Removal of arsenic from water: Effect of calcium ions on As(III) removal in the KMnO4–Fe(II) process

Журнал(книга):

Раздел науки:

Читать онлайн: 
Текст статьи: 
Xiaohong Guana,*, Jun Maa,b,*, Haoran Donga, Li Jianga aState Key Lab of Urban Water Resource and Environment (HIT), Harbin Institute of Technology, Harbin, PR China bNational Engineering Research Center of Urban Water Resources, Harbin Institute of Technology, Harbin 150090, PR China a r t i c l e i n f o Article history: Received 23 September 2008 Received in revised form 5 December 2008 Accepted 30 December 2008 Available online 20 January 2009 Keywords: As(III) Drinking water Permanganate Fe(II) Fe(III) formed in situ Calcium a b s t r a c t A novel KMnO4–Fe(II) process was developed in this study for As(III) removal. The optimum As(III) removal was achieved at a permanganate dosage of 18.6 mM. At the optimum dosage of permanganate, the KMnO4–Fe(II) process was much more efficient than the KMnO4– Fe(III) process for As(III) removal by 15–38% at pH 5–9. The great difference in As(III) removal in these two processes was not ascribed to the uptake of arsenic by the MnO2 formed in situ but to the different properties of conventional Fe(III) and the Fe(III) formed in situ. It was found that the presence of Ca2þ had limited effects on As(III) removal under acidic conditions but resulted in a significant increase in As(III) removal under neutral and alkaline conditions in the KMnO4–Fe(II) process. Moreover, the effects of Ca2þ on As(III) removal in the KMnO4–Fe(II) process were greater at lower permanganate dosage when Fe(II) was not completely oxidized by permanganate. This study revealed that the improvement of As(III) removal at pH 7–9 in the KMnO4–Fe(II) process by Ca2þ was associated with three reasons: (1) the specific adsorption of Ca2þ increased the surface charge; (2) the formation of amorphous calcium carbonate and calcite precipitate that could coprecipitate arsenate; (3) the introduction of calcium resulted in more precipitated ferrous hydroxide or ferric hydroxide. On the other hand, the enhancement of arsenic removal by Ca2þ under acidic conditions was ascribed to the increase of Fe retained in the precipitate. FTIR tests demonstrated that As(III) was removed as arsenate by forming monodentate complex with Fe(III) formed in situ in the KMnO4–Fe(II) process when KMnO4 was applied at 18.6 mM. The strength of the ‘‘non-surface complexed’’ As–O bonds of the precipitated arsenate species was enhanced by the presence of Ca2þ and the complexation reactions of arsenate with Fe(III) formed in situ in the presence or absence of Ca2þ were proposed. ª 2009 Elsevier Ltd. All rights reserved. 1. Introduction In recent years, arsenic has become a priority contaminant of concern due to its toxicological and carcinogenic effects on humans. Drinking water regulations have continued to lower the maximum contaminant level (MCL) for pollutants based on more information about health effects. The World Health Organization, United States Environmental Protection Agency (USEPA), and the Ministry of Health of PR China have revised the MCL for As in drinking water from 50 to 10 mg/L (WHO, 1993; USEPA, 2004; MHPRC, 2007). Existing treatments may not be able to attain the new MCL, or the cost for water treatment * Corresponding authors. Harbin Institute of Technology, Department of Municipal Engineering, 202#, Haihe Road, Harbin, China. Tel.: þ86 451 8628 3010; fax: þ86 451 82368074. E-mail addresses: hitgxh@126.com (X. Guan), majun@hit.edu.cn (J. Ma), dhrrhd@163.com (H. Dong), jiangli_vip@163.com (L. Jiang). Available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/watres 0043-1354/$ – see front matter ª 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2008.12.054 w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 to meet the new MCL may be beyond the acceptable level. Therefore the stiffening of regulations generates strong demands to improve methods for removing arsenic from the water and controlling water treatment residuals. Coagulation–flocculation, oxidation/reduction reactions, ion exchange, membrane processes, and adsorption are the most common methods for arsenic removal (USEPA, 2002; Garelick et al., 2005; Jiang, 2001). However, these methods have some limitations for As(V) removal and especially for the removal of As(III). Coagulation–flocculation was found to be not so efficient for As(III) as for As(V) and the production of sludge containing arsenic is not desirable. Arsenic removal by ion exchange was reported to be seriously influenced by the competing ions and ion exchange cannot remove As(III) effectively because H3AsO3 is not dissociated in the relevant pH range for drinking water (Karcher et al., 1999). The membrane process for arsenic removal was unattractive for most practical cases because of its high cost (USEPA, 2002). Shannon et al. (2008) reviewed the science and technology for water purification in the coming decades and pointed out that affordably reducing As(III)/As(V) concentrations to levels currently thought of as safe (<10 ppb), without producing toxic waste disposal issues has proved to be a major challenge. Moreover, when developing an arsenic remediation strategy for developing countries, one must consider the economic feasibility and simplicity of the system (Ciardelli et al., 2008). Arsenic occurs in waters in several different forms depending on the pH and redox potential Eh (Lytle et al., 2005). Arsenate (As(V)) and arsenite (As(III)) are the primary forms of arsenic found in natural waters. The thermodynamically stable forms of arsenic are As(V) in oxygenated surface water and As(III) in reducing groundwater (Lytle et al., 2005). The latter is of concern in this study because As(III) is much more difficult to remove than As(V). If only As(III) is present, consideration should be given to oxidation prior to coagulation to convert As(III) to As(V) species. Therefore, Lee et al. (2003) employed iron(VI) to oxidize As(III) to As(V) and then removed As(V) by Fe(III) coagulation. Liu (2005) applied potassium permanganate to oxidize As(III) to enhance the removal of As(III) by ferric sulfate and the removal efficiency of As(III) was enhanced by about 20%. Borho and Wilderer (1996) carried out pilot-scale experiments with a very low initial As(III) concentration varying from 0.034 to 0.044 mg L1 and revealed that the coupling of manganese dioxide coated quartz sand for As(III) oxidation with iron(II)/oxygen to bind the generated As(V) molecules was very effective for removing As(III). However, their investigation was a very preliminary one and to our knowledge, the combined effects of As(III) oxidation by a strong oxidant and subsequent removal of As(V) by Fe(II) have never been evaluated extensively. Because permanganate produces no harmful by-products, is easy to apply and affordable, this study aims at evaluating the synergetic performance of permanganate and Fe(II) in removing As(III) and investigating the mechanism of As(III) removal in the KMnO4–Fe(II) process. In this process, permanganate and Fe(II) were dosed to the As(III)-bearing water sequentially and permanganate was applied to induce the oxidation of both As(III) and Fe(II). Fe(III) derived from Fe(II) (Fe(III) formed in situ) worked as the coagulant to remove arsenic in this process. Calcium is ubiquitously present in natural water bodies including surface water, groundwater and seawater (Anazawa and Ohmori, 2001; Liu et al., 2007) and it is well known that calcium has great impacts on adsorption of anions. Many studies have been carried out to examine the effects of calcium on removing arsenate by coagulation or adsorption (Hering et al., 1996; Smith and Edwards, 2005; Liu et al., 2007). However, most of the studies investigating the effects of calcium on arsenate adsorption or removal focused on alkaline conditions. Few studies have systematically examined the effects of calcium on arsenic removal by coagulation over a wide pH range. Moreover, carbonate alkalinity is ubiquitous in water and groundwater (Stumm and Morgan, 1996). Thus, this study investigated the effects of calcium ions on As(III) removal in the KMnO4–Fe(II) process over a wide pH range and at different KMnO4 dosages with carbonate as the co-existing alkalinity. 2. Methods and materials All chemicals were reagent-grade and were used without any purification. All solutions were prepared with distilled water. The stock solutions of As(III) and As(V) were prepared from reagent-grade NaAsO2 and Na3AsO4$7H2O, respectively, and the stock solutions were disposed of in 1 week if they were not used up. CaCl2 stock solutions containing 25 mM Ca2þ was prepared every 2 weeks. Ferrous sulfate or ferric sulfate was employed as the coagulant in this study and was freshly prepared for each set of experiments. Background electrolyte solutions were prepared from the reagent-grade salts NaNO3 and NaHCO3. Jar tests were performed with a standard jar testing device (Stuart Scientific) to simulate a conventional coagulation/ flocculation process. If not otherwise specified, the jar testing procedure was initiated with rapid mixing at 300 rpm for 2 min for As(III) oxidation by permanganate followed by 120 rpm for 1 min, 40 rpm for 30 min, consecutively, and finally there was a 30 min settling. Rapid mixing was started immediately after dosing the pre-determined aliquots of potassium permanganate into 1 L synthetic water containing As(III) and ferrous sulfate or ferric sulfate was dosed after the rapid mixing. The initial As(III) or As(V) concentration was 13.3 mM and the coagulant was applied at 44.6 mM, if not otherwise specified. All jar tests were carried out in a temperature controlled room at 20–23 C and each experiment was carried out in (at least) duplicate. All experiments were performed with a constant ionic strength of 0.01 M NaNO3. As previous study revealed that alkalinity had negligible effect on arsenate removal over a wide pH range (Meng et al., 2000), 0.001 M NaHCO3 was added to provide necessary alkalinity and all experiments were carried out open to the atmosphere. Sodium hydroxide and hydrochloric acid were used to adjust the pH of the solutions, which was kept constant throughout the jar tests. All glassware was cleaned by soaking in 10% HNO3 and rinsed three times with distilled water. After each test, the supernatant was sampled and filtered immediately through a cellulose acetate membrane (MFS) of 0.45 mm pore size for the determination of total As, Fe and Mn 5120 w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 by inductively coupled plasma mass spectrometry (ICP-MS) method. In the ICP-MS method, all samples and standards were acidified according to the standard methods (APHA, 1995). The precipitates were collected, washed with distilled water and then freeze-dried before being subjected to SEM and FTIR analysis. The SEM images of the precipitates were collected using the scanning electron microscope with an EDAX (FE-SEM, S-4700, HITACHI, Japan) at an accelerating voltage of 20.0 kV. Diffuse reflectance FTIR spectra of the precipitates were recorded on a PerkinElmer Spectrum One FTIR. The samples were diluted to a concentration of 2% with IR-grade KBr. Sixty-four signal-averaged scans were collected at 2 cm1 resolution in the mid-IR region (4000–400 cm1) for pure KBr and for each KBr-mixed sample. Vibrational spectra of each sample were obtained by subtraction of the background spectra (pure KBr) from the spectra of KBr-mixed sample. A high performance pH meter with a saturated KCl solution as electrolyte (Corning 350) was used to measure solution pH. Daily calibration with proper buffer solutions (pH 4.00, 6.86 and 9.18) was performed to ensure its accuracy. The electrophoretic mobility of the flocs formed in the KMnO4– Fe(II) process was investigated at room temperature with a Zetasizer 3000HSA (Malvern Instruments Ltd, UK). 3. Results and discussion 3.1. As(III) removal in the KMnO4–Fe(II) process as functions of pH and permanganate dosages Different doses of KMnO4 as oxidant and ferrous sulfate applied at 44.6 mM as coagulant were applied to synthetic water containing 13.3 mM As(III) at different pH to investigate the effects of pH and permanganate dosages on As(III) removal in the KMnO4–Fe(II) process, as shown in Fig. 1a. When permanganate was applied at 9.3 mM, arsenic removal decreased from 81% at pH 4 to 57% at pH 6 and arsenic removal was almost negligible at pH 7–9. Increasing the dosage of permanganate from 9.3 mM to 14.3 mM significantly enhanced arsenic removal at pH 4–9. Arsenic removal at pH 7–9 was considerably improved by increasing permanganate dosage from 14.3 mM to 18.6 mM. However, further increase in permanganate dosage from 18.6 mM to 23.8 mM led to a notable decrease in arsenic removal in the KMnO4–Fe(II) process at pH 6–9. Thus the optimum As(III) removal in the KMnO4–Fe(II) process was achieved at permanganate dosage of 18.6 mM. It is well known that the oxidation potential and the reduction products of potassium permanganate are pHdependent. Under acidic conditions the oxidation half-reaction is (Weast, 1990): MnO 4 þ 8Hþ þ 5e ¼ Mn2þ þ 4H2O E0 ¼ þ1:51 V (1) Under weak acidic, neutral and weak alkaline conditions, the half-reaction is [21]: MnO 4 þ 2H2O þ 3e ¼ MnO2ðSÞ þ 4OH E0 ¼ þ0:588 V (2) Therefore oxidation reactions of As(III) and Fe(II) by permanganate are possibly to be as follows: 5H3AsO3 þ 2MnO 4 ¼ 5AsO3 4 þ 2Mn2þ þ 3H2O þ 9Hþ (3) 3H3AsO3 þ 2MnO 4 þ 7OH ¼ 3AsO3 4 þ 2MnO2 þ 8H2O (4) 5Fe2þ þ MnO 4 þ 11H2O ¼ 5FeðOHÞ3Y þ Mn2þ þ 7Hþ (5) 3Fe2þ þ MnO 4 þ 5H2O ¼ 3FeðOHÞ3Y þ MnO2 þ Hþ (6) According to Eqs. (3) and (4), 5.4 mM and 8.9 mM permanganate are required to oxidize 13.3 mM As(III) completely when the reduction products are Mn2þ and MnO2, respectively. For the complete oxidization of 44.6 mM Fe(II), 8.9 mM and 14.9 mM permanganate are required when the reduction products are Mn2þ and MnO2, respectively. Because As(III) could be completely oxidized under all conditions investigated in this study, the difference in arsenic removal efficiency in the KMnO4–Fe(II) process at different KMnO4 dosages could be ascribed to the different fractions of Fe(II) oxidized by KMnO4 and the properties of Fe(III) formed in situ generated under different conditions. Because only the precipitated Fe could mediate the removal of arsenic (McNeill and Edwards, 1997), the amounts of Fe retained in the precipitate were investigated as functions of pH and permanganate dosages, as illustrated in Fig. 1b. When permanganate dosage was increased from 9.3 mM to 14.3 mM, the solubility of iron decreased significantly. The Fig. 1 – (a) Effect of pH on As(III) removal and (b) the amount of Fe retained in the precipitate in the KMnO4-Fe(II) process at different KMnO4 dosages (As(III)=13.3 mM, Fe(II)=44.6 mM). w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 5121 amount of Fe retained in the precipitate was further increased as the permanganate dosage was increased from 14.3 mM to 18.6 mM at pH 7–9. However, it was found that the solubility of Fe was increased at pH 7–9 by increasing KMnO4 dosage from 18.6 mM to 23.8 mM. The minimum solubility of Fe(III) formed in situ achieved at permanganate dosage of 18.6 mM suggested that permanganate was reduced to MnO2 and Mn2þ, respectively, to oxidize As(III) and Fe(II) according to Eqs. (3–6). The analysis of As(III) removal in the KMnO4–Fe(II) process and the amounts of Fe retained in the precipitates as functions of permanganate dosages and pH implied that the removal efficiency of As(III) was strongly dependent on pH and the amount of Fe entrapped in the precipitate. The removal of arsenic in the KMnO4–Fe(II) process decreased with increasing pH at various permanganate dosages, which may be associated with the variation of arsenate species and the surface charge of ferric hydroxide or ferrous hydroxide with pH. As(III) was oxidized to As(V) in the oxidation process and arsenate species become more negatively charged with increasing pH (Guan et al., 2008). Fe(III) formed in situ and Fe(II) hydrolyzed in water to form Fe(OH)3 and Fe(OH)2, respectively. The surface charge of Fe(OH)3 and Fe(OH)2 particles decreases as pH increased from 4 to 9, as shown in Fig. 2. Therefore, the removal efficiency of As(III) in the KMnO4–Fe(II) process was determined by the following reasons: the amount of Fe in the precipitate, arsenate speciation and surface charge of the precipitates. As(III) removal in the KMnO4–Fe(III) process was also investigated at permanganate dosage of 9.3 mM, which was enough to oxidize 13.3 mM As(III), as demonstrated in Fig. S1. The percentage of arsenic removal is almost identical in the KMnO4–Fe(II) and KMnO4–Fe(III) processes at pH 4 but arsenic removal in the KMnO4–Fe(II) process (KMnO4 ¼ 18.6 mM) is higher than that in the KMnO4–Fe(III) (KMnO4 ¼ 9.3 mM) process by 15, 25, 38, 38 and 32% at pH 5, 6, 7, 8, and 9, respectively. Therefore higher arsenic removal efficiency can be achieved at the same iron dose in the KMnO4–Fe(II) process than in the KMnO4–Fe(III) process. Consequently, much lower iron doses than those required when using ferric iron are needed to obtain similar performance and much less arsenic-bearing sludge will be produced accordingly. In the KMnO4–Fe(III) process, As(III) was oxidized by KMnO4 and then removed by Fe(III) coagulation. However, As(III) was oxidized by KMnO4 and then removed by the Fe(III) formed in situ in the KMnO4–Fe(II) process. The great difference may be ascribed to the different properties of conventional Fe(III) and the Fe(III) formed in situ and the different amounts of MnO2 formed in situ. As a result, the uptake of As(V) by MnO2 formed in situ was examined, as shown in Fig. S2. It was found that only 1.4–3.2% and 2.0–5.6% of arsenate was removed by 10 mM and 18.6 mM MnO2 formed in situ, respectively. Therefore, the great difference in As(III) removal in these two processes was not ascribed to the uptake of arsenic by the MnO2 formed in situ but to the different properties of conventional Fe(III) and the Fe(III) formed in situ. It is possible that Fe(III) formed in situ is more effective than the conventional Fe(III) salts in removing other contaminants, such as phosphate, NOM and so on. 3.2. The interaction of As(V) with the Fe(III) formed in situ Many studies have investigated the ATR-FTIR spectra of free arsenate ions at various pH levels (Roddick-Lanzilotta et al., 2002; Pena et al., 2006; Guan et al., 2008). A strong peak appearing at 792 cm1 was observed in the IR spectrum of AsO4 3 ions (Roddick-Lanzilotta et al., 2002; Pena et al., 2006). A band at 859 cm1 was observed for the spectrum of HAsO4 2 and was assigned as vas(As–O) in HAsO4 2 (Guan et al., 2008). Two peaks at 907 and 877 cm1 were observed in the spectrum of H2AsO4 at pH 4.2 due to the splitting of the v3 mode. These two bands were assigned to vas(As–O) and vs(As–O), respectively. The vas(As–O) in H2AsO4 appearing at higher frequency than that in HAsO4 2 could be ascribed to the different modes of electron delocalization. Electron delocalization occurs on the two oxygen atoms in H2AsO4 and on the three oxygen atoms in HAsO4 2, which makes the bond order of the As–O bonds in these two moieties 3/2 and 4/3, respectively. At pH 2.1 another new band at 930 cm1 appeared and could be assigned to the vas(As–O) in H3AsO4. The infrared spectra of arsenic in precipitate collected in the KMnO4–Fe(II) (KMnO4 ¼ 18.6 mM) process at pH 5 and pH 7 are demonstrated in Fig. 3. A strong, well-resolved band at 835–834 cm1 appeared in the As–O stretching vibration region, indicating that As(III) was oxidized completely and removed as As(V). The peak positions of the precipitated samples were significantly different from those of the Fig. 2 – Effect of calcium on the Zeta potential of flocs formed in the KMnO4-Fe(II) process for As(III) removal (As(III)=13.3 mM, Fe(II)=44.6 mM). 5122 w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 dissolved As(V) species, which may be attributable to symmetry reduction. If the symmetry reduction were caused by protonation, as would be the case for outer-sphere adsorption, the bands would exhibit at the similar positions as the corresponding dissolved As species. Therefore the band shift observed in this study indicated the formation of innersphere complexes. Because metal ions are generally less electronegative than H atoms, metal ions are not as strongly coordinated to oxygen as H atoms (Tejedor-Tejedor and Aderson, 1990; Myneni et al., 1998a; Guan et al., 2005). The O atom binding with Fe has an empty orbit that partially participates in electron delocalization and in turn the strength of the As–O uncomplexed bond is reduced. Therefore, the As– O uncomplexed bond in (FeO)2AsO2 would be weaker than that in (HO)2AsO2  and the As–Ouncomplexed bond in (FeO)AsO3 would be weaker than that in (HO)AsO3 2. Consequently, red-shifts in the IR stretch frequencies would be predicted as a result of arsenate complexation to Fe(III) formed in situ. As discussed above, the peak at 835–832 cm1, which appears at a lower frequency than the vas(As–O)uncomplexed in HAsO4 2, was assigned to v(As–O)uncomplexed in monodentate complex (FeO)AsO3 and the precipitated arsenate was deprotonated at pH 5 and 7 (Myneni et al., 1998a,b). 3.3. Effect of Ca2þ ions on As(III) removal and the amount of precipitated Fe(III) formed in situ The effects of 2.5 mM Ca2þ ions on As(III) removal in the KMnO4–Fe(II) process were examined at different pH and two different KMnO4 dosages, 9.3 mM and 18.6 mM. As shown in Fig. 4a, when KMnO4 was dosed at 18.6 mM, the dosing of 2.5 mM Ca2þ did not affect the removal of arsenic at pH 4–6 while it enhanced arsenic removal by 5–25% at pH 7–9 and the enhancement was more significant at higher pH. Arsenic removal was increased by 4–19% at pH 4–6 and by 67–75% at pH 7–9 due to the presence of 2.5 mM Ca2þ when KMnO4 was applied at 9.3 mM, as illustrated in Fig. 5a. Enhancement of arsenic removal by calcium at high pH when arsenic was removed in the coagulation or adsorption process was also observed by many other researchers and it was attributable to the fact that the specific adsorption of Ca2þ increased the surface positive charge and led to the enhanced retention of arsenate (Hering et al., 1996; Wilkie and Hering, 1996; Smith et al., 2002; Smith and Edwards, 2005). The Ca2þ ions can compress the double layer and reduce the characteristic double layer thickness for an anionic surface from Fig. 3 – FTIR spectra of clean precipitates, precipitates bearing arsenic and precipitates bearing arsenic in the presence of calcium in the KMnO4-Fe(II) process (a) at pH 5 (b) at pH 7 (As(III)=13.3 mM, KMnO4=18.6 mM, Fe(II)=44.6 mM, Ca2+=2.5 mM). Fig. 4 – Effect of calcium on (a) As(III) removal and (b) the amount of Fe in the precipitate in the KMnO4-Fe(II) process (As(III)=13.3 mM, KMnO4=18.6 mM, Fe(II)=44.6 mM). w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 5123 9 nm to about 1.5 nm, which can help to reduce charge exclusion and make internal pores more accessible and enhance the removal of arsenate accordingly (Smith and Edwards, 2005; Masue et al., 2007). Moreover, adsorption of Ca2þ might contribute to a reduced negative charge on the flocs surface and hence provide conditions more favorable for arsenate removal (Masue et al., 2007). Liu et al. (2007) studied the effectiveness of calcium in enhancing arsenic removal by ferric chloride in the presence of silicate and showed that calcium increased the arsenic removal through the following ways: the introduction of calcium increased the surface charge of ferric hydroxide, facilitated the formation of larger ferric hydroxide flocs and increased the amount of precipitated solids. Jia and Demopoulos (2008) investigated the coprecipitation of arsenate with iron(III) with sodium hydroxide as the base and Ca2þ as the co-ions and found that the residual arsenate in the solution was reduced significantly. They concluded that the calcium ions were involved in direct association with arsenate in the precipitate in addition to adsorption by ferrihydrite. As illustrated in Fig. 2, the presence of 2.5 mM Ca2þ had a minor effect on the Zeta potential of the flocs at pH 4 and 5 but it enhanced the surface charge of the precipitates at pH 6–9 and the enhancement was greater under more alkaline conditions when permanganate was applied at 9.3 mM or 18.6 mM. This confirmed that the specific adsorption of Ca2þ and the compression of the double layer increased the surface positive charge and resulted in the enhanced retention of arsenate in the KMnO4–Fe(II) process under alkaline conditions. Fig. 4b shows that Ca2þ had almost no influence on the amount of Fe in the precipitate when KMnO4 was dosed at 18.6 mM, which indicated that the improvement of arsenic removal by Ca2þ at pH 7–9 was not ascribed to the formation of more ferric hydroxide precipitates in the presence of Ca2þ. On the other hand, the introduction of calcium increased the amount of Fe retained in the precipitate slightly and significantly at pH 4–6 and at pH 7–9, respectively, at a permanganate dosage of 9.3 mM, as demonstrated in Fig. 5b. Therefore, the improvement of As(III) removal in the KMnO4–Fe(II) process when KMnO4 was dosed at 9.3 mM due to the presence of 2.5 mM Ca2þ at pH 4 and 5 could be ascribed to the formation of more precipitates and the improvement of As(III) removal at pH 6 was associated with the formation of more precipitates and the increase of surface charge of the precipitates (Hering et al., 1996; Wilkie and Hering, 1996; Smith et al., 2002). The presenceof 2.5 mM Ca2þ resulted in theenhancement of As(III) removal by 74.4, 73.1 and 66.6%, respectively, at pH 7, 8, and 9 when KMnO4 was dosed at 9.3 mM. In addition, the introduction of 2.5 mM Ca2þ increased the amount of Fe retained in theprecipitatesby 30.4, 37.5 and 35.7 mM,respectively, atpH 7, 8, and 9. The significant enhancement in the formation of precipitates and the removal of As(III) at pH 7–9 could be associated with the formation of CaCO3 precipitates, which could entrap Fe(II) ions and As(V). The explanation is as follows. It was observed in our study that no obvious flocs were formed at pH 7–9 in the absence of calcium when KMnO4 was dosed at 9.3 mM. However, large flocs were formed under these conditions in the presence of calcium, which may be due to the formation of calcium carbonate precipitate. As sodium bicarbonate was employed in this study to simulate the alkalinity in water and the species distribution of H2CO3, HCO3 and CO3 2 as a function of pH was calculated and plotted in the inlay of Fig. 4. The concentrations of CO3 2 are 4.53  107, 5.46  106, and 5.31  105 M, respectively, at pH 7, 8, and 9. As the experiments in this study were carried out open to the atmosphere, the concentrations of CO3 2 at pH 7, 8, and 9 should be 2.51  108, 2.51  106, and 2.51  104 M, respectively, considering the equilibrium of CO2 in the solution and the atmosphere (Stumm and Morgan, 1996). The concentration of Ca2þ employed in this study was 2.5  103 M (2.5 mM) and the Ksp of CaCO3 at 25 C is 108.4 (Stumm and Morgan, 1996; Li et al., 2007). Therefore the solution was undersaturated with respect to calcium carbonate at pH 7 and saturated at pH 8 and 9. However, the formation of calcium carbonate at pH 7 cannot be ruled out. It was reported previously that surface precipitation of ferric phosphate on goethite (Ler and Stanforth, 2003), ferric arsenate on ferrihydrite (Jia et al., 2006) occurred under apparently undersaturated conditions. Therefore, the FTIR and SEM/ EDAX analyses were carried out for the precipitates collected at pH 5 and 7 in the presence or absence of 2.5 mM Ca2þ when permanganate was dosed at 18.6 mM or 9.3 mM, as demonstrated in Figs. 3, 6 and 7, respectively. The FTIR spectrum of the precipitates collected in the KMnO4–Fe(II) process in the presence of Ca2þ ions at pH 5 resembled that of the precipitates collected in the absence of Ca2þ ions at pH 5, indicating that calcium carbonate was not Fig. 5 – Effect of calcium on (a) As(III) removal and (b) the amount of Fe in the precipitate in the KMnO4-Fe(II) process (As(III)=13.3 mM, KMnO4=9.3 mM, Fe(II)=44.6 mM). 5124 w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 formed under this condition. However, strong and well resolved bands appearing at 1473 cm1, 1417 cm1, 1376 cm1, 874 cm1, 712 cm1 were observed in the FTIR spectrum of precipitates collected in the KMnO4–Fe(II) process in the presence of Ca2þ ions at pH 7. The peaks at 1473 cm1, 1417 cm1, and 1376 cm1 could be assigned to the formation of amorphous calcium carbonate while those at 874 cm1 and 712 cm1 might be assigned to the characteristic vibration of a calcite phase (Shen et al., 2006). Therefore, the formation of MnO2 and Fe(OH)3 facilitated the nucleating precipitation of calcium carbonate and generated two phases of calcium carbonate, amorphous calcium carbonate and calcite. The SEM images shown in Figs. 6 and 7 illustrate that the presence of calcium changed the morphology of the precipitate significantly. The presence of 2.5 mM calcium resulted in much more heterogeneous and rough surface of precipitates collected from the KMnO4–Fe(II) process for As(III) removal. Furthermore, much larger precipitate particles were formed in the presence of 2.5 mM Ca2þ compared to those formed in the absence of Ca2þ, which could facilitate the solid–liquid separation. The EDAX analysis revealed that many more calcium ions were entrapped in the precipitate at pH 7 than at pH 5, consistent with the FTIR analysis. Therefore, the improvement of arsenic removal at pH 7–9 in the KMnO4–Fe(II) process caused by the introduction of calcium Fig. 6 – SEM-EDAX analysis for the precipitates collected in the KMnO4-Fe(II) process for As(III) removal (As(III)=13.3 mM, KMnO4=18.6 mM, Fe(II)=44.6 mM) under various conditions (a) pH=5, Ca2+=0 mM; (b) pH=5, Ca2+=2.5 mM; (c) pH=7, Ca2+=0 mM; (d) pH=7, Ca2+=2.5 mM. The magnification factor is 100 and the inserts show the atomic percentage of the elements. w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 5125 may be associated with three reasons: (1) the specific adsorption of Ca2þ increases the surface charge; (2) the formation of calcium carbonate/calcite precipitates which can co-precipitate arsenate; (3) the introduction of calcium results in more precipitated ferrous hydroxide or ferric hydroxide. It was the first two reasons that contribute to the improvement of arsenic removal at pH 7–9 when KMnO4 was applied at 18.6 mM. The results of this study are significant for practical water treatment and for the transport of arsenic in natural environments. The concentration ranges for total carbonate and calcium in the groundwater are 0.5–8 mM (Stumm and Morgan, 1996) and 0.04–5.28 mM (Smith and Edwards, 2005), respectively. Moreover, the average Bangladesh groundwater pH is above 7 (Ciardelli et al., 2008). When coagulation is employed to remove arsenic from groundwater, it is highly possible that CaCO3 can form under slightly alkaline conditions and contribute to the removal of arsenic. In natural systems, the presence of calcium and alkalinity should be taken into account simultaneously for modeling or predicting the transport of arsenic under neutral and alkaline conditions. 3.4. The interaction of As(V) with the Fe(III) formed in situ in the presence of Ca2þ ions The interaction of As(V) with the Fe(III) formed in situ in the presence of Ca2þ ions was investigated by collecting the FTIR spectra of the precipitates generated in the KMnO4–Fe(II) process with the presence of 2.5 mM Ca2þ at pH 5 and pH 7, as shown in Fig. 3. Fig. 3 shows that calcium ions had minor effect on the interaction of As(V) with the Fe(III) formed in situ at pH 5 because the peak corresponding to the stretching vibration of As–O uncomplexed in the precipitate appeared at Fig. 7 – SEM-EDAX analysis for the precipitates collected in the KMnO4-Fe(II) process for As(III) removal (As(III)=13.3 mM, KMnO4=9.3 mM, Fe(II)=44.6 mM) under various conditions (a) pH=5, Ca2+=0 mM; (b) pH=5, Ca2+=2.5 mM; (c) pH=7, Ca2+=0 mM; (d) pH=7, Ca2+=2.5 mM. The magnification factor is 100 and the inserts show the atomic percentage of the elements. 5126 w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 similar positions (832 cm1 and 835 cm1). However, the stretching vibration of As–Ouncomplexed in the precipitate at pH 7 shifted from 834 cm1 to 844 cm1, suggesting that the strength of the ‘‘non-surface complexed’’ As–O bonds of the precipitated arsenate species was enhanced by the presence of Ca2þ. Since the As–O uncomplexed appeared at a lower frequency than the vas(As–O)uncomplexed in HAsO4 2, it was proposed that monodentate arsenate complexes were formed in the KMnO4–Fe(II) process in the presence of Ca2þ at pH 5 and 7. Due to the poorer electron affinity of Ca2þ compared to Fe(III), the Ca–As bond is longer than the Fe–As bond (Jing et al., 2003). Accordingly, vas(As–O)uncomplexed in arsenate complexed with one Ca2þ ion would appear at a lower wavenumber than that in arsenate complexed with one Fe(III) ion. Considering the increase in Zeta potential of the precipitates under neutral and alkaline conditions caused by the introduction of Ca2þ ions and shift of the vibration of the As–O uncomplexed bond to higher frequency, the arsenate retained in the precipitate under neutral and alkaline conditions should coordinate with one Fe(III) ion and one Ca2þ ion or with two Ca2þ ions simultaneously. Accordingly, the interaction of arsenate with Fe(III) formed in situ in the presence of Ca2þ was proposed, as illustrated in Fig. 8. 4. Conclusions and environmental implications This study examined As(III) removal in a new process, KMnO4–Fe(II) process,andtheeffectofCa2þ onAs(III)removalinthisprocess.The following conclusions could be drawn from this study: (1) The removal of As(III) in the KMnO4–Fe(II) process decreased with increasing pH at various permanganate dosages and optimum As(III) removal in the KMnO4–Fe(II) process was achieved at a permanganate dosage of 18.6 mM; (2) FTIR analyses revealed that As(III) was oxidized completely and removed as As(V) by forming a monodentate complex (FeO)AsO3 with Fe(III) formed in situ and the precipitated arsenate was deprotonated at pH 5 and 7; (3) The percentage of arsenic removal is almost identical in the KMnO4–Fe(II) and KMnO4–Fe(III) processes at pH 4 but arsenic removal in the KMnO4–Fe(II) process (KMnO4 ¼ 18.6 mM) is much higher than that in the KMnO4–Fe(III) (KMnO4 ¼ 9.3 mM) process by 15–38% at pH 5–9. The great difference in As(III) removal in these two processes was not ascribed to the uptake of arsenic by the MnO2 formed in situ but to the different properties of conventional Fe(III) and the Fe(III) formed in situ; (4) When KMnO4 was dosed at 18.6 mM, the dosing of 2.5 mM Ca2þ did not affect the removal of arsenic at pH 4–6 while it enhanced arsenic removal by 5–25% at pH 7–9 in the KMnO4–Fe(II) process and the enhancement was more significant at higher pH. (5) The presence of 2.5 mM Ca2þ increased the removal of As(III) in the KMnO4–Fe(II) process by 4–19% at pH 4–6 and by 67–75% at pH 7–9, when KMnO4 was applied at 9.3 mM. (6) Zeta potential analysis revealed that Ca2þ had a minor effect on the Zeta potential of the flocs at pH 4–5 while it enhanced the surface charge of the flocs at pH 6–9 and the enhancement was greater at higher pH. FTIR and SEM/ EDAX analyses showed that very few Ca2þ ions were retained in the precipitates at pH 5 but many more Ca2þ ions were incorporated in the precipitates and CaCO3 was formed at pH 7. Moreover, the introduction of 2.5 mM Ca2þ increased the amount of Fe retained in the precipitates significantly at pH 7–9 when KMnO4 was dosed at 9.3 mM. The complexation reactions of arsenate with Fe(III) formed in situ in the presence or absence of Ca2þ were proposed. Acknowledgements This work was supported by the Natural Scientific Research Innovation Foundation at Harbin Institute of Technology Fig. 8 – Possible complexation reactions of As(V) with Fe(III) formed in situ in the presence Ca2D ions (As(III) [ 13.3 mM, KMnO4 [ 18.6 mM, Fe(II) [ 44.6 mM). w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8 5127 (HIT.NSRIF.2008.65) and the Development Program for Outstanding Young Teachers at Harbin Institute of Technology (HITQNJS.2007.038). Supplemental material Supplementary information for this manuscript can be downloaded at doi: 10.1016/j.watres.2008.12.054. r e f e r e n c e s Anazawa, K., Ohmori, H., 2001. Chemistry of surface water at a volcanic summit area, Norikura, central Japan: multivariate statistical approach. Chemosphere 45, 807–816. APHA, 1995. Standard Methods for the Examination of Water and Wastewater, nineteenth ed. American Public Health Association, Washington, DC. Borho, M., Wilderer, P., 1996. Optimized removal of arsenate(III) by adaptation of oxidation and precipitation processes to the filtration step. Water Sci. Technol. 34 (9), 25–31. Ciardelli, M.C., Xu, H.F., Sahai, N., 2008. Role of Fe(II), phosphate, silicate, sulfate, and carbonate in arsenic uptake by coprecipitation in synthetic and natural groundwater. Water Res. 42, 615–624. Garelick, H., Dybowska, A., Valsami-Jones, E., Priest, N., 2005. Remediation technologies for arsenic contaminated drinking waters. J. Soils Sediments 5, 182–190. Guan, X.H., Liu, Q., Chen, G.H., Shang, C., 2005. Surface complexation of condensed phosphate to aluminum hydroxide: an ATR-FTIR spectroscopic investigation. J. Colloid Interface Sci. 289, 319–327. Guan, X.H., Wang, J.M., Chusuei, C.C., 2008. Removal of arsenic from water with granular ferric hydroxide: macroscopic and microscopic studies. J. Hazard. Mater. 156, 178–185. Hering, J.G., Chen, P.Y., Wilkie, J.A., Elimelech, M., Liang, S., 1996. Arsenic removal by ferric chloride. J. Am. Water Works Assoc. 88, 155–167. Jia, Y.F., Demopoulos, G.P., 2008. Coprecipitation of arsenate with iron(III) in aqueous sulfate media: effect of time, lime as base and co-ions on arsenic retention. Water Res. 42, 661–668. Jia, Y.F., Xu, L., Fang, Z., Demopoulos, G.P., 2006. Observation of surface precipitation of arsenate on ferrihydrite. Environ. Sci. Technol. 40, 3248–3253. Jiang, Q., 2001. Removing arsenic from groundwater for the developing world – a review. Water Sci. Technol. 44, 89–98. Jing, C.Y., Korfiatis, G.P., Meng, X.G., 2003. Immobilization mechanisms of arsenate in iron hydroxide sludge stabilized with cement. Environ. Sci. Technol. 37 (21), 5050–5056. Karcher, S., Ca ´ceres, L., Jekel, M., Contreras, R., 1999. Arsenic removal from water supplies in northern Chile using ferric chloride coagulation. Water Environ. J. 13 (3), 164–169. Lee, Y., Um, I.K., Yoon, J., 2003. Arsenic(III) oxidation by iron(VI) (ferrate) and subsequent removal of arsenic(V) by iron(III) coagulation. Environ. Sci. Technol. 37, 5750–5756. Ler, A., Stanforth, R., 2003. Evidence of surface precipitation of phosphate on goethite. Environ. Sci. Technol. 37, 2694–2700. Li, X.D., Lei, Y.Z., Gao, X.D., Ma, C.Y., Dong, S.L., 2007. Calcium carbonate supersaturation and precipitation in Chinese mitten crab (Eriocheir japonica sinensis) larval ponds in China: mass mortality, crystal form analysis, and safety saturation index. Aquaculture 272 (1-4), 361–369. Liu, R.P., 2005. Effectiveness and mechanisms of permanganate and permanganate composites removing pollutants through integrated oxidation and adsorption. Dissertation, Harbin Institute of Technology, Harbin. Liu, R.P., Li, X., Xia, S.J., Yang, Y.L., Wu, R.C., Li, G.B., 2007. Calcium-enhanced ferric hydroxide co-precipitation of arsenic in the presence of silicate. Water Environ. Res. 79, 2260–2264. Lytle, D.A., Sorg, T.J., Snoeyink, V.L., 2005. Optimizing arsenic removal during iron removal: theoretical and practical considerations. J. Water Supply Res. Technol. 54, 545–560. Masue, Y., Loeppert, R.H., Kramer, T.A., 2007. Arsenate and arsenite adsorption and desorption behavior on coprecipitated aluminum:iron hydroxides. Environ. Sci. Technol. 41, 837–842. McNeill, L.S., Edwards, M., 1997. Predicting As removal during metal hydroxide precipitation. J. Am. Water Works Assoc. 89, 75–86. Meng, X.G., Bang, S., Korfiatis, G.P., 2000. Effect of silicate, sulfate, and carbonate on arsenic removal by ferric chloride. Water Res. 34, 1255–1261. Ministry of Health of the People’s Republic of China (MHPRC), 2007. http://www.moh.gov.cn/open/db_query/new/ 20070628143525.pdf. Myneni, S.C.B., Traina, S.J., Waychunas, G.A., Logan, T.J., 1998a. Experimental and theoretical vibrational spectroscopic evaluation of arsenate coordination in aqueous solutions, solids, and at mineral–water interfaces. Geochim. Cosmochim. Acta 62, 3285–3300. Myneni, S.C.B., Traina, S.J., Waychunas, G.A., Logan, T.J., 1998b. Vibrational spectroscopy of functional group chemistry and arsenate coordination in ettringite. Geochim. Cosmochim. Acta 62, 3499–3514. Pena, M., Meng, X.G., Korfiatis, G.P., Jing, C.Y., 2006. Adsorption mechanism of arsenic on nanocrystalline titanium dioxide. Environ. Sci. Technol. 40, 1257–1262. Roddick-Lanzilotta, A.J., McQuillan, A.J., Craw, D., 2002. Infrared spectroscopic characterization of arsenate(V) ion adsorption from mine waters, Macraes mine, New Zealand. Appl. Geochem. 17, 445–454. Shannon, M.A., Bohn, P.W., Elimelech, M., Georgiadis, J.G., Marin ´ as, B.J., Mayes, A.M., 2008. Science and technology for water purification in the coming decades. Nature 452, 301–320. Shen, Q., Wei, H., Zhou, Y., Huang, Y.P., Yang, H.R., Wang, D.J., Xu, D.F., 2006. Properties of amorphous calcium carbonate and the template action of vaterite spheres. J. Phys. Chem. B 110, 2994–3000. Smith, E., Naidu, R., Alston, A.M., 2002. Chemistry of inorganic arsenic in soils: II. Effect of phosphorus, sodium, and calcium on arsenic sorption. J. Environ. Qual. 31, 557–563. Smith, S.D., Edwards, M., 2005. The influence of silica and calcium on arsenate sorption to oxide surfaces. J. Water Supply Res. Technol. 54, 201–211. Stumm, W., Morgan, J.J., 1996. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters. Wiley-Interscience, New York. Tejedor-Tejedor, M.I., Aderson, M.A., 1990. Protonation of phosphate on the surface of goethite as studied by CIR-FTIR and electrophoretic mobility. Langmuir 6, 602–613. USEPA, 2002. Arsenic treatment technologies for soil, waste and water. Tech. Report No. EPA-542-R-02–004. USEPA, 2004. http://www.epa.gov/safewater/arsenic.html. Weast, R.C., 1990. Handbook of Chemistry and Physics. CRC Press, Boca Raton, FL. WHO, 1993. Guidelines for Drinking Water Quality: 1. Recommendations. World Health Organization, Geneva. Wilkie, J.A., Hering, J.G., 1996. Adsorption of arsenic onto hydrous ferric oxide: effects of adsorbate/adsorbent ratios and cooccurring solutes. Colloids Surf. A 107, 97–110. 5128 w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 5 1 1 9 – 5 1 2 8

Комментировать