Adsorption and degradation of Congo red on a jarosite-type compound

Abstract

Natrojarosite particles were prepared by forced hydrolysis. X-ray diffraction and field-emission scanning electron microscopy were used to characterize the resulting products. Degradation of the azo dye Congo red (CR) by natrojarosite was investigated under various conditions, such as in the presence or absence of visible-light irradiation, catalyst loading, H₂O₂ concentration, and initial pH. Total organic carbon determination, UV-visible spectroscopy, and direct infusion-electrospray ionization mass spectrometry in the negative ion mode provided insight into the nature of the degradation products. Moreover, a complete degradation mechanism of CR on natrojarosite was presented. The degradation of CR in the current system occurred even at neutral pH, and the total degradation rate was close to 99.1% for a 30 mg L⁻¹ CR solution. Approximately 80% of the samples were completely mineralized and the other 20% were degraded to small-molecule products. The novel natrojarosite catalysts are potentially valuable for industrial applications because of their high activity, low iron leaching, and low cost.

---

1. Introduction

Jarosite-type compound, a member of the jarosite–alunite group of isostructural minerals, is described by the formula AFe₃(SO₄)₂(OH)₆, where A is usually Na⁺, K⁺, NH₄⁺ or H₃O⁺. This compound is a secondary iron sulfate mineral often found in acid rock or acid mine drainage environments, particularly in mining wastes from polymetallic sulfide ore deposits.¹ The mineral is exploited in many fields, such as physics,^2,3 hydrometallurgy,^4,5 geochemistry,^6,7 and environmental sciences.^8–12 Physicists usually use jarosite-type compound as a model system to investigate the magnetic properties of the Kagomé lattice.^2,3 In the hydrometallurgical field, jarosite is the product of acid-yielding precipitation reaction, and its presence importantly influence the regulation of pH and dissolved ferric iron in aqueous environments.^4,5 Therefore, jarosite is widely used as an iron scavenger in the hydrometallurgical process.¹² To compensate for the high cost of iron removal by jarosite precipitation, researchers and engineers focused on the use of jarosite precipitates as raw material in producing pigments and construction materials.^10,13 Moreover, both jarosite and schwertmannite are iron-oxyhydroxysulfate materials. These materials are increasingly attracting research interest in the geological fields,^6,14 especially because of their effective scavenging of heavy metals, production of low-toxicity ions, high biocompatibility, and substantial application potential in the environment.^8,11,12,15 As a constituent on jarosite surfaces, activated Fe(III) possibly helps facilitate Cr(VI) reduction by sulfide.¹ Adding jarosite significantly enhanced Cr(VI) reduction by small-molecular-weight organic acids under illumination by mimic solar light.¹¹ In a study by Grafe et al., jarosite appeared to be effective in sequestering Cu(II) and As(V) because of the presence of structural SO₄²⁻ ions.⁹ Meanwhile, other scholars reported the degradation of methyl orange in a photo-Fenton-like system using ammonium and hydronium jarosite as catalyst, and study results suggested the potential application of jarosite materials in wastewater treatment.¹²

Recently, we described a simple and rapid method for synthesizing trigonal antiprismatic natrojarosite-type compound particles by the forced hydrolysis of sulfuric acid pickling wastewater and clarified the possible formation mechanism.¹⁶ Elemental analysis of the as-prepared jarosite showed that the preparation process of jarosite can be simulated using FeSO₄·7H₂O and Fe₂(SO₄)₃·xH₂O (analytical-grade reagents) as raw materials. The main advantage of preparing natrojarosite is its tolerance of slightly acidic media, which then disregards the need for a neutralization step. Further study indicated that the as-prepared jarosite can catalyze the degradation of azo dye Congo red (CR) in the presence of H₂O₂ under visible light irradiation and even at neutral pH. In this regard, we formulated a natrojarosite/H₂O₂/visible light system for investigating the adsorption and degradation of the azo dye CR. Interestingly, compared with the traditional Fenton-like system, the adsorption and degradation of CR on natrojarosite exhibited some novel characteristics.

2. Experimental details

2.1. Materials

Analytically pure ferric sulfate (Fe₂(SO₄)₃·xH₂O), ferrous sulfate (FeSO₄·7H₂O), and sodium carbonate (Na₂CO₃) and distilled water were used. The ferric salt solution was filtered through a 0.22 μm Millipore filter to remove any particulate contaminants prior to use.

2.2. Synthesis

In the typical synthesis procedure, FeSO₄·7H₂O and Fe₂(SO₄)₃·xH₂O were dissolved in distilled water and then filtered to remove impurities. The total concentration of Fe²⁺ and Fe³⁺ was 0.8 mol L⁻¹, in which the ratio of n_Fe(III) to n_Fe(II) was 1 : 1. The pH was adjusted to 1.5 with Na₂CO₃ solution (2.0 mol L⁻¹). The resulting solution was stirred for 20 min at room temperature and then heated and maintained boiling and refluxing for 2 h. The precipitates were centrifuged, rinsed extensively with distilled water, and dried for 2 days at about 70–80 °C.

2.3. Sample characterization

X-ray diffraction (XRD) was carried out at room temperature using a D8ADVANCE diffractometer with CuKα radiation (λ = 0.15418 nm). Scanning electron microscopy (SEM, S-4800 SEM, Japan Hitachi Ltd.) images were obtained. Electrospray ionization mass spectrometry (ESI, AB SCIEX 3200Q-TRAP mass analyzer) was conducted at typical ESI conditions. Total organic carbon (TOC) was analyzed with a lquiTOC II analyzer (Elementar Analytik, Germany).

2.4. Adsorption and degradation experiments

The degradation experiments were conducted in a 250 mL photoreactor. The temperature inside the reactor was kept constant (20 °C) by circulating water within the jacket surrounding the reactor. Irradiation experiments were performed using a fluorescent lamp (10 W, λ > 400 nm, Philips, China), and the light intensity at the solution surface was 199 mW cm⁻². The scheme of the photoreactor is shown in Fig. S1.† Photocatalytic degradation of CR was carried out in a cylindrical water-jacketed glass reactor using a constant solution volume of 150 mL. The initial concentration of CR was 30 mg L⁻¹, and the dosage of natrojarosite was 2 g L⁻¹. During irradiation, the solution was stirred at a constant rate. The CR concentrations in the solution before and during the degradation were determined using a UV-visible (UV-vis) spectrophotometer at the maximum wavelength. In the degradation experiment, the pH of CR (30 mg L⁻¹) was 7.3, and the maximum wavelength of CR at pH 7.3 was 497 nm.

Natrojarosite (0.3 g) was added to 150 mL of CR solution (30 mg L⁻¹). The system was stirred at room temperature for 10 h. Then, the solution was centrifuged and the concentration was determined using a UV-vis spectrophotometer at the maximum absorbing wavelength. Natrojarosite was then rinsed extensively before and after adsorption with distilled water and dried for 1–2 days at about 70–80 °C.

3. Results and discussion

3.1. Sample characterization

Fig. 1 presents the XRD pattern and SEM image of the as-prepared sample. All diffraction peaks can be indexed to pure hexagonal natrojarosite (JCPDF 00-036-0425) (Fig. 1a). The narrow, sharp peaks suggest that the natrojarosite product was crystalline. No impurities can be detected from this pattern. The corresponding SEM image (Fig. 1b) shows that the product was compact.

Fig. 1 XRD pattern (a) and SEM image (b) of the as-prepared sample.

3.2. Selection of conditions for the photocatalytic degradation of CR

To investigate the catalytic activity of natrojarosite on CR degradation, five control experiments were designed. The first one was a blank experiment without natrojarosite under visible light (Fig. 2a). Results show that the CR concentration did not change with time, suggesting that the CR in solution was stable during the study period. In the second control experiment, 0.01 mol L⁻¹ H₂O₂ was added to the reaction system based on the first control experiment. CR (8.77%) was degraded over 4 h of irradiation. In the third control experiment, 2 g L⁻¹ of natrojarosite was added to the reaction system on the basis of the first control experiment. The degradation of 6.53% of CR was observed. In the fourth control experiment, 2 g L⁻¹ of natrojarosite and 0.01 mol L⁻¹ of H₂O₂ were added to the reaction system, and the reaction was conducted in the dark. In this case, about 23.36% of CR was degraded. In the fifth control experiment, when natrojarosite, H₂O₂, and irradiation of visible light were both introduced into the reaction system, about 46.62% of CR was degraded (Fig. 2a). The above-mentioned results indicate that although both natrojarosite and H₂O₂ can degrade CR, a higher degradation rate was achieved in the presence of both substances than individually, especially under visible light irradiation. This result suggests that these components can facilitate each other. As such, the coexistence system of natrojarosite, H₂O₂, and visible light was chosen for investigation in the present paper.

Fig. 2 (a) Changes of the degradation rate of CR with time under different conditions. (b) Changes of the degradation rate of CR at different pHs with time. (c) Effect of H₂O₂ addition on the degradation rate of CR. (d) Effect of dose of natrojarosite on the degradation rate of CR.

Subsequently, the changes in degradation rate with time under different pH values for 30 mg L⁻¹ CR were determined (Fig. 2b). Various iron oxides are often used in wastewater remediation because of the efficient Fenton-like reaction system. However, a high degradation rate for pollutants can be usually exhibited in an acidic pH range. When the pH is close to neutral, the degradation rate of pollutants becomes extremely low. Fig. 2b indicates the CR degradation at pH 3 is faster than at other pH values, whereas a high CR degradation rate can be observed at neutral pH. For example, the CR degradation rate reaches 94% at the normal pH (e.g., 6.85). Thus, when natrojarosite is used as catalyst to degrade pollutants, the pH of the reaction system does not require adjustment. As such, natrojarosite is highly beneficial for such application.

To ensure H₂O₂ addition in the current system, a set of degradation experiments were carried out under varying H₂O₂ concentrations from 2.5 mmol L⁻¹ to 15 mmol L⁻¹ (Fig. 2c). The CR degradation rate initially increased and then decreased with the increase in H₂O₂ concentration. When the H₂O₂ concentration was 10 mmol L⁻¹, the degradation rate peaked. The effect of natrojarosite dose on CR degradation was also investigated (Fig. 2d). When the dose of natrojarosite was 2.0 g L⁻¹, the degradation rate peaked. The reaction condition of the degradation system of CR was ascertained (Fig. 2). Particularly, CR degradation experiment was carried out under visible light irradiation and the CR and H₂O₂ concentrations were 30 mg L⁻¹ and 10 mmol L⁻¹, respectively. The natrojarosite dose was 2 g L⁻¹, and the solution pH was not adjusted.

Fig. 2a reveals that a higher degradation rate was achieved in the presence of both natrojarosite and H₂O₂. To compare the relative contribution of the two substances, the increase in the degradation rate with per mmol of natrojarosite catalyst or H₂O₂ was calculated based on the data in Fig. 2c and d. The degradation rate of CR increased 18.0% and 2.7% respectively for per mmol increase of natrojarosite or H₂O₂, indicating that the natrojarosite catalyst contributed more to the improved degradation rate than H₂O₂.

Moreover, the as-prepared natrojarosite was subjected to successive degradation cycles under the optimal condition. Results indicate that the recovery degradation efficiency was more than 95% after three regenerations, suggesting that the as-prepared sample holds good suitability and stability.

3.3. Degradation mechanism of CR

To elucidate the degradation mechanism of CR on natrojarosite, the by-products formed after the CR degradation are monitored by different techniques. The digital photos and UV-vis spectra during the degradation process of CR are displayed in Fig. 3. Before degradation, the color of CR was red (Fig. 3a). At this time, CR shows that the maximum absorption bands were at ∼336 and ∼497 nm, respectively (Fig. 3b). As time progresses, the color of the system lightened, and the two absorption peaks weakened. In particular, the system exhibited a light-red color at 6 h and was completely discolored at 10 h. The peaks at both ∼336 and ∼497 nm disappeared after 10 h. This result implies that the structure of CR was destroyed by the powerful oxidation ability of natrojarosite/H₂O₂/visible light. According to the data obtained at ∼497 nm, the degradation rate reached 99.1% at 10 h. To evaluate whether CR simply lost its stain, a total organic carbon (TOC) analysis was performed, and the result is shown in Fig. 4. We then found that the highest TOC degradation rate reached about 82.5%, and another 17.5% of CR was oxidized to other small-molecule products.

Fig. 3 Digital photos (a) and UV-vis absorption spectra (b) of the system at different times.

Fig. 4 The changes of degradation rate of CR with time. (a) UV-vis absorption spectra (b) TOC.

Electrospray ionization mass spectrometry in the negative ion mode (ESI(−)-MS) was used to characterize the by-products formed after CR degradation. The obtained results, and the structural formulas of the products are shown in Fig. 5. In Fig. 5a, for pure CR (sample of 0 h), the presence of the anion of m/z 697 was detected (deprotonated form of CR). For the sample obtained at 30 h, the anion was no longer detected, indicating that the dye was completely consumed. By contrast, other anions of m/z 149, 65.1, 75.2, 89.3, 95.1, 103.2, and 117.2 can be observed. Given these data, the possible products were inferred, and the results are listed in Table 1. The possible degradation path is shown in Fig. 6.

Fig. 5 ESI(−)-MS of CR, samples withdrawn at (a) 0 and (b) 30 h.

Table 1 The degradation products of CR in current system

m/z	Structural formula
61.1	CH3COOH
75.2	CH3CH2COOH
89.3	CH3CH2CH2COOH
95.1
103.2	CH3CH2CH2CH2COOH
117.2	CH3CH2CH2CH2CH2COOH
149	HOOCCH2CH2CH2CH2COOH

---

Fig. 6 A possible degradation path of CR in natrojarosite system.

In the beginning stage of degradation, the degradation rate determined by UV-vis spectrophotometer was larger than that determined by TOC. Meanwhile, the color of CR solution lightened (Fig. 3a) but the degradation rate calculated by TOC was less than that obtained by UV-vis spectroscopy (Fig. 4). This result implies that the –N=N– in the CR structure was destroyed first because the large π–π electron conjugation system, especially the –N=N– in the azo dye structure, is considered the basic structure entailed in azo dye color formation. Subsequently, the –C–C– bond linking two benzene rings was broken, and aminobenzene was generated. With time, the structures of both naphthalene and benzene rings were broken. Most of these groups (about 82.5%) were completely oxidized to form CO₂ and H₂O (Fig. 4), whereas other structures were partly oxidized to form some aliphatic acids (Fig. 5 and 6).

H₂O₂ decomposition rate in the presence of natrojarosite was determined to confirm the CR degradation mechanism in the current system. Fig. 7 shows that, similar to iron oxides (goethite, ferrihydrite, and hematite),^17–19 natrojarosite can catalyze H₂O₂ decomposition. The hydroxyl radicals (˙OH) produced in this process can oxidize CR, owing to its high oxidation potential.²⁰ The decomposition rate of H₂O₂ in the natrojarosite system is lower than that in the iron oxide system;^21,22 the superiority of the former lies in the fact that the degradation of pollutants occurs at a neutral pH, whereas pollutant degradation only occurs at acidic pH (e.g., pH 3) in the latter system.

Fig. 7 The catalytic decomposition of H₂O₂.

3.4. CR adsorption on natrojarosite

Usually, prior to irradiation, the suspension was magnetically stirred in the dark for about 1 h to establish the adsorption/desorption equilibrium between catalyst and pollutant.^22–25 However, in the current system, when this step was carried out, only about 6.5% of CR was adsorbed onto natrojarosite at 4 h. In fact, all the results in Fig. 2–5 indicate that CR was actually degraded in the system of H₂O₂/visible light. By chance, the adsorption reaction was carried out for about 10 h. The solid part of natrojarosite, which was naturally colored yellow, unexpectedly changed to black green. The pH of the adsorption system decreased from pH 6.85 to 4.06. After the CR-adsorbed natrojarosite was washed under ultrasonic conditions, the color remained black green. To further understand the adsorption behavior of CR onto natrojarosite, a time-dependence experiment was performed. In this experiment, 0.3 g of natrojarosite was added to 150 mL 30 mg L⁻¹ of CR solution. At given time intervals, about 8 mL aliquots of suspension were sampled and centrifuged to remove the natrojarosite. The CR concentration in the filtrate was analyzed by UV-vis spectroscopy (Fig. 8). At the same time, pH changes of the adsorption system with time were determined, and digital photos of the reaction system were obtained (Fig. 8).

Fig. 8 (a) UV-vis absorption spectra of the adsorption system at different times. (b) Changes of pH of the adsorption system with time. (c) Digital photos of the system at different time (a–g: 0 h, 2 h, 4 h, 5 h, 6 h, 7 h, 10 h).

In the beginning stage of adsorption reaction, two absorption peaks were obtained at 336 and 497 nm, respectively (Fig. 8a). With time, the intensity of the two peaks gradually weakened. At 4 h, the CR adsorption rate onto natrojarosite was only about 6.5%. At this time, the color of the solution was still red. At 5 h, the color of the solution became light blue. In the corresponding UV-vis spectrum, a new absorption peak at 646 nm appeared. At this time, the pH of the system was 4.5 (Fig. 8b), which means that the change in color of the reaction system did not result from the pH decrease. Actually, the color of CR solution was red (Fig. 8c). Natrojarosite was a yellow powder before adsorption, but the color became dark green after adsorption (Fig. S2a and S2b†). Although the compound was rinsed extensively with distilled water under ultrasonic exposure, the color did not change (Fig. S2c†), suggesting that the affinity between CR and natrojarosite was sufficiently large. If either red or yellow is one of the tricolor, then the mixture of two colors is well known to produce orange. Fig. S2a and S2b† suggest that the chromophore of CR (e.g., π–π conjugated system) was affected because of its adsorption onto natrojarosite. Perhaps, a new conjugated system was probably formed after adsorption. At 10 h, the solution became colorless, and the peaks at 336 and 497 nm disappeared (Fig. 8c), suggesting that all of CR molecules were adsorbed. To understand the change in natrojarosite before and after adsorption, XRD patterns were determined. At the same time, TopasP2-1 software was used to calculate the d-values of typical peaks (e.g., (113), (021), and (012)) of natrojarosite, and the parameters of crystal cell (a, b, and c). The results are shown in Fig. 9 and Table 2, respectively.

Fig. 9 XRD patterns natrojarosite (a) before adsorption and (b) after adsorption.

Table 2 The parameters of crystal cell and d-values of typical peaks of natrojarosite before and after adsorption

		Before adsorption	After adsorption
a (Å)		7.3434	7.3431
c (Å)		16.6590	16.6392
d (Å)	(113)	3.06263	3.06162
	(021)	3.12326	3.12327
	(012)	5.05457	5.05250

---

Results of XRD patterns indicated that all diffraction peaks can be indexed to pure hexagonal natrojarosite (JCPDF 00-036-0425); however, the intensities of the three main diffraction peaks ((021), (012), and (113)) became weaker in various degrees. Especially for (021) and (113) peaks, the relative strength reversed. The parameter of structure cell a almost did not change, whereas c decreased from 16.6590 to 16.6392 (Table 2). The d-values of the peaks (113) and (012) of natrojarosite decreased after adsorption. All above-mentioned results denote that the crystal structure of natrojarosite may change after adsorption reaction to some extent.

To date, synthetic routes to jarosites have exclusively relied on the precipitation of solids from hydrolyzed acidic solutions of sulfate anions and monovalent and trivalent cations. Under these conditions, the monovalent cations are susceptible to replacement by hydronium ions; the occupancy of the M³⁺ lattice sites is typically only 83–94%, which leads to the presence of the site defects.²⁶

The results in Fig. 9 and Table 2, as well as the analysis in literature,²⁶ suggest a reasonable explanation for the adsorption process. In the current system, the preparation of natrojarosite was completed in water solution, and the pH of the system was adjusted only by Na₂CO₃. Under these conditions, some hydronium ions entered inevitably into the crystal structure of natrojarosite. The CR used in the current system is a compound including sodium sulfonate. The results in Fig. 9 and Table 2 indicate that the as-prepared natrojarosite holds strong capture ability for Na⁺ ions. Thus, in the adsorption process of CR, the Na⁺ ions in CR can enter into the natrojarosite structure and replace some hydronium ions. When adsorption occurs, the pH decrease of the system supports the above-mentioned conclusion. The radius of Na⁺ is less than that of H₃O⁺; hence, the parameters of the crystal cell and d-values of typical peaks of natrojarosite decrease after adsorption. The change in color of natrojarosite powder from yellow to dark green probably means that a new conjugated system was formed after CR was adsorbed.

3.5. Repeat recognition for CR degradation on natrojarosite

To further understand the degradation of CR in the natrojarosite/H₂O₂/visible, a series of experiments was carried out. The pH changes with time were determined in the degradation process (Fig. 10). The digital photo of natrojarosite powder was obtained after the degradation reaction was completed (Fig. S3†). The proportion of various elements in natrojarosite before and after degradation was determined by EDS (Table S1†). The parameters of crystal cell and d-values of typical peaks of natrojarosite before and after adsorption were compared (Table S2†).

Fig. 10 The changes of pH with time in the degradation process.

Fig. 10 shows that the pH of the reaction system decreased with time, suggesting that some H₃O⁺ ions were released into solution. Fig. S3† reveals that the color of natrojarosite particles did not change after degradation completion, which suggests that the adsorbed CR molecules were degraded fully. In Table S1,† both quality and mole percentages of Na⁺ ions increased, indicating that some Na⁺ ions entered into the natrojarosite structure. Table S2† further supports this conclusion; in the CR degradation process, the Na⁺ ions in CR entered into the natrojarosite structure and replaced some hydronium ions. This phenomenon resulted in the contraction of the crystal cell of natrojarosite and reduction of interplanar spacing.

4. Conclusions

In summary, a system of natrojarosite/H₂O₂/visible light was built to investigate the adsorption and degradation of the azo dye CR. The results indicated that the degradation of CR in the current system occurred even at neutral pH, which is an advantage over the traditional Fenton-like system. For a 30 ppm solution of CR, 80% of CR was completely mineralized, and the other 20% was oxidized to small-molecule products. TOC, UV-vis spectroscopy, and direct infusion-ESI(−)-MS revealed the complete degradation mechanism of CR on natrojarosite. Comparing the degradation of CR in the current system with previous results under the use of iron oxides (e.g., goethite, hematite, and ferrihydrite) as catalyst showed three characteristics. (1) Both the adsorption of CR onto natrojarosite and the decomposition of H₂O₂ catalyzed by natrojarosite occur slowly, but the speeds of the two processes match and result in a high CR degradation rate. (2) The degradation reaction of CR in the natrojarosite system occurs even in a neutral-pH environment. (3) After degradation, Na⁺ ions in solution enter into the crystal structure of natrojarosite, which leads to contraction of crystal cell and reduction of interplanar spacing.

Acknowledgements

This work was supported by a grant from the Natural Science Foundation of China (21277040, 21477032) and Key Programs of Hebei Normal University (L2015Z03).

References

1. Z. H. Xu, B. Lu, J. Y. Wu, L. X. Zhou and Y. Q. Lan, Mater. Sci. Eng., C, 2013, 33, 3723–3729.
2. A. S. Wills, A. Harrison, C. Ritter and R. I. Smith, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 6156–6169.
3. T. Inami, M. Nishiyama, S. Maegawa and Y. Oka, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 12181–12186.
4. L. Castro, C. Garcia-Balboa, F. González, A. Ballester, M. L. Blazquez and J. A. Muñoz, Hydrometallurgy, 2013, 131–132, 29–33.
5. Y. W. Song, M. Wang, J. R. Liang and L. X. Zhou, Hydrometallurgy, 2014, 143, 23–27.
6. P. V. Burger, J. J. Papike, C. K. Shearer and J. M. Karner, Geochim. Cosmochim. Acta, 2009, 73, 3248–3259.
7. S. A. Welch, D. Kirste, A. G. Christy, F. R. Beavis and S. G. Beavis, Chem. Geol., 2008, 254, 73–86.
8. M. P. Asta, J. Cama, M. Martínez and J. Giménez, J. Hazard. Mater., 2009, 171, 965–972.
9. M. Grafe, D. A. Beattie, E. Smith, W. M. Skinner and B. Singh, J. Colloid Interface Sci., 2008, 322, 399–413.
10. P. Asokan, M. Saxena and S. R. Asolekar, J. Hazard. Mater., 2006, 137, 1589–1599.
11. Z. H. Xu, S. Y. Bai, J. R. Liang, L. X. Zhou and Y. Q. Lan, Mater. Sci. Eng., C, 2013, 33, 2192–2196.
12. Z. H. Xu, J. R. Liang and L. X. Zhou, J. Alloys Compd., 2013, 546, 112–118.
13. P. Asokan, M. Saxena and S. R. Asolekar, Mater. Charact., 2010, 61, 1342–1355.
14. J. J. Papike, J. M. Karner and C. K. Shearer, Geochim. Cosmochim. Acta, 2006, 70, 1309–1321.
15. Y. H. Liao, J. R. Liang and L. X. Zhou, Chemosphere, 2011, 83, 295–301.
16. X. Yang, M. Y. Zhu, F. F. Kang, S. S. Cao, R. F. Chen, H. Liu and Y. Wei, J. Cryst. Growth, 2015, 429, 49–55.
17. S. S. Lin and M. D. Gurol, Environ. Sci. Technol., 1998, 32, 1417–1423.
18. J. Filip, R. Zboril, O. Schneeweiss, J. Zeman, M. Cernik, P. Kvapil and M. Otyepka, Environ. Sci. Technol., 2007, 41, 4367–4374.
19. M. Hermanek, R. Zboril, I. Medrik, J. Pechousek and C. Gregor, J. Am. Chem. Soc., 2007, 129, 10929–10936.
20. W. R. Haeg and C. C. D. Yao, Environ. Sci. Technol., 1992, 26, 1005–1013.
21. H. Liu, X. L. Li, Y. Wang, X. Yang, Z. Zhen, R. F. Chen, D. L. Hou and Y. Wei, RSC Adv., 2014, 4, 11451–11458.
22. S. S. Cao, F. F. Kang, P. Li, R. F. Chen, H. Liu and Y. Wei, RSC Adv., 2015, 5, 66231–66238.
23. H. Liu, W. R. Cao, Y. Su, Y. Wang and X. H. Wang, Appl. Catal., B, 2012, 111–112, 271–279.
24. R. Ramakrishnan, S. Kalaivani, J. A. I. Joice and T. Sivakumar, Appl. Surf. Sci., 2012, 258, 2515–2521.
25. P. A. K. Reddy, B. Srinivas, P. Kala, V. D. Kumari and M. Subrahmanyam, Mater. Res. Bull., 2011, 46, 1766–1771.
26. D. Papoutsakis, D. Grohol and D. G. Nocera, J. Am. Chem. Soc., 2002, 124, 2647–2656.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19125h

---