Phosphate Separation and Recovery from Wastewater by Novel

Article pubs.acs.org/est

Phosphate Separation and Recovery from Wastewater by Novel Electrodialysis Yang Zhang,*,†,⊥ Evelyn Desmidt,‡ Arnaud Van Looveren,† Luc Pinoy,§,† Boudewijn Meesschaert,‡,∥ and Bart Van der Bruggen*,† †

Department of Chemical Engineering, Process Engineering for Sustainable Systems, KU Leuven, W. de Croylaan 46, B-3001 Leuven, Belgium ‡ Department of Industrial Sciences and Technology, Katholieke Hogeschool Brugge-Oostende, Associated to the KU Leuven as Faculty of Industrial Sciences, Zeedijk 101, B-8400 Oostende, Belgium § Department of Industrial Engineering, Laboratory for Chemical Process Technology, KaHo St.-Lieven, Associated to the KU Leuven as Faculty of Industrial Sciences, Technologie Campus, Gebroeders Desmetstraat 1, B-9000 Gent, Belgium ∥ Department of Microbial and Molecular Systems, Centre for Surface Chemistry and Catalysis, KU Leuven, Kasteelpark Arenberg 23, B-3001 Leuven, Belgium ⊥ Separation and Conversion Technology, Flemish Institute for Technological Research (VITO), Boeretang 200, B-2400 Mol, Belgium S Supporting Information *

ABSTRACT: Stimulated by the depletion of phosphate resources, phosphate recovery systems have been studied in recent years. The use of struvite reactors has proven to be an effective phosphate recovery process. However, the struvite reactor effluent still consists of an excessive amount of phosphate that cannot be recovered nor can be directly discharged. In this study, selectrodialysis (SED) was used to improve the efficiency of phosphate recovery from a struvite reactor: SED was implemented in such a way that phosphate from the effluent of an USAB (upflow anaerobic sludge blanket) reactor was transferred to the recycled effluent of a struvite reactor. Prior to the experiments, synthetic water with chloride and phosphate was used to characterize the efficiency of SED for phosphate separation. Results indicate that SED was successful in concentrating phosphate from the feed stream. The initial current efficiency reached 72%, with a satisfying (9 mmol L−1) phosphate concentration. In the experiments with the anaerobic effluent as the phosphate source for enrichment of the effluent of the struvite reactor, the phosphate flux was 16 mmol m−2 h−1. A cost evaluation shows that 1 kWh electricity can produce 60 g of phosphate by using a full scale stack, with a desalination rate of 95% on the feed wastewater. Finally, a struvite precipitation experiment shows that 93% of phosphate can be recovered. Thus, an integrated SED-struvite reactor process can be used to improve phosphate recovery from wastewater.

■

INTRODUCTION

Several studies investigated the phosphorus depletion rate in view of phosphorus mining. The results are ranging from 100 years to 400 years.4,5 It is important to consider that all studies show that there will be a shortage/depletion of phosphate rock, which means that doing nothing is not sustainable. Meanwhile, phosphorus containing sludge and wastewater pollute surface/ groundwater and result in water deterioration and eutrophication. Due to the phosphorus resources depletion and the eutrophication problems to the environment, phosphorus recovery in industrial and municipal waste streams drew more

Phosphorus is one of the essential elements of all living organisms and accounts for around 2−4% of the dry weight of most cells.1 It has several principal roles in cell formation and activity: it combines lipids to make cell membranes, maintains the structure of DNA and RNA, and transports energy within the cell through the molecule adenosine triphosphate (ATP).2 On the other hand, phosphorus is also essential to maintain the development of societies since a huge amount of phosphorus is used in agriculture and in some industries. During the past decade, the demand for phosphorus increased globally. This is due to an increase per capita and overall demand for food and global trends toward more meat- and dairy-based diets. Phosphorus has no substitute in food production and cannot be manufactured or synthesized.3 © XXXX American Chemical Society

Received: January 30, 2013 Revised: May 6, 2013 Accepted: May 7, 2013

A

dx.doi.org/10.1021/es4004476 | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

and more attention during the past decade.3 Phosphorus recovery techniques are regarded as important to close the phosphorus loop in human society. Guest et al. 6 proposed a new paradigm, which focuses on what can be recovered, so that wastewater can be regarded as a renewable resource by using resource recovery systems to selectively extract valuable waste materials or generate energy. Recently, several techniques have been investigated for phosphate recovery from wastewater and other aqueous solutions. In water with a low phosphate concentration, adsorption or membrane filtration is applied;7−9 while in water with a higher phosphate concentration, precipitation and crystallization is used.10−12 One of the most notable and promising techniques is struvite formation and precipitation. Struvite precipitation in wastewater treatment plants was discovered in the early 1960s, and it was regarded as an operational problem.13 At the end of the last century, the potential markets of struvite were explored to be used as a slow release fertilizer.14,15 Precipitation of struvite requires that its components, magnesium, ammonium, and phosphate, are available simultaneously in the wastewater and the pH should be kept above 8.5.16−18 Ma and Rouff19 investigated struvite precipitation by using the same concentration (17 mmol.L−1) of (NH4)2HPO4 and MgCl2 with 0.1 g L−1 commercial struvite as seed. pH values between 8 and 11 were chosen for the experiments, and it was found that the highest struvite precipitation yield was obtained at pH 10. Low influent phosphate-P concentrations (20−30 mg L−1, or lower than 1 mmol L−1) require pH higher than 9.020,21 for recovery. Therefore, phosphate-P recovery through struvite crystallization is probably not a cost efficient process for low phosphate-P concentrations (0.96 >0.93 >0.97

SO3Na NR4Cl N/A

0.75−3 1−1.5 N/A

three cell trios in the stack, with each cell trio containing a feed, a product, and a brine compartment. Three pieces of AM, three pieces of MVA, and four pieces of CM were used in total. All the membranes were produced by PCA GmbH, Germany. The membrane properties provided by the manufacturer are given in Table 1. The total active membrane surface area was 0.0192 m2. The ED-64 004 stack (Figure 3) was supplied by PCCell

GmbH, Germany. A dc adjustable power supply (CNRood 0− 20 V, 10A, Zellik, Belgium) was used to keep a constant current during the experiments. The water composition and volume of initial feed, product, and brine used in the SED experiments is given in Table 2. A 3 L portion of 0.1 M Na2SO4 was used as the electrode rinsing solution (ERS) in both experiments. C



Article

applied current being constant, the current efficiency of ion A (ηA) can be calculated as

Table 2. Initial Feed, Product, and Brine Used in the SED Experiments synthetic water experiment content

wastewater experiment

volume

content

volume

10 L

brine

NaCl 20 mmol L−1

10 L

anaerobic effluent struvite reactor effluent NaCl 8 mmol L−1

20 L

product

NaCl 8 mmol L−1, Na2HPO4 8 mmol L−1 NaCl 30 mmol L−1

feed

3L

z ηA =

ΔmA (t ) F MA

nIt

× 100(%)

(1)

where ΔmA(t) is the weight of transferred ion, z is the charge number of ion A, MA is molar mass of ion A, F is the Faraday constant, I is the applied current, t is the time period, n is the number of cell trios in the ED stack. A more detailed explanation on the current efficiency and the current efficiency of a specific ion species can be found in Zhang et al.26 The normalized current efficiency (N-CE) is introduced to illustrate in particular the efficiency of the current to transfer HPO42‑ from the feed compartment and to retain it in the product compartment, or to exclude Cl− from the product compartment toward the brine. The N-CE of ion A during a time period i (τiA) was calculated as

5L 20 L

The experiments were first carried out on synthetic water, which only contained chloride and phosphate. All the salts in this investigation were analytical grade, and the salt solutions were made by deionized water. The pH in the product compartment was adjusted to 10 by adding 1.0 M NaOH before the experiment started, to improve membrane selectivity toward monovalent anions.32 The pH was monitored and was kept constant by adding 1.0 M NaOH during the experiment. The applied current density was 31.25 A m−2 (0.2 A) and was kept constant during the entire experiment. The anaerobic effluent and struvite reactor effluent were both obtained from Agristo NV, Harelbeke, Belgium, and was filtrated by 0.5 μm cartridge filter for half an hour before use. The flow rates for the different compartments are 30 L h−1. The measurement of the flow rates is accomplished by rotameters for the feed, product, and brine compartments. Samples of 10 mL are taken from the feed, product, and brine every hour. The pH and conductivity measurements are performed on every sample by using a Crison MM40 m, produced by Crison (Spain). Possible adjustment of the pH was done by adding the appropriate amount of 1 M NaOH or HCl. Struvite Reactor Experiment. A schematic diagram on the SED-struvite reactor combined process is shown in Figure 1. In the struvite formation experiment, the struvite reactor was stabilized before carrying out the experiment. The influent and effluent flow rate of the struvite reactor was 0.528 L h−1 with a hydraulic retention time (HRT) of 7.6 h. The anaerobic effluent was used as the feed stream, and the struvite reactor effluent acts as the initial product stream after filtration by a 0.5 μm cartridge filter. In the struvite formation experiment, NaOH and KH2PO4 were added to the anaerobic effluent until a pH of 9 and a phosphate concentration of approx 6.8 mmol.L−1 was reached. During the operation, MgCl2 was added to precipitate struvite. The struvite formation experiment was running for one week to collect enough information on the phosphate recovery and the effluent quality. Sample Analysis and Data Processing. Sample Analysis. Chloride phosphate and carbonate concentrations were measured with ion chromatography (DX-120 Ion chromatography with IONPAC AS11-HC Analytical Column, DIONEX). The sample preparation for IC consisted of diluting the experiment sample with Millipore Milli-Q deionized water according to the expected concentration. A 20 mM NaOH eluent was used for the measurements. Data Processing. The current density (A m−2) was calculated as the applied current (I) divided by the membrane surface area (S). The current efficiency (CE) of ion A was calculated as the ratio of the electrical charge used for the transport of ion A to the total electrical current charge. The

τAi =

zVp(|ΔcAi |)F nIt

× 100(%)

(2)

|ΔciA|

where Vp denotes to the volume of the product, is the absolute value of the concentration difference at a period i of ion A in the product vessel. In this study, i refers to a time interval between two samples. Due to electroneutralization in the product compartment, the concentration differences of Cl− and HPO42‑ follow the correlation i 0 0 − − c −| = 2|c i |cCl − c HPO 2 −| Cl HPO2 − 4

4

That is τCl− = τHPO24−

(3)

(4)

The membrane selectivity, i.e., membrane separation efficiency, was calculated by the method introduced by Van der Bruggen et al.33 In this method, the separation efficiency S between ion A and B is evaluated as SBA(t ) =

(cA(t )/cA(0)) − (c B(t )/c B(0)) (1 − cA(t )/cA(0)) + (1 − (c B(t )/c B(0))

(5)

The range of SAB is from −1 to 1. If ion A is transported slower than ion B, the SAB value is between 0 to 1; if ion B is transported slower, then the SAB is between −1 to 0. In this work, A and B refer to phosphate (HxPO4y‑) and chloride (Cl−), respectively.

■

RESULTS AND DISCUSSION Selectrodialysis of Reference Water. A synthetic water which only contained chloride and phosphate was used as the feed. As shown in Figure 4, the concentration profiles of chloride and phosphate in the diluate are very similar, and the standard anion exchange membrane separation efficiency was between −0.1 to 0.2 during the experiment. The AM membrane was slightly selective to the chloride ions, but the selectivity did not result in an obvious difference to the final concentration of chloride and phosphate: both concentrations were reduced to around 0.5 mmol L−1. Figure 5 shows the concentration profile of chloride and phosphate in the product stream, the normalized current efficiency of chloride, and the MVA membrane selectivity as a function of time. It can be seen that, unlike the AM membrane, the MVA membrane had a high selectivity toward chloride ions, and the separation efficiency ranged from 0.5 to 0.8 during the D



Article

Figure 4. Concentration profiles of chloride and phosphate, and standard anion exchange membrane selectivity as a function of time.

Figure 6. Concentration profile of chloride and phosphate in the brine stream and the normalized current efficiency to transport chloride and phosphate to the brine.

On the other hand, the current efficiency to transport phosphate has a different profile: the value was only 6% in the first hour, but it increased as a function of time until 6 h, with the value of 36%; finally it dropped to 17% in the last hour. The different profile of the current efficiency of chloride and the current efficiency of phosphate correspond with the depletion of chloride and the accumulation of phosphate in the product stream. During the first 5−6 h, in the product stream, the chloride ions were depleted and were replaced by the phosphate ions. Thus, much more chloride than phosphate was transported to the brine stream during the initial period. On one hand, both chloride and phosphate were depleting in the feed stream as the experiment went on; on the other hand, because of the depletion of chloride and the accumulation of phosphate in the product stream, more phosphate became available, and thus, the current efficiency of chloride decreased and the current efficiency of phosphate increased as a function of time. Another important parameter is the overall current efficiency. In this experiment, the overall current efficiency is the total of the current efficiency of chloride and of phosphate since the chloride and the phosphate were the only anions. As can be seen, the overall current efficiency was around 100% during the first 300 min, and then decreased to 78% after 360 min. At the final stage of the experiment, the overall current efficiency decreased to a low level due to the ion depletion in the feed stream, and water splitting was getting more severe (the applied current was higher than the limiting current). Selectrodialysis of Struvite Reactor Effluent. SED was performed using an industrial anaerobic effluent as the feed and the phosphate source while a struvite reactor effluent was used as the initial product and the phosphate acceptor. Besides phosphate and chloride, around 21 and 27 mmol L−1 of bicarbonate was present in the feed and initial product, respectively. The experiment was operated for 62 h to obtain a maximum concentration of phosphate in the product stream. Figure 7 shows the concentration profile of (a) phosphate, (b) chloride, and (c) bicarbonate as a function of time. It can be seen from Figure 7a that the phosphate concentration in the product increased over 7 times, from 0.93 to 6.64 mmol L−1 after 62 h. Considering that a good

Figure 5. Concentration profile of chloride and phosphate in the product stream, the current efficiency of chloride, and the MVA membrane selectivity.

experiment. As a result, the concentration of chloride decreased from 29 to 14 mmol L−1, and the concentration of phosphate increased from 0 to 9 mmol L−1 in 5 h. However, the concentration profile of both chloride and phosphate reached a plateau after 5 h; since then, a steady state was formed between the incoming and the outgoing chloride and phosphate ions. This conclusion can also be drawn from the current efficiency of chloride: 72% of the applied current was used to transport the excessive chloride ions from the product stream to the brine during the first hour, and 56% of the current was used during the second hour; after 6 h, the current efficiency value for transporting excessive chloride was only 3%, and finally it decreased to around zero, meaning that same amount of incoming chloride from the feed to the product compartment as the amount of outgoing chloride from the product compartment to the brine at this period. Figure 6 shows the ion profiles in the brine compartment during the experiment. The concentration of chloride and phosphate both increased as a function of time; however, their rate was different. The current efficiency to transport chloride in the first hour was 93% and was around 100% in the second hour, and it decreased to only 42% after 6 h. E



Article

phosphate-P) after 14 h, which can be further diluted or treated for discharge. The long operational period is due to four facts: (1) Iit was intended to identify the maximum phosphate concentration that can be reached by SED. In practice, such a high concentration of phosphate is not necessary (1.5 mmol L−1 is sufficient). (2) Phosphate concentration was only 2% of the whole ion concentration, leading to a low normalized current efficiency toward phosphate (current efficiency was only 2% in average). (3) The membrane surface area was low, only 192 cm2 was used in this experiment. (4) The applied current density was low; a current density of 31.25 A m−2 (0.2 A) was used during the entire experiment. A rough economical evaluation can be made on the basis of these results. During the process, a constant current of 0.2 A was used, and the voltage was around 5 V. The phosphate flux was 16 mmol m−2 h−1, and the energy consumption was roughly 1 W h. This means that 1 kWh electricity can produce around 307 mmol HxPO4y‑ (0.0064 m2 × 3 cells × 16 mmol m−2 h−1 × 1000). This value is derived from a 3 cell-pair stack, where the anode and cathode consumes most of the energy (around 50%). If the stack consists of enough cell pairs in which the energy spent in anode and cathode reactions can be ignored (when 400 cells are installed, the electrode reactions consume less than 1% of the total energy34). In this case, 1 kWh electricity produces around 60 g of phosphate. In Figure 7b, a linear correlation with respect to time can be observed for the chloride concentration in the feed. However, the chloride concentration in the product compartment increased during the initial 16 h (960 min). This resulted in a slower concentration build-up during the first 16 h in the brine. This is due to the fact that the pH was adjusted by adding NaOH in the initial hours. Concentrations of bicarbonate are shown in Figure 7c. In the feed, the bicarbonate concentration decreases slower during the initial 28 h (1680 min). Simultaneously, the concentration in the product also decreases. After 28 h the concentration in the product starts to increase; in the brine it is decreasing. The final concentration of chloride and bicarbonate in the product has the same level compared to the struvite reactor effluent (SED initial product). Meanwhile, the concentration of these ions in the feed decreased to 0.5 and 2 mmol L−1, respectively. This means that the feedwater had a desalination rate of over 95%. Struvite Formation by Using SED Effluent. As shown in Figure 1, it is the final aim to use the SED product as the feed of the struvite reactor. With the lab-scale reactor, not enough SED-product was obtained, and so a synthetic influent was prepared and fed to the struvite reactor. This influent was derived from the effluent of an UASB (upflow anaerobic sludge blanket) reactor by correcting for pH and increasing the phosphate concentration to about 7.23 mmol L−1. The added molar ratio Mg2:PO4-P during the experiment was 1.05 ± 0.15, resulting in a total molar ratio Mg2:PO4-P of 1.15 ± 0.15, when one takes in to account the endogenous Mg2+-ions present in the effluent. The pH in the reactor dropped from 9 to 8.6 due to the precipitation of struvite according to the following reaction:

Figure 7. Concentration of (a) phosphate, (b) chloride, and (c) carbonate in the feed (anaerobic effluent), product, and brine streams as a function of time.

production efficiency of struvite can be obtained when the phosphate concentration of 1.5 mmol L−1 (145 mg L−1), this was reached after 14 h of operation. The phosphate concentration in the brine was 0.1 mmol L−1 (3 mg L−1

Mg 2 + + NH4 + + OH− + HPO4 2 − + 5H 2O → MgNH4PO4 · 6H 2O F



Article

The phosphate concentration decreased from 7.23 ± 0.13 to 0.52 ± 0.16 mmol L−1, resulting in an average phosphate removal efficiency of 93%. A wet chemical analysis of the precipitate gave a molar ratio Mg2+:NH4+:PO4-P of 0.95 ± 0.04:0.84 ± 0.07:1.00 ± 0.01. Traces of calcium, potassium, sulfate, and sodium were also found in the precipitate. Table 3 shows the phosphate concentration in the UASB effluent before SED (SED initial feed) and the accumulation of

(3) Cordell, D.; Rosemarin, A.; Schröder, J. J.; Smit, A. L. Towards global phosphorus security: A systems framework for phosphorus recovery and reuse options. Chemosphere 2011, 84, 747−758. (4) Cisse, L.; Mrabet, T. World phosphate production: Overview and prospects. Phosphorus Res. Bull. 2004, 15, 21−25. (5) Van Vuuren, D. P.; Bouwman, A. F.; Beusen, A. H. W. Phosphorus demand for the 1970−2100 period: A scenario analysis of resource depletion. Global Environ. Change 2010, 20, 428−439. (6) Guest, J. S.; Skerlos, S. J.; Barnard, J. L.; Beck, M. B.; Daigger, G. T.; Hilger, H.; Jackson, S. J.; Karvazy, K.; Kelly, L.; Macpherson, L.; Mihelcic, J. R.; Pramanik, A.; Raskin, L.; van Loosdrecht, M. C. M.; Yeh, D.; Love, N. G. A new planning and design paradigm to achieve sustainable resource recovery from wastewater. Environ. Sci. Technol. 2009, 43 (16), 6126−6130. (7) Kumar, M.; Badruzzaman, M.; Adham, S.; Oppenheimer, J. Beneficial phosphate recovery from reverse osmosis (RO) concentrate of an integrated membrane system using polymeric ligand exchanger (PLE). Water Res. 2007, 41, 2211−2219. (8) Hong, S. U.; Ouyang, L.; Bruening, M. L. Recovery of phosphate using multilayer polyelectrolyte nanofiltration membranes. J. Membr. Sci. 2009, 327 (1−2), 2−5. (9) Ogata, T.; Morisada, S.; Oinuma, Y.; Seida, Y.; Nakano, Y. Preparation of adsorbent for phosphate recovery from aqueous solutions based on condensed tannin gel. J. Hazard. Mater. 2011, 192 (2), 698−703. (10) Wilsenach, J. A.; Schuurbiers, C. A. H.; van Loosdrecht, M. C. M. Phosphate and potassium recovery from source separated urine through struvite precipitation. Water Res. 2007, 41, 458−466. (11) Ueno, Y.; Fujii, M. Three years experience of operating and selling recovered struvite from full-scale plant. Environ. Technol. 2011, 22, 1373−1381. (12) Karapinar, N.; Hoffmann, E.; Hahn, H. H. P-recovery by secondary nucleation and growth of calcium phosphates on magnetite mineral. Water Res. 2006, 40 (6), 1210−1216. (13) Borgerding, J. Phosphate deposits in digestion systems. J. Water Pollut. Control Fed. 1972, 44, 813. (14) Schuiling, R. D.; Andrade, A. Recovery of struvite from calf manure. Environ. Technol. 1999, 20, 765−768. (15) Stratful, I.; Scrimshaw, M. D.; Lester, N. J. Conditions influencing the precipitation of magnesium ammonium phosphate. Water Res. 2001, 35, 4191−4199. (16) Doyle, J. D.; Parsons, S. A. Struvite formation, control and recovery. Water Res. 2002, 36, 3925−3940. (17) Desmidt, E.; Verstraete, W.; Dick, J.; Meesschaert, B.; Carballa, M. Ureolytic phosphate precipitation from anaerobic effluent. Water Sci. Technol. 2009, 59 (10), 1983−1988. (18) Desmidt, E.; Ghyselbrecht, K.; Monballiu, A.; Verstraete, W.; Meesschaert, B. Evaluation and thermodynamic calculation of ureolytic magnesium ammonium phosphate precipitation from UASB effluent at pilot scale. Water Sci. Technol. 2012, 65 (11), 1954−1962. (19) Ma, N.; Rouff, A. A. Influence of pH and oxidation state on the interaction of arsenic with struvite during mineral formation. Environ. Sci. Technol. 2012, 46, 8791−8798. (20) Adnan, A.; Dastur, M.; Mavinic, D. S.; Koch, F. A. Preliminary investigation into factors affecting controlled struvite crystallization at the bench scale. J. Environ. Eng. Sci. 2004, 3, 195−202. (21) Adnan, A.; Mavinic, D. S.; Koch, F. A. Pilot-scale study of phosphorus recovery through struvite crystallizationexamining the process feasibility. J. Environ. Eng. Sci. 2003, 2, 315−324. (22) Battistoni, P.; Fava, G.; Pavan, P.; Musacco, A.; Cecchi, F. Phosphate removal in anaerobic liquors by struvite crystallization without addition of chemicals: Preliminary results. Water Res. 1997, 11 (31), 2925−2929. (23) Munch, E. V.; Barr, K. Controlled struvite crystallisation for removing phosphorus from anaerobic digester sidestreams. Water Res. 2000, 34, 1868−1880. (24) Jaffer, Y.; Clark, T. A.; Pearce, P.; Parsons, S. A. Potential phosphorus recovery by struvite formation. Water Res. 2002, 36, 1834−1842.

Table 3. Comparison of Phosphate Concentration on Different Streams of SED-Struvite Reactor SED initial feed

SED initial product

SED product

struvite reactor effluent

2.5

0.8

6.8

0.5

phosphate concentration (mmol L−1)

phosphate during SED in the struvite reactor effluent: from 0.8 to 6.8 mmol L−1. After precipitation of the phosphate as struvite from “the synthetic SED-effluent” the remaining concentration was 0.5 mmol L−1, which is in good agreement with the phosphate concentration of the initial SED product stream. These results show that the concentrated phosphate from SED can be well precipitated and recovered with high efficiency (93%).

■

ASSOCIATED CONTENT

* Supporting Information S

The characteristics of the anaerobic effluent of a potato processing wastewater from Agristo NV (Table S1) and the concentration of components in SED product and the synthetic SED product by using anaerobic effluent with a dosage of NaOH and KH2PO4 (Table S2). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: + 32 14 335744 (Y.Z.); +32 16 322340 (B.V.d.B.). Fax: +32 14 321186 (Y.Z.); +32 16 322991 (B.V.d.B.). E-mail: [email protected] (Y.Z.); [email protected]. be (B.V.d.B.). Author Contributions

The manuscript was written through contributions from all of the authors. All of the authors have approved the final version of the manuscript. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS KU Leuven is gratefully acknowledged for funding through Project IOF HB/10/023. REFERENCES

(1) Karl, D. M. Aquatic ecology: Phosphorus, the staff of life. Nature 2000, 406, 31−33. (2) Wolfe-Simon, F.; Blum, J. S.; Kulp, T. R.; Gordon, G. W.; Hoeft, S. E.; Pett-Ridge, J.; Stolz, J. F.; Webb, S. M.; Weber, P. K.; Davies, P. C. W.; Anbar, A. D.; Oremland, R. S. A bacterium that can grow by using arsenic instead of phosphorus. Science 2011, 332, 1163−1166. G



Article

(25) Cao, X.; Harris, W. Carbonate and magnesium interactive effect on calcium phosphate precipitation. Environ. Sci. Technol. 2008, 42, 436−442. (26) Zhang, Y.; Paepen, S.; Pinoy, L.; Meesschaert, B.; Van Der Bruggen, B. Selectrodialysis: Fractionation of divalent ions from monovalent ions in a novel electrodialysis stack. Sep. Purif. Technol. 2012, 88, 191−201. (27) Sata, T.; Mine, K.; Higa, M. Change in permselectivity between sulfate and chloride ions through anion exchange membrane with hydrophilicity of the membrane. J. Membr. Sci. 1998, 141 (1), 137− 144. (28) Elattar, A.; Elmidaoui, A.; Pismenskaia, N.; Gavach, C.; Pourcelly, G. Comparison of transport properties of monovalent anions through anion-exchange membranes. J. Membr. Sci. 1998, 143 (1−2), 249−261. (29) Xu, T.; Hu, K. A simple determination of counter-ionic permselectivity in an ion exchange membrane using of bi-ionic membrane potential: permselectivity of anionic species in a novel anion exchange membrane. Sep. Purif. Technol. 2004, 40, 231−236. (30) Bazinet, L.; Moalic, M. Coupling of porous filtration and ionexchange membranes in an electrodialysis stack and impact on cation selectivity: A novel approach for sea water demineralization and the production of physiological water. Desalination 2011, 277, 356−363. (31) Banasiak, L.; Schäfer, A. I. The removal of boron, fluoride and nitrate in electrodialysis in the presence of bulk organic matter. J. Membr. Sci. 2009, 344 (1), 101−109. (32) Zhang, Y.; Van der Bruggen, B.; Pinoy, L.; Meesschaert, B. Separation of nutrient ions and organic compounds from salts in RO concentrates by standard and monovalent selective ion-exchange membranes used in electrodialysis. J. Membr. Sci. 2009, 332, 104−112. (33) Van der Bruggen, B.; Koninckx, A.; Vandecasteele, C. Separation of monovalent and divalent ions from aqueous solution by electrodialysis and nanofiltration. Water Res. 2004, 38, 1347−1353. (34) Strathmann, H. Ion-exchange membrane separation processes. Membrane Science Technology; Elsevier: Amsterdam, 2004; Vol. 9.

H


Phosphate Separation and Recovery from Wastewater by Novel

Recommend Documents