An initial study of EDTA complex based draw solutes

Desalination 378 (2016) 28–36

Contents lists available at ScienceDirect

Desalination journal homepage: www.elsevier.com/locate/desal

An initial study of EDTA complex based draw solutes in forward osmosis process Yuntao Zhao, Yiwei Ren ⁎, Xingzu Wang, Ping Xiao, Enling Tian, Xiao Wang, Jing Li Center of Membrane Technology and Application Engineering, Key Laboratory of Reservoir Aquatic Environment, Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, No. 266 Fangzheng Avenue, Shuitu Hi-tech Industrial Park, Shuitu Town, Beibei District, Chongqing, 400714, China

H I G H L I G H T S • • • •

Zinc, manganese, calcium, and magnesium complexes with EDTA are employed as draw solutions for FO. EDTA complexes are characterized for pH, conductivity, solution viscosity and osmotic pressure. EDTA complexes show superiority in water flux and reverse draw solute flux compared with NaCl. The diluted draw solution can be regenerated using a pressure-driven nano-filtration process.

a r t i c l e

i n f o

Article history: Received 29 May 2015 Received in revised form 28 July 2015 Accepted 6 September 2015 Available online 30 September 2015 Keywords: Forward osmosis Draw solute EDTA complex

a b s t r a c t The selection of appropriate draw solutes is a critical component for the development of FO technologies. Two significant concerns related to draw solution are the draw solute leakage in FO process and the intensive energy consumption in regenerating draw solute. In this study, a series of EDTA complexes (EDTA-MgNa2, EDTA-CaNa2, EDTA-MnNa2, and EDTA-ZnNa2) were systemically investigated as draw solutes for FO. Their characteristics of high solubility in water, moderate molecular size, expanded molecular structure, nontoxicity, low viscosity, and relatively high osmotic pressure can provide favorable FO performance and easy approaches in regeneration, which ensure the suitability of EDTA complexes as a new class of competent draw solutes. All EDTA complexes demonstrated higher water fluxes in PRO mode and much lower salt leakage in both FO and PRO modes when compared with conventional draw solute of NaCl, while EDTA-ZnNa2 possessed the best performance. The NF regeneration of EDTA complexes at relatively high initial concentration of 0.25 M indicated that all NF membranes performed well with the special water flux and rejection rate ranging between 0.96 and 2.0 LMH/ bar, and 96 and 98%. The overall performance proves that the new concept of using EDTA complex based draw solutes in FO process is applicable. © 2015 Elsevier B.V. All rights reserved.

1. Introduction As an emerging membrane separation technology, forward osmosis (FO) has recently received growing attention in the context of clean water scarcity and energy crisis [1–3]. In FO process, clean water permeates from the feed solution (FS) to the draw solution (DS) driven by the osmotic pressure difference across a semi-permeable membrane which serves as a separation medium; meanwhile, other ions or molecules in the FS are rejected by the semi-permeable membrane. Unlike typical pressure-driven membrane processes, FO holds the advantages of low energy consumption, reduced membrane fouling propensity and easy fouling removal due to no or low hydraulic pressure required [4–6]. Consequently, FO process has found wide applications in seawater/brackish water desalination [7–9], treatment ⁎ Corresponding author. E-mail address: [email protected] (Y. Ren).

http://dx.doi.org/10.1016/j.desal.2015.09.006 0011-9164/© 2015 Elsevier B.V. All rights reserved.

of complex and impaired liquid streams [10], power generation [11–13], food processing [14,15], and pharmaceutical concentration [16,17]. To date, most studies on FO have been focused on membrane development [18–20], mass transfer analysis [21,22], process design and optimization [6,8,23], fouling phenomena and control [24,25], potential applications [26–28], and exploration of novel draw solutes [29–31]. Excellent FO membrane and efficient draw solute are two key factors for the further advancement of FO process [32]. However, in contrast to continuous development and even commercialization of FO membrane [33], to find or synthesize suitable draw solutes becomes particularly crucial. The ideal draw solute should possess the following principal traits: high solubility, minimal reverse draw solute leakage, easy regeneration, nontoxicity, low cost, and compatibility with membrane [3,31]. Many different draw solutes have been studied over the past decades. Achilli et al. [34] developed a protocol for the selection of optimal inorganic-based draw solutions for specific FO applications.

Y. Zhao et al. / Desalination 378 (2016) 28–36

Among 14 candidates, CaCl2, KHCO3, MgCl2, MgSO4, NaHCO3, NaCl, and NaSO4 were proved to be promising from the analysis of FO performance and replenishment cost; when taking the potential of scaling into account simultaneously, MgCl2 may be the best draw solute for most water and wastewater applications. NaCl appears to be the most employed draw solute due to its high solubility, low cost and relatively high osmotic potential [35]. However, it is difficult to separate the water from those diluted inorganic salt solutions. McCutcheon et al. [7] have studied ammonium bicarbonate as draw solute for seawater desalination. The draw solute could be recovered easily by distillation at around 60 °C where it decomposed to ammonia and carbon dioxide. This draw solution has been tested and demonstrated in pilot scale desalination processes [36,37]. But it suffered serious reverse diffusion when compared to other draw solutions such as NaCl and MgCl2 solutions [34]. Moreover, a variety of inorganic salts functioned as fertilizers were explored as draw solutes [38]. Despite the fact that the diluted draw solutions can be directly utilized for agricultural irrigation and thus be free from reconcentration, their applications are limited. Meanwhile, the content of final diluted fertilizer draw solution is generally higher than acceptable limits for fertigation, which requires additional process to reduce the fertilizer concentration [39]. Most recently, there has been a significant progress in the development of synthetic materials as draw solutes, such as magnetic nanoparticles, polyelectrolytes, polymer hydrogels, and stimuli-responsive polymer [29,40–42]. Although there is a considerable advancement in regeneration and reverse leakage for those innovative synthetic draw solutes, they suffer problems of poor repeatability, insufficient water flux, inability to generate high osmotic pressures at low viscosity, serious concentration polarization (CP) exacerbated by their low diffusivity, and complexity in the synthesis [3]. When choosing draw solutes, a tradeoff exists between small molecular size to be highly mobile, reduce solution viscosity, and mitigate CP and large molecular size to be easily separated and decrease reverse draw solute flux [32]. The variety of organic compounds and their chemical reactions enables us to screen or freely design organicbased draw solutes to obtain the suitable size. Hau et al. [43] studied EDTA sodium salt as draw solute in hybrid FO–NF process for dewatering high nutrient containing sludge. The results show that EDTA sodium salt had better FO performance relative to common inorganic salt (NaCl) and could be recovered through NF with the highest rejection of 93%. Nevertheless, EDTA sodium salt exhibits good solubility merely at high pH which may be detrimental for membrane. Ge and Chung [44,45] proposed a new class of draw solutes from hydroacid complexes. Superior performance was achieved in terms of high water fluxes and negligible reverse solute fluxes using hydroacid complexes as draw solutes in FO, but its synthesis is relatively complicated. In this work, EDTA complexes with different central metal ions are employed as draw solutes in FO for the first time. The related characteristics, FO performance, regenerability by NF, and FO applications of these EDTA complexes are systematically investigated. EDTA complexes are selected as draw solutes for the following reasons: (i) these compounds give good solubility without adjustment of pH, which is different from EDTA sodium salt; (ii) they are commercially available avoiding the complex synthesis; (iii) their molecular sizes are identified as being moderate (between 358 and 399); (iv) a series of EDTA complexes with different central metal ions help to comprehensively study the relation between properties of complexes and system performance of FO; (v) it is essential to further broaden the exploration of potential draw solutes. 2. Materials and methods

29

98%), and EDTA magnesium disodium (C10H12MgN2Na2O8, 98%) were supplied by Nanchang Changmao Chemical Industry Co., Ltd. (China) and used following the measurement of crystal water content by thermogravimetric analysis. Sodium chloride (NaCl, 99.5%) was obtained from Chengdu KeLong Chemical Co., Ltd. (China) and used as received. Deionized (DI) water with a resistivity of 18.25 MΩ·cm was produced from an ultrapure water system (MolecularΣH2O®, China) All solutions of different concentrations were prepared in deionized water. 2.2. Characterizations of EDTA complexes The pH values of draw solutions at different concentrations were determined using a pH meter (PB-10, Sartorius, China). All electric conductivities were measured with a conductivity meter (DDSJ-308A, Rex Electric Chemical, China) at 25 ± 1 °C. The viscosities of metal– EDTA complex solutions were obtained by a rotary viscometer (LVDVII + Pro, Brookfield, USA) equipped with the enhanced UL Adapter at 25 ± 0.5 °C. The osmotic pressure differences between NaCl and EDTA complex solutions under the same concentrations were reflected by the direction of water flux in FO [8,13,39]. During the comparison, NaCl solution was placed against the active layer of FO membrane and EDTA complex solution with the same concentration flowed on the other side of the membrane. The osmolality of the EDTA complex solutions was measured with a freezing point osmometer (STY-1A, Tianda Tianfa, China), which can be converted to osmotic pressure using van't Hoff equation. 2.3. FO process FO experiments were carried out through a lab-scale system as depicted in Fig. 1. Commercially available thin film composite (TFC) FO membrane from HTI (Hydration Technologies Inc., OR, USA) was employed. The physical and chemical properties of this FO membrane have been presented in various literatures [46,47]. Before testing, the FO membrane was immersed into DI water overnight to ensure that the membrane's porous support layer is fully water saturated [48]. In the experiments, FO membrane was held in a customized cross-flow permeation cell which was designed in a plate-and-frame configuration with a rectangular channel (10 cm in length, 4.5 cm in width and 0.2 cm in height) on each side of the membrane. DI water and simulated seawater (0.5 M NaCl solution) were used as feed solutions. Draw solutions were prepared from EDTA-MgNa2, EDTA-CaNa2, EDTA-MnNa2, EDTA-ZnNa2, and NaCl. The initial volumes of feed solutions and draw solutions were 1000 mL and 500 mL, respectively. Both feed and draw solutions flowed concurrently through respective cell channels at the same flow rates of 6.4 cm/s, which can reduce strain on the suspended membrane [7]. The temperatures of the feed and draw solutions were maintained at 25 ± 1 °C. Two membrane orientations of FO and pressure retarded osmosis (PRO) mode were applied in this work. PRO mode indicates that the draw solution is against the selective layer of FO membrane, while FO mode means the draw solution is on the support layer. The integrity of the FO membrane was assessed at the beginning of each experiment using the draw solution of 0.5 M NaCl to confirm that the water flux and reverse solute flux were in a reasonable range. Each test was measured for 30 min after removing air bubbles in the pipeline. The dilution of draw solution was ignored due to a small ratio of permeate water flow to the overall volume. The water flux, Jw (Lm −2 h−1, abbreviated as LMH) was calculated from the mass change of the draw solution using Eq. (1): Jw ¼

Δm A  Δt  1000

ð1Þ

2.1. Materials Zinc disodium EDTA (C10H12N2Na2O8Zn, 98%), manganese disodium EDTA (C10H12MnN2Na2O8, 98%), EDTA calcium disodium (C10H12CaN2Na2O8,

where Δm (g) is the mass change of the draw solution determined by a digital balance over a given period of time Δt (h), assuming the density of water is 1000 g/L; A is the effective area of FO membrane.

30

Y. Zhao et al. / Desalination 378 (2016) 28–36

Fig. 1. The schematic of bench-scale FO apparatus.

The amount of draw solute leaked reversely into the feed solution (DI water) was measured based on the calibration curve of the conductivity versus the concentration of draw solute. The reverse solute flux, Js (gm−2 h−1, abbreviated as gMH) was obtained according to Eq. (2): Jw ¼

CV A  Δt

ð2Þ

rate of 2.5 m/s, and temperature of 27–28 °C. The solute rejection was measured via Eq. (3): R¼

  CP 1−  100% CF

ð3Þ

where R is the solute rejection, CP (g/L) and CF (g/L) are the solute concentrations in the permeate and feed solution, respectively.

where C (g/L) and V (L) are the reverse solute concentration and feed volume over a predetermined time Δt (h).

3. Results and discussion

2.4. Reconcentration of draw solution by nanofiltration

3.1. Properties of EDTA complexes

The NF experiments were performed with a laboratory scale crossflow filtration unit (FlowMem-0021-HP, China) as presented schematically in Fig. 2. The effective filtration area is 70 cm2. Four commercial NF membranes listed in Table 1 were used, taking into account the molecular weight of EDTA complexes. Prior to the experiments, Each new membrane was soaked in DI water overnight and compacted with DI water at 20 bar until a constant value of flux was obtained [49]. The pure water permeability of the NF membranes was determined via Eq. (1) when using pure water as the feed. The reconcentration of EDTA-ZnNa2 and EDTA-MgNa2 solutions with relatively high concentration of 0.25 M was conducted at an operating pressure of 20 bar, flow

Ethylenediaminetetraacetic acid (EDTA), frequently produced in the form of its alkali metal salts, can act as a haxadentate ligand to form stable complexes with most metal ions. The strong metal–ligand interactions in the metal–EDTA complexes originate from two sources: the donor–acceptor interactions between the six donor atoms (two nitrogens and four oxygens) and metal cation; the electrostatic attractions between the four carboxylate oxygens carrying formally − 1 charge and metal cation [50]. The basic properties of the studied EDTA complexes can be found in Table 2. It is observed that by binding divalent metal ions of Mg2 +, Ca2 +, Mn2 +, and Zn2 +, the four EDTA complexes have some significant improvement in solubility over EDTA

Fig. 2. The schematic of experimental setup for nanofiltration.

Y. Zhao et al. / Desalination 378 (2016) 28–36

31

Table 1 Properties of four commercial NF membranes. Property

Membrane

Pure water permeability a (LMH/bar) Top layer material Feedwater pH range Maximum applied pressure (psi) MWCO (Da) Average MgSO4 rejection

DK (GE)

DL (GE)

NF-90 (DOW)

TS-80 (TriSep)

7.87

10.22

6.68

9.64

Polyamide Polyamide Polyamide Polyamide 2–11 2–11 2–11 2–11 600 psi 600 psi 600 psi 600 psi 150–300 98%

150–300 96%

– N97%

100–200 99%

All other data are from the manufacturers of NF membranes. a The water flux was determined according to experimental analysis.

disodium salt. The feasibility of EDTA sodium salt as draw solute and its recovery by NF from the diluted solution after FO have been proved [43]. The better solubility and slight higher molecular weight of the four EDTA complexes than EDTA sodium salt may enable them as new competent draw solutes accordingly. The degree of dissociation for coordination ion or coordination entity can be quantitatively described by stability constant. The stability of complex ion increases with the rise in stability constant. The stability constants of the four EDTA complexes were obtained according to G. Anderegg [51]. As shown in Table 2, EDTA complex of zinc is most stable among the four investigated complexes. The chemical structures of EDTA complexes and EDTA disodium salt are given in Fig. 3. The representative configuration of metal–EDTA complex is octahedral. This expanded structure may make for a low reverse flux in FO and easy solute regeneration in post-treatment when metal–EDTA complex is used as draw solute. Furthermore, EDTA complexes are nontoxic and appear as diagnostic and therapeutic agents for treating metabolic disorders and disease, like CaNa2EDTA [52]. In conclusion, the basic properties of EDTA complexes, excellent solubility, moderate molecular weight, expanded molecular configuration, and nontoxicity, well meet the demands of draw solute. 3.2. Characteristics for draw solutions of EDTA complexes To further check whether the four EDTA complexes could act as draw solutes, their solution properties with different concentrations were characterized and recorded in Fig. 4, including pH, conductivity, viscosity, and relative osmotic pressure versus NaCl. Fig. 4(a) shows that there is a little effect of concentration on the pH for the four EDTA complexes and the solution pH of EDTA-MgNa2, EDTA-MnNa2, and EDTA-ZnNa2 at different concentrations is similar (pH ~ 6.2), while EDTA-CaNa2 solution is slightly alkaline with pH ~ 7.9. The recommended operational pH ranges of the current commercial CTA and TFC forward osmosis membranes are from 4.0 to 8.0 and 2.0 to 12.0, respectively. Therefore, the FO membrane will not undergo hydrolysis and structure change when it is tested in the draw solutions of the four EDTA complexes, which can ensure the consistent FO performance. As shown in Fig. 4(b), the solution conductivity of the four EDTA complexes is not directly proportional to that of the increment in concentration, and appears to approach plateaus or even decrease Table 2 Basic properties of EDTA complexes and EDTA disodium salt. Compound

Molecular weight (g/mol)

Solubility at 20 °C (g/L)

Stability constant

EDTA-MgNa2 EDTA-CaNa2 EDTA-MnNa2 EDTA-ZnNa2 EDTA-Na2

358 374 389 399 336

~900a ~800a ~800a ~1000a 96

8.83 10.6 14.04 16.44 –

a

The data is gained from the manufacturers of metal–EDTA complexes.

Fig. 3. Structures of EDTA disodium salt (left) and metal–EDTA complexes (right).

slightly after an initial increase. It is noted that this is comparable to previous study using the draw solutes of organic salts [53]. This phenomenon may be attributed to the decrease in the dissociation and the increase in solution resistivity when solution concentration rises. In addition, there exists difference in conductivity for the studied EDTA complexes, following the order of EDTA-ZnNa2 N EDTAMnNa2 ~ = EDTA-MgNa2 N EDTA-CaNa2. The higher conductivity may lead to higher osmotic pressure [29]. The electrical conductivity in the metal–EDTA complex solution mainly derives from two sources: the dissociation of metal ion and ligand which is affected by the property of metal ion and external conditions; the dissociation of complex ion and outer coordination sphere. Hence, the measured conductivity of metal–EDTA complex solution results from the joint effects. The viscosity of EDTA-MgNa2, EDTA-CaNa2, EDTA-MnNa2, and EDTA-ZnNa2 solutions at different concentrations is shown in Fig. 4(c). The results present that for each complex, the viscosity increases with an increase in its concentration. At the same concentration, the viscosity of EDTA-MgNa2, EDTA-MnNa2, and EDTA-ZnNa2 is comparable, but lower than that of EDTA-CaNa2. A high viscosity of the draw solution not only leads to high energy consumption for fluid transport through the membrane but also results in severe internal polarization concentration [54]. However, compared with the draw solutes of polyelectrolytes, these EDTA complexes exhibit insignificant viscosity [29]. This may indicate that the adverse effect of viscosity on the FO performance should be reduced when the latter are used as draw solutes. The conductivity of NaCl is generally higher than the investigated EDTA complexes and shows no saturation in higher concentration (Fig. 4(b)). This implies that NaCl may generate higher osmotic pressure in water and perform better in FO than the four EDTA complexes when they are used as draw solutes at the same concentration. This hypothesis would not be confirmed later in osmotic pressure comparisons and FO performance tests under the same operating conditions. Osmosis refers to the spontaneous flow of the water from a semipermeable membrane from a low osmotic pressure solution to a high osmotic pressure solution. The same principle is adopted in the FO process where the draw solution with high osmotic pressure is usually used a driving force. Consequently, based on the direction of water flux in FO process, the comparison of osmotic pressure between feed solution and draw solution is determined, as illustrated in Fig. 5. It is noteworthy that the osmotic pressure range of draw solution is easily obtained via the direction and value of water flux in FO when the osmotic pressure of feed solution can be simply calculated using OLI Stream Analyzer™. This method facilitates the determination of osmotic pressure for new developed draw solution whose experimental properties have yet to be included in the database of OLI system and limit the usefulness of osmometer [55]. Fig. 4(d) shows the water flux in FO mode where the solutions of NaCl and EDTA complexes with equal bulk concentrations serve as the feed solution and draw solution, respectively. It is observed that all the measured water fluxes are positive, indicating that the four EDTA complexes produce higher osmotic pressure than NaCl in water when their molar concentrations are equal. Furthermore, there is

32

Y. Zhao et al. / Desalination 378 (2016) 28–36

Fig. 4. The comparisons of characteristics for draw solutions of the four metal–EDTA complexes: (a) pH, (b) conductivity, (c) viscosity, and (d) relative osmotic pressure versus NaCl reflected by experimental water flux in the FO process.

water flux in FO only when the osmotic pressure difference across the membrane is sufficient to overcome the mass transfer resistance for water movement and concentration polarization. The experiments verify that the solution conductivity is not a good indicator for its osmotic pressure, especially for the draw solutes of different types.

Fig. 5. The relationship between direction of water flux and osmotic pressure difference across an ideal semipermeable membrane modified from [8,13,39], where πD is the osmotic pressure of draw solution and πF is the osmotic pressure of feed solution. The draw solution and feed solution in the same concentration are facing membrane support layer and active layer, respectively. The flux direction from feed side to draw side is supposed to be positive.

As osmotic pressure is considered as the driving force for FO process, the accurate osmotic pressure of EDTA complex solution was also measured in terms of osmolality. As shown in Fig. 6, the osmolality for all complexes is comparable under the same concentration and has an almost linear correlation with concentration from 0.25 M to 0.75 M. The osmolality of the four EDTA complex solutions at the concentration of 1 M is beyond the upper range value of 3000 mOsm/kg for the osmometer. Extrapolating from the trends observed in Fig. 6, the investigated EDTA complex solutions of 1 M may have an osmolality of approximately 3600 mOsm/kg. According to previous studies, 33 wt.% thermoresponsive magnetic nanoparticles and 0.4 g/mL Na+-functionalized carbon quantum dots can produce an osmolality of 2250 mOsm/kg and 3140

Fig. 6. Osmolality of metal–EDTA complex solution as a function of concentration.

Y. Zhao et al. / Desalination 378 (2016) 28–36

mOsm/kg, respectively [42,56]. In comparison with these draw solutes of synthetic material, the EDTA complex generates higher osmolality and is commercially available, indicating its great potential as an alternative draw solute. 3.3. FO performance The aforementioned characteristics of good solubility, expanded configuration, and befitting solution properties make EDTA complex a good candidate for draw solute in FO process. The suitability of EDTAMgNa2, EDTA-CaNa2, EDTA-MnNa2, and EDTA-ZnNa2 as draw solutes was evaluated by quantifying the water flux and reverse draw solute flux in FO tests. As NaCl has been extensively employed to be a draw solute, it is chosen as a benchmark for comparison study. Results in Fig. 7 show that the experimental water flux and reverse solute flux as a function of molar concentration in both FO and PRO mode where DI water is employed as the feed solution. It is seen that, for EDTA complex and NaCl, the water flux increases with increasing solute concentration under both FO and PRO mode. This is directly related to the fact that the higher draw solute concentration can generate greater osmotic driving force for water transport through the membrane. It is also observed that the water flux in the PRO mode consistently outperforms that in the FO mode at the same concentration. This can be attributed to the occurrence of the dilutive internal concentration polarization (ICP) under FO mode, which is a primary obstacle to use asymmetric membrane for osmotic process [57]. In FO mode, water permeating through the membrane will dilute the draw solution in the porous supporting layer, significantly reducing the available osmotic driving force. Fig. 7(a) shows that NaCl produce higher water fluxes than the four

33

EDTA complexes under the FO mode, and the correlation between molar concentration and water fluxes for NaCl is almost linear whereas water fluxes of EDTA complexes increase slowly beyond 0.5 M. It is supposed that the draw solutes of larger molecular weight will suffer from more severe ICP in FO mode because of their relatively low diffusivity and thus cannot generate higher water flux [57,58]. To further improve their performance under FO mode, the novel FO membrane with reduced ICP effects still needs to be worked out in the future study. All the water fluxes induced by the four EDTA complexes are better than those by NaCl under the PRO mode (Fig. 7(b)), which is quite different from the results in the FO mode. This is yet consistent with the observation in their osmotic pressure comparisons discussed in Section 3.2. Since the DI water is used as feed solution, there only exists dilutive external concentration polarization (ECP) on the draw solution side under the PRO mode, and its adverse effects can be mitigated or even eliminated by a certain crossflow velocity of draw solution [59]. Therefore, the higher osmotic pressure of metal–EDTA complex leads to a higher water flux. Besides, EDTA-ZnNa2 appears to exhibit the highest pure water flux followed by EDTA-MnNa2, EDTAMgNa2, and EDTA-CaNa2 (Fig. 7(a–b)). Their viscosity and conductivity could be the contributing factors to the efficiency of draw solution. As shown in Fig. 7(c–d), the reverse solute flux also increases when draw solute concentration increases from 0.25 M to 1 M, which is mainly due to the increased concentration gradient across the membrane. For the four EDTA complexes, the increment of reverse solute flux with the increase in solute concentration is smaller than NaCl in both FO and PRO modes. And the four EDTA complexes have significantly lower reverse fluxes than NaCl under the same conditions. This could be due to the larger molecular size of metal–EDTA complex than that

Fig. 7. The performance comparison of metal-EDTA complex and NaCl draw solutes: (a) water flux in FO mode (b) water flux in PRO mode (c) reverse solute flux in FO mode (d) reverse solute flux in PRO mode.

34

Y. Zhao et al. / Desalination 378 (2016) 28–36

of NaCl and its expanded octahedral configuration when metal center coordinates with EDTA ligand. The above two characteristics results in a large barrier for the reverse diffusion of draw solute across the FO membrane. This reveals that the reverse solute leakage in FO, the replenishment cost to maintain a constant concentration, and the feed contamination are reduced when employing the four EDTA complexes as draw solutes. Moreover, Fig. 7(c) and (d) show that the four EDTA complexes give comparable reverse solute flux at the same conditions. This could be explained by their similar molecular weight (between 358 and 399) and the same molecular structure. For a specific comparison in economic perspective, the FO operating cost and specific cost of draw solution were summarized in Table 3. The operating cost was calculated as the product of specific reverse solute flux (Js/Jw) (which represents draw solute loss per volume of water permeation through FO membrane) and the unit cost of draw solute. As shown in Table 3, the FO operating cost for EDTA complex is remarkably lower than that of NaCl. Although much more draw solute of EDTA complex is needed to express a comparable osmotic pressure of around 1500 mOsm/kg with NaCl, which results in higher specific cost, a more important cost associated with the draw solution is the operating cost. On the basis of the above results, we summarize that the four EDTA complexes are well suited as draw solutes and show the advantages of comparable or higher water flux and much lower reverse solute flux over NaCl in FO process.

3.4. Recovery of metal–EDTA complex draw solution via NF Considering the large molecular weight and expanded configuration of the four metal–EDTA complexes, they can be readily separated from water with a relatively larger pore size membrane than that of RO. The recovery convenience of metal–EDTA complex draw solution was studied through a pressure-driven NF process. Fig. 8 displays the water fluxes and solute rejections for different kinds of NF membranes using 0.25 M EDTA-ZnNa2 (molecular weight of 399) and EDTAMgNa2 (molecular weight of 358) as diluted draw solutions. The water transfer in FO process occurs until the point at which the osmotic pressure of the draw solution reaches equilibrium with the feed solution [60]. Once osmotic equilibrium is attained, the water flux becomes zero and thus further dilution of the draw solution beyond this equilibrium concentration is not possible. Consequently, adopting the diluted draw solution with relatively high initial concentration of 0.25 M as the NF feed solution accords better with the actual application. From Fig. 8, it can be observed that the four commercial NF membranes perform well with a high solute retention of more than 96% for both EDTA-ZnNa2 and EDTA-MgNa2, while TS-80 achieves the highest rejection of more than 98%. This is because the molecular weights of EDTA-ZnNa2 and EDTA-MgNa2 are larger than the molecular weight cut-off (MWCO) of the four NF membranes and TS-80 has the smallest MWCO. Meanwhile, Fig. 8 shows that DL generates the highest water flux for both EDTA-ZnNa2 and EDTA-MgNa2, followed by DK, TS-80,

Table 3 Concentration (CD), unit cost, specific cost, specific reverse solute flux (Js/Jw), and FO operating cost for each draw solution. Draw solute

CD, g/L

Cost, $/kg

Specific cost, $/L

Js/Jw, g/L

Operating cost, $/m3

NaCl EDTA-MgNa2 EDTA-CaNa2 EDTA-MnNa2 EDTA-ZnNa2

43.8 179 187 194.5 199.5

1.6a 3.9a 3.6a 3.7a 3.7a

0.07 0.70 0.67 0.72 0.74

0.90 0.12 0.09 0.11 0.08

1.44 0.47 0.32 0.41 0.30

The specific cost is defined as the cost of solute needed to produce 1 L of draw solution. a The unit cost is from the respective manufacturer.

Fig. 8. The recovery of (a) EDTA-ZnNa2 and (b) EDTA-MgNa2 with different kinds of NF membranes.

and NF-90. This indicates that DL is the loosest membrane and the production of permeate across the NF membrane was easier. It is noted that under the same conditions, the permeate water flux for EDTA-MgNa2 is generally higher than that for EDTA-ZnNa2. This behavior could be attributed to the difference in feed osmotic pressure, causing a difference in driving force accordingly at the fixed hydraulic pressure of NF process. In addition, since the recovery of EDTA sodium salt from its diluted draw solution via NF has been established, it was used a benchmark to evaluate the NF recovery performance of metal–EDTA complex. According to Hau et al. [43], when adopting 0.07 M EDTA sodium salt as NF feed solution, the specific water flux and rejection rate were between 0.7 and 1.0 LMH/bar, and 80 and 93%, respectively. The NF recovery of 0.25 M EDTA-ZnNa2 and EDTA-MgNa2, by contrast, has a better performance with the special water flux and rejection rate ranging between 0.96 and 2.0 LMH/bar, and 96 and 98%. To improve the recovery efficiency, other recycles systems such as membrane distillation should be further explored. 3.5. Application of metal–EDTA complex in FO In view of the best FO performance among the four EDTA complexes, EDTA-ZnNa2 was chosen as the draw solute to study seawater desalination by using model seawater as the feed solution. As shown in Fig. 9, the water flux increases with increasing EDTA-ZnNa2 concentration and the PRO mode always gives a higher water flux in comparison to the FO mode, which is consistent with the observation using DI water as the feed solution. However, it is noted that the water flux decreases

Y. Zhao et al. / Desalination 378 (2016) 28–36

35

Acknowledgments The authors would like to thank Chongqing Commission of Science and Technology for funding this research project with a Grant Number of cstc2012ggC20001 entitled “The development and application of high-throughput nanofibers forward osmosis membranes and membrane modules”. The research was also funded by the National Natural Science Foundation of China for the project entitled, “Research on the anti-corrosion, anti-heat mechanism and air filtration performance of graphene oxide enhanced aramid nanofibers” (No. 51478452). References

Fig. 9. Water fluxes under (a) FO and (b) PRO modes using EDTA-ZnNa2 (0.5 M, 0.75 M, and 1 M) as the draw solution and simulated seawater (0.5 M NaCl) as the feed solution.

considerably when simulated seawater replaces DI water as the feed solution, as observed elsewhere [29,54]. This is due to the fact that the osmotic pressure of model seawater at the feed side greatly reduces the net driving force for water transfer. Simultaneously, on the basis of the results in Fig. 4(d), it is concluded that the application of metal-EDTA complex in FO to desalinate seawater has certain feasibility.

4. Conclusions In this work, zinc, manganese, calcium, and magnesium EDTA complexes were initially evaluated as draw solutes in FO process. Their characteristics of high solubility in water, moderate molecular size, expanded molecular configuration, nontoxicity, low viscosity, relatively high osmotic pressure not only ensure favorable FO performance, but also provide relatively easy reconcentration. FO experiments prove that compared with conventional inorganic salt NaCl, the four EDTA complexes show superiority in terms of comparable or higher water fluxes and much lower reverse solute leakages. The NF regeneration of metal-EDTA complex from its diluted draw solution was conducted via a nano-filtration process. The results indicate that the commercial NF membranes (DK, DL, TS-80, and NF-90) perform well with the special water flux and rejection rate ranging between 0.96 and 2.0 LMH/bar, and 96 and 98%. Moreover, the feasibility of the application of metal-EDTA complex in forward osmosis desalination was verified by using model seawater as feed solution. This work may further broaden the exploration of potential draw solutes.

[1] T.Y. Cath, A.E. Childress, M. Elimelech, Forward osmosis: principles, applications, and recent developments, J. Membr. Sci. 281 (2006) 70–87. [2] S.F. Zhao, L. Zou, C.Y.Y. Tang, D. Mulcahy, Recent developments in forward osmosis: opportunities and challenges, J. Membr. Sci. 396 (2012) 1–21. [3] C. Klaysom, T.Y. Cath, T. Depuydt, I.F.J. Vankelecom, Forward and pressure retarded osmosis: potential solutions for global challenges in energy and water supply, Chem. Soc. Rev. 42 (2013) 6959–6989. [4] R.W. Holloway, A.E. Childress, K.E. Dennett, T.Y. Cath, Forward osmosis for concentration of anaerobic digester centrate, Water Res. 41 (2007) 4005–4014. [5] W.C.L. Lay, T.H. Chong, C.Y.Y. Tang, A.G. Fane, J.S. Zhang, Y. Liu, Fouling propensity of forward osmosis: investigation of the slower flux decline phenomenon, Water Sci. Technol. 61 (2010) 927–936. [6] Y.J. Choi, J.S. Choi, H.J. Oh, S. Lee, D.R. Yang, J.H. Kim, Toward a combined system of forward osmosis and reverse osmosis for seawater desalination, Desalination 247 (2009) 239–246. [7] J.R. McCutcheon, R.L. McGinnis, M. Elimelech, Desalination by ammonia-carbon dioxide forward osmosis: influence of draw and feed solution concentrations on process performance, J. Membr. Sci. 278 (2006) 114–123. [8] S.F. Zhao, L.D. Zou, D. Mulcahy, Brackish water desalination by a hybrid forward osmosis-nanofiltration system using divalent draw solute, Desalination 284 (2012) 175–181. [9] S. Phuntsho, S. Hong, M. Elimelech, H.K. Shon, Forward osmosis desalination of brackish groundwater: meeting water quality requirements for fertigation by integrating nanofiltration, J. Membr. Sci. 436 (2013) 1–15. [10] B.D. Coday, P. Xu, E.G. Beaudry, J. Herron, K. Lampi, N.T. Hancock, T.Y. Cath, The sweet spot of forward osmosis: treatment of produced water, drilling wastewater, and other complex and difficult liquid streams, Desalination 333 (2014) 23–35. [11] K. Saito, M. Irie, S. Zaitsu, H. Sakai, H. Hayashi, A. Tanioka, Power generation with salinity gradient by pressure retarded osmosis using concentrated brine from SWRO system and treated sewage as pure water, Desalin. Water Treat. 41 (2012) 114–121. [12] G. Han, P. Wang, T.S. Chung, Highly robust thin-film composite pressure retarded osmosis (PRO) hollow fiber membranes with high power densities for renewable salinity-gradient energy generation, Environ. Sci. Technol. 47 (2013) 8070–8077. [13] K.L. Lee, R.W. Baker, H.K. Lonsdale, Membranes for power generation by pressureretarded osmosis, J. Membr. Sci. 8 (1981) 141–171. [14] K.B. Petrotos, H.N. Lazarides, Osmotic concentration of liquid foods, J. Food Eng. 49 (2001) 201–206. [15] M. Wong, R.J. Winger, Direct osmotic concentration for concentrating liquid foods, Food Aust. 51 (1999) 200–205. [16] K.Y. Wang, M.M. Teoh, A. Nugroho, T.S. Chung, Integrated forward osmosis–membrane distillation (FO–MD) hybrid system for the concentration of protein solutions, Chem. Eng. Sci. 66 (2011) 2421–2430. [17] M.M. Ling, T.S. Chung, Novel dual-stage FO system for sustainable protein enrichment using nanoparticles as intermediate draw solutes, J. Membr. Sci. 372 (2011) 201–209. [18] N.Y. Yip, A. Tiraferri, W.A. Phillip, J.D. Schiffman, M. Elimelech, High performance thin-film composite forward osmosis membrane, Environ. Sci. Technol. 44 (2010) 3812–3818. [19] S. Zhang, K.Y. Wang, T.S. Chung, H.M. Chen, Y.C. Jean, G. Amy, Well-constructed cellulose acetate membranes for forward osmosis: minimized internal concentration polarization with an ultra-thin selective layer, J. Membr. Sci. 360 (2010) 522–535. [20] F. Lotfi, S. Phuntsho, T. Majeed, K. Kim, D.S. Han, A. Abdel-Wahab, H.K. Shon, Thin film composite hollow fibre forward osmosis membrane module for the desalination of brackish groundwater for fertigation, Desalination 364 (2015) 108–118. [21] W.A. Phillip, J.S. Yong, M. Elimelech, Reverse draw solute permeation in forward osmosis: modeling and experiments, Environ. Sci. Technol. 44 (2010) 5170–5176. [22] J.S. Yong, W.A. Phillip, M. Elimelech, Coupled reverse draw solute permeation and water flux in forward osmosis with neutral draw solutes, J. Membr. Sci. 392 (2012) 9–17. [23] V. Yangali-Quintanilla, Z.Y. Li, R. Valladares, Q.Y. Li, G. Amy, Indirect desalination of Red Sea water with forward osmosis and low pressure reverse osmosis for water reuse, Desalination 280 (2011) 160–166. [24] S. Zou, Y.N. Wang, F. Wicaksana, T. Aung, P.C.Y. Wong, A.G. Fane, C.Y. Tang, Direct microscopic observation of forward osmosis membrane fouling by microalgae: critical flux and the role of operational conditions, J. Membr. Sci. 436 (2013) 174–185. [25] S.R.V. Castrillon, X.L. Lu, D.L. Shaffer, M. Elimelech, Amine enrichment and poly(ethylene glycol) (PEG) surface modification of thin-film composite forward osmosis membranes for organic fouling control, J. Membr. Sci. 450 (2014) 331–339.

36

Y. Zhao et al. / Desalination 378 (2016) 28–36

[26] P.H.H. Duong, T.S. Chung, Application of thin film composite membranes with forward osmosis technology for the separation of emulsified oil–water, J. Membr. Sci. 452 (2014) 117–126. [27] K.L. Hickenbottom, N.T. Hancock, N.R. Hutchings, E.W. Appleton, E.G. Beaudry, P. Xu, T.Y. Cath, Forward osmosis treatment of drilling mud and fracturing wastewater from oil and gas operations, Desalination 312 (2013) 60–66. [28] Y. Dong, Z.W. Wang, C.W. Zhu, Q.Y. Wang, J.X. Tang, Z.C. Wu, A forward osmosis membrane system for the post-treatment of MBR-treated landfill leachate, J. Membr. Sci. 471 (2014) 192–200. [29] Q.C. Ge, J.C. Su, G.L. Amy, T.S. Chung, Exploration of polyelectrolytes as draw solutes in forward osmosis processes, Water Res. 46 (2012) 1318–1326. [30] M.L. Stone, C. Rae, F.F. Stewart, A.D. Wilson, Switchable polarity solvents as draw solutes for forward osmosis, Desalination 312 (2013) 124–129. [31] Q.C. Ge, M.M. Ling, T.S. Chung, Draw solutions for forward osmosis processes: developments, challenges, and prospects for the future, J. Membr. Sci. 442 (2013) 225–237. [32] D.L. Shaffer, J.R. Werber, H. Jaramillo, S.H. Lin, M. Elimelech, Forward osmosis: where are we now? Desalination 356 (2015) 271–284. [33] J. Ren, J.R. McCutcheon, A new commercial thin film composite membrane for forward osmosis, Desalination 343 (2014) 187–193. [34] A. Achilli, T.Y. Cath, A.E. Childress, Selection of inorganic-based draw solutions for forward osmosis applications, J. Membr. Sci. 364 (2010) 233–241. [35] K. Lutchmiah, A.R.D. Verliefde, K. Roest, L.C. Rietveld, E.R. Cornelissen, Forward osmosis for application in wastewater treatment: a review, Water Res. 58 (2014) 179–197. [36] R.L. McGinnis, N.T. Hancock, M.S. Nowosielski-Slepowron, G.D. McGurgan, Pilot demonstration of the NH3/CO2 forward osmosis desalination process on high salinity brines, Desalination 312 (2013) 67–74. [37] Y. Kim, J.H. Lee, Y.C. Kim, K.H. Lee, I.S. Park, S.J. Park, Operation and simulation of pilot-scale forward osmosis desalination with ammonium bicarbonate, Chem. Eng. Res. Des. 94 (2015) 390–395. [38] S. Phuntsho, H.K. Shon, S. Hong, S. Lee, S. Vigneswaran, A novel low energy fertilizer driven forward osmosis desalination for direct fertigation: evaluating the performance of fertilizer draw solutions, J. Membr. Sci. 375 (2011) 172–181. [39] S. Sahebi, S. Phuntsho, J.E. Kim, S. Hong, H.K. Shon, Pressure assisted fertiliser drawn osmosis process to enhance final dilution of the fertiliser draw solution beyond osmotic equilibrium, J. Membr. Sci. 481 (2015) 63–72. [40] D. Li, X.Y. Zhang, G.P. Simon, H.T. Wang, Forward osmosis desalination using polymer hydrogels as a draw agent: influence of draw agent, feed solution and membrane on process performance, Water Res. 47 (2013) 209–215. [41] D.L. Zhao, P. Wang, Q.P. Zhao, N.P. Chen, X.M. Lu, Thermoresponsive copolymerbased draw solution for seawater desalination in a combined process of forward osmosis and membrane distillation, Desalination 348 (2014) 26–32. [42] Q.P. Zhao, N.P. Chen, D.L. Zhao, X.M. Lu, Thermoresponsive magnetic nanoparticles for seawater desalination, ACS Appl. Mater. Interfaces 5 (2013) 11453–11461. [43] N.T. Hau, S.S. Chen, N.C. Nguyen, K.Z. Huang, H.H. Ngo, W.S. Guo, Exploration of EDTA sodium salt as novel draw solution in forward osmosis process for dewatering of high nutrient sludge, J. Membr. Sci. 455 (2014) 305–311.

[44] Q.C. Ge, F.J. Fu, T.S. Chung, Ferric and cobaltous hydroacid complexes for forward osmosis (FO) processes, Water Res. 58 (2014) 230–238. [45] Q.C. Ge, T.S. Chung, Hydroacid complexes: a new class of draw solutes to promote forward osmosis (FO) processes, Chem. Commun. 49 (2013) 8471–8473. [46] X. Jin, Q.H. She, X.L. Ang, C.Y.Y. Tang, Removal of boron and arsenic by forward osmosis membrane: influence of membrane orientation and organic fouling, J. Membr. Sci. 389 (2012) 182–187. [47] P. Zhao, B.Y. Gao, S.P. Xu, J.J. Kong, D.F. Ma, H.K. Shon, Q.Y. Yue, P. Liu, Polyelectrolyte-promoted forward osmosis process for dye wastewater treatment — exploring the feasibility of using polyacrylamide as draw solute, Chem. Eng. J. 264 (2015) 32–38. [48] T.Y. Cath, M. Elimelech, J.R. McCutcheon, R.L. McGinnis, A. Achilli, D. Anastasio, A.R. Brady, A.E. Childress, I.V. Farr, N.T. Hancock, J. Lampi, L.D. Nghiem, M. Xie, N.Y. Yip, Standard methodology for evaluating membrane performance in osmotically driven membrane processes, Desalination 312 (2013) 31–38. [49] T. Bras, V. Guerra, I. Torrado, P. Lourenco, F. Carvalheiro, L.C. Duarte, L.A. Neves, Detoxification of hemicellulosic hydrolysates from extracted olive pomace by diananofiltration, Process Biochem. 49 (2014) 173–180. [50] A. Kovacs, D.S. Nemcsok, T. Kocsis, Bonding interactions in EDTA complexes, J. Mol. Struct. (THEOCHEM) 950 (2010) 93–97. [51] G. Anderegg, Critical survey of stability constants of EDTA complexes, in: G. Anderegg (Ed.), Critical Survey of Stability Constants of EDTA Complexes, Pergamon 1977, pp. 1–36. [52] H.V. Aposhian, M.M. Aposhian, Meso-2,3-dimercaptosuccinic acid — chemical, pharmacological and toxicological properties of an orally effective metal chelating agent, Annu. Rev. Pharmacol. 30 (1990) 279–306. [53] S.K. Yen, F.M. Haja, M.L. Su, K.Y. Wang, T.S. Chung, Study of draw solutes using 2-methylimidazole-based compounds in forward osmosis, J. Membr. Sci. 364 (2010) 242–252. [54] D.L. Zhao, S.C. Chen, P. Wang, Q.P. Zhao, X.M. Lu, A dendrimer-based forward osmosis draw solute for seawater desalination, Ind. Eng. Chem. Res. 53 (2014) 16170–16175. [55] A. Grattoni, G. Canavese, F.M. Montevecchi, M. Ferrari, Fast membrane osmometer as alternative to freezing point and vapor pressure osmometry, Anal. Chem. 80 (2008) 2617–2622. [56] C.X. Guo, D.L. Zhao, Q.P. Zhao, P. Wang, X.M. Lu, Na +-functionalized carbon quantum dots: a new draw solute in forward osmosis for seawater desalination, Chem. Commun. 50 (2014) 7318–7321. [57] J.R. McCutcheon, M. Elimelech, Influence of concentrative and dilutive internal concentration polarization on flux behavior in forward osmosis, J. Membr. Sci. 284 (2006) 237–247. [58] G.T. Gray, J.R. McCutcheon, M. Elimelech, Internal concentration polarization in forward osmosis: role of membrane orientation, Desalination 197 (2006) 1–8. [59] H.M. Zhang, S.Y. Cheng, F.L. Yang, Use of a spacer to mitigate concentration polarization during forward osmosis process, Desalination 347 (2014) 112–119. [60] S. Phuntsho, S. Hong, M. Elimelech, H.K. Shon, Osmotic equilibrium in the forward osmosis process: modelling, experiments and implications for process performance, J. Membr. Sci. 453 (2014) 240–252.