- Email: [email protected]
Removal of phosphate by layered double hydroxides containing iron
Water Research 36 (2002) 1306–1312
Removal of phosphate by layered double hydroxides containing iron Yoshimi Seida*, Yoshio Nakano Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Nagatsuta 4259, Midori-ku, Yokohama, 226-8502 Japan Received 9 November 2000; received in revised form 8 June 2001; accepted 2 July 2001
Abstract 2 3þ Iron-based layered double hydroxides ðM2þ a Feb ðOHÞ2ðaþbÞ CO3 b=2 mH2 OÞ were synthesized. Removal of phosphate by the compounds was studied from the viewpoint of buffering pH effect of the compounds and buffering capacity of solution. The compounds released metal cations (Mg2+, Ca2+, Fe3+) and/or their hydroxides responding to various water environments due to their buffering pH function. The released cations and/or hydroxides worked effectively as coagulants for the phosphate removal. The removal of phosphate depended on the buffering capacity of the solution that is the function of the solution pH and the concentration of phosphate. The removal of phosphate from the solution with small buffering capacity followed a Langmuir-type isotherm. The removal of phosphate from the solution with larger buffering capacity was largely increased. The removal of phosphate by the compounds was analyzed based on the model describing the buffering pH effect of the compounds from which the amount of released cations (CB ) can be determined. The removal was well correlated with the amount of dissolution of the compounds, CB : The mechanism of phosphate removal was examined based on the removal efficiency (mol of removed phosphate/mol of released alkali). The efficiencies showed below one in the solution with large buffering capacity and above one in the solution with small buffering capacity. The efficiency below one showed the removal of phosphate through coagulation by the released metal cations and hydroxides. The successful removal of dilute phosphate (0.2 mg P/l) from the drain water was also demonstrated. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Buffering pH; Coagulation; Ion exchange; Layered double hydroxide; Phosphate removal; Precipitation; Water treatment
1. Introduction Phosphate is recognized as being one of the resources that will be lost in near future [1]. A large amount of used phosphate finally reaches water environment as diluted waste, which often leads to pollution of the water environment. It is of value to collect the finally disposed phosphates from effluents and drain water before further dispersion and dilution of them in the water environment. Many techniques have been proposed for the *Corresponding author. Institute of Research and Innovation, 1201 Takada, Kashiwa, Chiba. Tel.: +81-471-44-8845; fax: +81-471-44-7602. E-mail address: [email protected] (Y. Seida).
removal of phosphate from wastewater. Coagulation– precipitation and biological methods are widely accepted methods of phosphate removal at industrial level. Extensive research has also been carried out to produce simplification of maintenance, stable running and removal efficiency. Among these researches, many researchers have promoted development of adsorbents with high selectivity and removal capacity for phosphate ([1–6], 1997). Recent recognition of the risk of very dilute contaminants in water makes it important to produce a new kind of adsorbents and/or coagulants to avoid any risk that may be produced such as contamination by the agents through their dissolution (Research report by Japan Water Research Centre [7] and Seida and Nakano [8]). Even the use of aluminum-based
0043-1354/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 3 - 1 3 5 4 ( 0 1 ) 0 0 3 4 0 - 2
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
agents has attracted attention because of suspicion that it is involved in brain disease [9]. It is of value to develop safe water treatment agents and removal methods based upon their use. In the present study, iron-based layered double hydroxides (LDH; the clay-type anion exchanger; 2+ 2 3þ M2þ ; Mg2+ or Ca2+, a Feb ðOHÞ2ðaþbÞ CO3 b=2 mH2 O (M 2 CO3 is exchangeable intercalated anion) were synthesized. The phosphate removal property of the compounds was examined as a function of the buffering pH effect of the compounds and buffering capacity of solution that is the function of solution pH and concentration of phosphate. The removal of phosphate by the compounds was analyzed based on the model describing the buffering effect of the compounds. Phosphate removal from an effluent of drain water was also studied to confirm the practical effectiveness of the compound for the removal of dilute phosphate.
1307
2.3. Phosphate removal from effluent Effluent sample solution was collected from a drain into the Sakai River in Kanagawa Prefecture in Japan. The concentration of total phosphate in the sample was 0.2 mg P/l. The water quality of the sample is listed in Table 1. Mg/Fe compound (0.2 g) was packed into a column with a packing height of 1 mm. The solution was introduced into the column with a flow rate of 20 ml/h. The flow rate was determined based on the space velocity operated in a facility of drain water treatment [12]. Effluent from the column was collected at regular time interval. The pH and the concentration of metal ions in the effluent from the column were measured by pH meter and the ICP spectrophotometer, respectively. The concentration of phosphate in the effluent was analyzed by the molybdenum-blue method.
3. Results 2. Materials and methods 2.1. Sample preparation Iron-based compounds were synthesized following a procedure reported [10]. The compounds are abbreviated hereafter as M2+/Fe=x based on the constituent metal ions (x is the molar ratio between divalent alkaline earth metal and iron at the preparation stage). The anions intercalated in the compounds were confirmed to be carbonate ions by temperature programmed desorption analysis [11]. The compounds dried at 353 K for one night and crushed below 250 mm in diameter were used.
2.2. Removal property A mass of 0.05 g of each compound was placed in a series of 10 ml solution of Na2HPO4 (Solution A; solution pH=8.4) or equimolar mixture of KH2PO4 and Na2HPO4 (Solution B; solution pH=6.86). The concentrations of phosphate in the solutions were varied from 1 to 1500 mg P/l. Another set of the compounds was placed in Na2HPO4 solutions with a series of solution pH (pH-varied solution; the concentration of phosphate was 100 mg P/l). The pH of the solutions was adjusted with HCl. The solutions were kept at room temperature for 24 h with shaking. The concentrations of both phosphate and metal cations in the solutions were measured by ICP emission spectrophotometer (Seiko Instrument Co. Ltd., SSC8000). The amount of removed phosphate was calculated by mass balance.
3.1. Removal property Figs. 1(a) and (b) show the effect of the concentration of phosphate on the removal by both Mg/Fe=3 and Ca/Fe=3 compounds in Solution A (pH=8.4) and Solution B (pH=6.86), respectively. The removal curves obtained from Solution A (pH=8.4) followed a Langmuir-type isotherm in both compounds. The mechanism of phosphate removal in this system is considered to be ion exchange between phosphate and carbonate ions in the compounds. The maximum removal, qmax ; by the compounds for phosphate in Solution A was calculated from a Langmuir plot as 15.5 and 28.8 mg/g, respectively. On the contrary, the amount of removed phosphate, q; increased largely in Solution B in both compounds. The removal was increased with increase in the concentration of phosphate. The pH of the solutions increased during the removal as shown in the final pH of the solution in Figs. 1(a) and (b) (initial pHs of the solutions are 8.4 (Solution A) and 6.86 (Solution B). The increase of pH
Table 1 Water qualities of the effluent sample collected from a drain into the Sakai river in Kanagawa prefecture of Japan Analytical factor pH BOD COD T-P T-N NO 3 Cl
7.4 4.0 mg/l 8.3 mg/l 0.2 mg/l 5.7 mg/l 3.5 mg/l 26.0 mg/l
1308
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
Fig. 1. (a) Removal property of the Mg/Fe compound. (b) Removal property of the Ca/Fe compound. (c) Concentration of Mg2+ and Fe3+ ions after the removal (Mg/Fe compound). (d) Concentration of Ca2+ and Fe3+ ions after the removal (Ca/Fe compound).
occurs through dissolution of the compounds with releasing metal cations and their hydroxides due to their buffering pH function. The concentrations of the metal cations in the solutions after the removal are shown in Figs. 1(c) and (d), respectively. Compared with the concentrations of the metal cations in the absence of the phosphate (the values at concentration of phosphate is zero), the concentrations of the metal cations decreased due to the presence of phosphate in the region of small concentration of phosphate (o100 mg P/l). It means that some amount of the released cations was consumed for the phosphate removal. The final concentrations of divalent cations were increased with increasing the concentration of phosphate in Solution B, a process that did not occur in Solution A (Figs. 1(c) and (d)). The amount of dissolved cations increased by more than the amount of cations consumed in the coagulation of phosphate due to the buffering effect of Solution B. The concentrations of the metal cations are smaller in the Ca/Fe system than in the Mg/Fe system. It is considered that the dissolved cations in Solution B were used for the removal of phosphate in the Ca/Fe compound.
The results of the solution pH and the concentration of metal cations in Figs. 1(a)–(d) mean that larger amounts of metal cations are released from the compounds in Solution B during the buffering of the solution pH by the compounds and the amount of the release depends on the concentration of phosphate. The dissolved metal cations also form their hydroxides in the solutions at the final pH [13]. The increase of phosphate removal in Solution B should be due to the contribution of the released cations and hydroxides as coagulants and/or precipitants to the removal. Floc formation was not observed in the final solutions in the present system. The effect of buffering pH function of the compounds on the removal of phosphate was clarified further through the study of the effect of solution pH on the removal, given in the following section. 3.2. Effect of solution pH Fig. 2(a) shows the effect of initial pH of the solution on both the phosphate removal and the concentration of dissolved metal cations for the Mg/Fe compound. The
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
final pH of the solutions reached up to around nine in every solution. The amount of removed phosphate, q; was increased with decrease in the initial pH of the solution. A larger amount of magnesium ion was released from the compounds in the solution with lower initial pH, resulting in the larger removal of phosphate. Fig. 2(b) shows the results in the case of the Ca/Fe type compound. Contrary to the results in the Mg/Fe system (Fig. 2(a)), the phosphate removal from lower pH solutions decreased in this system. Large amounts of dissolved metal cations were detected in the solutions with low pH (opH=3), which means that the released metal cations did not work effectively to produce sufficient coagulation and/or precipitation of phosphate in the solution due to the low final pH.
1309
From the final pH and the concentration of the metal cations in the solutions in Figs. 2(a) and (b), release of the metal cations and hydroxides from the compounds and their contribution to the phosphate removal were clarified. These results in Figs. 2(a) and (b) imply that the phosphate can be removed continuously while solution with a pH below the buffering pH of the compounds is fed to the system. There was also no floc formation observed in this system. 3.3. Removal from effluent The potential of the compounds for the removal of dilute phosphate from the drain water sample was tested and the result is shown in Fig. 3. Above 80% of the
Fig. 2. (a) Effect of pH on the removal of phosphate in the Mg/Fe compound, initial concentration of phosphate is 100 mg P/l. (b) Effect of pH on the removal of phosphate in the Ca/Fe compound, initial concentration of phosphate is 100 mg P/l.
1310
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
Fig. 3. Removal of phosphate from the effluent by the Mg/Fe compound, sample: collected from a drain of the Sakai river in Kanagawa Prefecture of Japan (0.2 mg P/l).
phosphate in the effluent was successfully removed by the compounds during the treatment for 3000 times larger volume of the solution to the compound. The concentrations of the dissolved metal cations during the removal was not so high as shown in Fig. 3, which would be within an acceptable extent based on the criteria of drinking water.
where ki and Kw show the ith dissociation constant of phosphoric acid and that of water, respectively. The Cp shows the analytical concentration of phosphoric acid in the solution. The amount of alkali, CB ; required to attain a solution pH from an initial one is calculated by integrating Eq. (1) between the initial pH and the final one, and is shown by Eq. (3).
4. Discussion
CB ¼
The amount of phosphate removal by the compounds is considered to depend on both the ion exchange capacity and the buffering property of the compounds as stated in the above sections. The buffering property of the compounds can be examined from the viewpoint of the following buffering property of solutions. The buffering property of a solution is shown by the buffer index, b; defined by Eq. (1) [14]. b¼
dCB : dpH
ð1Þ
Eq. (1) shows the amount of alkali (NaOH), CB ; required to raise the solution pH from some initial pH. The buffer index, b; is shown by Eq. (2) for phosphoric acid solution. Kw b ¼ 2:3 þ ½Hþ ½Hþ ð2Þ Cp k1 g½Hþ ; þ ð½Hþ 3 þ k1 ½Hþ 2 þ k1 k2 ½Hþ þ k1 k2 k3 Þ2 where 2 3 CB ¼ ½OH ½Hþ þ ½H2 PO 4 þ 2½HPO4 þ 3½PO4 ;
g ¼ ½Hþ 4 þ 4k2 ½Hþ 3 þ ðk1 k2 þ 9k2 k3 Þ½Hþ 2 þ 4k1 k2 k3 ½Hþ þ k1 k22 k3 ;
Z¼
B¼ x¼
KW ½Hþ þ Z þ 2B þ 3x; ½Hþ þ 3
½H þ
Cp k1 ½Hþ 2 þ 2 k1 ½H þ k1 k2 ½Hþ
þ k1 k2 k3
;
Cp k1 k2 ½Hþ ; ½H þ k1 ½Hþ 2 þ k1 k2 ½Hþ þ k1 k2 k3 þ 3
þ 3
½H þ
C p k1 k2 k3 þ 2 k1 ½H þ k1 k2 ½Hþ
þ k1 k2 k3
:
ð3Þ
Figs. 4(a) and (b) show the buffer index, b; and the amount of alkali, CB ; required to change the solution pH for phosphoric acid solutions with a series of concentrations, Cp ; respectively (CB was calculated using a final pH of nine). The amount of alkali, CB ; required to raise the solution pH to the final one is consistent with the area under the b-curve between the initial pH and the final one of the solution in Fig. 4(a). Buffering a solution by the compounds under study is basically equivalent to neutralization of the solution by adding alkali to the solution. Thus, the amount of alkali released (dissolved) from the compounds during the buffering is equivalent to the amount of alkali, CB ; required to change the solution pH. The buffer index of the solution increases with increase in the phosphoric acid concentration (Fig. 4(a)). The alkali required to change a solution pH to a final one increases with increasing the concentration of phosphoric acid and
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
1311
Fig. 4. (a) Relationship between solution pH and buffer index for phosphoric acid. (b) Relationship between the initial solution pH and the amount of released alkali.
Fig. 5. Relationship between CB and q:
with decreasing the solution pH (Fig. 4(b)). The amount of released alkali showed a plateau region between pH=3.5 and 5.5. This pattern was experimentally observed in the phosphate removal from pH-varied solutions as shown in Fig. 2(a). The alkaline earth metal and iron ions released from the compounds during the buffering potentially work as coagulants and/or precipitants for the phosphate in the present system as stated above. The amounts of phosphate removal shown in Figs. 1 and 2 are plotted against CB in Fig. 5. CB was calculated by Eq. (3) using the experimentally measured solution pH. The removal of phosphate increased linearly with increasing CB in the larger CB region (CB > 0:005). The increase of the removal should be achieved by the contribution of coagulation and/or precipitation. The removal performed through ion exchange (in Solution A) results in the steep increase of the removal (q) against CB in spite of the small CB : The mechanism of the phosphate removal by the compounds can be evaluated based on the CB calculated from the buffering capacity of the solutions as shown in Fig. 5. The removal efficiencies (mol of removed phosphate/ mol of released alkali, CB ) calculated from the results in
Fig. 6. Removal efficiency (mol of removed phosphate/mol of released alkali).
Fig. 5 are shown in Fig. 6. The removal efficiency decreased with increasing CB and shows a constant value in the large CB region. The efficiencies in the region below CB ¼ 0:005 show a high value of more than one. The large value of the efficiency is compatible with the removal by ion exchange. The efficiencies in the large CB region show values of 0.15–0.5. In the case of coagulation system using MgCl2, CaCl2 and mixture of MgCl2 and FeCl3 as coagulants, the removal efficiencies for phosphate were around 0.5–0.7 (experimental conditions: 10 ml of 100 mg P/l Na2HPO4 solution, 20B70 mg/l of coagulant, coagulation pH is 9.0 (the pH was adjusted by NaOH after adding the coagulant into the solution)). The values of the efficiencies in the high CB region show the removal by coagulation. The Ca/Fe compound removes phosphate effectively in the large CB region based on the efficiency. The compounds remove phosphate through coagulation as well as ion exchange. The enhanced removal of phosphate can be achieved due to the buffering pH function of the compound by which the metal cations and their hydroxides are released to work as coagulants. Much more phosphate is removed during the dissolution–coagulation process of the compound in the
1312
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
solution with larger buffering capacity. It can be also said that diluted phosphate with its concentration below the concentration of the released species is removable by the dissolution–coagulation process of the compound. The result in Fig. 3 shows the successful example of this idea. More than 2.5 times larger amount of phosphate than the ion exchange capacity (0.17 mg/g) of the Mg/Fe compound in the solution with 0.2 mg P/l phosphate was removed in the treatment (Fig. 3) (the removal capacity was calculated based on the isotherm in Fig. 1(a)). The dissolution–coagulation process of the compound should achieve excess amount of the removal.
5. Conclusion The removal of phosphate by iron-based layered double hydroxides was studied from the viewpoint of the buffering pH effect of the compounds and buffering capacity of solution. The compounds release metal cations and/or hydroxides to create a coagulation and/ or precipitation atmosphere by themselves responding to various water environments. The released cations and/or hydroxides work effectively as coagulants and/or precipitants for phosphate removal. The removal of phosphate depends on the buffering capacity of the solution. The removal from the solution with larger buffering capacity is largely increased. The removal was correlated well with the amount of dissolution of the compounds. The dissolution–coagulation process of the compounds based on the buffering pH function is the key process for the enhanced phosphate removal. The effectiveness of the compounds in the removal of diluted phosphate from the drain water was confirmed.
References [1] Itasaka N, Takanashi H, Hirata M, Hano T. Suicyu Teinodo Rin no Jyokyo Kaisyu you Kyucyakuzai no Kaihatsujyokyou to Kadai. Yosui to Haisui 1999;41(3): 5–13. [2] Haron MJ, Wasay SA, Tokunaga S. Preparation of basic yttrium carbonate for phosphate removal. Water Environ Res 1997;69(5):1047–51. [3] Nishiguchi T, Takahashi T, Haruta S. Phosphate removal using iron. Yosui to Haisui 1989;31(11):11–8. [4] Ookubo A, Ooi K, Hayashi H. Preparation and phosphate ion-exchange properties of a hydrotalcite-like compound. Langmuir 1993;9:1418–22. [5] Shin H-S, Kim M-J, Nam S-Y, Moon H-C. Phosphorus removal by hydrotalcite-like compounds (HTLcs). Water Sci Technol 1996;34(1–2):161–8. [6] Zhang W-X, Sakane H, Hatsushika T, Kinomura N, Suzuki T. Phosphorus anion-exchange characteristics of a pyroaurite-like compound. Inorg Mater 1997;4:132–8. [7] Japan Water Research Center. Suidou-yo Gyoushuzai no Tayouka Kento Chosa, 2000. p. 124. [8] Seida Y, Nakano Y. Rapid removal of dilute lead by pyroaurite-like compounds. Water Res 2001;35(10): 2341–6. [9] Guidelines for drinking water quality, 2nd edition, Addendum to volume 1, Recommendations, 3–4. Geneva, World Health Organization, 1998. [10] Seida Y, Nakano Y. Removal of humic substances by layered double hydroxides containing iron. Water Res 2000;34(5):1487–94. [11] Seida Y, Nakano Y. Removal of phosphate through dissolution–coagulation process of layered double hydroxide. J Chem Eng Japan 2001;34(7):906–11. [12] Hashimoto K, Nishihori M, Yamauchi T, Sekine K. Tetsu-sutenresu Sessyokuzai wo mochiita Haisuiro karano Rin Jyokyo. Mizusyori Gijutsu 1996;37(11):19–25. [13] Nihon Bunseki Kagaku-kai, Bunseki Kagaku Hannou no Kiso, Nihon Bunseki Kagaku Kai, Hokkaido Shibu-hen Fujimoto et al., editors. Baifukan, Tokyo, 1980. p. 67–9. [14] Freiser H, Fernando Q. Ionic equilibria in analytical chemistry. New York: Wiley, 1963.
Removal of phosphate by layered double hydroxides containing iron Yoshimi Seida*, Yoshio Nakano Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Nagatsuta 4259, Midori-ku, Yokohama, 226-8502 Japan Received 9 November 2000; received in revised form 8 June 2001; accepted 2 July 2001
Abstract 2 3þ Iron-based layered double hydroxides ðM2þ a Feb ðOHÞ2ðaþbÞ CO3 b=2 mH2 OÞ were synthesized. Removal of phosphate by the compounds was studied from the viewpoint of buffering pH effect of the compounds and buffering capacity of solution. The compounds released metal cations (Mg2+, Ca2+, Fe3+) and/or their hydroxides responding to various water environments due to their buffering pH function. The released cations and/or hydroxides worked effectively as coagulants for the phosphate removal. The removal of phosphate depended on the buffering capacity of the solution that is the function of the solution pH and the concentration of phosphate. The removal of phosphate from the solution with small buffering capacity followed a Langmuir-type isotherm. The removal of phosphate from the solution with larger buffering capacity was largely increased. The removal of phosphate by the compounds was analyzed based on the model describing the buffering pH effect of the compounds from which the amount of released cations (CB ) can be determined. The removal was well correlated with the amount of dissolution of the compounds, CB : The mechanism of phosphate removal was examined based on the removal efficiency (mol of removed phosphate/mol of released alkali). The efficiencies showed below one in the solution with large buffering capacity and above one in the solution with small buffering capacity. The efficiency below one showed the removal of phosphate through coagulation by the released metal cations and hydroxides. The successful removal of dilute phosphate (0.2 mg P/l) from the drain water was also demonstrated. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Buffering pH; Coagulation; Ion exchange; Layered double hydroxide; Phosphate removal; Precipitation; Water treatment
1. Introduction Phosphate is recognized as being one of the resources that will be lost in near future [1]. A large amount of used phosphate finally reaches water environment as diluted waste, which often leads to pollution of the water environment. It is of value to collect the finally disposed phosphates from effluents and drain water before further dispersion and dilution of them in the water environment. Many techniques have been proposed for the *Corresponding author. Institute of Research and Innovation, 1201 Takada, Kashiwa, Chiba. Tel.: +81-471-44-8845; fax: +81-471-44-7602. E-mail address: [email protected] (Y. Seida).
removal of phosphate from wastewater. Coagulation– precipitation and biological methods are widely accepted methods of phosphate removal at industrial level. Extensive research has also been carried out to produce simplification of maintenance, stable running and removal efficiency. Among these researches, many researchers have promoted development of adsorbents with high selectivity and removal capacity for phosphate ([1–6], 1997). Recent recognition of the risk of very dilute contaminants in water makes it important to produce a new kind of adsorbents and/or coagulants to avoid any risk that may be produced such as contamination by the agents through their dissolution (Research report by Japan Water Research Centre [7] and Seida and Nakano [8]). Even the use of aluminum-based
0043-1354/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 3 - 1 3 5 4 ( 0 1 ) 0 0 3 4 0 - 2
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
agents has attracted attention because of suspicion that it is involved in brain disease [9]. It is of value to develop safe water treatment agents and removal methods based upon their use. In the present study, iron-based layered double hydroxides (LDH; the clay-type anion exchanger; 2+ 2 3þ M2þ ; Mg2+ or Ca2+, a Feb ðOHÞ2ðaþbÞ CO3 b=2 mH2 O (M 2 CO3 is exchangeable intercalated anion) were synthesized. The phosphate removal property of the compounds was examined as a function of the buffering pH effect of the compounds and buffering capacity of solution that is the function of solution pH and concentration of phosphate. The removal of phosphate by the compounds was analyzed based on the model describing the buffering effect of the compounds. Phosphate removal from an effluent of drain water was also studied to confirm the practical effectiveness of the compound for the removal of dilute phosphate.
1307
2.3. Phosphate removal from effluent Effluent sample solution was collected from a drain into the Sakai River in Kanagawa Prefecture in Japan. The concentration of total phosphate in the sample was 0.2 mg P/l. The water quality of the sample is listed in Table 1. Mg/Fe compound (0.2 g) was packed into a column with a packing height of 1 mm. The solution was introduced into the column with a flow rate of 20 ml/h. The flow rate was determined based on the space velocity operated in a facility of drain water treatment [12]. Effluent from the column was collected at regular time interval. The pH and the concentration of metal ions in the effluent from the column were measured by pH meter and the ICP spectrophotometer, respectively. The concentration of phosphate in the effluent was analyzed by the molybdenum-blue method.
3. Results 2. Materials and methods 2.1. Sample preparation Iron-based compounds were synthesized following a procedure reported [10]. The compounds are abbreviated hereafter as M2+/Fe=x based on the constituent metal ions (x is the molar ratio between divalent alkaline earth metal and iron at the preparation stage). The anions intercalated in the compounds were confirmed to be carbonate ions by temperature programmed desorption analysis [11]. The compounds dried at 353 K for one night and crushed below 250 mm in diameter were used.
2.2. Removal property A mass of 0.05 g of each compound was placed in a series of 10 ml solution of Na2HPO4 (Solution A; solution pH=8.4) or equimolar mixture of KH2PO4 and Na2HPO4 (Solution B; solution pH=6.86). The concentrations of phosphate in the solutions were varied from 1 to 1500 mg P/l. Another set of the compounds was placed in Na2HPO4 solutions with a series of solution pH (pH-varied solution; the concentration of phosphate was 100 mg P/l). The pH of the solutions was adjusted with HCl. The solutions were kept at room temperature for 24 h with shaking. The concentrations of both phosphate and metal cations in the solutions were measured by ICP emission spectrophotometer (Seiko Instrument Co. Ltd., SSC8000). The amount of removed phosphate was calculated by mass balance.
3.1. Removal property Figs. 1(a) and (b) show the effect of the concentration of phosphate on the removal by both Mg/Fe=3 and Ca/Fe=3 compounds in Solution A (pH=8.4) and Solution B (pH=6.86), respectively. The removal curves obtained from Solution A (pH=8.4) followed a Langmuir-type isotherm in both compounds. The mechanism of phosphate removal in this system is considered to be ion exchange between phosphate and carbonate ions in the compounds. The maximum removal, qmax ; by the compounds for phosphate in Solution A was calculated from a Langmuir plot as 15.5 and 28.8 mg/g, respectively. On the contrary, the amount of removed phosphate, q; increased largely in Solution B in both compounds. The removal was increased with increase in the concentration of phosphate. The pH of the solutions increased during the removal as shown in the final pH of the solution in Figs. 1(a) and (b) (initial pHs of the solutions are 8.4 (Solution A) and 6.86 (Solution B). The increase of pH
Table 1 Water qualities of the effluent sample collected from a drain into the Sakai river in Kanagawa prefecture of Japan Analytical factor pH BOD COD T-P T-N NO 3 Cl
7.4 4.0 mg/l 8.3 mg/l 0.2 mg/l 5.7 mg/l 3.5 mg/l 26.0 mg/l
1308
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
Fig. 1. (a) Removal property of the Mg/Fe compound. (b) Removal property of the Ca/Fe compound. (c) Concentration of Mg2+ and Fe3+ ions after the removal (Mg/Fe compound). (d) Concentration of Ca2+ and Fe3+ ions after the removal (Ca/Fe compound).
occurs through dissolution of the compounds with releasing metal cations and their hydroxides due to their buffering pH function. The concentrations of the metal cations in the solutions after the removal are shown in Figs. 1(c) and (d), respectively. Compared with the concentrations of the metal cations in the absence of the phosphate (the values at concentration of phosphate is zero), the concentrations of the metal cations decreased due to the presence of phosphate in the region of small concentration of phosphate (o100 mg P/l). It means that some amount of the released cations was consumed for the phosphate removal. The final concentrations of divalent cations were increased with increasing the concentration of phosphate in Solution B, a process that did not occur in Solution A (Figs. 1(c) and (d)). The amount of dissolved cations increased by more than the amount of cations consumed in the coagulation of phosphate due to the buffering effect of Solution B. The concentrations of the metal cations are smaller in the Ca/Fe system than in the Mg/Fe system. It is considered that the dissolved cations in Solution B were used for the removal of phosphate in the Ca/Fe compound.
The results of the solution pH and the concentration of metal cations in Figs. 1(a)–(d) mean that larger amounts of metal cations are released from the compounds in Solution B during the buffering of the solution pH by the compounds and the amount of the release depends on the concentration of phosphate. The dissolved metal cations also form their hydroxides in the solutions at the final pH [13]. The increase of phosphate removal in Solution B should be due to the contribution of the released cations and hydroxides as coagulants and/or precipitants to the removal. Floc formation was not observed in the final solutions in the present system. The effect of buffering pH function of the compounds on the removal of phosphate was clarified further through the study of the effect of solution pH on the removal, given in the following section. 3.2. Effect of solution pH Fig. 2(a) shows the effect of initial pH of the solution on both the phosphate removal and the concentration of dissolved metal cations for the Mg/Fe compound. The
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
final pH of the solutions reached up to around nine in every solution. The amount of removed phosphate, q; was increased with decrease in the initial pH of the solution. A larger amount of magnesium ion was released from the compounds in the solution with lower initial pH, resulting in the larger removal of phosphate. Fig. 2(b) shows the results in the case of the Ca/Fe type compound. Contrary to the results in the Mg/Fe system (Fig. 2(a)), the phosphate removal from lower pH solutions decreased in this system. Large amounts of dissolved metal cations were detected in the solutions with low pH (opH=3), which means that the released metal cations did not work effectively to produce sufficient coagulation and/or precipitation of phosphate in the solution due to the low final pH.
1309
From the final pH and the concentration of the metal cations in the solutions in Figs. 2(a) and (b), release of the metal cations and hydroxides from the compounds and their contribution to the phosphate removal were clarified. These results in Figs. 2(a) and (b) imply that the phosphate can be removed continuously while solution with a pH below the buffering pH of the compounds is fed to the system. There was also no floc formation observed in this system. 3.3. Removal from effluent The potential of the compounds for the removal of dilute phosphate from the drain water sample was tested and the result is shown in Fig. 3. Above 80% of the
Fig. 2. (a) Effect of pH on the removal of phosphate in the Mg/Fe compound, initial concentration of phosphate is 100 mg P/l. (b) Effect of pH on the removal of phosphate in the Ca/Fe compound, initial concentration of phosphate is 100 mg P/l.
1310
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
Fig. 3. Removal of phosphate from the effluent by the Mg/Fe compound, sample: collected from a drain of the Sakai river in Kanagawa Prefecture of Japan (0.2 mg P/l).
phosphate in the effluent was successfully removed by the compounds during the treatment for 3000 times larger volume of the solution to the compound. The concentrations of the dissolved metal cations during the removal was not so high as shown in Fig. 3, which would be within an acceptable extent based on the criteria of drinking water.
where ki and Kw show the ith dissociation constant of phosphoric acid and that of water, respectively. The Cp shows the analytical concentration of phosphoric acid in the solution. The amount of alkali, CB ; required to attain a solution pH from an initial one is calculated by integrating Eq. (1) between the initial pH and the final one, and is shown by Eq. (3).
4. Discussion
CB ¼
The amount of phosphate removal by the compounds is considered to depend on both the ion exchange capacity and the buffering property of the compounds as stated in the above sections. The buffering property of the compounds can be examined from the viewpoint of the following buffering property of solutions. The buffering property of a solution is shown by the buffer index, b; defined by Eq. (1) [14]. b¼
dCB : dpH
ð1Þ
Eq. (1) shows the amount of alkali (NaOH), CB ; required to raise the solution pH from some initial pH. The buffer index, b; is shown by Eq. (2) for phosphoric acid solution. Kw b ¼ 2:3 þ ½Hþ ½Hþ ð2Þ Cp k1 g½Hþ ; þ ð½Hþ 3 þ k1 ½Hþ 2 þ k1 k2 ½Hþ þ k1 k2 k3 Þ2 where 2 3 CB ¼ ½OH ½Hþ þ ½H2 PO 4 þ 2½HPO4 þ 3½PO4 ;
g ¼ ½Hþ 4 þ 4k2 ½Hþ 3 þ ðk1 k2 þ 9k2 k3 Þ½Hþ 2 þ 4k1 k2 k3 ½Hþ þ k1 k22 k3 ;
Z¼
B¼ x¼
KW ½Hþ þ Z þ 2B þ 3x; ½Hþ þ 3
½H þ
Cp k1 ½Hþ 2 þ 2 k1 ½H þ k1 k2 ½Hþ
þ k1 k2 k3
;
Cp k1 k2 ½Hþ ; ½H þ k1 ½Hþ 2 þ k1 k2 ½Hþ þ k1 k2 k3 þ 3
þ 3
½H þ
C p k1 k2 k3 þ 2 k1 ½H þ k1 k2 ½Hþ
þ k1 k2 k3
:
ð3Þ
Figs. 4(a) and (b) show the buffer index, b; and the amount of alkali, CB ; required to change the solution pH for phosphoric acid solutions with a series of concentrations, Cp ; respectively (CB was calculated using a final pH of nine). The amount of alkali, CB ; required to raise the solution pH to the final one is consistent with the area under the b-curve between the initial pH and the final one of the solution in Fig. 4(a). Buffering a solution by the compounds under study is basically equivalent to neutralization of the solution by adding alkali to the solution. Thus, the amount of alkali released (dissolved) from the compounds during the buffering is equivalent to the amount of alkali, CB ; required to change the solution pH. The buffer index of the solution increases with increase in the phosphoric acid concentration (Fig. 4(a)). The alkali required to change a solution pH to a final one increases with increasing the concentration of phosphoric acid and
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
1311
Fig. 4. (a) Relationship between solution pH and buffer index for phosphoric acid. (b) Relationship between the initial solution pH and the amount of released alkali.
Fig. 5. Relationship between CB and q:
with decreasing the solution pH (Fig. 4(b)). The amount of released alkali showed a plateau region between pH=3.5 and 5.5. This pattern was experimentally observed in the phosphate removal from pH-varied solutions as shown in Fig. 2(a). The alkaline earth metal and iron ions released from the compounds during the buffering potentially work as coagulants and/or precipitants for the phosphate in the present system as stated above. The amounts of phosphate removal shown in Figs. 1 and 2 are plotted against CB in Fig. 5. CB was calculated by Eq. (3) using the experimentally measured solution pH. The removal of phosphate increased linearly with increasing CB in the larger CB region (CB > 0:005). The increase of the removal should be achieved by the contribution of coagulation and/or precipitation. The removal performed through ion exchange (in Solution A) results in the steep increase of the removal (q) against CB in spite of the small CB : The mechanism of the phosphate removal by the compounds can be evaluated based on the CB calculated from the buffering capacity of the solutions as shown in Fig. 5. The removal efficiencies (mol of removed phosphate/ mol of released alkali, CB ) calculated from the results in
Fig. 6. Removal efficiency (mol of removed phosphate/mol of released alkali).
Fig. 5 are shown in Fig. 6. The removal efficiency decreased with increasing CB and shows a constant value in the large CB region. The efficiencies in the region below CB ¼ 0:005 show a high value of more than one. The large value of the efficiency is compatible with the removal by ion exchange. The efficiencies in the large CB region show values of 0.15–0.5. In the case of coagulation system using MgCl2, CaCl2 and mixture of MgCl2 and FeCl3 as coagulants, the removal efficiencies for phosphate were around 0.5–0.7 (experimental conditions: 10 ml of 100 mg P/l Na2HPO4 solution, 20B70 mg/l of coagulant, coagulation pH is 9.0 (the pH was adjusted by NaOH after adding the coagulant into the solution)). The values of the efficiencies in the high CB region show the removal by coagulation. The Ca/Fe compound removes phosphate effectively in the large CB region based on the efficiency. The compounds remove phosphate through coagulation as well as ion exchange. The enhanced removal of phosphate can be achieved due to the buffering pH function of the compound by which the metal cations and their hydroxides are released to work as coagulants. Much more phosphate is removed during the dissolution–coagulation process of the compound in the
1312
Y. Seida, Y. Nakano / Water Research 36 (2002) 1306–1312
solution with larger buffering capacity. It can be also said that diluted phosphate with its concentration below the concentration of the released species is removable by the dissolution–coagulation process of the compound. The result in Fig. 3 shows the successful example of this idea. More than 2.5 times larger amount of phosphate than the ion exchange capacity (0.17 mg/g) of the Mg/Fe compound in the solution with 0.2 mg P/l phosphate was removed in the treatment (Fig. 3) (the removal capacity was calculated based on the isotherm in Fig. 1(a)). The dissolution–coagulation process of the compound should achieve excess amount of the removal.
5. Conclusion The removal of phosphate by iron-based layered double hydroxides was studied from the viewpoint of the buffering pH effect of the compounds and buffering capacity of solution. The compounds release metal cations and/or hydroxides to create a coagulation and/ or precipitation atmosphere by themselves responding to various water environments. The released cations and/or hydroxides work effectively as coagulants and/or precipitants for phosphate removal. The removal of phosphate depends on the buffering capacity of the solution. The removal from the solution with larger buffering capacity is largely increased. The removal was correlated well with the amount of dissolution of the compounds. The dissolution–coagulation process of the compounds based on the buffering pH function is the key process for the enhanced phosphate removal. The effectiveness of the compounds in the removal of diluted phosphate from the drain water was confirmed.
References [1] Itasaka N, Takanashi H, Hirata M, Hano T. Suicyu Teinodo Rin no Jyokyo Kaisyu you Kyucyakuzai no Kaihatsujyokyou to Kadai. Yosui to Haisui 1999;41(3): 5–13. [2] Haron MJ, Wasay SA, Tokunaga S. Preparation of basic yttrium carbonate for phosphate removal. Water Environ Res 1997;69(5):1047–51. [3] Nishiguchi T, Takahashi T, Haruta S. Phosphate removal using iron. Yosui to Haisui 1989;31(11):11–8. [4] Ookubo A, Ooi K, Hayashi H. Preparation and phosphate ion-exchange properties of a hydrotalcite-like compound. Langmuir 1993;9:1418–22. [5] Shin H-S, Kim M-J, Nam S-Y, Moon H-C. Phosphorus removal by hydrotalcite-like compounds (HTLcs). Water Sci Technol 1996;34(1–2):161–8. [6] Zhang W-X, Sakane H, Hatsushika T, Kinomura N, Suzuki T. Phosphorus anion-exchange characteristics of a pyroaurite-like compound. Inorg Mater 1997;4:132–8. [7] Japan Water Research Center. Suidou-yo Gyoushuzai no Tayouka Kento Chosa, 2000. p. 124. [8] Seida Y, Nakano Y. Rapid removal of dilute lead by pyroaurite-like compounds. Water Res 2001;35(10): 2341–6. [9] Guidelines for drinking water quality, 2nd edition, Addendum to volume 1, Recommendations, 3–4. Geneva, World Health Organization, 1998. [10] Seida Y, Nakano Y. Removal of humic substances by layered double hydroxides containing iron. Water Res 2000;34(5):1487–94. [11] Seida Y, Nakano Y. Removal of phosphate through dissolution–coagulation process of layered double hydroxide. J Chem Eng Japan 2001;34(7):906–11. [12] Hashimoto K, Nishihori M, Yamauchi T, Sekine K. Tetsu-sutenresu Sessyokuzai wo mochiita Haisuiro karano Rin Jyokyo. Mizusyori Gijutsu 1996;37(11):19–25. [13] Nihon Bunseki Kagaku-kai, Bunseki Kagaku Hannou no Kiso, Nihon Bunseki Kagaku Kai, Hokkaido Shibu-hen Fujimoto et al., editors. Baifukan, Tokyo, 1980. p. 67–9. [14] Freiser H, Fernando Q. Ionic equilibria in analytical chemistry. New York: Wiley, 1963.