Journal of Hazardous Materials 300 (2015) 272–280
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Incorporation of arsenic into gypsum: Relevant to arsenic removal and immobilization process in hydrometallurgical industry Danni Zhang a , Zidan Yuan a , Shaofeng Wang a,∗ , Yongfeng Jia a,∗ , George P. Demopoulos b a b
Key Laboratory of Pollution Ecology and Environmental Engineering, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang 110016, China Department of Mining and Materials Engineering, McGill University, Montreal, QC H3A 2B2, Canada
h i g h l i g h t s • • • • •
Quantitatively studied the incorporation of arsenic into the structure of gypsum. Arsenic content in the solid increased with pH and initial arsenic concentration. Calcium arsenate phase precipitated in addition to gypsum at higher pH values. The structure of gypsum and its morphology was altered by the incorporated arsenate. The incorporated arsenate formed sainfeldite-like local structure in gypsum.
a r t i c l e
i n f o
Article history: Received 17 April 2015 Received in revised form 11 June 2015 Accepted 7 July 2015 Available online 9 July 2015 Keywords: Arsenic Incorporation Gypsum
a b s t r a c t Gypsum precipitates as a major secondary mineral during the iron-arsenate coprecipitation process for the removal of arsenic from hydrometallurgical effluents. However, its role in the fixation of arsenic is still unknown. This work investigated the incorporation of arsenic into gypsum quantitatively during the crystallization process at various pHs and the initial arsenic concentrations. X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), X-ray absorption near edge spectroscopy (XANES) and scanning electron microscopy (SEM) were employed to characterize the coprecipitated solids. The results showed that arsenate was measurably removed from solution during gypsum crystallization and the removal increased with increasing pH. At lower pH where the system was undersaturated with respect to calcium arsenate, arsenate ions were incorporated into gypsum structure, whereas at higher pH, calcium arsenate was formed and constituted the major arsenate bearing species in the precipitated solids. The findings may have important implications for arsenic speciation and stability of the hydrometallurgical solid wastes. © 2015 Published by Elsevier B.V.
1. Introduction Arsenic is commonly present in most base metal and precious metal ores and concentrates as co-occurring minerals in various forms such as sulfides and arsenides [1]. It is liberated from the concentrate during pyrometallurgy and hydrometallurgy operations. In pyrometallurgy operations, such as roasting and converting, most arsenic in the concentrate is volatilized and collected as As2 O3 and As2 S3 [1]. They are unstable forms of arsenic hence not suitable for direct disposal as solid wastes. Aqueous oxidation of arsenic trioxide or sulfide to arsenate followed by precipitation operations is
∗ Corresponding authors. Fax: +86 24 8397 0503. E-mail addresses: [email protected] (S. Wang), [email protected] (Y. Jia). http://dx.doi.org/10.1016/j.jhazmat.2015.07.015 0304-3894/© 2015 Published by Elsevier B.V.
needed to immobilize the arsenic. In hydrometallurgy operations, arsenic is released into mineral processing solutions and effluents during extraction of metals by oxidation and acid dissolution of the arsenic-containing minerals. Tens of thousands tons of arsenic are liberated every year in non-ferrous metal industry around the world. Because of its well known toxicity and limited market, most of arsenic in the industrial effluents must be removed and immobilized as a stable solid and disposed safely for the prevention of contamination to surrounding soils and waters [1–3]. The common methods employed in industrial practice to remove the arsenic from mineral processing and metallurgical operation solutions include scorodite precipitation, iron-arsenic coprecipitation etc [1–3]. Among these technologies, lime neutralization accompanied by coprecipitation of arsenic with ferric iron is widely practiced in most hydrometallurgical operations worldwide and regarded by USEPA as the best demonstrated available
D. Zhang et al. / Journal of Hazardous Materials 300 (2015) 272–280
technology (BDAT) [4,5]. In the coprecipitation process, arsenate can be efficiently removed from the solution to sub-ppm level in the presence of excess ferric iron (Fe/As ≥ 3) [3]. The main views in the past two decades regarding the mechanism of arsenic fixation in Fe(III)–As(V) coprecipitates include adsorption of arsenate via surface complex on ferrihydrite [6–10] and formation of poorly crystalline ferric arsenate (i.e. Fe/As ∼ 1 compounds) [11–14]. Mineral processing effluents are usually sulfate-rich solutions due to the use of sulfuric acid as leaching agent or addition of ferric sulfate to coprecipitate arsenic [1]. Gypsum (CaSO4 ·2H2 O) is generated during Fe(III)–As(V) coprecipitation process because of the use of lime as base for neutralization [15]. Arsenic can be incorporated into gypsum lattice structure via isomorphic substitution for sulfate as unambiguously evidenced by neutron diffraction [16,17]. This finding is confirmed by a latest study using singlecrystal electron paramagnetic resonance spectroscopy (EPR) and pulsed electron nuclear double resonance spectroscopy (ENDOR) [18]. This is similar to the incorporation of arsenate ions into jarosite structure by substitution of sulfate [19] and vice versa incorporation of sulfate ions into scorodite structure by substitution of arsenate [20,21]. Fernández-Martínez et al. [16,17] estimated the concentration of arsenic incorporated into the bulk of gypsum by interpolating the experimental and theoretical volume expansion data. The results indicated that more arsenic could be incorporated into gypsum at higher pH. However, due to the lack of direct experimental evidence, the ability of gypsum in the fixation of arsenic in the coprecipitation process as well as the speciation of arsenic in the structure of gypsum still remains unclear. As mentioned above, a great amount of gypsum could be produced during the hydrometallurgical arsenic removal process. Therefore, a clear understanding of quantity of As fixed by gypsum is crucial for evaluating the environmental risk of arsenic bearing industrial solid waste due to its high solubility. A systematic and quantitative investigation on the coprecipitation of arsenic with gypsum is necessary in order to shed more light on the role of gypsum in arsenic removal/fixation in hydrometallurgical arsenic removal process. The objective of this work was to qualitatively and quantitatively study the incorporation of arsenate and arsenite into gypsum structure in arsenic-gypsum coprecipitation process at various pHs and arsenic concentrations. Relatively high arsenic concentration solutions were used in this study since this is the case in hydrometallurgical operations where arsenic removal is practiced. 2. Materials and methods All chemicals were of analytical grade and used without further purification. Deionized (DI) water was used for all experiments. All glassware was cleaned by soaking in 5% HNO3 for at least 24 h and rinsed three times with DI-water before use. The applied concentrations of arsenic in solutions covered the relevant levels in hydrometallurgical operations where arsenic removal is practiced. 2.1. Coprecipitation of arsenic with gypsum The coprecipitation experiments were performed by introducing calcium solution (from Ca(NO3 )2 ·4H2 O) into a mixture of sulfate solution (from Na2 SO4 ) and arsenic solution (from Na3 AsO4 ·12H2 O or NaAsO2 ). Briefly, 42 mL of 0.5 mol L−1 sulfate solution and 16 mL of various concentration of arsenic solution were mixed in a beaker and the pH was adjusted to the desired value (i.e. pH 3–10), which was followed by addition of 42 mL of 0.5 mol L−1 sulfate solution (i.e. equimolar calcium and sulfate ions) in a titration mode. The mixtures were allowed to stabilize for 30 min with the pH controlled at constant value. Mechanical stirring was applied
273
throughout the experiment. The volume of the final solution was ∼100 mL and the arsenic concentration in the system based on the final volume was reported (i.e. arsenate: 50, 150, 500, 1000, 2000, 5000, 10,000 mg L−1 ; arsenite: 5000 mg L−1 ). Arsenic-free gypsum was also precipitated at pH 3 and 9 as the reference material for characterization. After precipitation, the solids were separated by filtration, rinsed 5 times with 20 mL of DI-water each time, dried in a vacuum oven at 60 ◦ C overnight and stored at room temperature in a desiccator for further analysis. The filtrates were analyzed for concentration of calcium, sulfate and arsenic. To analyze the lattice incorporated arsenate, the gypsumarsenate coprecipitates synthesized at higher pH values were treated two times with ascorbic acid solutions for 1 h in order to remove the calcium arsenate phases that possibly precipitated during the coprecipitation process. Gypsum-saturated ascorbic acid solution was used in order to minimize the effect of washing out of the lattice incorporated arsenic. The pH of the mixtures of coprecipitates and ascorbic acid solutions was ∼2.5 at which calcium arsenates are not stable [22]. Our pre-experiment clearly showed that amorphous calcium arsenate can be completely dissolved (Fig. S1). 2.2. Determination of As, SO4 , and Ca concentrations in the liquid/solid phase The ascorbic acid treated and untreated solid samples were digested in 1 mol L−1 HCl solution and diluted for the analysis of As, SO4 , and Ca concentrations. The concentration of arsenic in the digests and filtrates was determined on an atomic fluorescence spectrophotometer coupled with a hydride generator (HG-AFS) [23,24]. The detection limit of the instrument was 0.1 g L−1 . The testing solution was pretreated with a reducing agent containing 5% thiourea and 5% ascorbic acid. During hydride generation and AFS measurement, a solution containing 2% KBH4 and 0.3% NaOH was used as reducing solution, and 5% HCl was used as carrier solution. The concentrations of Ca and SO4 were determined by using inductively coupled plasma—atomic emission spectroscopy (ICPAES, Thermo-6300) with the detection limit of 0.02 and 0.03 mg L−1 , respectively. 2.3. XRD analysis The mineralogical characteristics of the precipitates were characterized by using a Rigaku D/max 2400 X-ray diffractometer (Rigaku Corporation, Japan) equipped with a copper target (CuK␣1 radiation, = 1.5418 Å), a crystal graphite monochromator and a scintillation detector. The equipment was run at 56 kV and 182 mA by step scanning from 10◦ to 100◦ 2 with increments of 0.02◦ 2. The X-ray diffraction data were analyzed by Rietveld refinement using a Pseudo-Voigt function incorporated in the Fullprof software. The monoclinic unit cell of gypsum with C 1 2/c 1 space group was used as starting refinement model [25]. For the refinements of the arsenate incorporated gypsum structures, the sulfur position was assumed to be shared with the arsenic atom and the occupancy was calculated from the concentration of arsenate in the ascorbic acid treated solids. 2.4. FTIR analysis The infrared spectra of the powdered samples were collected on a Thermo Nicolet 6700 Fourier transform infrared spectrometer. The KBr/sample discs were prepared by mixing 0.5% of finely ground samples in KBr. The samples were scanned in mid-IR range (400–4000 cm−1 ) with the resolution of 4 cm−1 .
274
D. Zhang et al. / Journal of Hazardous Materials 300 (2015) 272–280
2.5. SEM analysis
1000
The morphologies of the solid samples were analyzed on a FEI Quanta 250 scanning electron microscope. All the samples were mounted on pin stubs by use of double-sided carbon tape and sputter-coated with gold, and imaged at 30 KV.
100 10 1
2.6. X-ray absorption near edge spectroscopy (XANES) analysis
3. Results and discussion 3.1. Arsenic concentrations in solid phase Bulk chemical analysis results of the arsenic-gypsum solids coprecipitated at different pHs and initial arsenic concentrations before and after ascorbic acid treatments are shown in Fig. 1. Arsenic content in the coprecipitated solid (black legends) was highly dependent on the precipitation pH and aqueous arsenic concentration. The concentration of arsenate in the solid phase increased exponentially with increasing pH. For example, at initial arsenate concentration of 5000 mg L−1 , As content was ∼0.53 mg g−1 in the solid coprecipitated at pH 3 and increased to ∼13.2 mg g−1 and ∼252 mg g−1 for the solids coprecipitated at pH 6 and 10, respectively. Arsenate content in the solid phase also increased significantly when higher initial arsenate concentration was employed. For the coprecipitates at pH 7, its value was only ∼0.41 mg g−1 at initial arsenate concentration of 50 mg L−1 , but rising to ∼38.5 mg g−1 at 10,000 mg L−1 , which was 93 times that of the former. Coprecipitation of arsenite with gypsum was also conducted at pH 3–10 at initial arsenite concentration of 5000 mg L−1 (Fig. 1).
Arsenic concentration in solid (mg/g)
0.1 The Ca K-edge XANES of the arsenate-gypsum samples coprecipitated at initial arsenic concentration of 10,000 mg L−1 at various pHs as well as the reference materials (gypsum and amorphous calcium arsenate) were collected on the station 4B7A at the Beijing Synchrotron Radiation Facilities (BSRF) in China with a storage ring of 2.5 GeV and 250 mA. The white beam was monochromatized with a fixed Si(311) double crystal monochromator. The spectra were collected in total electron yield (TEY) mode for all of the samples and reference materials at room temperature without dilution of the samples. A small portion of the sample powder was mounted on a conducting carbon double-faced tape, which was attached to an electrode. The signals were measured with a pico-ampere meter. Repeat scans were acquired to improve the signal-to-noise ratios of the scans where necessary. Before measurements, the energy was calibrated by defining the peak top of XANES of CaCl2 ·2H2 O at 4038.1 eV. The scan started at 150 eV before the absorption edge and finished at 400 eV after the absorption edge. All the carbon tape and the filters did not produce any absorption in the Ca K-edge XANES regions. Athena in the Demeter software package (Version 0.9.20) was used to process background removal and normalization. The Autobk algorithm was applied for background removal using a linear pre-edge line between 113 and 30 eV before the edge, and normalization range from 150 to 330 eV. Linear combination fit (LCF) for both normalized Ca K-edge spectra and first derivatives of spectra was performed to distinguish the percentage of different Ca species in the arsenate-gypsum coprecipitates. Amorphous calcium arsenate was synthesized according to the methods reported by Bothe and Brown [26]. Briefly, the pre-set pH 11 As(V) solution (from Na3 AsO4 ·12H2 O) was titrated by the 0.5 mol L−1 calcium solution (from Ca(NO3 )2 ·4H2 O) within 20 min (Ca/As = 1.7) at the initial arsenic concentration of 5000 mg L−1 . Arsenic-free gypsum synthesized at pH 3 was also used as a reference material.
As (V) 2000 As (V) 2000, VC As (V) 5000 As (V) 5000, VC As (V) 10000 As (V) 10000, VC As (III) 5000 As (III) 5000, VC
0.01 1E-3 1E-4 1E-5 10
1
As (V) 50 As (V) 50, VC As (V) 150 As (V) 150, VC As (V) 500 As (V) 500, VC As (V) 1000 As (V) 1000, VC
0.1
0.01
1E-3 2
3
4
5
6
7
8
9
10
11
pH Fig. 1. Arsenic content in the arsenic-gypsum coprecipitates (black legends) and gypsum separated by ascorbic leaching (red legends) as a function of precipitation pH and initial arsenic concentrations (50–10,000 mg L−1 ). “VC” stands for ascorbic acid.
The content of arsenic in the solid phase also increased significantly with pH and the results suggested that arsenite can also be fixed in the solid phase during the precipitation of gypsum. However, its value was much lower than that of the arsenate-gypsum coprecipitation system at the same pH and initial arsenic concentration. The contents of arsenate, sulfate and calcium in the coprecipitated solids and the residual solutions at various conditions are shown in Tables S1 and S2. The SO4 /Ca molar ratios in the solid phases precipitated at different initial arsenate concentrations as a function of pH are plotted in Fig. S2. Gypsum has stoichiometrically equimolar sulfate and calcium and the results showed that the molar ratio was around 1.0 at the initial arsenate concentration of 1000 mg L−1 and less. At higher initial arsenic concentrations, SO4 /Ca molar ratio decreased significantly with increasing pH, especially at pH > 6, implying the formation of other Ca-containing phases. The saturation state with respect to calcium arsenate in the arsenate-gypsum coprecipitation systems at different pH and initial arsenate concentrations was estimated using PhreePlot (Fig. 2) [27]. Various types of calcium arsenate with different chemical formula and logKsp were considered. The solubility products as well as the input file of Phreeplot were presented in Tables S3 and S4. According to the estimation, calcium arsenate precipitated in
D. Zhang et al. / Journal of Hazardous Materials 300 (2015) 272–280
275
Table 1 The results of LCF of Ca K-edge XANES and goodness of fitting for the arsenategypsum coprecipitates formed at various pHs with initial arsenate concentration of 10,000 mg L−1 . Data in parenthesis stands for three times the estimated standard deviation for each component from LCF fit.
Fig. 2. Estimation of the saturation states with respect to calcium arsenate in the arsenate-gypsum coprecipitation systems at pH 3–10 and initial arsenate concentrations of 50 (䊉), 150 (), 500 (䊏), 1000 (䊐), 2000 (), 5000 () and 10,000 () mg L−1 respectively, using the geochemical modeling program PhreePlot and the WATEQ4F database.
some of the coprecipitation systems, especially at higher pH values, suggesting that the coprecipitated solids probably contain both gypsum and calcium arsenate. This result was further confirmed by Ca K-edge XANES analysis (Fig. 3). Gypsum was the major Cacontaining species in the coprecipitates formed at pH 3–5, initial arsenate concentration of 10,000 mg L−1 , whereas, amorphous calcium arsenate became the dominant phase at pH 8–10 (Table 1).
A
B
pH 10
pH 9
pH 9
pH 8
pH 8
pH 5 pH 4 pH 3 GY
derivatives of normalized spectra
Normalized absorbance
pH 10
pH 5
pH 4
pH 3
GY
CA CA
4020
4050
4080
4110
Photon energy (eV)
4140
4020
4050
4080
4110
4140
Photon energy (eV)
Fig. 3. The normalized (A) and first derivative (B) of calcium K-edge XANES spectra (black lines) of reference materials and linear combination fit (red lines) for the arsenate-gypsum coprecipitates formed at various pHs with initial arsenate concentration of 10,000 mg L−1 . “GY” and “CA” stand for arsenic-free gypsum and amorphous calcium arsenate, respectively.
pH
Percent of gypsum (%)
Percent of amorphous calcium arsenate (%)
Reduced chi-square
3 4 5 8 9 10
100 (0.00) 97.4 (0.05) 92.3 (0.04) 15.0 (0.05) 8.50 (0.04) 4.00 (0.00)
0.00 (0.00) 0.00 (0.05) 8.30 (0.04) 89.2 (0.05) 92.2 (0.04) 96.6 (0.00)
0.000103 0.000240 0.000188 0.000235 0.000142 0.000074
Hence, to determine the amount of As incorporated into the gypsum crystal, the arsenate-gypsum coprecipitates synthesized at higher pH values were subjected to ascorbic acid treatment to remove the amorphous calcium arsenate phases. The results (red legends in Fig. 1) showed that at low pH, arsenic concentration in the separated gypsum was equal to that in the untreated samples, indicating that almost all the arsenic in the solids was incorporated into the bulk lattice of gypsum. At the pH where calcium arsenate precipitated, the arsenic contents in the coprecipitates (black legends) were higher than those in the ascorbic acid treated samples (red legends). This difference represented the content of calcium arsenate in the coprecipitates, which rose as the pH and the initial As concentration increased. The content of arsenic in the gypsum for each initial arsenate concentration increased with increasing pH until the formation of calcium arsenate phases, but decreased with increasing pH after that due to the preferential precipitation of calcium arsenate. For example, at initial arsenate concentration of 5000 mg L−1 , As content in the gypsum increased from
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Incorporation of arsenic into gypsum: Relevant to arsenic removal and immobilization process in hydrometallurgical industry Danni Zhang a , Zidan Yuan a , Shaofeng Wang a,∗ , Yongfeng Jia a,∗ , George P. Demopoulos b a b
Key Laboratory of Pollution Ecology and Environmental Engineering, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang 110016, China Department of Mining and Materials Engineering, McGill University, Montreal, QC H3A 2B2, Canada
h i g h l i g h t s • • • • •
Quantitatively studied the incorporation of arsenic into the structure of gypsum. Arsenic content in the solid increased with pH and initial arsenic concentration. Calcium arsenate phase precipitated in addition to gypsum at higher pH values. The structure of gypsum and its morphology was altered by the incorporated arsenate. The incorporated arsenate formed sainfeldite-like local structure in gypsum.
a r t i c l e
i n f o
Article history: Received 17 April 2015 Received in revised form 11 June 2015 Accepted 7 July 2015 Available online 9 July 2015 Keywords: Arsenic Incorporation Gypsum
a b s t r a c t Gypsum precipitates as a major secondary mineral during the iron-arsenate coprecipitation process for the removal of arsenic from hydrometallurgical effluents. However, its role in the fixation of arsenic is still unknown. This work investigated the incorporation of arsenic into gypsum quantitatively during the crystallization process at various pHs and the initial arsenic concentrations. X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), X-ray absorption near edge spectroscopy (XANES) and scanning electron microscopy (SEM) were employed to characterize the coprecipitated solids. The results showed that arsenate was measurably removed from solution during gypsum crystallization and the removal increased with increasing pH. At lower pH where the system was undersaturated with respect to calcium arsenate, arsenate ions were incorporated into gypsum structure, whereas at higher pH, calcium arsenate was formed and constituted the major arsenate bearing species in the precipitated solids. The findings may have important implications for arsenic speciation and stability of the hydrometallurgical solid wastes. © 2015 Published by Elsevier B.V.
1. Introduction Arsenic is commonly present in most base metal and precious metal ores and concentrates as co-occurring minerals in various forms such as sulfides and arsenides [1]. It is liberated from the concentrate during pyrometallurgy and hydrometallurgy operations. In pyrometallurgy operations, such as roasting and converting, most arsenic in the concentrate is volatilized and collected as As2 O3 and As2 S3 [1]. They are unstable forms of arsenic hence not suitable for direct disposal as solid wastes. Aqueous oxidation of arsenic trioxide or sulfide to arsenate followed by precipitation operations is
∗ Corresponding authors. Fax: +86 24 8397 0503. E-mail addresses: [email protected] (S. Wang), [email protected] (Y. Jia). http://dx.doi.org/10.1016/j.jhazmat.2015.07.015 0304-3894/© 2015 Published by Elsevier B.V.
needed to immobilize the arsenic. In hydrometallurgy operations, arsenic is released into mineral processing solutions and effluents during extraction of metals by oxidation and acid dissolution of the arsenic-containing minerals. Tens of thousands tons of arsenic are liberated every year in non-ferrous metal industry around the world. Because of its well known toxicity and limited market, most of arsenic in the industrial effluents must be removed and immobilized as a stable solid and disposed safely for the prevention of contamination to surrounding soils and waters [1–3]. The common methods employed in industrial practice to remove the arsenic from mineral processing and metallurgical operation solutions include scorodite precipitation, iron-arsenic coprecipitation etc [1–3]. Among these technologies, lime neutralization accompanied by coprecipitation of arsenic with ferric iron is widely practiced in most hydrometallurgical operations worldwide and regarded by USEPA as the best demonstrated available
D. Zhang et al. / Journal of Hazardous Materials 300 (2015) 272–280
technology (BDAT) [4,5]. In the coprecipitation process, arsenate can be efficiently removed from the solution to sub-ppm level in the presence of excess ferric iron (Fe/As ≥ 3) [3]. The main views in the past two decades regarding the mechanism of arsenic fixation in Fe(III)–As(V) coprecipitates include adsorption of arsenate via surface complex on ferrihydrite [6–10] and formation of poorly crystalline ferric arsenate (i.e. Fe/As ∼ 1 compounds) [11–14]. Mineral processing effluents are usually sulfate-rich solutions due to the use of sulfuric acid as leaching agent or addition of ferric sulfate to coprecipitate arsenic [1]. Gypsum (CaSO4 ·2H2 O) is generated during Fe(III)–As(V) coprecipitation process because of the use of lime as base for neutralization [15]. Arsenic can be incorporated into gypsum lattice structure via isomorphic substitution for sulfate as unambiguously evidenced by neutron diffraction [16,17]. This finding is confirmed by a latest study using singlecrystal electron paramagnetic resonance spectroscopy (EPR) and pulsed electron nuclear double resonance spectroscopy (ENDOR) [18]. This is similar to the incorporation of arsenate ions into jarosite structure by substitution of sulfate [19] and vice versa incorporation of sulfate ions into scorodite structure by substitution of arsenate [20,21]. Fernández-Martínez et al. [16,17] estimated the concentration of arsenic incorporated into the bulk of gypsum by interpolating the experimental and theoretical volume expansion data. The results indicated that more arsenic could be incorporated into gypsum at higher pH. However, due to the lack of direct experimental evidence, the ability of gypsum in the fixation of arsenic in the coprecipitation process as well as the speciation of arsenic in the structure of gypsum still remains unclear. As mentioned above, a great amount of gypsum could be produced during the hydrometallurgical arsenic removal process. Therefore, a clear understanding of quantity of As fixed by gypsum is crucial for evaluating the environmental risk of arsenic bearing industrial solid waste due to its high solubility. A systematic and quantitative investigation on the coprecipitation of arsenic with gypsum is necessary in order to shed more light on the role of gypsum in arsenic removal/fixation in hydrometallurgical arsenic removal process. The objective of this work was to qualitatively and quantitatively study the incorporation of arsenate and arsenite into gypsum structure in arsenic-gypsum coprecipitation process at various pHs and arsenic concentrations. Relatively high arsenic concentration solutions were used in this study since this is the case in hydrometallurgical operations where arsenic removal is practiced. 2. Materials and methods All chemicals were of analytical grade and used without further purification. Deionized (DI) water was used for all experiments. All glassware was cleaned by soaking in 5% HNO3 for at least 24 h and rinsed three times with DI-water before use. The applied concentrations of arsenic in solutions covered the relevant levels in hydrometallurgical operations where arsenic removal is practiced. 2.1. Coprecipitation of arsenic with gypsum The coprecipitation experiments were performed by introducing calcium solution (from Ca(NO3 )2 ·4H2 O) into a mixture of sulfate solution (from Na2 SO4 ) and arsenic solution (from Na3 AsO4 ·12H2 O or NaAsO2 ). Briefly, 42 mL of 0.5 mol L−1 sulfate solution and 16 mL of various concentration of arsenic solution were mixed in a beaker and the pH was adjusted to the desired value (i.e. pH 3–10), which was followed by addition of 42 mL of 0.5 mol L−1 sulfate solution (i.e. equimolar calcium and sulfate ions) in a titration mode. The mixtures were allowed to stabilize for 30 min with the pH controlled at constant value. Mechanical stirring was applied
273
throughout the experiment. The volume of the final solution was ∼100 mL and the arsenic concentration in the system based on the final volume was reported (i.e. arsenate: 50, 150, 500, 1000, 2000, 5000, 10,000 mg L−1 ; arsenite: 5000 mg L−1 ). Arsenic-free gypsum was also precipitated at pH 3 and 9 as the reference material for characterization. After precipitation, the solids were separated by filtration, rinsed 5 times with 20 mL of DI-water each time, dried in a vacuum oven at 60 ◦ C overnight and stored at room temperature in a desiccator for further analysis. The filtrates were analyzed for concentration of calcium, sulfate and arsenic. To analyze the lattice incorporated arsenate, the gypsumarsenate coprecipitates synthesized at higher pH values were treated two times with ascorbic acid solutions for 1 h in order to remove the calcium arsenate phases that possibly precipitated during the coprecipitation process. Gypsum-saturated ascorbic acid solution was used in order to minimize the effect of washing out of the lattice incorporated arsenic. The pH of the mixtures of coprecipitates and ascorbic acid solutions was ∼2.5 at which calcium arsenates are not stable [22]. Our pre-experiment clearly showed that amorphous calcium arsenate can be completely dissolved (Fig. S1). 2.2. Determination of As, SO4 , and Ca concentrations in the liquid/solid phase The ascorbic acid treated and untreated solid samples were digested in 1 mol L−1 HCl solution and diluted for the analysis of As, SO4 , and Ca concentrations. The concentration of arsenic in the digests and filtrates was determined on an atomic fluorescence spectrophotometer coupled with a hydride generator (HG-AFS) [23,24]. The detection limit of the instrument was 0.1 g L−1 . The testing solution was pretreated with a reducing agent containing 5% thiourea and 5% ascorbic acid. During hydride generation and AFS measurement, a solution containing 2% KBH4 and 0.3% NaOH was used as reducing solution, and 5% HCl was used as carrier solution. The concentrations of Ca and SO4 were determined by using inductively coupled plasma—atomic emission spectroscopy (ICPAES, Thermo-6300) with the detection limit of 0.02 and 0.03 mg L−1 , respectively. 2.3. XRD analysis The mineralogical characteristics of the precipitates were characterized by using a Rigaku D/max 2400 X-ray diffractometer (Rigaku Corporation, Japan) equipped with a copper target (CuK␣1 radiation, = 1.5418 Å), a crystal graphite monochromator and a scintillation detector. The equipment was run at 56 kV and 182 mA by step scanning from 10◦ to 100◦ 2 with increments of 0.02◦ 2. The X-ray diffraction data were analyzed by Rietveld refinement using a Pseudo-Voigt function incorporated in the Fullprof software. The monoclinic unit cell of gypsum with C 1 2/c 1 space group was used as starting refinement model [25]. For the refinements of the arsenate incorporated gypsum structures, the sulfur position was assumed to be shared with the arsenic atom and the occupancy was calculated from the concentration of arsenate in the ascorbic acid treated solids. 2.4. FTIR analysis The infrared spectra of the powdered samples were collected on a Thermo Nicolet 6700 Fourier transform infrared spectrometer. The KBr/sample discs were prepared by mixing 0.5% of finely ground samples in KBr. The samples were scanned in mid-IR range (400–4000 cm−1 ) with the resolution of 4 cm−1 .
274
D. Zhang et al. / Journal of Hazardous Materials 300 (2015) 272–280
2.5. SEM analysis
1000
The morphologies of the solid samples were analyzed on a FEI Quanta 250 scanning electron microscope. All the samples were mounted on pin stubs by use of double-sided carbon tape and sputter-coated with gold, and imaged at 30 KV.
100 10 1
2.6. X-ray absorption near edge spectroscopy (XANES) analysis
3. Results and discussion 3.1. Arsenic concentrations in solid phase Bulk chemical analysis results of the arsenic-gypsum solids coprecipitated at different pHs and initial arsenic concentrations before and after ascorbic acid treatments are shown in Fig. 1. Arsenic content in the coprecipitated solid (black legends) was highly dependent on the precipitation pH and aqueous arsenic concentration. The concentration of arsenate in the solid phase increased exponentially with increasing pH. For example, at initial arsenate concentration of 5000 mg L−1 , As content was ∼0.53 mg g−1 in the solid coprecipitated at pH 3 and increased to ∼13.2 mg g−1 and ∼252 mg g−1 for the solids coprecipitated at pH 6 and 10, respectively. Arsenate content in the solid phase also increased significantly when higher initial arsenate concentration was employed. For the coprecipitates at pH 7, its value was only ∼0.41 mg g−1 at initial arsenate concentration of 50 mg L−1 , but rising to ∼38.5 mg g−1 at 10,000 mg L−1 , which was 93 times that of the former. Coprecipitation of arsenite with gypsum was also conducted at pH 3–10 at initial arsenite concentration of 5000 mg L−1 (Fig. 1).
Arsenic concentration in solid (mg/g)
0.1 The Ca K-edge XANES of the arsenate-gypsum samples coprecipitated at initial arsenic concentration of 10,000 mg L−1 at various pHs as well as the reference materials (gypsum and amorphous calcium arsenate) were collected on the station 4B7A at the Beijing Synchrotron Radiation Facilities (BSRF) in China with a storage ring of 2.5 GeV and 250 mA. The white beam was monochromatized with a fixed Si(311) double crystal monochromator. The spectra were collected in total electron yield (TEY) mode for all of the samples and reference materials at room temperature without dilution of the samples. A small portion of the sample powder was mounted on a conducting carbon double-faced tape, which was attached to an electrode. The signals were measured with a pico-ampere meter. Repeat scans were acquired to improve the signal-to-noise ratios of the scans where necessary. Before measurements, the energy was calibrated by defining the peak top of XANES of CaCl2 ·2H2 O at 4038.1 eV. The scan started at 150 eV before the absorption edge and finished at 400 eV after the absorption edge. All the carbon tape and the filters did not produce any absorption in the Ca K-edge XANES regions. Athena in the Demeter software package (Version 0.9.20) was used to process background removal and normalization. The Autobk algorithm was applied for background removal using a linear pre-edge line between 113 and 30 eV before the edge, and normalization range from 150 to 330 eV. Linear combination fit (LCF) for both normalized Ca K-edge spectra and first derivatives of spectra was performed to distinguish the percentage of different Ca species in the arsenate-gypsum coprecipitates. Amorphous calcium arsenate was synthesized according to the methods reported by Bothe and Brown [26]. Briefly, the pre-set pH 11 As(V) solution (from Na3 AsO4 ·12H2 O) was titrated by the 0.5 mol L−1 calcium solution (from Ca(NO3 )2 ·4H2 O) within 20 min (Ca/As = 1.7) at the initial arsenic concentration of 5000 mg L−1 . Arsenic-free gypsum synthesized at pH 3 was also used as a reference material.
As (V) 2000 As (V) 2000, VC As (V) 5000 As (V) 5000, VC As (V) 10000 As (V) 10000, VC As (III) 5000 As (III) 5000, VC
0.01 1E-3 1E-4 1E-5 10
1
As (V) 50 As (V) 50, VC As (V) 150 As (V) 150, VC As (V) 500 As (V) 500, VC As (V) 1000 As (V) 1000, VC
0.1
0.01
1E-3 2
3
4
5
6
7
8
9
10
11
pH Fig. 1. Arsenic content in the arsenic-gypsum coprecipitates (black legends) and gypsum separated by ascorbic leaching (red legends) as a function of precipitation pH and initial arsenic concentrations (50–10,000 mg L−1 ). “VC” stands for ascorbic acid.
The content of arsenic in the solid phase also increased significantly with pH and the results suggested that arsenite can also be fixed in the solid phase during the precipitation of gypsum. However, its value was much lower than that of the arsenate-gypsum coprecipitation system at the same pH and initial arsenic concentration. The contents of arsenate, sulfate and calcium in the coprecipitated solids and the residual solutions at various conditions are shown in Tables S1 and S2. The SO4 /Ca molar ratios in the solid phases precipitated at different initial arsenate concentrations as a function of pH are plotted in Fig. S2. Gypsum has stoichiometrically equimolar sulfate and calcium and the results showed that the molar ratio was around 1.0 at the initial arsenate concentration of 1000 mg L−1 and less. At higher initial arsenic concentrations, SO4 /Ca molar ratio decreased significantly with increasing pH, especially at pH > 6, implying the formation of other Ca-containing phases. The saturation state with respect to calcium arsenate in the arsenate-gypsum coprecipitation systems at different pH and initial arsenate concentrations was estimated using PhreePlot (Fig. 2) [27]. Various types of calcium arsenate with different chemical formula and logKsp were considered. The solubility products as well as the input file of Phreeplot were presented in Tables S3 and S4. According to the estimation, calcium arsenate precipitated in
D. Zhang et al. / Journal of Hazardous Materials 300 (2015) 272–280
275
Table 1 The results of LCF of Ca K-edge XANES and goodness of fitting for the arsenategypsum coprecipitates formed at various pHs with initial arsenate concentration of 10,000 mg L−1 . Data in parenthesis stands for three times the estimated standard deviation for each component from LCF fit.
Fig. 2. Estimation of the saturation states with respect to calcium arsenate in the arsenate-gypsum coprecipitation systems at pH 3–10 and initial arsenate concentrations of 50 (䊉), 150 (), 500 (䊏), 1000 (䊐), 2000 (), 5000 () and 10,000 () mg L−1 respectively, using the geochemical modeling program PhreePlot and the WATEQ4F database.
some of the coprecipitation systems, especially at higher pH values, suggesting that the coprecipitated solids probably contain both gypsum and calcium arsenate. This result was further confirmed by Ca K-edge XANES analysis (Fig. 3). Gypsum was the major Cacontaining species in the coprecipitates formed at pH 3–5, initial arsenate concentration of 10,000 mg L−1 , whereas, amorphous calcium arsenate became the dominant phase at pH 8–10 (Table 1).
A
B
pH 10
pH 9
pH 9
pH 8
pH 8
pH 5 pH 4 pH 3 GY
derivatives of normalized spectra
Normalized absorbance
pH 10
pH 5
pH 4
pH 3
GY
CA CA
4020
4050
4080
4110
Photon energy (eV)
4140
4020
4050
4080
4110
4140
Photon energy (eV)
Fig. 3. The normalized (A) and first derivative (B) of calcium K-edge XANES spectra (black lines) of reference materials and linear combination fit (red lines) for the arsenate-gypsum coprecipitates formed at various pHs with initial arsenate concentration of 10,000 mg L−1 . “GY” and “CA” stand for arsenic-free gypsum and amorphous calcium arsenate, respectively.
pH
Percent of gypsum (%)
Percent of amorphous calcium arsenate (%)
Reduced chi-square
3 4 5 8 9 10
100 (0.00) 97.4 (0.05) 92.3 (0.04) 15.0 (0.05) 8.50 (0.04) 4.00 (0.00)
0.00 (0.00) 0.00 (0.05) 8.30 (0.04) 89.2 (0.05) 92.2 (0.04) 96.6 (0.00)
0.000103 0.000240 0.000188 0.000235 0.000142 0.000074
Hence, to determine the amount of As incorporated into the gypsum crystal, the arsenate-gypsum coprecipitates synthesized at higher pH values were subjected to ascorbic acid treatment to remove the amorphous calcium arsenate phases. The results (red legends in Fig. 1) showed that at low pH, arsenic concentration in the separated gypsum was equal to that in the untreated samples, indicating that almost all the arsenic in the solids was incorporated into the bulk lattice of gypsum. At the pH where calcium arsenate precipitated, the arsenic contents in the coprecipitates (black legends) were higher than those in the ascorbic acid treated samples (red legends). This difference represented the content of calcium arsenate in the coprecipitates, which rose as the pH and the initial As concentration increased. The content of arsenic in the gypsum for each initial arsenate concentration increased with increasing pH until the formation of calcium arsenate phases, but decreased with increasing pH after that due to the preferential precipitation of calcium arsenate. For example, at initial arsenate concentration of 5000 mg L−1 , As content in the gypsum increased from
