Aqueous nickel sequestration and release during structural Fe(II) hydroxide remediation: the roles of coprecipitation, reduction and substitution

Abstract

Aqueous Ni²⁺ removal by structural Fe(II) hydroxides (SFH) under well-controlled experimental conditions was investigated in this study, and a possible mechanism for Ni²⁺ release from solid products was revealed. Experiments with SFH and Ni²⁺ showed the excellent reactivity of SFH towards aqueous Ni²⁺ when the molar ratio of [Fe(II)]/[OH⁻] was below 1 : 2. The reaction started with adsorption and precipitation of Ni²⁺, followed by reduction of Ni(II) and substitution of Fe(II) with the formation of Ni(II)/Fe(III) layered double hydroxides. However, at long reaction time, Ni(II) release from Ni(OH)₂ and Ni_xFe_(1−x)(OH)₂ precipitates was observed due to the delivery of dissolved Fe²⁺ and Fe³⁺, which were determined to substitute Ni(II) by forming Fe(OH)₂, Fe(OH)₃ and Fe^III_(1+2x/3)Fe^II_(1−x)(OH)₅. The presence of O₂ and NO₃⁻ reduced the removal efficiency of Ni²⁺ and promoted its release by consuming ≡Fe(II) and promoting Fe³⁺ delivery. However, CO₃²⁻ and PO₄³⁻ might enhance the removal of Ni²⁺ and inhibit its release. For wastewater containing a high Ni²⁺ concentration, using SFH is also beneficial to Ni²⁺ recycling, as removed Ni(II) could be released and enriched, followed by future utilization.

---

1. Introduction

Nickel (Ni) is a ubiquitous element and high levels of Ni may cause lung and kidney damage as well as gastrointestinal distress, pulmonary fibrosis and dermatitis.¹ Potentially toxic metals including Ni could be removed effectively by various alternatives, such as electro-coagulation,² co-precipitation,³ bio-sorption,⁴ and membrane separation.⁵

Zerovalent iron (ZVI) has been generally recognized as highly effective for the removal of persistent pollutants, including potentially toxic metals, in the last two decades, due to its high efficiency and relatively low economic and environmental cost.^6,7 Mechanisms including surface complexation, adsorption, reduction and precipitation contributed to the overall potentially toxic metals removal.^7–9 Due to its reductive nature and the fact that the corrosion of the material is thermodynamically favored in the presence of oxygen, ZVI particles unavoidably develop a thin film of oxides on the surface, being extensively accepted that this external film has a fundamental role in potentially toxic metals sequestration.¹⁰ However, corrosive passivation on Fe⁰ surface results in slow kinetics of corrosion by hindering both electrons and Fe(II) transfer from the underlying structural or absorbed Fe(II) to the aqueous interface. Therefore, the subsequent iron efficiency on contaminants is weakened.^11–14 The structural and absorbed Fe(II) on the Fe⁰ (E^θ = −0.44 V) surface generates powerful reducing agents (E^θ = −0.65 to −0.34 V), which are responsible for contaminant reductive transformation when Fe⁰ is covered by an oxide scale.^13,15,16 Therefore, Fe(II)-bearing minerals were supposed to present excellent property in potentially toxic metals removal. Green rust (GR), as a mixture of ferrous/ferric hydroxides, was an outstanding representative. Similarly, the high reactivity of GR was ascribed to relatively high content of ≡Fe(II).¹⁷ Based on the analysis above, Fe(II)-bearing minerals with higher ≡Fe(II) content would greatly enhance its reactivity. Thus, the creation of well-controlled structural Fe(II) hydroxides (SFH) might be an alternative for avoiding passivation and serve for toxic metals removal. Our previous studies showed the high reactivity of SFH towards organic pollutants,¹⁸ nitrite¹⁹ and toxic metals, such as Cr(VI)²⁰ and Cu(II).²¹ Both rapid reaction rate and overall high removal efficiency were presented, which may be ascribed to its high reduction potential and relatively high surface area. Therefore, its positive effect on the Ni(II) removal could be forecasted.

Potentially toxic metals remediation by ZVI and GR mainly focused on assessing significance of chemical or structural differences and reaction kinetics. Particularly, the Ni(II)-contained solid products were characterized to reveal possible mechanisms convincingly.^22,23 There is limited research dealing with Ni(II) release controlling. Two examples are the works of Calderon and Fullana²⁴ and Belebchouche et al.²⁵ about metals release, in which they proved that Ni(II) releasing has shown a remarkable dependence on pH decreasing. For example, in the study of Calderon and Fullana,²⁴ a delivery percentage to the water of 44% and 65% in 24 h and 21 days, respectively, was determined during long reactions of Ni(II) with nZVI. In that case, it is suggested that remediation should be performed with a pH higher than 7.5 in order to avoid the delivery of the metal ions from nZVI. Nevertheless, studies are lacking in Fe²⁺ or Fe³⁺ effect on continuous experiments. When iron-based materials are applied for Ni(II) removal, it is important to determine whether Fe²⁺ or Fe³⁺ will affect Ni(II) release and subsequent removal efficiency.

Accordingly, the aim of this work is to study the removal of aqueous Ni²⁺ by SFH and provide a new sight to the release of removed Ni(II) during the process. This study will (1) investigate the roles of OH⁻ and ≡Fe(II) in the removal process and chemical behaviors of Ni(II) to reveal the removal mechanisms; (2) determine the effect of released Fe²⁺ and Fe³⁺ on Ni(II) release during reaction time, which has not been studied before; (3) assesses whether Ni(II) removal are less impacted by oxyacids (PO₄³⁻, CO₃²⁻, NO₃⁻), and thereby help to better design ground and wastewater treatments.

2. Experimental

2.1 Material

All reagents (Sinopharm Chemical Reagent) were analytical grade and used without further purification. Ni(II) stock solution was prepared with NiSO₄ in an amber bottle with deoxygenated ultrapure water. Ni(II)-citrate and Ni(II)-tartaric acid complexes were prepared by mixing the solution of Ni(II) with excess citrate and tartaric acid to obtain a 1 : 2 molar ratio of [Ni²⁺]/[citrate] and [Ni²⁺]/[tartaric acid]. All solutions were prepared with 18 MΩ cm⁻¹ ultrapure water that was pumped by N₂ flow to blow off dissolved O₂. Ultrapure water was obtained from a Milli-Q water purification system (Millipore, USA). High-purity (99.999%) N₂ was purchased from Spring Specialty Gases (Shanghai, China). PTFE syringe filters (0.22 μm × 50 mm) were purchased from ANPEL Scientific Instrument (Shanghai, China).

2.2 Batch experiments

Structural Fe(II) hydroxides, as the above mentioned ≡Fe(II)-bearing mineral, was synthesized by precipitation from a mixture of FeSO₄ and NaOH.^18–21 Firstly, FeSO₄·7H₂O was dissolved in 100 mL of deoxygenated ultrapure water in a 200 mL flask under magnetic stirring. Afterward, 100 mL of NaOH, corresponding to the [Fe(II)]/[OH⁻] ratio of 2 : 1, 1 : 1, 1 : 2, 1 : 3, 1 : 4, respectively, was added to the solution under the same condition of magnetic stirring. Finally, a green suspension with pH of approximately 6.63, 7.87, 9.05, 11.18 and 11.65 was generated, which was referred as SFH (2 : 1, 1 : 1, 1 : 2, 1 : 3 and 1 : 4, respectively). All preparation works were performed in an anoxic glove box under N₂ atmosphere to avoid possible oxidation, and SFH was freshly prepared to avoid oxidation by oxygen.

For the study of Ni(II) removal from the solution by SFH, batch experiments were performed in 100 mL headspace vials with Teflon-lined caps and wrapped with aluminum foil. Reactions were initiated by adding SFH suspensions to the solution (90 mL) containing Ni(II), Ni(II)-citrate and Ni(II)-tartaric acid complexes. Ni(II)-citrate and Ni(II)-tartaric acid complexes were employed to investigate the reducing capacity of SFH under alkaline circumstance. During the reaction, the headspace vials were exposed to the ambient air to investigate the role of O₂ in Ni(II) removal. To assess the roles of NO₃⁻,CO₃²⁻ and PO₄³ in the reaction, experiments were conducted in headspace vials sealed with caps and the solution was purged by nitrogen gas for 30 min prior to the addition of SFH to initiate the reaction and for the whole reaction to exclude dissolved oxygen (DO). This condition was termed “oxygen-limited” in contrast to the “oxygen-rich” condition, in which the vials were open to the air. The initial pH values were adjusted with NaOH and H₂SO₄, and no attempt was made to maintain a constant pH during the reaction. The variation of pH values are shown in Table S1 (ESI†). At given time intervals, approximately 1.5 mL of suspension was sampled, filtered through a 0.45 μm membrane filter, and acidified for analysis. All experiments were run in duplicates or triplicates, and the data were averaged. The error bars for experimental data points were noted unless the standard deviations were within 5%.

2.3 Analytical methods and characterization

Residual Ni(II) concentration was measured using inductively coupled plasma optical emission spectrometry (ICP-OES; Agilent 720ES, USA). The pH was measured using pHSJ-3F (Shanghai Precision & Scientific Instrument). After specific tests, the precipitates were collected for further analysis. Morphology of SFH solid phases was obtained using a field emission scanning electron microscope (SEM) (Hitachi S-4800, Japan). X-ray Diffraction (XRD) patterns were collected using a D8 Advance Powder X-ray Diffractometer (WXRD, Germany) with a Cu Kα (λ = 1.5406 Å) radiation source (40 kV, 40 mA); the diffraction angle (2θ) was recorded from 10° to 80°, with a scanning speed of 1°min⁻¹ and a step size of 0.02°. Related oxidation states on the surface of Ni-containing solid products were analyzed via X-ray photoelectron spectroscopy (XPS) by using a PHI-5000 Versaprobe spectrometer equipped with a rotating Al anode generating Al Kα X-ray radiation at 1486.6 eV. Fourier transform infrared (FT-IR) spectroscopy spectra of samples were recorded to monitor changes in the functional group of SFH, SFH–CO₃²⁻ and SFH–PO₄³⁻ after the Ni(II) removal tests on a Nicolet 5700 spectrometer using KBr pellets in the range of 4000–400 cm⁻¹.

3. Results and discussion

3.1 Sequestration of Ni(II) by SFH

The reactivity of SFH, with various [Fe(II)]/OH⁻ molar ratio, on Ni²⁺ sequestration was investigated in Fig. 1. Fe²⁺ alone showed negligible effect on Ni²⁺ sequestration, as approximately 95% Ni²⁺ was detected at 60 min. However, simultaneous addition of OH⁻ was found to enhance the sequestration. When [Fe(II)]/[OH⁻] decreased to 1 : 2, only 22.4% Ni²⁺ was remained within first 20 min, which means 78.6% Ni²⁺ was effectively sequestrated. 100% Ni²⁺ sequestration was observed with further decrease of [Fe(II)]/[OH⁻] to 1 : 3 and 1 : 4. However, at [Fe(II)]/[OH⁻] of 1 : 2, Ni²⁺ concentration at 60 min was noticed to be obviously higher than that of 20 min, indicating the release of sequestrated Ni(II). The overall reaction of Ni²⁺ and SFH was proposed to be divided into three stages. The first stage was responsible for Ni²⁺ concentration decrease and immediate pH decline (Fig. 1a and b), in which amorphous Ni–Fe complexes were formed due to adsorption and co-precipitation, and its possible molecular formula was proposed as Ni_xFe_(1−x)(OH)₂ (reaction 1). The generation of H⁺ was favored by such a rapid reaction. Followed by the first stage, highest Ni²⁺ removal efficiency was observed and no obvious Ni(II) release was detected in the second stage (within initial 20 min). Finally, the last stage was responsible for Ni(II) release from the solid products. Hydrolysis of Fe(II) and Fe(III) in solution generated protons, Fe²⁺ and Fe³⁺, and resulted in pH decrease (Fig. 1b and Table S1†). However, the pH varies within a small range at higher initial pH (e.g., SFH(1 : 3), SFH(1 : 4)), due to great buffer ability of SFH. No release of Ni(II) was observed by using SFH(1 : 3), SFH(1 : 4). Thereafter, generated H⁺ could lead to dissolution of precipitated Ni(OH)₂ (reaction 2).

xNi²⁺ + (1 − x)≡Fe²⁺ + 2H₂O → Ni_xFe_(1−x)(OH)₂ + 2H⁺ (1)

Ni(OH)₂ + 2H⁺ → Ni²⁺ + 2H₂O (2)

Fig. 1 Influence of Fe(II)/OH⁻ molar ratio on removal of aqueous Ni²⁺ (a) and the corresponding variations in dissolved iron and solution pH (b) (SFH = 2.0 mM, initial Ni²⁺ = 1.0 mM, initial pH of Ni²⁺ solution with Fe²⁺, SFH(2 : 1), SFH(1 : 1), SFH(1 : 2), SFH(1 : 3) and SFH(1 : 4) were 5.31, 6.63, 7.87, 9.05, 11.18 and 11.65, respectively).

Besides, the reactivity of SFH(1 : 2) and the release of Ni(II) from Ni–Fe contained solid products were influenced by co-present cations. The K_sp of Ni(OH)₂, Fe(OH)₂ and Fe(OH)₃ were 2.0 × 10⁻¹⁵, 8.0 × 10⁻¹⁶ and 4.0 × 10⁻³⁸, respectively. Fe(II) oxidation generated Fe(III), which combined with the hydroxyls from Ni(OH)₂ forming Fe(OH)₃, as K_sp of Fe(OH)₃ was obviously lower than that of Ni(OH)₂. Also, the oxidation of SFH resulted in the decrease of adsorption capacity of Ni²⁺. Therefore, Ni(II) release was observed from Ni(II)-contained solid product. Experiments were conducted to confirm that Ni(II) in Ni(OH)₂ precipitation can be substituted by Fe(III), as showed in Fig. 2. It could be clearly seen that aqueous Ni²⁺ concentration increased rapidly after adding Fe³⁺ immediately. Therefore, the Ni(II) in Ni(OH)₂ could be quickly and effectively substituted by Fe(III), and Ni(II) was released completely within a shorter time. The substitution of Ni(II) by Fe³⁺ can be described by reaction 3. Due to similar K_sp of Fe(OH)₂ (8.0 × 10⁻¹⁶) and Ni(OH)₂ (2.0 × 10⁻¹⁵), the substitution of Ni(II) by Fe²⁺ is less efficient before being oxidized into Fe³⁺, as shown in Fig. 2. Meanwhile, Ni(II) in the Ni–Fe solid phases was presumably substituted by Fe³⁺ and formed Fe^II–Fe^III species (reaction 4).

Fig. 2 Evolution of aqueous Ni²⁺ concentration when Fe²⁺ or Fe³⁺ was added to Ni(OH)₂ (initial Ni²⁺ = 1.0 mM, Fe²⁺ or Fe³⁺ = 2.0 mM).

In order to verify the assumption, the K_sp of Ni_xFe_(1−x)(OH)₂ and Fe^III_(1+2x/3)Fe^II_(1−x)(OH)₅ were determined (Text S1†). Based on the results:

8.0 × 10⁻¹⁶ < K_sp[Ni_xFe_(1−x)(OH)₂] < 2.0 × 10⁻¹⁵ and 4.7 × 10⁻⁶³ < K_sp[Fe^III_(1+2x/3)Fe^II_(1−x)(OH)₅] < 1.7 × 10⁻³⁸,

it can be seen that K_sp[Ni_xFe_(1−x)(OH)₂] is much larger than K_sp[Fe^III_(1+2x/3)Fe^II_(1−x)(OH)₅], which indicated that the substitution in reaction 4 is favorable.

3Ni(OH)₂ + 2Fe³⁺ → 2Fe(OH)₃ + 3Ni²⁺ (3)

Ni_xFe_(1−x)(OH)₂ + (1 + x)Fe³⁺ + 3H₂O → Fe^III_(1+2x/3)Fe^II_(1−x)(OH)₅ + xNi²⁺ + 3H⁺ (4)

The above investigations revealed that Ni(II) release from solid product could be controlled by keeping the solution pH above 8.0. Under this circumstance, highest Ni(II) removal efficiency was acquired without possible Ni(II) release. Alkaline environment was also beneficial to keep a relatively stable structure of SFH. While the pH decline may lead to the change of inner structure, thus resulting in delivery of Fe²⁺ or Fe³⁺ and subsequent Ni(II) release, which has never been considered before. Meanwhile, Ni(II) release could also be controlled by employing an appropriate reaction time. For example, the reaction time for SFH(1 : 2) should be no more than 50 min, and no less than 20 min.

3.2 Comparison between OH⁻ and ≡Fe(II) on Ni²⁺ removal

Sequestration of Ni²⁺ could obviously be caused by precipitation under higher OH⁻ concentration by forming Ni(OH)₂. Thus, higher Ni(II) sequestration efficiency by SFH was presumably associated with Ni(OH)₂ formation to a great extent. Fig. 3 showed the Ni²⁺ sequestration induced by Ni(OH)₂ precipitation under various pH. The pH₀ of Ni²⁺ solution with SFH(1 : 1), SFH(1 : 2), SFH(1 : 3) and SFH(1 : 4) was 5.31, 6.63, 7.87, 9.05, 11.18 and 11.65, respectively. It can be seen that Ni²⁺ was completely precipitated at pH values of 11.18 and 11.65, implying that strong alkaline condition played a significant role in the sequestration. At pH of 9.05, only 29.9% Ni²⁺ was precipitated by OH⁻ (Fig. 3). However, up to 78.6% Ni²⁺ was sequestrated by SFH(1 : 2) under the same pH condition (Fig. 1). The results indicated limited precipitation capacity of OH⁻ in SFH(1 : 2). Similar to our previous conclusion, indispensible role of ≡Fe(II) that generated in the simultaneous interaction between aqueous Fe²⁺ and OH⁻ was reasonably proposed in enhancing Ni(II) sequestration efficiency.²¹ The complexes of Ni(II)-citrate and Ni(II)-tartaric acid were applied to avoid the formation of insoluble Ni(OH)₂ compounds and thus to evaluate the role of ≡Fe(II) on Ni²⁺ removal, as shown in Fig. 4.

Fig. 3 Evolution of Ni²⁺ concentration at various pH values (initial Ni²⁺ = 1.0 mM).

Fig. 4 Influence of Fe(II)/OH⁻ molar ratio on removal of chelated Ni(II) (initial Ni²⁺ = 1.0 mM, SFH = 9.0 mM).

The residual Ni²⁺ decreased with the increase of [Fe(II)]/[OH⁻] and ≡Fe(II) content, indicating that ≡Fe(II) played a crucial role during the entire sequestration process, which was presumably correlated with its high adsorption ability and reductive potential. It is well known that Ni in complex state is more difficult to treat than dissociative Ni²⁺. In general, oxidation was used to break the complexants in previous studies. This study provides a new strategy to efficiently sequestrate refractory chelated Ni²⁺ through reduction. In this study, the high efficiency of ≡Fe(II) on Ni²⁺ removal may be due to several possible reasons: (1) large surface area of SFH results in efficient adsorption; (2) Ni(II) substitutes ≡Fe(II) in SFH, forming layered double hydroxides (LDHs); (3) ≡Fe(II) is better electron donor and thus oxidizes Ni(II).

3.3 The roles of adsorption, substitution and reduction in Ni(II) sequestration

Ni(II) reduction by ≡Fe(II) occurs at surface sites or after diffusion into interlayer spaces of SFH, thus the removal efficiency depends on ≡Fe(II) content, specific surface area, and type of interlayer anion.^15,26 The morphology of the solid phase of SFH(1 : 1), SFH(1 : 2) and SFH(1 : 3) was acquired by SEM, as showed in Fig. S1.†

Small particles with rough and porous surfaces along with undefined shape crystals are seen. However, the grain size of particles changed with the decrease of [Fe(II)]/[OH⁻], which might exert influence on the surface specific areas. In particular, particle agglomeration was observed in SFH(1 : 3), which may be not beneficial to reaction. The BET of SFH(1 : 2) solid phase was determined to be 88.6 m² g⁻¹, lager than that of GR-CO₃ (30.1 m² g⁻¹),¹⁷ which was an evidence of high adsorption reactivity of SFH(1 : 2). Adsorption data obtained were fitted to Langmuir and Freundlich isotherm models as shown in Fig. S2.† Correlation coefficients of fittings showed that Ni²⁺ adsorption was not well fitted by both Langmuir isotherm (R² = 0.81) and Freundlich isotherm (R² = 0.68). It indicated that multiple mechanisms would be involved in the process of Ni²⁺ removal.

Fig. 5 showed the XRD pattern of the reaction products of Ni(II) and SFH(1 : 2). The result revealed a high degree of crystallinity with various characteristic peaks, i.e. (003), (006), (009), (012), (015), (018), (110) and (113), which are similar to those previously found in layered double hydroxides.^27–29 Furthermore, peaks at 2θ of 11.3° and 22.7° indicated the presence of Ni(OH)₂·0.75H₂O, but also showed overlap with the (003) and (006) reflection.³⁰ It has been confirmed that the interlayer Fe(II) in LDH could be substituted by Mg(II) or Al(III) forming a hydrotalcite or pyroaurite.³¹ Due to similar LDH structure of SFH,¹⁸ ≡Fe(II) in SFH(1 : 2) could be possibly substituted by aqueous Ni²⁺ forming Ni(II)/Fe(III) LDH. Therefore, Ni(II) adsorption on the surface of SFH(1 : 2) and subsequent substitution for ≡Fe(II) was confirmed, which was consistent with the result above.

Fig. 5 XRD pattern (Cu Kα radiation) of reaction products between SFH (1 : 2) and Ni(II).

The structural and absorbed Fe(II) generates powerful reducing agents (E^θ = −0.65 to −0.34 V), making ≡Fe(II) a stronger reducing agent than aqueous Fe(II).³² Therefore, XPS analysis was conducted to reveal the potential ≡Fe(II)-induced valence change of Ni(II) during the reaction, as shown in Fig. 6. The major elements of the reacted SFH(1 : 2) were detected to be iron (Fe), nickel (Ni), and oxygen (O) (Fig. 6a). Binding energies of 709.8 eV and 710.8 eV in Fe 2p spectra (Fig. 6b) indicated the presence of Fe(II) and Fe(III),³² confirming Fe(II) oxidation by Ni(II). Ni 2p spectra was shown in Fig. 6c, and the binding energies at 852.4 eV and 853.8 eV indicated the presence of Ni(0) and Ni(II), respectively.³³ Therefore, Ni(0) generation from Ni(II) by ≡Fe(II) reduction was confirmed as reaction 5:

2≡Fe(II) + Ni(II) → 2Fe(III) + Ni(0) (5)

Fig. 6 (a) Full-range XPS spectra of fresh SFH and Ni-loaded spent SFH; (b) XPS survey on Fe 2p of spent SFH (c) XPS survey on Ni 2p of spent SFH (Fe²⁺/OH⁻ = 1 : 2, initial pH = 9.05).

Fig. 7 showed the SEM images of solid products after the reaction of SFH(1 : 2) and Ni(II) within different reaction time. Irregular solid products with rough surface in smaller diameters was observed at 20 min, as showed in Fig. 7b. Besides, small flocs in considerable content were also observed on the layer surface, which might be Ni–Fe precipitations (Ni_xFe_(1−x)(OH)₂), Ni–Fe LDHs and precipitated Ni(OH)₂. However, Ni(II) release was observed at a longer reaction time of 60 min as mentioned above, which was relevant to the SFH(1 : 2) oxidation. At 60 min, complexes surface was found to be platelike (Fig. 7c), which might be ascribed to amorphous Ni–Fe complexes dissolution and Ni(II) release. Flocs disappeared completely at 120 min and smooth surface was observed, indicating that almost all the removed Ni(II) was released into the solution.

Fig. 7 SEM images of solid product samples collected from the reaction between SFH(1 : 2) and Ni(II) at different reaction time: (a) 0 min, (b) 20 min, (c) 60 min and (d) 120 min.

Overall, the above results demonstrated that multiple mechanisms were involved in the removal of aqueous Ni²⁺, including adsorption, co-precipitation, substitution, and reduction. Particularly, the reaction started with adsorption and precipitation of Ni²⁺, followed by reduction of aqueous Ni²⁺ and substitution of Fe(II). SFH has a strong potential for reducing aqueous Ni²⁺ and even chelated Ni²⁺ to Ni⁰. The adsorbed Ni(II) could substitute Fe(II) with the formation of Ni(II)/Fe(III) layered double hydroxides at short time. However, with the decrease of pH and subsequent delivery of Fe²⁺ or Fe³⁺, Ni(II) began to release, as discussed above.

3.4 Influence of anions on Ni(II) removal and release

Several studies have shown the effect of dissolved inorganic anions on the structure of Fe(II)–Fe(III) species and the reaction between SFH and metals.^20,34–36 It is proposed that CO₃²⁻ was incorporated in the structure of Fe precipitates, PO₄³⁻ prevented the formation of crystalline Fe oxide phase and NO₃⁻ consumed Fe(II) during reaction. In this study, the effects of above three anions on Ni(II) removal were investigated (Fig. 8).

Fig. 8 Effects of NO₃⁻, CO₃²⁻ and PO₄³⁻ on Ni²⁺ sequestration by SFH(1 : 2) (initial Ni²⁺ = 1.0 mM, SFH = 2.0 mM. Note that anions were first mixed with SFH for 2 min and then the mixture was applied for Ni²⁺ removal).

Fig. 8a showed the effect of NO₃⁻ on the reaction between SFH(1 : 2) and Ni(II) with [NO₃⁻]/[Fe(II)] of 1/10, 1/1, 10/1. It is found that NO₃⁻ exerted an obvious inhibition on Ni(II) removal and more Ni(II) was released into the solution. And the results were similar to that in the presence of O₂ (Fig. S3†), with which SFH could be oxidized quickly. On the basis of current conditions, we determined E(Ni²⁺/Ni) and E(NO₃⁻/NO) to be −0.398 V and ≈0.216 V, respectively (Text S2†). Besides, Fe₃O₄ was found in the reaction of SFH with NO₃⁻ (Fig. S4†). These results indicated that NO₃⁻ would be easier to be reduced by SFH than Ni(II), although NO₃⁻ and Ni(II) were present in this system. The presence of NO₃⁻ leads to the consumption of Fe(II) and inhibits Ni(II) reduction. Besides, the reaction between Fe(II) and NO₃⁻ generates H⁺. So the oxidation of Fe(II) and subsequent decrease of pH might be the major factors for inefficient Ni(II) removal and fast Ni(II) release.

CO₃²⁻ and PO₄³⁻ was found to shown entirely different influence on Ni(II) removal. CO₃²⁻ could enhance the Ni(II) sequestration efficiency at all three [CO₃²⁻]/[Fe(II)] ratios (Fig. 8b). While PO₄³⁻ acted as an inhibitor at a lower concentration ([PO₄³⁻]/[Fe(II)] = 1 : 10) (Fig. 8c). Previous study found that the number of OH⁻ ions per Fe atom introduced decreased from 2 in the absence of P to about 1.5 ± 0.1 in the presence of P (P/Fe = 0.86). It is inferred that PO₄³⁻ decreased the content of ≡Fe(II) and subsequent Ni(II) removal efficiency by SFH. However, PO₄³⁻ acted as an accelerator at [PO₄³⁻]/[Fe(II)] = 1 : 1 and 10 : 1 as a result of high pH (9.6 and 10.9, respectively). Due to the positive effect of SFH–PO₄³⁻ and SFH–CO₃²⁻ on pH increase, Ni(II) sequestration by SFH was investigated under identical initial pH value, which was adjusted by NaOH, as shown in Table 1. Nearly complete sequestration of Ni²⁺ was observed under initial pH ∼9.5 at [anions]/[SFH] of 1 : 1, while 18.99 mg L⁻¹ Ni²⁺ was detected under the almost same pH adjusted by NaOH. Therefore, it was quite evident that CO₃²⁻ played a crucial role in enhancing Ni²⁺ sequestration efficiency, in addition to positive effect on pH increase. Previous study¹⁸ showed that the introduction of CO₃²⁻ resulted in the formation of layered hydroxide metal complex (Fe₆(OH)₁₂CO₃·2H₂O) and the presence of anionic interlayer in the complex could contribute to the electron transfer and presumably benefit the reduction of Ni(II). Moreover, with CO₃²⁻ as anionic interlayer, the layered hydroxide metal complex has high specific area, which is reported to be 110.4 m² g⁻¹ by Legrand et al.³⁷

Table 1 Comparison of SFH-anion with SFH–NaOH on Ni²⁺ removal under various pH conditions^a

[Anion]/[SFH]	Initial pH		Residual Ni^2+ (mg L^−1) (20 min)
	SFH–CO3^2−	SFH–PO4^3−	SFH–CO3^2−	SFH–PO4^3−	SFH–NaOH^a
a The pH values adjusted by NaOH without anions were 9.0, 9.5, 10.0 and 10.5,respectively.
0	9.0	9.0	12.21	12.21	12.21
1 : 10	9.2	9.1	3.09	25.65	9.72
1 : 1	9.9	9.6	0.27	0.35	4.83
10 : 1	10.5	10.9	0	0	00.68

---

It was found that approximately 100% Ni(II) could be sequestrated within the initial 10 min, followed by slight change of Ni(II) concentration. On the one hand, final pH of the suspensions was higher than 8.0, which inhibited Ni(II) release (Table S1†). On the other, the introduction of PO₄³⁻ and CO₃²⁻ inhibited the release of Ni(II) from solid phases. FT-IR spectra of solids was found to change significantly (Fig. S5†), indicating the probably formation of vivianite (Fe₃(PO₄)₂(H₂O)₈) and FeCO₃, respectively. Thus the concentration of Fe²⁺ and Fe³⁺ decreased with the increase of PO₄³⁻ and CO₃²⁻. Overall, not only the removal efficiency was promoted but also the release of Ni(II) was inhibited.

4. Conclusions

In conclusion, SFH is effective in Ni(II) sequestration and the role of structural Fe(II) is indispensable. In addition to the precipitation by OH⁻, multiple mechanisms involved in Ni(II) sequestration include reduction, adsorption, co-precipitation and substitution. Ni(II) release should be taken into consideration in an extended reaction. It is determined that substitution of Ni(II) by Fe²⁺ and Fe³⁺ in Ni(OH)₂ and Ni–Fe complex (e.g. Ni_xFe_(1−x)(OH)₂) is favorable. pH is a critical factor affecting both aqueous Ni(II) removal and precipitated Ni(II) release. When the pH is higher than 8.0, highest Ni(II) removal efficiency was acquired and maintained without Ni(II) release. However, considerable Ni(II) release was observed at pH <8.0 in a longer reaction time, due to the delivery of iron into solution. Therefore, shortening reaction time was another alternative in Ni(II) release controlling. In the presence of CO₃²⁻ and PO₄³⁻, not only the removal efficiency was promoted but also the release of Ni(II) was inhibited by impeding the delivery of Fe²⁺ or Fe³⁺ to Ni–Fe products. In addition, the fact that Ni(II) release could be controlled provides a practicable alternative of resource recycling in high Ni(II) concentration wastewater. This study presents a novel point of view on Ni(II) removal by SFH and the release of removed Ni(II), which should be taken into consideration for the design environmental remediation process.

Acknowledgements

This study was financially supported by the Natural Science Foundation of China (Grant No. 41572211), State Key Laboratory of Pollution Control and Resource Reuse Foundation(PCRRT16001), and the Fundamental Research Funds for the Central Universities.

References

1. C. E. Borba, R. Guirardello, E. A. Silva, M. T. Veit and C. R. G. Tavares, Biochem. Eng. J., 2006, 30, 184–191.
2. A. de Mello Ferreira, M. Marchesiello and P. X. Thivel, Sep. Purif. Technol., 2013, 107, 109–117.
3. A. Miller, T. Wildeman and L. Figueroa, Appl. Geochem., 2013, 37, 57–63.
4. R. Gill, A. Mahmood and R. Nazir, J. Mater. Cycles Waste Manage., 2013, 15, 115–121.
5. B. Tanhaei, M. Pourafshari Chenar, N. Saghatoleslami, M. Hesampour, T. Laakso, M. Kallioinen, M. Sillanpää and M. Mänttäri, Sep. Purif. Technol., 2014, 124, 26–35.
6. N. Melitas, O. Chuffe-Moscoso and J. Farrell, Environ. Sci. Technol., 2001, 35, 3948–3953.
7. T. Satapanajaru, P. J. Shea, S. D. Comfort and Y. Roh, Environ. Sci. Technol., 2003, 37, 5219–5227.
8. X. Q. Li and W. X. Zhang, Langmuir, 2006, 22, 4638–4642.
9. M. J. Alowitz and M. M. Scherer, Environ. Sci. Technol., 2002, 36, 299–306.
10. V. Nahuel Montesinos, N. Quici, E. Beatriz Halac, A. G. Leyva, G. Custo, S. Bengio, G. Zampieri and M. I. Litter, Chem. Eng. J., 2014, 244, 569–575.
11. S. Tang, X. M. Wang, Y. Q. Mao, Y. Zhao, H. W. Yang and Y. F. Xie, Water Res., 2015, 73, 342–352.
12. Y. Sun, X. Guan, J. Wang, X. Meng, C. Xu and G. Zhou, Environ. Sci. Technol., 2014, 48, 6850–6858.
13. C. Noubactep, Chemosphere, 2016, 153, 528–530.
14. X. Guan, Y. Sun, H. Qin, J. Li, I. M. Lo, D. He and H. Dong, Water Res., 2015, 75, 224–248.
15. A. F. White and M. L. Peterson, Geochim. Cosmochim. Acta, 1996, 60, 3799–3814.
16. C. Noubactep, Korean J. Chem. Eng., 2012, 29, 1050–1056.
17. D. L. Bond and S. Fendorf, Environ. Sci. Technol., 2003, 37, 2750–2757.
18. D. L. Wu, B. B. Shao, Y. Feng and L. M. Ma, Environ. Technol., 2014, 36, 901–908.
19. D. L. Wu, B. B. Shao, M. Y. Fu, C. Luo and Z. G. Liu, Chem. Eng. J., 2015, 279, 149–155.
20. H. P. He, D. L. Wu, Q. M. Wang, C. Luo and N. Duan, Chem. Eng. J., 2016, 283, 948–955.
21. H. P. He, D. L. Wu, L. H. Zhao, C. Luo, C. M. Dai and Y. L. Zhang, J. Hazard. Mater., 2016, 309, 116–125.
22. C. M. Dai, Z. Zhou, X. F. Zhou and Y. L. Zhang, Water, Air, Soil Pollut., 2014, 225, 1799.
23. Y. L. Zhang, Y. M. Su, X. F. Zhou, C. M. Dai and A. A. Keller, J. Hazard. Mater., 2013, 263, 685–693.
24. B. Calderon and A. Fullana, Water Res., 2015, 83, 1–9.
25. C. Belebchouche, K. Moussaceb, A. Tahakourt and A. Aït-Mokhtar, Mater. Struct., 2014, 48, 2323–2338.
26. A. E. Schellenger and P. Larese-Casanova, Environ. Sci. Technol., 2013, 47, 6254–6262.
27. T. Iwasaki, H. Yoshii, H. Nakamura and S. Watano, Appl. Clay Sci., 2012, 58, 120–124.
28. G. Fan, X. Xiang, J. Fan and F. Li, J. Mater. Chem., 2010, 20, 7378.
29. F. Millange, R. I. Walton and D. ÓHare, J. Mater. Chem., 2000, 10, 1713–1720.
30. I. Giannopoulou and D. Panias, Hydrometallurgy, 2008, 90, 137–146.
31. K. Rozov, U. Berner, C. Taviot-Gueho, F. Leroux, G. Renaudin, D. Kulik and L. W. Diamond, Cem. Concr. Res., 2010, 40, 1248–1254.
32. P. Keller and H.-H. Strehblow, Corros. Sci., 2004, 46, 1939–1952.
33. Y. Jie, H. Fan, J. R. Niskala and W. You, Colloids Surf., A, 2013, 434, 194–199.
34. X. Châtellier, M. Grybos, M. Abdelmoula, K. M. Kemner, G. G. Leppard, C. Mustin, M. M. West and D. Paktunc, Appl. Geochem., 2013, 35, 325–339.
35. F. Bocher, A. Géhin, C. Ruby, J. Ghanbaja, M. Abdelmoula and J.-M. R. Génin, Solid State Sci., 2004, 6, 117–124.
36. K. Barthelemy, S. Naille, C. Despas, C. Ruby and M. Mallet, J. Colloid Interface Sci., 2012, 384, 121–127.
37. L. Legrand, A. EL Figuigui, F. Mercier and A. Chausse, Environ. Sci. Technol., 2004, 38, 4587–4595.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra16299a

---