Environment
Investigation of the Scorodite Formation Mechanism in As(V) Solution Containing Fe(II) with Hematite Addition Using a Stable Iron Isotope
2022 Volume 63 Issue 4 Pages 655-661
Details
2022 Volume 63 Issue 4 Pages 655-661
Arsenic is a toxic element, and development of effective methods for As removal and stabilization are necessary. Removal and stabilization of pentavalent As as crystalline scorodite (FeAsO4·2H2O) is a promising method for As treatment of industrial byproducts. When hematite (Fe(III)2O3) powder is added to As(V) solution containing Fe(II) under appropriate conditions, scorodite crystals form from the gel-like precursor. When Fe(II) is not contained in the solution, the formation reaction does not proceed, even if hematite is added. Therefore, it is considered that the Fe(II) in the solution is heavily involved in the formation mechanism. Here, to elucidate the scorodite crystal formation mechanism in the hematite addition method, the origin of the iron in scorodite was investigated through scorodite synthesis experiments with addition of a stable iron isotope (54Fe) to the reaction solution. It was estimated that the Fe(III) constituting the scorodite crystals was mainly (more than about 80 atom%) derived from the Fe(II) in the solution. From this result, scorodite formation from the reaction solution with solid hematite can be considered to occur as follows. The gel-like precursor is composed of Fe(II) from the reaction solution together with Fe(III) from hematite. During conversion of the gel-like precursor to scorodite crystals, Fe(II) in the precursor is oxidized by Fe(III), and it then combines with the arsenate ion to form scorodite. With conversion to scorodite crystals, the Fe(II) generated by this electron exchange (originally from hematite) dissolves in the reaction solution. It is speculated that electron exchange between solution-derived Fe(II) and hematite-derived Fe(III) plays an important role in formation of scorodite crystals in this synthetic method.
Arsenic (As) is toxic to most plants and animals, and long-term exposure to arsenic causes various adverse health effects, including skin cancer.1) To prevent its distribution in the environment, development of effective methods for As removal and its stabilization is necessary. Removal and stabilization of pentavalent As as scorodite (FeAsO4·2H2O) crystals is a promising method to treat As in industrial byproducts. Coarse and crystalline scorodite particles can suppress dissolution of As.2) The overall reaction of scorodite formation can be expressed as follows:
| \begin{equation} \text{Fe$^{3+}$} + \text{H$_{3}$AsO$_{4}$} + \text{2H$_{2}$O} \to \text{FeAsO$_{4}{\cdot}$2H$_{2}$O} + \text{3H$^{+}$}. \end{equation} | (1) |
The DOWA Metals & Mining Scorodite Process (DMSP) method to produce large crystalline scorodite particles at normal atmospheric pressure and temperatures below 100°C with O2 gas blowing has been developed by DOWA Metals & Mining Co., Ltd.20–26) Combined use of ultrasound irradiation in this method has been also reported.27–30) It has been reported that the gel-like precursor in the DMSP method mainly consists of Fe(II) with a small amount of Fe(III) by X-ray absorption near-edge structure spectroscopy.31) This indicates that Fe(II) in solution is essential to form the gel-like precursor.
Direct supply of Fe(III) to As(V) solution containing Fe(II) has been investigated. Direct addition of Fe(III) as Fe2(SO4)3 solution to an As(V) solution containing Fe(II), without oxidation by O2 gas, has also been investigated.32,33) The results indicated that a high initial Fe(II) concentration is necessary to produce faceted scorodite crystals through conversion of the gel-like precursor by this method. Furthermore, direct addition of Fe(III) as solid Fe(III) compounds, such as hematite,33–36) magnetite,37–39) ferrihydrite,40) and limonite,41) has also been reported. For the hematite addition method, the overall scorodite formation reaction can be expressed as
| \begin{equation} \text{Fe$_{2}$O$_{3}$(s)} + \text{2H$_{3}$AsO$_{4}$(aq)} + \text{H$_{2}$O(l)} \to \text{2FeAsO$_{4}{\cdot}$2H$_{2}$O(s)}. \end{equation} | (2) |
In this study, to elucidate the scorodite crystal formation mechanism in the hematite addition method, the origin of the iron in scorodite was investigated through scorodite synthesis experiments with addition of a stable iron isotope (54Fe) to the reaction solution. A preliminary study of investigation of the origin of the iron in scorodite in the hematite addition method using a stable iron isotope has been reported.36) The concept to determine the origin of the iron in scorodite formed by the hematite addition method (from the iron in solution or that in hematite) is shown in Fig. 1. The stable isotope of iron (54Fe) was added to increase its fraction in the reaction solution. The variation of the iron isotope fraction in the generated scorodite was then measured and the origin of the iron in scorodite was quantitatively estimated.
Schematic of the method to determine the origin of the Fe in scorodite.
Metal powder of 54Fe with purity of 99.90% was purchased from the Center of Molecular Research (Moscow, Russia). The contents of 56Fe and the other isotopes in the metal powder were 0.08 and 0.02 atom%, respectively. Note that the percentages of 54Fe, 56Fe, and the other isotopes in nature are 5.845, 91.754, and 2.401 atom%, respectively.42) Hematite powder (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) was used as an additive to trigger scorodite formation. Ferrous sulfate heptahydrate (FeSO4·7H2O) and 60% arsenic acid were purchased from FUJIFILM Wako Pure Chemical Corporation. Ultrapure-grade sulfuric acid for trace element determination (Kanto Chemical Co., Inc., Tokyo, Japan) was used. Ultrapure water with electrical resistivity of more than 18.4 MΩ·cm prepared with a Q-POD Element unit (Merck, KGaA Darmstadt, Germany) was used in the experiments.
Ultrapure-grade HCl, HNO3 (FUJIFILM Wako Pure Chemical Corporation), HClO4, and H2SO4 (Kanto Chemical Co., Inc.) were used for dissolution of the solid samples.
2.2 Scorodite formation procedureA certain amount of 54Fe metal powder (100 or 300 mg) was dissolved with a stoichiometric amount of sulfuric acid and ultrapure water to prepare FeSO4 solution. During the dissolution step, Ar gas was blown against the gas–solution interface at 50 mL/min to prevent iron oxidation. The solution was stirred with a magnetic stirrer overnight. The solution temperature was maintained at 40°C. Solid Fe(II)SO4·7H2O and 60% arsenic acid were then added to the obtained solution with a total Fe(II) concentration of 25 g/L (0.45 mol/L) and an As(V) concentration of 50 g/L (0.67 mol/L). The Fe isotope composition in the initial reaction solution for each experiment is summarized in Table 1. The values were estimated from the measured 54Fe/56Fe atomic ratios in the initial solutions and the isotope compositions in the 54Fe metal powder and nature. Note that these four experiments were conducted under the same conditions, except for the 54Fe addition amount, to confirm the contribution of Fe in the reaction solution to Fe in the scorodite formed.
The experimental apparatus for scorodite crystal formation using a stable Fe isotope is shown in Fig. 2. Before the reaction using hematite, the solution was heated to 75°C and the precipitated solid was removed using a membrane filter with a pore size of 0.45 µm (C045A047A, cellulose acetate, φ = 47 mm, ADVANTEC Toyo Roshi Kaisha, Ltd., Tokyo, Japan). The filtrate (25 mL) was used as the reaction solution. The reaction solution was immediately heated to 83°C, and then 0.348 g of hematite powder was added to make the Fe(III) content in the solution 9.734 g/L (0.174 mol/L) as solid (t = 0). In the reaction system, there was excess arsenic to minimize the unreacted hematite amount. Ar gas was blown against the gas–solution interface at 50 mL/min to remove air, and the reactant was stirred at 300 rpm. During the reaction, the solution temperature was maintained at 83–84°C.
Experimental apparatus for scorodite crystal formation using a stable Fe isotope.
The reaction time was set to 5 h to obtain faceted coarse scorodite crystals. After the reaction, the reaction solution was filtered and the remaining solid was washed with ultrapure water. The solid part was then freeze-dried overnight with a freeze drier (FD-1000, EYELA, Tokyo Rikakikai Co., Ltd. Tokyo, Japan). The obtained solid, that is, the mixture of unreacted hematite and generated scorodite, was examined by X-ray diffraction (XRD; D2PHASER, Bruker Corp., MA. USA; radiation source Cu Kα) and observed by scanning electron microscopy (SEM; VE-9800, KEYENCE Corporation, Osaka, Japan). The obtained solid (100 mg) was dissolved in 19 mL of an acid mixture [10 mL of HCl (concentration of 36 mass%), 2 mL of HClO4 (60 mass%), 5.83 mL of HNO3 (70 mass%), and 1.17 mL of H2SO4 (96 mass%)] for 30 min at 200°C using a graphite digestion system (ECOPRE, ACTAC, Tokyo, Japan). HCl (10 mL, 36 mass%) was then added to the solution and diluted with ultrapure water to 50 mL. This procedure for solid dissolution was performed twice to obtain duplicate solution samples (n = 2). The prepared solutions were then analyzed by inductively coupled plasma–atomic emission spectrometry (ICP-AES; Spectro Arcos, Spectro Analytical Instruments, Kleve, Germany) to determine the concentration of As and by inductively coupled plasma–mass spectrometry (ICP-MS; ELEMENT 2, Thermo Fisher Scientific K.K., Tokyo, Japan) to analyze the Fe isotopes. The As concentration in the solution (mg/L) was then converted to the mass percentage of As in the original solid (mass%).
In the ICP-MS measurements, measurements to determine the calibration curves of 54Fe/56Fe were performed before and after the measurements of the samples. The range of the 54Fe/56Fe molar fraction was 0.05–0.6, and a linear function was used to fit the plots. Note that the relative atomic masses of 54Fe and 56Fe (i.e., 53.93960899 and 55.93493633)43) were used to calculate the 54Fe/56Fe molar fractions in the calibration solutions. The obtained calibration curves were used for result calibration. The average values of the duplicate solution samples (n = 2) were used as the 54Fe/56Fe values for the sample solutions. In ICP-MS analysis, the ion current value for each isotope of each sample was repeatedly measured 1200 times. The variability information of the repeated measurements in ICP-MS for 54Fe and 56Fe was obtained and used to evaluate the variability of the fraction estimation of Fe in scorodite, which originates from the Fe in solution (see Section 2.3). The measurement variability of the Fe measurement in ICP-MS was less than 10% in relative standard deviation (RSD). Note that the average RSD of the duplicate solution samples (n = 2) was used for the evaluation. The reaction solutions before and after scorodite formation were also analyzed by ICP-AES and ICP-MS.
2.3 Estimation of the Fe fraction derived from solutionThe final solid products of the scorodite formation experiments were a mixture of scorodite, unreacted hematite, and a trace amount of the gel-like precursor. To estimate the origin of the Fe in the scorodite formed, the molar fractions of these components should first be determined. The following assumptions were used for estimation:
Considering the overall reaction of hematite with arsenic acid (eq. (2)), when the initial amount of hematite is mh,i [mol] and conversion of hematite is X [-], 2mh,iX [mol] of scorodite is generated and mh,i(1 − X) [mol] of unreacted hematite remains under the above described assumptions. The amount of As in the final solid after scorodite formation mAs,f [mol] is equal to 2mh,iX, which can be easily calculated from elemental analysis of the final solid. Thus, X can be estimated by
| \begin{equation} X = \frac{m_{\text{As,f}}}{2m_{\text{h,i}}}. \end{equation} | (3) |
The origin of the Fe in the scorodite formed was then estimated from the values and 54Fe/56Fe analysis of the solid. When the fraction of Fe derived from Fe(II)SO4 solution is equal to Y [-], the amounts of 54Fe and 56Fe in the final solid (m54,f [mol] and m56,f [mol], respectively) can be expressed as
| \begin{align} m_{\text{54,f}} & = (2m_{\text{h,i}} - m_{\text{As,f}})F_{\text{54,h}} + m_{\text{As,f}}(1 - Y) F_{\text{54,h}} \\ & \quad + m_{\text{As,f}}YF_{\text{54,s}}\\ & = 2m_{\text{h,i}}F_{\text{54,h}} + m_{\text{As,f}}Y (F_{\text{54,s}} - F_{\text{54,h}}), \end{align} | (4) |
| \begin{align} m_{\text{56,f}} & = (2m_{\text{h,i}} - m_{\text{As,f}}) F_{\text{56,h}} + m_{\text{As,f}} (1 - Y) F_{\text{56,h}} \\ & \quad + m_{\text{As,f}}YF_{\text{56,s}}\\ & = 2m_{\text{h,i}}F_{\text{56,h}} + m_{\text{As,f}}Y(F_{\text{56,s}} - F_{\text{56,h}}), \end{align} | (5) |
Solving m54,f/m56,f for Y gives
| \begin{equation} Y = \frac{2m_{\text{h,i}}}{m_{\text{As,f}}}\cdot\frac{F_{54,h} - F_{56,h} (m_{54,f}/m_{56,f})}{(F_{56,s} - F_{56,h})(m_{54,f}/m_{56,f}) - (F_{54,s} - F_{54,h})}. \end{equation} | (6) |
It should be noted that F54,s and F56,s can vary with time when Y is not equal to 0 because the ratio of the Fe amount in the reaction solution to that in the scorodite formed is not infinite (about 2.5–2.7:1 in our experiments) owing to the high cost of the Fe isotope. In this study, Y was estimated to be about 0.8, which will be described later. In this case, Fe in the gel-like precursor derived from hematite can move to the reaction solution accompanied by gel-like precursor conversion to scorodite. This can result in a decrease of F54,s and an increase of F56,s in the reaction solution because we added 54Fe to the initial reaction solution. If the Fe in the reaction solution with lower F54,s and higher F56,s values is involved in gel-like precursor formation, the estimated Y must decrease. In other words, the Y value will be underestimated by the estimation method. Note that sequential estimation of Y considering this influence is difficult because the reaction must pass the intermediate situation with the presence of much gel-like precursor, which can change the total molar mass of Fe in the reaction solution.
Evaluation of the variation of the estimated values was performed as follows. The measurement variability information in the ICP-MS measurements of 54Fe and 56Fe was obtained and the propagation of the variability was calculated for 54Fe/56Fe. The variability influence was then considered in eq. (6). It was found that the upper and lower variation ranges were different, and the 68% and 95% confidence intervals for 54Fe/56Fe were calculated by the average ±1SD and ±2SD, respectively.
SEM images of the obtained solids after the reaction are shown in Fig. 3. For all of the trials, faceted scorodite crystals were observed. In the SEM images, unreacted hematite particles were not observed. Unreacted hematite would be included in the scorodite crystals and would not be observed from the surface. Trace amounts of the fibrous precursors were observed in the SEM images.
SEM image of the solid obtained after the scorodite formation reaction for each trial.
The XRD patterns of the obtained solids after the reaction are shown in Fig. 4. Only peaks that can be assigned to scorodite or hematite were observed in the XRD patterns. The peaks for hematite were very weak compared with those for scorodite, and thus it was estimated that the unreacted hematite amount was smaller than the amount of scorodite formed.
XRD patterns of the solids obtained after the scorodite formation reaction: hematite (PDF No. 00-033-0664) and scorodite (PDF No. 00-037-0468).
For experimental trial 4, it was confirmed that the Fe(II) concentration in the reaction solution did not obviously change, even after scorodite synthesis.
The recovered solid amounts and their compositions after the scorodite formation experiments are given in Table 3. The 54Fe/56Fe ratios of the recovered solids (scorodite and unreacted hematite) were much larger than that in nature (0.0637). Thus, the Fe originally contained in the solution had an influence on scorodite formation. Hematite conversion was then estimated by the method described in Section 2.3. The results are summarized in Table 4. The estimated conversion ratios were about 0.8 for all of the experimental trials, which means that about 80% of hematite was consumed by scorodite formation in the experiments.
The origin of the Fe in the scorodite was then estimated from the conversion values and 54Fe/56Fe analysis of the solid. The results are summarized in Table 5. The fraction of Fe in scorodite derived from the Fe(II)SO4 solution (Y) was estimated to be about 0.8 for all four experimental trials. This means that the Fe(III) contained in the scorodite crystals was mainly (more than about 80 atom%) derived from the Fe(II) in the solution. From this result, scorodite formation from the reaction solution with solid hematite can be considered to occur as follows. The gel-like precursor is composed of Fe(II) from the reaction solution together with Fe(III) from hematite. During conversion of the gel-like precursor to scorodite crystals, the Fe(II) in the precursor is oxidized by Fe(III) and then combines with the arsenate ion to form scorodite. With conversion to scorodite crystals, the Fe(II) generated by this electron exchange (originally from hematite) dissolves in the reaction solution. It is speculated that electron exchange between solution-derived Fe(II) and hematite-derived Fe(III) plays an important role in formation of scorodite crystals in this synthetic method. These reactions occur in the potential range from 0 to +200 mV vs. Ag|AgCl.34) In addition, they are conducted under Ar-gas flow to exclude the influence of dissolved oxygen, and thus the oxidant is only Fe(III) from hematite. Considering the results, the possible mechanism of generation of scorodite from hematite is shown in Fig. 5.
Possible mechanism of generation of scorodite from hematite.
The origin of the Fe in scorodite may be influenced by the reaction conditions, such as the Fe(II) concentration in the solution, the ratio of Fe(II) in the solution to Fe(III) in solid hematite, and the temperature. It has been reported that as with Fe(II) concentration of 50 g/L, scorodite also forms with lower Fe(II) concentration of 25 g/L, but it doesn’t form without Fe(II).35) It has also been reported that the scorodite formation rate decreases at lower temperature in the range from 50 to 95°C.35) Thus, further investigation under different conditions is required. Furthermore, the reason why Fe(III) in scorodite is dominantly derived from Fe(II) in solution must be investigated.
To elucidate the scorodite crystal formation mechanism in the hematite addition method, the origin of the Fe in scorodite has been investigated through scorodite synthesis experiments with addition of a stable Fe isotope (54Fe) to the reaction solution. It was estimated that most of the Fe(III) in the scorodite crystals (more than about 80%) was derived from Fe(II) in the solution. From the results, scorodite formation from the reaction solution with solid hematite can be considered to occur as follows. The gel-like precursor is composed of Fe(II) from the reaction solution together with Fe(III) from hematite. During conversion of the gel-like precursor to scorodite crystals, the Fe(II) in the precursor is oxidized by Fe(III), and it then combines with the arsenate ion to form scorodite.
With conversion to scorodite crystals, the Fe(II) generated by this electron exchange (originally from hematite) dissolves in the reaction solution. It is speculated that electron exchange between solution-derived Fe(II) and hematite-derived Fe(III) plays an important role in formation of scorodite crystals by this method. In the process, it is considered that hematite acts as an oxidizer rather than an Fe source.
The origin of the Fe in the scorodite may be influenced by the reaction conditions, such as the Fe(II) concentration in the solution, the ratio of Fe(II) in the solution to Fe(III) in solid hematite, and the temperature. Thus, further investigation under different conditions is required. Furthermore, the mechanism for how Fe(III) in scorodite is dominantly derived from Fe(II) in solution must be investigated.
This work was supported by JSPS KAKENHI (Grant Number JP19H02753).
