optimal polymer concentration and kept for the following experiments.

gelation rate was so fast that the gelation process completed instantaneously in one or two seconds which was hardly to describe.) and the pore formation of the membrane improved. The number of the membrane surface pore increased, and the distance between pores became smaller which resulted in connecting pores, the pore size became larger, so water flux increased and rejection declined [7]. However, on the other hand, with the increase of PEO–PPO–PEO concentration, the viscosity increased (Fig. 5), the membrane should have thicker skin layer, more compact sublayer

3.2. Effects of PEO–PPO–PEO on membrane properties With the PEO–PPO–PEO concentration increasing from 1.0 to 8.0 wt.%, the water flux and the rejection were basically stable at 210 L/m2 h and 90.0%, respectively (Fig. 5). This could be explained with following two effects. On one hand, when the PEO–PPO–PEO concentration increased, the thermodynamic stability of the casting polymer solution reduced because of the hydrogen bonding force between PEO–PPO–PEO and NMP increasing, thus the phase inversion process accelerated shown in Fig. 6 (When the PEO–PPO–PEO concentration increased from 3.0 to 8.0 wt.%, the

Fig. 4. Effects of PPESK concentration on gelation kinetics curve.

718

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

Fig. 7. Effects of oxalic acid concentrations with 3.0 wt.% PEO–PPO–PEO on membrane performance and casting solution viscosity.

Fig. 5. Effects of PEO–PPO–PEO concentrations on membrane performance and casting solution viscosity.

and smaller membrane pores, so water flux decreased and rejection increased. These two impacts could mutually offset each other. Thus, the water flux and the rejection were stable around a certain value. Compared to PPESK membrane with no additive, PEO–PPO–PEO could decrease the water flux and increase the rejection (Fig. 5), which is different from the results displayed in the previous references [18]. The reason maybe is that the solid content

(PPESK + PEO–PPO–PEO) of the casting solution increased with the addition of PEO–PPO–PEO, which resulted in smaller pores forming. With the PEO–PPO–PEO concentration increasing from 0 to 8.0 wt.%, the contact angle of the membrane declined from 88.6◦ to 55.7◦ (shown in Fig. 5). Since PEO–PPO–PEO is an amphiphilic copolymer, the hydrophobic PPO segments of PEO–PPO–PEO are miscible with PPESK and close to the polymer core, the hydrophilic PEO segments of PEO–PPO–PEO extend from the surface of substrates, providing a PEO-rich surface and enhancing the membrane wettability [8]. Meanwhile, the results were same with Part 3.1, the arsenic (V) adsorbance of the PPESK membranes with different PEO–PPO–PEO concentrations was zero. As a result, 3.0 wt.% PEO–PPO–PEO was selected as the optimal concentration by contemplating the membrane water flux, rejection and hydrophilic performance. 3.3. Effects of oxalic acid/3.0 wt.% PEO–PPO–PEO on membrane properties

Fig. 6. Effects of PEO–PPO–PEO concentration on gelation kinetics curve.

In this part, all the casting solution included PEO–PPO–PEO whose concentration was 3.0 wt.%. Fig. 7 shows that with the increase of oxalic acid concentration ranged from 0 to 4.0 wt.%, the water flux increased from 219 to 510 L/m2 h, and the rejection decreased from 91.9 to 75.8%. This could emerge the following influences. On one hand, oxalic acid could combine NMP, which decreased the thermodynamic stability of the PPESK solution and increased the viscosity (Fig. 7). On the other hand, oxalic acid could combine water as well by hydrogen bonds, which increased the phase separation velocity displayed in Fig. 8 (As the oxalic acid concentration increased from 2.0 to 4.0 wt.%, since the gelation process completed instantaneously in one or two seconds, the gelation rate was too fast to describe.) and promoted the ability of pore formation [9]. As a result, the pore number of the membrane increased, the distance between pores became smaller which

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

Fig. 8. Effects of oxalic acid concentration with 3.0 wt.% PEO–PPO–PEO on gelation kinetics curve.

resulted in connecting pores, and the pore size became bigger, so the water flux increased and the rejection decreased. PPESK membranes (14.0 wt.% PPESK, 3.0 wt.% PEO–PPO–PEO and different oxalic acid concentrations) had no adsorption ability to arsenic (V), this result was similar to Part 3.1 and Part 3.2. By considering the properties of the membranes, 3.0 wt.% of oxalic acid/3.0 wt.% PEO–PPO–PEO was selected as the optimal additive concentration.

Fig. 9. Effects of nano-TiO2 concentrations with 3.0 wt.% PEO–PPO–PEO and 3.0 wt.% oxalic acid on membrane performance and casting solution viscosity.

719

Fig. 10. Effects of nano-TiO2 concentration with 3.0 wt.% PEO–PPO–PEO and 3.0 wt.% oxalic acid on gelation kinetics curve.

3.4. Effects of nano-TiO2 /3.0 wt.% oxalic acid/3.0 wt.% PEO–PPO–PEO on membrane properties 3.4.1. PPESK membrane characterization Combined the study results above, PEO–PPO–PEO and oxalic acid concentration were both set as 3.0 wt.%. In Fig. 9, with the increase of nano-TiO2 concentration ranged from 0 to 0.4 wt.%, the rejection decreased and the water flux increased, also the gelation rate increased mildly (Fig. 10). As the nano-TiO2 concentration increasing from 0.4 to 1.2 wt.%, the rejection increased from 42.2 to 74.6% while the water flux changed indistinctively which basically

Fig. 11. The X-ray diffraction patterns of nano-TiO2 particles, PPESK membrane with 3.0 wt.% PEO–PPO–PEO and 3.0 wt.% oxalic acid and the optimized PPESK membrane.

720

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

Fig. 12. Cross-sectional SEM micrographs of different magnifications with different nano-TiO2 concentrations and 3.0 wt.% PEO–PPO–PEO and 3.0 wt.% oxalic acid.

kept stable around 610 L/m2 h and viscosity increased slightly (Fig. 9), and the gelation rate was so fast that the gelation process completed instantaneously in one or two seconds and was hardly to depict which resulting in forming smaller membrane pores and increasing the pore number.

Fig. 9 proves that the contact angle basically remained at 76.0◦ with the nano-TiO2 concentration between 0 and 1.0 wt.%, however, the contact angle decreased from 76.0◦ to 60.0◦ as the nano-TiO2 concentration further increasing from 1.0 to 1.2 wt.%. This may because a proper amount of nano-TiO2 particles could

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

721

Fig. 12. (Continued ).

low the surface energy of the casting solution, which resulted in decreasing contact angle of the membranes’ surface. Also, Gao found the same rules [19]. By comparing the X-ray diffraction patterns in Fig. 11, it was proved that nano-TiO2 was assuredly loaded in the optimized PPESK membrane. Also, through SEM images, nano-TiO2 particles

could be seen distinctly (Fig. 12), and it was found that the structures of PPESK membrane cross-section changed indistinctively which were between finger-like and sponge structure with the nano-TiO2 concentration increasing. Through the measurements of adsorbability on arsenic (V), PPESK membranes with nano-TiO2 had the adsorbing ability to arsenic (V). Regarding the properties

722

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

Fig. 13. Adsorption kinetic curve of nano-TiO2 particles. Fig. 15. Effect of contact time on the efficiency of arsenic adsorption.

of membranes, the optimal preparation conditions were: 14.0 wt.% PPESK, NMP as solvent, 3.0 wt.% PEO–PPO–PEO as the first additive, 3.0 wt.% oxalic acid as the second additive, and 1.2 wt.% nano-TiO2 as the third additive. The adsorbability on arsenic (V) of the optimized PPESK membrane was evaluated in detail as follow. 3.4.2. Evaluation of adsorbability on arsenic (V) for optimized PPESK membrane Nano-TiO2 , with small particle diameter and relative large specific area, shows good adsorption ability to arsenic (V) [11–14]. Because the bond between Ti and O has strong polarity, the adsorbed water tends to form hydroxyl by polarization and dissolution. The hydroxyl on nano-TiO2 surface and the oxygen atom in HAsO4 2− incline to form hydrogen bonds which would attract the arsenic (V) closer to the nano-TiO2 surface [20]. Therefore, nano-TiO2 has a large adsorption capacity for arsenic (V) which makes the optimized PPESK membrane has a good adsorption ability to arsenic (V). Fig. 13 shows that the efficiency of arsenic adsorption for nano-TiO2 particles reached the balance of 85.3% in 1 min and then basically kept stable with the adsorption time further increased to 60 min. However, with the arsenic (V) concentration increased from 0 to 5000 ppb, the efficiency of arsenic adsorption for nano-TiO2 particles decreased from 100 to 72.9% (Fig. 14).

Fig. 14. Adsorption isotherm curve of nano-TiO2 particles.

3.4.2.1. Effects of contact time and operation pressure. The initial arsenic (V) concentration of simulated arsenic water sample was 50 ppb, operation pressure was 0.10 MPa and the feed temperature was 25 ◦ C. Fig. 15 shows that from 0 to 20 min, the arsenic adsorption efficiency decreased rapidly from 100 to 80.8%; from 20 to 90 min, the efficiency of arsenic adsorption reduced slightly from 80.8 to 73.7%. This phenomenon can be explained that with the contact time increasing, the arsenic adsorption of nano-TiO2 has tended to be saturated gradually, so the adsorption capacity of arsenic (V) decreased. Fig. 16 shows the effect of operation pressure on arsenic adsorption using 100 ppb feed concentration under the temperature of 25 ◦ C and contact time of 10 min. The efficiency of arsenic adsorption decreased slightly from 86.6 to 70.8% as operation pressure increased from 0.05 to 0.25 MPa. The reason is that with the operation pressure increasing, the permeate velocity increased which resulted in less contact time of arsenic (V) and nano-TiO2 . Thus, the efficiency of arsenic adsorption decreased. 3.4.2.2. Effects of arsenic (V) concentration, temperature and pH. The impacts of arsenic (V) concentration were studied under the operation pressure of 0.10 MPa, the feed temperature of 25 ◦ C and contact time of 10 min. Fig. 17 shows that with the arsenic (V) concentration increasing from 0 to 500 ppb, the efficiency of arsenic adsorption

Fig. 16. Effect of pressure on the efficiency of arsenic adsorption.

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

723

Fig. 17. Effect of arsenic (V) concentration on the efficiency of arsenic adsorption.

Fig. 19. Effect of pH on the efficiency of arsenic adsorption.

reduced from 100 to 8.3%. This can be interpreted that with the arsenic (V) concentration increasing, adsorption capacity of arsenic (V) for quantified nano-TiO2 tended toward saturation rapidly, so efficiency of arsenic adsorption declined. It was found that for 200 ppb arsenic (V) concentration, the efficiency of arsenic (V) adsorption of optimized PPESK membrane (35.1%) was lower than that of nano-TiO2 particles (95.8%). The reason is that the nano-TiO2 was enwrapped by macromolecules in optimized PPESK membrane which resulted in active sites of nano-TiO2 surface decreasing. Hence, the efficiency of arsenic adsorption for optimized PPESK membrane was lower. The influence of feed temperature (20–40 ◦ C) on arsenic adsorption using 100 ppb arsenic (V) concentration under operation pressure of 0.10 MPa and contact time of 10 min was as plotted in Fig. 18. The efficiency of arsenic adsorption basically kept stable at 78.5% with increasing feed temperature. This rule agreed with the influence of arsenic adsorption on temperature for nano-TiO2 particles in Ref. [12]. Under the operation pressure of 0.10 MPa, the temperature of 25 ◦ C and the contact time of 10 min, the effect of pH ranged from 3 to 11 was studied using 100 ppb arsenic (V) concentration. With the pH increasing from 3 to 5, the efficiency of arsenic adsorption basically remained at 85.0%, with the pH further increased to 11, the efficiency of arsenic adsorption decreased to 63.4% (Fig. 19). This could be attributed to the two factors. Firstly, when the feed

pH was in the range of 3–5, arsenic (V) predominately existes as H2 AsO4 − while the HAsO4 2− (whose hydrated radii is larger than that of H2 AsO4 − ) is dominant above pH 5 [21]. Meanwhile, the surface potential of nano-TiO2 is positive when pH is lower than 5 and negative when pH above 5. Thus, relatively higher adsorption of arsenic (V) up to pH 5 could be attributed to the electrostatic attraction between the oxyanions and positively charged surface of nano-TiO2 , whereas the decrease in adsorption when pH above 5 could be attributed to electrostatic repulsion between oxyanions and the negatively charged surface of nano-TiO2 [22]. Secondly, compared with acid solution, acidulous or alkaline solution has more hydroxyls, which could form hydrogen bonds with hydroxyls of nano-TiO2 surface, so the active sites of nano-TiO2 surface combined with more hydroxyls than HAsO4 2− . Hence, the efficiency of arsenic adsorption decreased. 4. Conclusions With the increase of PPESK concentration, the water flux increased and the rejection decreased. For PEO–PPO–PEO, the water flux and the rejection both led an indistinctively change as the PEO–PPO–PEO concentration increasing. For oxalic acid/3.0 wt.% PEO–PPO–PEO, with the oxalic acid concentration increasing, the water flux went up and the rejection diminished. For nanoTiO2 /3.0 wt.% oxalic acid/3.0 wt.% PEO–PPO–PEO, the water flux changed indistinctively and the rejection increased with the nanoTiO2 concentration increasing. It was found that PPESK membranes with additives of PEO–PPO–PEO and oxalic acid/PEO–PPO–PEO had no adsorption capability to arsenic (V). However, nano-TiO2 could adsorb arsenic (V) no matter as single particles or in the optimized PPESK membranes. For the optimized PPESK membrane, (1) with the contact time increasing, the arsenic adsorption efficiency led a decreasing trend; (2) as the arsenic (V) concentration increasing, the efficiency of arsenic adsorption reduced; (3) with the increase of operation pressure, the efficiency of arsenic adsorption led a light decrease; (4) the effect of feed temperature on arsenic adsorption was indistinctive; (5) with the pH increasing, the efficiency of arsenic adsorption first kept stable at strong acid condition, and then decreased slightly at the acidulous and basic conditions. Acknowledgements

Fig. 18. Effect of temperature on the efficiency of arsenic adsorption.

This work was financially supported by “the National Nature Science Foundation of China (Grant No. 21376030)” and “the

724

J. Ji et al. / Applied Surface Science 351 (2015) 715–724

technological innovation project of instrument and equipment function development of Chinese Academy of Sciences (Grant No. Y12A011FF5)”.

[11]

References

[12]

[1] E.O. Kartinen, C.J. Martin, An overview of arsenic removal processes, Desalination 103 (1995) 79–91. [2] E.M. Vrijenhoek, J.J. Waypa, Arsenic removal from drinking water by a loose nanofiltration membrane, Desalination 130 (2000) 265–278. [3] M.C. Shi, An overview of arsenic removal by pressure-driven membrane processes, Desalination 172 (2005) 85–97. [4] H. Gecol, E. Ergican, A. Fuchs, Molecular level separation of arsenic (V) from water using cationic surfactant micelles and ultrafiltration membrane, J. Membr. Sci. 241 (2004) 105–119. [5] Y.Z. Meng, A.S. Hay, X.G. Jian, S.C. Tong, Synthesis of novel poly(phthalazinone ether sulfone ketone)s and improvement of their melt flow properties, J. Appl. Polym. Sci. 66 (1997) 1425–1432. [6] R. Raslan, A.W. Mohammad, Polysulfone/Pluronic F127 blend ultrafiltration membrane: preparation and characterizations, J. Appl. Sci. 10 (2010) 2628–2632. [7] C. Liu, Y.B. Yun, N.J. Wu, Y. Hua, C.L. Li, Effects of amphiphilic additive Pluronic F127 on performance of poly (ether sulfone) ultrafiltration membrane, Desalination Water Treat. 51 (2013) 3776–3785. [8] Y. Jin, Y. Hua, P. Zhang, Y.B. Yun, P.Y. Zhang, C.L. Li, Preparation and characterization of poly (vinylidene fluoride) ultrafiltration membrane with organic and inorganic porogens, Desalination 336 (2014) 1–7; Y.B. Yun, P. Le-Clech, G.X. Dong, D.Z. Sun, Y.L. Wang, P.Y. Qin, Z. Chen, J.D. Li, C.X. Chen, Formation kinetics and characterization of polyphthalazinone ether ketone hollow fiber ultrafiltration membranes, J. Membr. Sci. 389 (2012) 416–423. [9] X. Zhao, S.Y. Liu, C.X. Chen, J.D. Li, The impact of additive oxalic acid to PEK-C microporous membrane, J. Appl. Polym. Sci. 92 (2004) 906–914. [10] P. Liang, R. Liu, Speciation analysis of inorganic arsenic in water samples by immobilized nanometer titanium dioxide separation and graphite furnace

[13] [14]

[15]

[16] [17]

[18]

[19]

[20]

[21]

[22]

atomic absorption spectrometric determination, Anal. Chim. Acta 602 (2007) 32–36. C.Y. Jing, X.G. Meng, S.Q. Liu, S. Baidas, R. Patraju, C. Christodoulatos, G. Korfiatis, Surface complexation of organic arsenic on nanocrystalline titanium oxide, J. Colloid Interface Sci. 290 (2005) 14–21. S.Q. Wang, D.D. Li, L.M. Song, Research on the treatment for wastewater containing arsenic by nano-titanium, Ind. Saf. Environ. Prot. 33 (2007) 14–15. M.E. Pena, G.P. Korfiatis, M. Patel, Adsorption of As (V) and As (III) by nanocrystalline titanium dioxide, Water Res. 39 (2005) 2327–2337. Y.N. Yang, P. Wang, Q.Z. Zheng, Preparation and properties of polysulfone/TiO2 composite ultrafiltration membranes, J. Polym. Sci. B: Polym. Phys. 44 (2006) 879–887. J.B. Li, J.W. Zhu, M.S. Zheng, Morphologies and properties of poly (phthalazinone ether sulfone ketone) matrix ultrafiltration membranes with entrapped TiO2 nanoparticles, J. Appl. Polym. Sci. 103 (2007) 3623–3629. P.P. Zhu, H.Y. Yang, P.S. He, The specialty and meaning of interactions between macromolecular, Polym. Bull. 5 (2002) 73–78. C.L. Lv, Y.L. Su, Y.Q. Wang, X.L. Ma, Q. Sun, Z.Y. Jiang, Enhanced permeation performance of cellulose acetate ultrafiltration membrane by incorporation of Pluronic F127, J. Membr. Sci. 294 (2007) 68–74. W.J. Gao, X.K. Bian, X.F. Lu, L.Q. Shi, H.J. Lin, The influence of Nano TiO2 on different material ultrafiltration membrane structure and performance, Membr. Sci. Technol. 6 (2008) 28–43. C.Y. Jing, X.G. Meng, S.Q. Liu, S. Baidas, R. Patraju, C. Christodoulatos, G.P. Korfiatis., Surface complexation of organic arsenic on nanocrystalline titanium oxide, J. Colloid Interface Sci. 290 (2005) 14–21. W. Fan, Z.Q. Bian, X.H. Tian, J. Zhang, Study on arsenic adsorption performance by sulfhydryl silane modified oxidation graphite, Environ. Chem. 32 (2013) 810–818. C.W. Zhao, S.S. Du, T. Wang, J. Zhang, Z.K. Luan, Arsenic removal from drinking water by self-made PMIA nanofiltration membrane, Adv. Chem. Eng. Sci. 2 (2012) 366–371. N. Deedar, A. Irfan, Q.A. Ishtiaq, Evaluation of the adsorption potential of titanium dioxide nanoparticles for arsenic removal, J. Environ. Sci. 21 (2009) 402–408.