Environ. Sci. Technol. 2007, 41, 8275–8280
Coprecipitation of Arsenate with Metal Oxides. 2. Nature, Mineralogy, and Reactivity of Iron(III) Precipitates A N T O N I O V I O L A N T E , * ,† STEFANIA DEL GAUDIO,† MASSIMO PIGNA,† MARIAROSARIA RICCIARDELLA,† AND DIPANJAN BANERJEE‡ Dipartimento di Scienze del Suolo, della Pianta e dell’Ambiente, Università di Napoli Federico II, Portici (Napoli), and Italy Department of Chemistry, Umeå University, 90187 Umea, Sweden
Received February 14, 2007. Revised manuscript received September 17, 2007. Accepted September 17, 2007.
Coprecipitation of arsenic with iron or aluminum occurs in natural environments and is a remediation technology used to remove this toxic metalloid from drinking water and hydrometallurgical solutions. In this work, we studied the nature, mineralogy, and reactivity toward phosphate of iron–arsenate coprecipitates formed at As(V)/Fe(III) molar ratios (R) of 0, 0.01, or 0.1 and at pH 4.0, 7.0, and 10.0 aged for 30 or 210 days at 50 °C and studied the desorption of arsenate. At R ) 0, goethite and hematite (with ferrihydrite at pH 4.0 and 7.0) crystallized, whereas at R ) 0.01, the formation of ferrihydrite increased and hematite crystallization was favored over goethite. In some samples, the morphology of hematite changed from rounded platy crystals to ellipsoids. At R ) 0.1, ferrihydrite formed in all the coprecipitates and remained unchanged even after 210 days of aging. The surface area and chemical composition of the precipitates were affected by pH, R, and aging. Chemical dissolution of the samples showed that arsenate was present mainly in ferrihydrite, but at R ) 0.01, it was partially incorporated into the structures of crystalline Fe oxides. The sorption of phosphate onto the coprecipitates was affected not only by the mineralogy and surface area of the samples but also by the amounts of arsenate present in the oxides. The samples formed at pH 4.0 and 7.0 and at R ) 0.1 sorbed lower amounts of phosphate than the precipitates obtained at R ) 0 or 0.01, despite the former having a larger surface area and showing only a presence of short-range ordered materials. This is mainly due to the fact that in the coprecipitates at R ) 0.1 arsenate occupied many sorption sites, thus preventing phosphate sorption. Less than 20% of the arsenate present in the coprecipitates formed at R ) 0.1 was removed by phosphate and more from the samples synthesized at pH 7.0 or 10.0 than at pH 4.0. Moreover, we found that more arsenate was desorbed by phosphate from a ferrihydrite on which arsenate was added than from an iron–arsenate coprecipitate,
* Corresponding author phone: 39 081 253 91 76; fax: 39 081 253 91 86; e-mail:
[email protected]. † Università di Napoli Federico II. ‡ Umeå University. 10.1021/es070382+ CCC: $37.00
Published on Web 11/13/2007
2007 American Chemical Society
attributed to the partial occlusion of some arsenate anions into the framework of the coprecipitate. XPS analyses confirmed these findings.
Introduction Arsenic is present in all natural environments and, being extremely toxic for living organisms, is of health and ecological concern. Its occurrence in soils, sediments, groundwaters, and surface waters may be due to natural processes as well as anthropogenic activities (1–5). The mobility of arsenic in soils, sediments, and waters is affected by sorption/desorption reactions, solid-phase precipitates, and coprecipitation with metal ions. The importance of oxides (mainly Fe oxides) in controlling the mobility and concentration of arsenic in natural environments has been well-studied (1, 3, 5–9). Recently, several studies have demonstrated that coprecipitation of arsenic with iron or aluminum occurs in natural environments (3, 7, 10–14). Rancourt et al. (11) reported that a natural As-rich hydrous ferric oxide was very similar to synthetic iron–arsenic coprecipitates, as opposed to sorbed arsenic on synthetic iron oxide, and was also different from all As-free synthetic and natural iron precipitates including ferrihydrite. Ford (13) demonstrated that arsenate coprecipitated with hydrous ferric oxides was stabilized against dissolution during the transformation of precipitates to more crystalline iron oxides and that crystallization of hematite or goethite was significantly inhibited at or above an arsenate loading of 390 mmol kg-1. The chemical coprecipitation of arsenic with iron or aluminum is one of the most practical and effective techniques for the removal of arsenic from hydrometallurgical solutions and from drinking waters (5, 15–19). It is generally regarded that As(V) is the oxidation state that is most efficiently removed by chemical coprecipitation, so if arsenite is present in a water source, a preoxidation process is recommended and usually applied (16, 18, 19). In the hydrometallurgical industry, the effluents containing arsenic are usually neutralized up to pH 7.0–9.0 to facilitate the removal of heavy metals as metal oxides (15–18). Despite the fact that coprecipitation of arsenic with iron or aluminum occurs in natural environments and is a remediation technology to remove arsenic from contaminated solutions, only a few detailed studies on the influence of pH, aging, and concentration of arsenic on the mineralogy, surface properties, chemical composition, and reactivity of metal–arsenic coprecipitates have been conducted (13, 20). In a previous study (20), the nature and reactivity of aluminum–arsenate precipitates were described, where it was observed that at a certain pH the crystallization of Al(OH)3 polymorphs was inhibited by arsenate, promoting the formation of very poorly crystalline Al-oxyhydroxides. Removal of arsenate by phosphate was affected by phosphate concentration, reaction time, and the nature and age of the precipitates. In this study, we present our findings on the mineralogy, nature, and reactivity of iron–arsenate coprecipitates formed at As/Fe(III) molar ratios (R) of 0, 0.01, or 0.1 and at pH 4.0, 7.0, or 10.0 and aged for 30 or 210 days at 50 °C. Desorption rates of arsenate through the addition of different amounts of phosphate to these coprecipitates are also described. Furthermore, a comparison of desorption kinetics of arsenate by phosphate where arsenic is coprecipitated with iron or added to a freshly synthesized iron oxide is also reported. VOL. 41, NO. 24, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8275
TABLE 1. Synthesis Conditions, Mineralogy, and Surface Area of the Precipitates Formed in the Absence or Presence of Arsenate after 30 or 210 Days of Aging at 50 °Ca synthesis conditions
surface area (m2 g-1)
mineralogy
sample
initial As/Fe molar ratio (R)
initial pH
aged 30 d
aged 210 d
aged 30 d
aged 210 d
4FR0 4FeR0.01 4FeR0.1 7FeR0 7FeR0.01 7FeAR0.01b 7FeR0.1 7FeAR0.1b 10FeR0 10FeR0.01 10FeR0.1
0 0.01 0.1 0 0.01 0.01 0.1 0.1 0 0.01 0.1
4.0 4.0 4.0 7.0 7.0 7.0 7.0 7.0 10.0 10.0 10.0
Go + Fh Fh + {H} Fh Go + (Fh) H + Fh + [Go] H + (Fh) Fh Fh H + Go H + [Go] Fh
Go + [Fh] H + (Go) + [Fh] Fh Go + [Fh] H + (Go) + [Fh]
85 ( 5 133 ( 5 116 ( 8 80 ( 5 152 ( 2 82 ( 4 147 ( 8 130 ( 6 25 ( 2 24 ( 3 154 ( 7
33 ( 2 46 ( 3 92 ( 5 53 ( 3 60 ( 1
Fh H + Go H + [Go] Fh
a Go, goethite; Fh, ferrihydrite; H, hematite; () not predominant; [] small amount; {}trace. immediately after the formation of iron precipitate.
Methods and Materials Preparation of Iron–Arsenate Coprecipitates. Nine iron precipitation products were prepared by slowly adding 0.5 mol L-1 NaOH under stirring up to a pH of 4.0, 7.0, or 10.0 to 1 L of 0.05 mol L-1 Fe(NO3)3 solutions in the absence or presence of arsenate (Na2HAsO4) at an initial As/Fe molar ratio (R) of 0, 0.01, or 0.1. The suspensions were aged at 50 °C for 30 or 210 days. Every 7–10 days, the pH of the suspensions was adjusted with 0.5 mol L-1 NaOH or HNO3 to the initial value (20). The synthesis conditions and symbols of the samples prepared are summarized in Table 1. Two samples (7FeAR0.01 and 7FeAR0.1) were obtained by adding arsenate (R ) 0.01 and 0.1, respectively) 30 min after the precipitation of iron at pH 7.0 and aged for 30 days at 50 °C. After aging, the suspensions were cooled and centrifuged at 10 000g min-1 for 30 min and washed five times with deionized water. The suspensions were translated to cellulose dialysis bags with a molecular weight cutoff of 15 000 and then dialyzed against deionized water, freeze-dried, and lightly ground to pass through a 0.315 mm sieve. Mineralogical, Physical, and Chemical Analyses. The X-ray diffraction (XRD) patterns of randomly oriented samples were obtained using a Rigaku diffractometer (Rigaku Co., Tokyo) equipped with Co KR radiation generated at 40 kV and 30 mA and a scan speed of 2° 2θ min-1. The XRD traces were the results of eight summed signals. The Fourier transform infrared spectroscopy (FT-IR) spectra of the samples were obtained using a Perkin-Elmer Spectrum One FT-IR Spectrophotometer (Perkin-Elmer USA), and transmission electron micrographs were obtained with a Philips EM 208S electron microscope operating at 120 kV (20). X-ray photoelectron spectroscopy (XPS) spectra were collected with a KRATOS Axis Ultra electron spectrometer under monochromatic Al KR radiation (1486.6 eV), equipped with a hybrid lens system with a magnetic lens and a charge neutralizer. An analysis pass energy of 160 eV with a step size of 1 eV was used for survey scans, while a pass energy of 20 eV with a step size of 0.1 eV was used for narrow scans. The binding-energy scale was referenced to the C 1s peak of aliphatic carbon contamination, which is ubiquitous in XPS analyses, at 285.0 eV. All samples were analyzed in duplicate and some in triplicate to check for reproducibility. The values are significant to one decimal place, and the estimated error range is about (5% of the reported values. The surface area of the samples was determined by H2O sorption at 20% relative humidity according to the method of Quirk (21). The point of zero charge (PZC) of selected samples (7FeR0, 10FeR0, 7FeR0.1, and 7FeAR0.1) was measured by laser doppler velocimetry–photon correlation 8276
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 24, 2007
b
127 ( 4 13 ( 1 14 ( 2 116 ( 5
Arsenate was added
spectroscopy using a DELSA 444 spectrometer (Beckham Coulter Electronics, Hialeah, FL). The chemical composition of the samples was determined by treating about 10 mg of the iron precipitation products aged for 30 or 210 days at 50 °C with 10 mL of 0.2 mol L-1 NH4-oxalate/oxalic acid at pH 3.0 or with 1 mL of 6 mol L-1 HCl (20, 22). The samples were kept to react for 10 h (20), since it has been demonstrated that the rate of dissolution of ferrihydrite precipitated in the presence of different concentrations of arsenate was strongly retarded by the presence of arsenate (23). Dissolved iron was subsequently determined by atomic absorption spectroscopy on a Perkin-Elmer AA700. Arsenate present in the samples after dissolution with HCl or with NH4-oxalate/oxalic acid at pH 3.0 was determined by inductively coupled plasma (ICPAES, Varian, Liberty 150). Phosphate Sorption Isotherms and Arsenate Removal. A total of 20–50 mg of each sample, in duplicate, was equilibrated at 20 °C with 19 mL of 0.05 mol L-1 KCl at pH 6.0. Suitable amounts of 0.01 mol L-1 solutions containing KH2PO4 were then added in order to have an initial phosphate concentration in the range 5 × 10-4 to 10-2 mol L-1. The pH of each suspension was kept constant for 24 h by adding 0.1 or 0.01 mol L-1 HCl or NaOH. The final suspensions (20 mL) were centrifuged at 10 000g for 20 min and filtered through a 0.22 µm membrane filter. Arsenate removed by different quantities of phosphate and phosphate in the final solutions were determined by ion chromatography, using a Dionex DX-300 Ion Chromatograph (Dionex Co, Sunnyvale, CA) (20, 22). The phosphate standard concentration was 0.1–2 mmol L-1; the arsenate standard concentration was 0.05–0.5 mmol L-1. The amount of phosphate sorbed was determined by the difference between the initial and final concentrations. Coefficients of variation among the replicates ranged from 2 to 4%. The sorption data of phosphate conformed to the Langmuir equation in the following form: X ) XmKc/(1 + Kc) where X is the amount of phosphate sorbed per unit mass of adsorbent (mmol kg-1), Xm is the maximum amount of phosphate that may be bound to the adsorbent (sorption capacity), c is the equilibrium solution concentration (mmol L-1), and K is a constant related to the binding energy. Kinetics of Desorption of Arsenate. A total of 25 mg of the iron–arsenate coprecipitates aged 30 or 210 days were equilibrated with 19 mL of 0.05 mol L-1 KCl. A large amount of phosphate (2000 mmol kg-1) was added to the suspensions to have a phosphate concentration in solution sufficiently high to facilitate arsenate desorption. The suspensions were allowed to react at pH 6.0 for 5, 24, 48, and 170 h. A comparison
of desorption kinetics of arsenate coprecipitated with iron (7FeR0.1) or sorbed onto the surfaces of a preformed iron precipitate (7FeAR0.1) aged 30 days was carried out by adding 500 mmol kg-1 of phosphate in order to have a final surface coverage of this anion onto the oxides of about 100%, calculated by the maximum value of phosphate sorbed on the samples (Xm) determined by sorption isotherms. Further experiments were also carried out on these two samples by adding 2000 mmol kg-1 of phosphate and keeping the suspensions to react for 100 days. The pH of the suspensions (pH 6.0) was kept constant for all of the experiments with 0.1 or 0.01 mol L-1 HCl or NaOH. Arsenate and phosphate were determined in the supernatants by ion chromatography, as discussed before. Coefficients of variation among the replicates ranged from 3 to 6%.
Results and Discussions Mineralogy and Surface Properties. The initial pH, As/Fe molar ratio (R), and aging influenced the final hydrolytic products of iron (Table 1; Figures S1–S3). In the absence of arsenate and at pH 4.0 (4FeR0), goethite and ferrihydrite formed after 30 days of aging, but further aging favored the transformation of ferrihydrite into goethite with a consequent decrease of the surface area of this sample from 85 to 33 m2/g (Table 1). In the presence of arsenate (samples 4FeR0.01 and 4FeR0.1), short-range ordered precipitates (ferrihydrite) formed after 30 days (Figures S1 and S2). After 210 days, hematite (predominant) and goethite crystallized in the sample 4FeR0.01 (the bands at 894 and 800 cm-1 characteristic of goethite appeared evident in the FT-IR spectrum; Figure S2), and its surface area decreased from 133 to 46 m2/g. In contrast, the sample 4FeR0.1 remained practically unchanged, and its surface area decreased by 21%, indicating that the crystallinity of the short-range ordered precipitate increased slightly after prolonged aging. The precipitation product formed at pH 7.0 in the absence of arsenate (7FeR0) showed the presence of goethite (predominant) and ferrihydrite after 30 days (Figure S3a). With further aging, the surface area of this sample decreased from 80 to 53 m2 g-1 (34%) due to partial transformation of ferrihydrite into goethite (Table 1). The sample formed at R ) 0.01 (7FeR0.01) showed the presence of hematite and ferrihydrite with a small amount of goethite after 30 days (Figures S1 and S3b), and hematite (predominant), goethite, and ferrihydrite after 210 days (Figures S2 and S3c). The crystals of hematite appeared subrounded (Figure S3b,c). The surface area of this iron precipitation product decreased substantially with aging (from 152 to 60 m2 g-1; 60%). It is interesting to note that a sample obtained at pH 7.0 and R ) 0.01 (7FeAR0.01), by adding arsenate immediately after the precipitation of iron, had a surface area of 82 m2 g-1 after 30 days and contained well-crystallized hematite (Table 1; Figure S3d), whereas the iron–arsenate coprecipitate 7FeR0.01 had a greater surface area (152 m2 g-1) and contained a large amount of ferrihydrite (compare Figure S3d with Figure S3b, Supporting Information). Evidently, arsenate added to preformed iron precipitates did not prevent, under certain conditions, the crystallization of iron oxides so effectively as when arsenate was coprecipitated with iron (Table 1). The samples obtained at R ) 0.1 and pH 7.0 (7FeR0.1) and 10.0 (10FeR0.1) showed the presence of ferrihydrite after 30 and 210 days of aging. The surface area of these materials slightly decreased after prolonged aging (Table 1). At pH 10, hematite and goethite formed at R ) 0 (10FeR0.1) and 0.01 (10FeR0.01) (Table 1; Figure S3e,f), but in the latter sample, hematite was predominant, indicating that the presence of arsenate promoted the crystallization of hematite over goethite (see also the sample 4FeR0.01 vs. 4FeR0 aged 210 days; Table 1). Hematite crystallization is favored by
TABLE 2. Amounts (mmol kg-1) of Iron and Arsenic Solubilized by 6 mol L-1 HCl (FeHCl, AsHCl) or 0.2 mol L-1 Oxalic Acid/NH4 Oxalate at pH 3.0 (FeOX, AsOX) from the Samples and Percentages of Iron Solubilized by NH4 Oxalate (FeOX) as Referred to the Total Iron Solubilized by HCl sample
FeOX % AsHCl mmol kg-1 AsOX AsOX/AsHCl Aged for 30 d
pH 4.0
4FeR0 4FeR0.01 4FeR0.1 pH 7.0 7FeR0 7FeR0.01 7FeR0.1 pH 10.0 10FeR0 10FeR0.01 10FeR0.1
77.5 98.0 94.7 66.5 75.0 89.0 7.1 9.3 95.3
140 980
120 1100
0.86 1.12
130 1020
100 1090
0.77 1.07
80 720
100 670
0.12 0.93
100 1020
50 1070
0.50 1.05
100 1050
30 1000
0.30 0.95
100 810
10 750
0.10 0.92
Aged for 210 d pH 4.0
4FeR0 4FeR0.01 4FeR0.1 pH 7.0 7FeR0 7FeR0.01 7FeR0.1 pH 10.0 10FeR0 10FeR0.01 10FeR0.1
34.5 50.0 85.1 21.9 46.0 84.5 4.2 7.5 83.0
conditions that promote coagulation of the ferrihydrite particles, whereas goethite forms readily at pH’s that promote dissolution of ferrihydrite (24–26). The influence of increasing concentrations of organic and inorganic ligands on the surface area of natural and synthetic aluminum or iron precipitates has been widely studied (ref 20 and references therein and refs 25–28). The morphology of hematites formed in the presence of arsenate also changed in some samples from rhombohedral or rounded platy crystals (10FeR0; Figure S3e) to ellipsoids (Figure S3f). Similar results were found by Galvez et al. (25, 26), who demonstrated that phosphate, an analog of arsenate, favored the formation of hematite over goethite and facilitated the formation of ellipsoidal or spindleshaped hematite. Points of zero charge of selected samples formed in the absence of arsenate (7FeR0 and 10FeR0) occurred at pH 8.0–8.5. The PZC of ferrihydrite and noncrystalline Fe oxides has been reported to range from 7.8 to 7.9 to 8.5 (24, 27). The PZC of ferrihydrite obtained from coprecipitating iron with arsenate at pH 7.0 (7FeR0.1) dropped to 6.3, whereas ferrihydrite obtained from the sample on which arsenate was added immediately after iron precipitation (7FeAR0.1) showed a PZC still lower (5.8), probably because more arsenate anions were present on the external surfaces as evidenced by XPS (as discussed below). These findings also demonstrate that arsenate added to an iron oxide or coprecipitated with iron forms inner-sphere complexes (3, 8, 9, 27–29). Chemical Composition of the Samples. The samples aged for 30 or 210 days at 50 °C were dissolved separately by 6 mol L-1 HCl, and by 0.2 mol L-1 NH4-oxalate/oxalic acid at pH 3.0. The oxalate method is widely used as a semiquantitative separation of ferrihydrite from more crystalline Fe oxides (25). The percentage of iron dissolved by NH4-oxalate/oxalic acid (FeOX) normalized to the total iron dissolved by HCl, the amounts of arsenate released after dissolution with HCl (AsHCl) or NH4-oxalate/oxalic acid (AsOX), and the AsOX/AsHCl molar ratios are reported in Table 2. The percentage of FeOX decreased with increasing the pH and aging time and increased with an increasing arsenate content. These findings VOL. 41, NO. 24, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8277
FIGURE 1. Sorption isotherms of phosphate onto the coprecipitates formed at pH 4.0 and at R ) 0 (4FeR0), 0.01 (4FeR0.01), and 0.1 (4FeR0.01) at pH 6.0, and 0.02 mol L-1 ionic strength KCl and at 25 °C (for sample symbol, see Table 1). strengthen the observation that the presence of arsenate strongly prevented the crystallization of goethite and hematite, promoting the formation of poorly crystalline and more soluble precipitates (Table 1 and Figures S1–S3). In the precipitates which showed only the presence of ferrihydrite (Table 1), the percentage of FeOX ranged from 83.0 to 98.0%. In the samples obtained at R ) 0.1, AsHCl and AsOX were substantially similar (AsHCl/AsOX molar ratios being practically equal to 1) and usually decreased with increasing pH. In contrast, in the samples formed at R ) 0.01, AsOX/AsHCl molar ratios were always lower than 1, particularly in the precipitates where crystalline Fe oxides (hematite and goethite) predominated over ferrihydrite (e.g., 7FeR0.01 and 10FeR0.01 aged both 30 and 210 days). In a previous work (20), while studying the influence of arsenate on aluminum precipitation products, we found that the AsOX and AsHCl values were always similar in all the samples formed either at R ) 0.01 or at R ) 0.1 and concluded that arsenate was mainly retained into the short-range ordered Al-oxyhydroxides [(pseudo)boehmites]. A possible explanation of our findings in this work is that in iron systems arsenate could be partly incorporated into the structures of crystalline Fe oxides, and it was completely released only after dissolution with HCl. Galvez et al. (25, 26) demonstrated that hematite and goethite particles incorporated phosphate in an occluded form which was not desorbed by repeated alkali treatments used to remove phosphate sorbed on the surfaces of Fe oxides. These authors demonstrated that crystalline Fe oxides (hematite and goethite) were more efficient than ferrihydrite in retaining significant amounts of phosphate in such irreversibly occluded forms. Arsenate seems to display a similar behavior. Sorption of Phosphate onto Iron Oxides. The sorption of phosphate onto iron–arsenate coprecipitates and the subsequent release of arsenate were studied. The sorption isotherms of phosphate at pH 6.0 onto Fe precipitation products formed at pH 4.0, 7.0, and 10.0 and aged 30 days are shown in Figure 1 and Figure S4 of the Supporting Information. Surprisingly, the samples 4FeR0.1 and 7FeR0.1, which showed only the presence of short-range ordered materials and a large surface area (Table 1), sorbed much lower amounts of phosphate than the precipitates formed at the same pH but at R ) 0 and 0.01 (Figure 1 and Figure S4A, Supporting Information). These findings may be explained by considering that the sorption of phosphate onto iron–arsenate coprecipitates is not only affected by the mineralogy and surface area of the samples but also by the amount of arsenate in the precipitates (8, 9, 21, 32, 33). Many 8278
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 24, 2007
FIGURE 2. Arsenate desorbed at pH 6.0 from 4FeR0.1, 7FeR0.1, and 10FeR0.1 after addition of 2000 mmol kg-1 of phosphate. Full symbols and continuous lines indicate the samples aged 30 days; open symbols and dashed lines indicate samples aged 210 days (for sample symbol, see Table 1). more arsenate anions were present in the precipitates formed at R ) 0.1 than at R ) 0.01 (Table 2), which strongly prevented the sorption of phosphate by occupying many sorption sites at the surfaces of the precipitates (8, 9, 20, 30, 31). Arsenate present in the precipitates also lowered the PZC of the samples, thus reducing phosphate sorption (8, 9, 24, 27). Arsenate inhibited phosphate sorption more on iron–arsenate than aluminum–arsenate coprecipitates, as demonstrated in a previous work (ref 20 and Figure 2 therein), evidently because arsenate competes with phosphate for sorption sites or inhibits phosphate sorption much more onto Fe oxides than Al oxides (1, 3, 5, 8, 20, 30, 31). Only on the samples formed at pH 10.0 were greater amounts of phosphate sorbed on 10FeR0.1 than on 10FeR0 and 10FeR0.01 (Figure S4B), evidently because these latter samples had surface area values that were particularly low (24–25 m2 g-1 for 10FeR0 and 10FeR0.01 vs 154 m2 g-1 for 10FeR0.1) due to the presence of very well crystallized hematite and goethite (Table 1; Figure S3e,f). The sorption of phosphate onto the samples aged for 210 days at 50 °C decreased substantially for the samples at R ) 0 and 0.01 (not shown) due to the improved crystallinity and lower surface area of the minerals present in these samples after prolonged aging (Table 1). Desorption of Arsenate by Phosphate from Iron–Arsenate Coprecipitates. Arsenate was partially released from the iron–arsenate coprecipitates formed at R ) 0.1. Similar results were also observed in aluminum systems (20). Figure 2 shows the effect of reaction time on the desorption of arsenate at pH 6.0 in the presence of a high concentration of phosphate (2000 mmol kg-1) from the samples formed at R ) 0.1 and pH 4.0, 7.0, or 10.0 and aged 30 or 210 days. The amounts of arsenate desorbed by phosphate from the iron–arsenate coprecipitates increased with reaction time, but greater amounts were desorbed from samples synthesized at pH 7.0 or 10.0 than from pH 4.0 samples. The quantities of arsenate removed from the coprecipitates aged 30 or 210 days were very similar. From the samples aged 210 days, 52, 92, and 95 mmol of arsenate kg-1 were desorbed after 170 h from 4FeR0.1, 7FeR0.1, and 10FeR0.1, respectively, which expressed as percentages of the total amounts of arsenate present in the samples ranged from 13% (10FeR0.1) to 9% (7FeR0.1) to 5% (4FeR0.1) (Figure 2). However, relatively lower amounts of arsenate were desorbed from iron–arsenate than from aluminum–arsenate coprecipitates (20), evidently because arsenate is held more strongly onto the surfaces of iron oxides (1, 3, 8, 9, 30–32).
FIGURE 3. Kinetics of desorption (Elovich model: refs 9, 33) of arsenate desorbed from 7FeR0.1 (arsenate coprecipitated with iron) and 7FeAR0.1 (arsenate sorbed onto a preformed iron oxide) after addition of phosphate. Phosphate was added at about 100% of surface coverage (500 mmol kg-1).
TABLE 3. Atomic Concentrations of Fe, O, As, and P at the Surface of 7FeR0.1 (As and Fe Coprecipitated) and 7FeAR0.1 (as Added to Preformed Ferrihydrite) and the Same Samples after Addition of Phosphate (2000 mmol kg-1, 24 h of Reaction) atomic concentration sample 7FeR0.1 7FeAR0.1 7FeR0.1 + PO4 7FeAR0.1 + PO4
total Fe total O total As total P (As + P)/Fe 36.0 34.4 26.1 28.1
60.3 61.7 67.5 66.1
3.6 4.0 3.4 3.0
3.0 3.0
0.24 0.21
Our experiments show that a very high amount of arsenate was irreversibly fixed in the precipitates. Such low amounts of arsenate desorbed by phosphate from iron–arsenate coprecipitates must be attributed to many factors, such as (i) the strong innersphere complexes that arsenate forms onto metal oxide surfaces (3, 5, 27–32), (ii) the possible formation of iron arsenate precipitates (33), and (iii) the partial occlusion of arsenate into the Fe oxides (ref 20; as discussed below; Figure 3 and Table 3). Particularly low amounts of arsenate were desorbed from 4FeR0.1, probably because, although the desorption experiments were carried out at pH 6.0 for all of the samples, arsenate could have been sorbed more strongly in the sample formed at pH 4.0 than in those formed at pH 7.0 or 10.0 (9, 27, 30–32). Violante and Pigna (30) have also shown that phosphate inhibits arsenate sorption on the surfaces of many sorbents more in neutral and alkaline systems than in acidic systems. Furthermore, the formation of iron arsenate precipitates onto the surfaces of iron oxides seems to be favored at low pH values (33) with a consequent reduction in the amount of available arsenate anions for desorption. Jia et al. (33) showed evidence of a ferric arsenate precipitate on synthetic ferrihydrite at acidic pH (3.0–5.0), whereas at pH 8.0, arsenate was sorbed predominantly via surface sorption. Precipitation of arsenate was not observed by Waychunas et al. (29) since all of their samples were prepared at pH 8.0. Violante et al. (20) also showed evidence of an aluminum arsenate precipitate only in the sample formed at pH 4.0. It is important to consider that other factors such as residence time, surface coverage, chemical and mineralogical properties of the sorbents (crystalline and short-range ordered Fe or Al oxides), and reaction time affect the removal of arsenate by phosphate (refs 8, 9, 30, and 31 and references
therein). Pigna et al. (31) found that, after a residence time of 15 days or longer on the Fe oxide surfaces, about 90% of the arsenate was not desorbed by phosphate after 24 h of reaction. Desorption Kinetics of Arsenate Sorbed or Coprecipitated by Fe Oxides. Kinetic studies on the sorption of phosphate and desorption of arsenate onto and from two samples formed at pH 7.0 and R ) 0.1 obtained by coprecipitating iron and arsenate (7FeR0.1) or by adding arsenate immediately after the precipitation of iron (7FeAR0.1) were performed. On these two samples, aged 30 days at 50 °C, which had similar surface areas (147 and 130 m2 g-1), phosphate was added in order to have a final surface coverage of about 100%. The fit for arsenate desorption data (Figure 3) was best obtained using the Elovich model. The amounts of arsenate desorbed from the two samples were greatly different; those desorbed after 48 h from 7FeAR0.1 were 2.5 times greater than those released from 7FeR0.1. The sorbed phosphate/ desorbed arsenate molar ratios during the reaction time ranged from 4.1 to 7.5 for 7FeAR0.1 and from 10 to 26 for 7FeR0.1. To facilitate desorption of arsenate, 2000 mmol kg-1 of phosphate was added, and these samples were kept to react up to 100 days. About 130 mmol kg-1 of arsenate was desorbed from 7FeR0.1 (14% of the total arsenate present in the coprecipitates) versus 190 mmol kg-1 from 7FeAR0.1 (19% of total arsenate). However, from 7FeAR0.1 only negligible amounts of arsenate were replaced by phosphate after more than 20 days of reaction. XPS analyses were also performed on these two samples before and after the addition of phosphate. The surface concentrations of elements (atomic %) for these samples are presented in Table 3. The results demonstrate that the addition of phosphate removed more arsenic (≈25%) from the surfaces of the sample where arsenate was added to a preformed ferrihydrite (7FeAR0.1), compared to the sample where arsenate was coprecipitated with Fe (≈5% As removed; 7FeR0.1). In addition, the total As + P values for these two samples (normalized to Fe) are almost identical, confirming that the removed arsenate from the surface has been replaced by phosphate. Furthermore, the phosphate sorbed/arsenate desorbed atomic ratio was 3 for 7FeAR0.1 and 15 for 7FeR0.1, which indicates that more sites were arsenate-free at the surfaces of the latter.
Acknowledgments The authors acknowledge Dr. M. Martin for the determination of the point of zero charge of selected samples. This work was supported by the Italian Research Program of National Interest (PRIN2006). DiSSPA contribution no 133.
Supporting Information Available Additional figures showing mineralogical properties of selected arsenate–iron coprecipitates by X-ray powder diffractograms (Figure S1), infrared spectra (Figure S2), and trasmission electron micrographs (Figure S3) and the sorption isotherms of phosphate onto the coprecipitates formed at pH 7.0 and 10.0 (Figure S4). This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Smith, E.; Naidu, R.; Alston, A. M. Arsenic in the soil environment: a review. Adv. Agron. 1998, 64, 149–195. (2) Berg, M.; Tran, H. C.; Nguyen, T. C.; Pham, H. V.; Schertenleib, R.; Giger, W. Arsenic contamination of groundwater and drinking water in Vietnam: A human health threat. Environ. Sci. Technol. 2001, 35, 2621–2626. (3) Frankenberger, W. T., Jr. Environmental Chemistry of Arsenic; Marcel Dekker: New York, 2002; pp 391. VOL. 41, NO. 24, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8279
(4) Mandal, B. K.; Suzuki, K. T. Arsenic round the world: a review. Talanta 2002, 58, 201–235. (5) Smedley, P. L.; Kinniburgh, D. G. A review of the source, behaviour and distribution of arsenic in natural waters. Appl. Geochem. 2002, 17, 517–568. (6) Livesey, N. T.; Huang, P. M. Adsorption of arsenate by soils and its relation to selected properties and anions. Soil Sci. 1981, 131, 8894. 28. (7) Cances, B.; Juillot, F.; Morin, G.; Laperche, V.; Alvarez, L.; Proux, O.; Hozemann, J. L.; Brown, G. E., Jr.; Calas, G. XAS evidence of As (V) association with iron oxyhydroxides in a contaminated soil at a former arsenical pesticide processing plant. Environ. Sci. Technol. 2005, 39, 9398–9405. (8) Violante, A.; Del Gaudio, S.; Pigna, M. Adsorption/desorption processes of arsenate in soil environments. In Soil abiotic and biotic interactions and impact on the ecosystem and human welfare; Huang, P. M., Violante, A., Bollag, J.-M., Vityakon, P., Eds.; Science Publishers: Enfield, NH, 2005. (9) Violante, A.; Krishnamurti, G. S. R.; Pigna, M. Factors affecting the sorption-desorption of trace elements in soil environments. In Biophysico-Chemical Processes of Metals and Metalloids in Soil Environments; Violante, A., Huang, P. M., Gadd, G., Eds.; John Wiley & Sons Ltd.: Hoboken, NJ, 2007; pp 169–213. (10) Pichler, T.; Veizer, J.; Hall, G. E. M. Natural input of arsenic into a coral-reef ecosystem by hydrothermal fluids an its removal by Fe(III) oxyhydroxides. Environ. Sci. Tecnol. 1999, 33, 1373– 1378. (11) Rancourt, D. G.; Fortin, F.; Pichler, T.; Thibault, P.- J.; Lamarche, G.; Morris, R. V.; Mercier, P. H. J. Mineralogy of a natural As-rich hydrous ferric oxide coprecipitate formed by mixing of hydrothermal fluid and seawater: Implications regarding surface complexation and color bonding in ferrihydrite deposits. Am. Mineral. 2001, 86, 834–851. (12) Carlson, L.; Bigham, J. M.; Schwertmann, U.; Kyek, A.; Wagner, F. Scavenging of As from acid mine drainage by schwertmannite and ferrihydrite; a comparison with synthetic analogues. Environ. Sci. Technol. 2002, 36, 1712–1719. (13) Ford, R. Rates of hydrous ferric oxide crystallization and the influence on coprecipitated arsenate. Environ. Sci. Techonol. 2002, 36, 2459–2463. (14) Liu, W. J.; Zhu, Y. G.; Hu, Y.; Williams, P. N.; Gault, A. G.; Meharg, A. A.; Charnock, J. M.; Smith, F. A. Arsenic sequestration in iron plaque, its accumulation and speciation in mature rice plants (Oryza Sativa L.). Environ. Sci. Technol. 2006, 40, 5730–5736. (15) Scott, N. K.; Green, F. J.; Do, D. H.; McLean, J. S. Arsenic removed by coagulation. J. Am. Water Works Assoc. 1995, 87, 114–126. (16) Viraraghavan, T.; Subramanian, K. S.; Aruldoss, J. A. Arsenic in drinking water-problems and solutions. Water Sci. Technol. 1999, 40, 69–76. (17) Clifford D. A.; Ghurye, G. L. Metal-oxide adsorption, ion exchange, and coagulation-microfiltration for arsenic removal from water. In Environmental chemistry of arsenic; Frankenberger, W. T., Jr., Ed.; Marcel Dekker: New York, 2002.
8280
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 24, 2007
(18) Moldovan, B. T.; Jiang, D. T.; Hendry, M. J. Mineralogical characterization of arsenic in uranium mine tailings precipitated from iron-rich hydrometallurgical solutions. Environ. Sci. Technol. 2003, 37, 873–879. (19) Roberts, L. C.; Hug, S. J.; Ruettimann, T.; Morsaline Billah, M. D.; Khan, W. A.; Rahman, M. T. Arsenic removal with iron(II) and iron(III) in waters with high silicate and phosphate concentrations. Environ. Sci. Technol. 2004, 38, 307–315. (20) Violante, A.; Ricciardella, M.; Del Gaudio, S.; Pigna, M. Coprecipitation of arsenate with metal oxides: Nature, mineralogy and reactivity of aluminum precipitates. Environ. Sci. Technol. 2006, 40, 4961–4967. (21) Quirk, J. P. Significance of surface area calculated from water vapour sorption isotherms by use of the B. E. T. equation. Soil Sci. 1955, 80, 423–430. (22) Liu, F.; De Cristofaro, A.; Violante, A. Effect of pH, phosphate and oxalate on the adsorption/desorption of arsenate on/from goethite. Soil Sci. 2001, 166, 197–208. (23) Paige, C. R.; Snodgrass, W. J.; Nicholson, R. V.; Scharer, J. M. An arsenate effect of ferrihydrite dissolution kinetics under acidic oxic conditions. Water Res. 1997, 31 (9), 2370–2382. (24) Cornell, R. M.; Schwertmann, U. The iron oxides. Structure, properties, reactions and uses; VCH Publishers: New York, 1996. (25) Galvez, N.; Barron, V.; Torrent, J. Effect of phosphate on the crystallization of hematite, goethite and lepidocrocite from ferrihydrite. Clays Clay Miner. 1999, 47, 304–311. (26) Galvez, N.; Barron, V.; Torrent, J. Preparation and properties of hematites with structural phosphorus. Clays Clay Miner. 1999, 47, 375–385. (27) Goldberg, S.; Johnston, C. T. Mechanisms of arsenic adsorption on amorphous oxides evaluated using macroscopic measurements, vibrational spectroscopy, and surface complexation modeling. J. Colloid Interface Sci. 2001, 234, 204–216. (28) Manceau, A. The mechanism of anion adsorption on iron oxides: Evidence for the bonding of arsenate tetrahedra on free Fe(O, OH)6 edges. Geochim. Cosmochim. Acta 1995, 59, 3647–3653. (29) Waychunas, G. A.; Rea, B. A.; Fuller, C. C.; Davis, J. A. Surface chemistry of ferrihydrite: Part 1. EXAFS studies on the geometry of coprecipitated and adsorbed arsenate. Geochim. Cosmochim. Acta 1993, 57, 2251–2269. (30) Violante, A; Pigna, M. Competitive sorption of arsenate and phosphate on different clay minerals and soils. Soil Sci. Soc. Am. J. 2002, 66, 1788–1796. (31) Pigna, M.; Krishnamurti, G. S. R.; Violante, A. Kinetics of arsenate sorption-desorption from metal oxides: Effect of residence time. Soil Sci. Soc. Am. J. 2006, 70, 2017–2027. (32) Dixit, S.; Hering, J. G. Comparison of arsenic(V) and arsenic(III) sorption onto iron oxide minerals; Implications for arsenic mobility. Environ. Sci. Technol. 2003, 37, 4182–4189. (33) Jia, Y. F.; Xu, L.; Fang, Z.; Demopoulos, G. P. Observation of surface precipitation of arsenate on ferrihydrite. Environ. Sci. Technol. 2006, 40, 3248–3253.
ES070382+
