High-dispersive FeS₂ on graphene oxide for effective degradation of 4-chlorophenol

Abstract

A high-dispersive FeS₂ micro-cube crystal on graphene oxide (FeS₂@GO) was fabricated by a one-pot hydrothermal method. The catalytic degradation of 4-chlorophenol (4-CP) and its mechanism in a FeS₂@GO-based Fenton system was investigated. Under acidic to slight alkaline conditions, FeS₂@GO demonstrated an excellent capacity to remove 4-CP. More than 97% of 4-CP was eliminated within 60 min in pH 7.0 reaction solutions initially containing 0.2 g L⁻¹ FeS₂@GO, 128.6 mg L⁻¹ 4-CP and 100 mM H₂O₂ at 25 ± 1 °C, and the removal of 4-CP was further enhanced with increasing FeS₂@GO loadings. In the meantime, the FeS₂@GO also achieved lower iron leaching and a more complete TOC removal compared with pure synthetic FeS₂ without graphene oxide. Furthermore, acetic acid and oxalic acid were identified as the primary products. The remarkable capacity of the FeS₂@GO-based Fenton system in removing 4-CP displays its potential application in the treatment of organic compound-contaminated water.

---

Introduction

Pyrite (FeS₂) is one of the most abundant metal sulfide minerals on Earth, possessing a degradation capacity for contaminants.¹ For example, degradation kinetics and the mechanism of aqueous trichloroethylene (TCE) in aerobic pyrite suspensions has been researched with O₂ as the common oxidant.^2–4 Liang et al.⁵ investigated oxidative degradation of methyl tert-butyl ether (MTBE) by activated persulfate, using pyrite as the source of ferrous ion activators. Recently, pyrite has been considered as a potential and promising heterogeneous iron source for Fenton-like systems to treat various environmental organic pollutants in wastewater and groundwater.^6–10 However, for the limitations of pool purity of natural pyrite, its Fenton-like catalytic capacity can't be further improved. It is known that the catalytic activity of catalysts depends on the size distributions and morphologies of the particles.¹¹ Hence, reducing the diameter of the material to the nanometer or micrometer scale may result in enhanced its efficacy. But, the difficulty in obtaining fine particles because of its high Mons' hardness scale is the other limitation for further improvement in efficiency.

The hydrothermal method was an alternative way for obtaining more pure and small particles. Up to now, FeS₂ nanocrystallines with different morphologies have been synthesized via various solventhermal methods. One-dimensional nanowires of FeS₂ were synthesized in large quantities by solvothermal process at relatively low temperature with different with morphologies by Kar and Chaudhuri,¹² and single phase FeS₂ nanocrystals with cubic shape were synthesized but in a relative complicated solventhermal process.¹³ Afterward well-defined FeS₂ micro-cubes and micro-octahedras with high-yield and good uniformity were synthesized by a more simple polymer-assisted hydrothermal method.^14,15 However, these FeS₂ particles are prone to aggregate and form large particles during the hydrothermal synthesis process, thus losing their dispersibility and specific area which eventually diminish their activity. Therefore, it is necessary to prepare high-dispersive FeS₂ on a suitable support to preserve or even improve their unique properties.^16–18

In the past decade, graphene and its derivatives are widely investigated as promising materials for the immobilization of nanoparticles. However, the lack of surface functionalities in graphene to directly immobilize the nanoparticles onto its surfaces has led to favorable utilization of graphene oxide (GO) as an alternative support for the assembly of graphene based nanocomposites.¹⁹ GO is fabricated by exfoliating of graphite oxide and is abundant of oxygenated functional groups, such as hydroxyl and epoxides on the plane with carbonyl and carboxyl groups at the edges. These oxygenated functional groups can serve as nucleation sites for metal ions to form GO/nanoparticles composites. As a result, GO used as an attractive material in this field owing to its unique two-dimensional lamellar structure, large surface area, and full surface accessibility.²⁰ The graphene or GO is not only able to prevent the aggregation of immobilized particles but also improve the overall catalytic activity owing to the synergistic effects between both components.^21,22 For example, several recent studies have been reported using GO for the support of Fe₃O₄ NPs in catalysis for the oxidation of cysteine²³ and 3,3,5,5-tetramethylbenzidine,²⁴ and the reduction of nitrobenzene.²⁵ Furthermore, the reported enhancement in catalytic activity was attributed to the synergistic effects between GO sheets and Fe₃O₄ nanoparticles.

The objective of the present work was to explore the degradation of 4-CP with FeS₂@GO by investigating removal efficacy and influencing factors such as FeS₂@GO loading, solution pH, and H₂O₂ concentration. The proposed mechanism was given and the principal reaction intermediates were also identified.

Experimental

1. Chemicals

The 4-chlorophenol (4-CP) standard (purity > 99%) was purchased from Aladdin Chemistry (Shanghai, China). Graphene oxide (purity > 99%, single layer ratio > 99%, diameter 1–5 μm, thickness 0.8–1.2 nm) was purchased from XFNano Inc. (Nanjing, China). Pyrite (purity 95%) was purchased from Strem Chemicals (Newburyport, MA, USA). Triton™ X-100 (TX-100), sulfur (purity 99.5–100.5%) and 2,9-dimethyl-1,10-phenanthroline (DMP) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Other chemicals and solvents used in this study were of analytical grade or high performance liquid chromatography (HPLC) grade. The concentration of the purchased hydrogen peroxide solution (30 wt%) was calibrated by titration with potassium permanganate²⁶ and the purchased pyrite was grounded, sieved through 150 μm mesh and washed with 0.1 M HCl prior to use. Other chemicals were used as received. Ultrapure water (18.2 MΩ cm resistivity) was prepared using a Millipore® purification system and used throughout the experiments. Stock solution of 1.0 M 4-CP was prepared in methanol. The 1% (w/v) DMP solution was prepared in ethanol in a brown bottle. A 0.01 M copper(II) sulfate (CuSO₄) solution was prepared by dissolving CuSO₄·5H₂O in ultrapure water. All solutions were stored at 4 °C prior to use.

2. Preparation of FeS₂ and FeS₂@GO

Iron disulfide microcube crystal (FeS₂) was synthesized using FeSO₄ and sulfur in alkaline solution based on Wang's method.¹⁴ Briefly, 16 mL of non-ionic surfactant TX-100 was added in 44 mL of ethylene glycol (EG) at room temperature followed by addition of 0.39 g of ferrous sulfate heptahydrate (FeSO₄·7H₂O), forming a homogenous solution under vigorous stirring. Then 0.4 g of sulfur was added to the solution under magnetic stirring for 1 h, after complete dispersion of sulfur powder, 10 mL of 1.5 M NaOH was added and stirred for 30 min. Then the final mixture was sealed in a 100 mL Teflon-lined stainless steel autoclave, and maintained at 180 °C for 12 h, then cooled to room temperature naturally. The resulting black solid was collected by centrifugation, washed alternately with ultrapure water and ethanol several times to remove the excess surfactant and finally dried in vacuum at 50 °C for 10 h before further use. Iron disulfide–graphene oxide composite (FeS₂@GO) was synthesized according to the above procedures with appropriate amount of GO (0.03, 0.06, 0.12 g) dispersed in EG by ultrasonication in advance.

3. Reaction setup

All degradation experiments of 4-CP were carried out in 50 mL glass flasks with a total solution volume of 40 mL under magnetic stirring (400 r min⁻¹) and 25 ± 1 °C. A 500 W xenon lamp (Trustech Inc., Beijing, China) was used as the light source for the experiments conducted in the presence of visible light. The light radiates to the solutions through a glass slide to obtain visible light without UV wave band. The initial pH (pH_i) of the solutions was adjusted to the designated value with 1 M H₂SO₄ and 1 M NaOH standard solution. A 40 μL aliquot of 1.0 M 4-CP stock solution was added to make a nominal initial concentration of 1.0 mM (128.6 mg L⁻¹) and appropriate volume of H₂O₂ solution was injected to make a demanded concentration. Reactions were initiated by adding a predetermined amount of pyrite, FeS₂ or FeS₂@GO into the pre-equilibrated and constantly stirred solutions. Aliquots of 1.0 mL sample were periodically withdrawn and filtrated through 0.45 μm syringe filter. The supernatant was transferred to 2 mL vials containing 10 μL of tert-butanol (as a radical scavenger) and subjected to HPLC analysis to determine the remaining 4-CP concentration and the formation of carbonyl acids. For the runs of cyclic reaction, the post-reaction FeS₂@GO was collected by filtration and dried in 50 °C. For TOC measurement, aliquots of 5 mL sample were periodically withdrawn and filtrated through 0.45 μm syringe filter, but no radical scavenger was added to avoid background TOC interference. All samples were stored at 4 °C and analyzed within 24 h.

To quantify the consumption of H₂O₂ and leaching of Fe ions in the reaction systems, an additional reaction solution without 4-CP was also prepared. For H₂O₂ consumption quantification, 100 μL samples were periodically withdrawn and filtrated through 0.45 μm syringe filter. Aliquots of 50 μL supernatant was transferred to 5 mL volumetric flask and diluted to 500 μL. The dilution was used for H₂O₂ measurement using a method according to Kasaka et al.²⁷ The determination of H₂O₂ is based on a spectrophotometric method via the stoichiometric reaction of H₂O₂ with copper(II) ion and DMP. Briefly, 0.5 mL each of DMP, CuSO₄ and phosphate buffer was added to a 5 mL volumetric flask containing 0.5 mL diluted sample supernatant and the flask was filled up to 5 mL with water. After mixing, the solution was transferred to 1 cm cells and the absorbance was measured at 454 nm on a Shimadzu UV-2401 PC UV/Vis spectrophotometer (Tokyo, Japan). While, for Fe ions leaching quantification, aliquots 2 mL sample were withdrawn and filtrated through 0.45 μm syringe filter. The supernatant was transferred to 5 mL vials containing 10 μL HNO₃ (65 wt%) and subjected to inductively coupled plasma mass spectrometry (ICP-MS) analysis to determine the Fe ions concentration.

4. Chemical analysis

The 4-chlorophenol and by-products concentrations in reaction samples were determined using a Waters® e2695 reverse-phase HPLC coupled with a Waters® 2998 photodiode array detector (Milford, MA, USA). The detection wavelength was both 210 nm. A Waters® XBridge™ Phenyl column (250 × 4.6 mm, 5 μm) was employed for the separation of 4-chlorophenol. The isocratic mobile phase consisted of 70% methanol and 30% water with a flow rate of 0.8 mL min⁻¹. The injection volume was 20 μL. Under these conditions, the typical retention time for 4-CP was 5.5 min. For the products such as carbonyl acids, a Waters® Atlantis™ T3 column (250 × 4.6 mm, 5 μm) was employed for the separation. The isocratic mobile phase was 10 mM NaH₂PO₄ aqueous solution (adjusting pH to 3.0 with H₃PO₄) with a flow rate of 0.5 mL min⁻¹. The injection volume was 50 μL. Under these conditions, the typical retention time for acetic acid, oxalic acid and chloroacetic were 6.1, 12.7 and 14.1 min, respectively.

5. Characterization

Surface morphology studies of FeS₂ and FeS₂@GO were achieved with a field emission scanning electron microscope (FEI, SIRON) at a voltage of 25.0 kV to test. The sample surfaces were gold-coated before analysis. The microstructures were further examined by transmission electron microscopy (TEM, JEM1200EX, JEOL). X-ray diffraction (XRD) measurements were conducted using a D8 Advance (Bruker, Germany) X-ray diffractometer with Cu Kα radiation (λ = 1.5418 Å).

Results and discussion

1. Characterization of FeS₂ and FeS₂@GO

The morphology and microstructures of synthetic FeS₂ and FeS₂@GO were investigated by a scanning electron microscope (SEM). Fig. 1a and b showed the SEM images of FeS₂ and FeS₂@GO, respectively. It illustrates that the shape of synthetic FeS₂ is irregular, and it is unhomologous in size. Besides, the FeS₂ crystal aggregates as clusters, making an obscure microcube structure and a low surface area (3.48 m² g⁻¹). On the other hand, GO facilitates the dispersion of FeS₂ during the synthesis process. The FeS₂ synthesized with GO sheets spiking was found to be single crystal with distinct microcubic structures. It is also homologous in size (800–1000 nm), which has a much larger surface area (23.26 m² g⁻¹) compared with FeS₂ synthesized without GO. The microstructures of FeS₂ and FeS₂@GO were further examined by transmission electron microscopy (TEM) images (Fig. 1c and d) and drawn a same result. Furthermore, the composition and phase purity of natural pyrite and synthetic FeS₂ were examined by XRD (Fig. 2). It can be seen that all reflection peaks in the red line can be readily indexed as a pure cubic phase of FeS₂, which is consistent with value given in Joint Committee on Powder Diffraction Standards (JCPDS) diffraction data files (no. 71-2219). However, a lower purity of pyrite can be easily observed in the black line with relatively poor clearness in reflection peaks.

Fig. 1 SEM image of FeS₂ (a) and FeS₂@GO (b), and TEM images of FeS₂ (c) and FeS₂@GO (d).

Fig. 2 X-ray diffraction patterns of natural pyrite and synthetic FeS₂.

2. Removal of 4-chlorophenol by FeS₂@GO

A typical degradation profile of 4-CP by FeS₂ and FeS₂@GO is given in Fig. 3a. For verifying and comparing the capacity of iron disulfide in organics removal, commercially available natural pyrite crystals (the major ingredient is iron disulfide) were also employed for the removal of 4-CP. According to the results, FeS₂@GO possesses a fairly high capacity in removing 4-CP in H₂O₂ solutions compared to FeS₂ particles and pyrite, both the synthesis and none-synthesis iron disulfide. For example, 86% of 4-CP was removed in 30 min in a pH 5.0 reaction system originally containing 0.8 g L⁻¹ FeS₂@GO, 50 mM H₂O₂ and 1.0 mM 4-CP, while only 58% and 19% 4-CP was removed by FeS₂ and pyrite under the same conditions, respectively. Up to 99% of 4-CP disappeared as the reaction was prolonged to 60 min as well as 97% for FeS₂. However, the inevitable aggregation of FeS₂ during the reaction tends to inhibit the removal rate and can't be ignored. As a result, natural pyrite have the lowest efficiency for 4-CP removal as compared to FeS₂ and FeS₂@GO. And FeS₂@GO possesses the most remarkable dispersibility as well as 4-CP removal efficiency in the reaction process. To study the effect of GO content on the efficiency of FeS₂@GO for removal of 4-CP and fix an appropriate GO amount during the synthesis process, FeS₂@GO with different GO content was employed for 4-CP removal. As shown in Fig. 3b, efficacy of 4-CP removal by FeS₂@GO in H₂O₂ solutions was slightly improved as the GO content of FeS₂@GO increased. Take both economy and efficiency into consideration, 0.03 g was chosen as the optimal amount for spiking in FeS₂@GO synthesis, and was employed throughout all experiments below.

Fig. 3 Time profiles of 4-chlorophenol (4-CP) removal by pyrite, FeS₂ FeS₂@GO (a), and time profiles of 4-CP removal by FeS₂@GO with different GO content (b) under the conditions of pH 5.0, 25 ± 1 °C, [4-CP] = 128.6 mg L⁻¹ (1.0 mM), [H₂O₂] = 50 mM and [catalyst] = 0.8 g L⁻¹. Data points are given as means ± standard deviations (n = 3).

3. Leaching of iron

The level on the iron leaching is another essential factor to evaluate the Fenton-like catalytic performance. The aqueous iron content in the FeS₂–H₂O₂ or FeS₂@GO–H₂O₂ suspension was detected by inductively coupled plasma mass spectrometry (ICP-MS) during a 2 h reaction. According to the results, the leaching rate of FeS₂ was a little faster than that of FeS₂@GO (Fig. 4a). For example, after 120 min reaction, the concentration of aqueous iron was 5.4 mg L⁻¹ for FeS₂, while it was 4.5 mg L⁻¹ for FeS₂@GO, equaling to only 0.81% and 0.68% of the initial iron content. Meanwhile, the total iron content in FeS₂@GO was 98.9% of that in FeS₂. It is reported that the charges on GO surface are highly negative when dispersed in water attribute to the ionization of the carboxylic acid and the phenolic hydroxyl groups.²⁸ Hence, the most plausible reason was that Fe³⁺/Fe²⁺ was adsorbed on the GO surface through static interaction or complex with oxygenated functional groups.²⁹ But the leaching iron makes fairly finite contribution to 4-CP removal. As shown in Fig. 4b, only 42% of 4-CP was eliminated within 60 min in a Fenton system with 5.4 mg L⁻¹ Fe²⁺ and 50 mM H₂O₂.

Fig. 4 Iron leaching of FeS₂ and FeS₂@GO during the reaction with 4-hlorophenol (4-CP) (a) and time profiles of 4-CP in classic Fenton system (b) under the conditions of pH 5.0, 25 ± 1 °C, [4-CP] = 128.6 mg L⁻¹ (1.0 mM), [H₂O₂] = 50 mM, [catalyst] = 0.8 g L⁻¹ and [Fe²⁺] = 5.4 mg L⁻¹. Data for classic Fenton treatment are from single measurement, while the other data points are given as means ± standard deviations (n = 3).

4. Effect of FeS₂@GO loading, pH and H₂O₂ concentration

The removal rate of 4-CP increased with the initial FeS₂@GO content (Fig. 5a). For example, 4-CP removal at 60 min increased from 76% for a solution with 0.2 g L⁻¹ FeS₂@GO to 100% with 0.8 g L⁻¹ FeS₂@GO. Two factors are responsible for the enhanced removal of 4-CP with increasing loading. It is reported by Bae et al.⁷ the hydroxyl radicals (˙OH), which are the dominantly oxidant for the decomposition of organic compounds in pyrite Fenton reaction, were generated through reduction of H₂O₂ by Fe²⁺ dissolved from FeS₂ under aerobic condition (eqn (1)).

2FeS₂ + 7O₂ + 2H₂O → 2Fe²⁺ + 4SO₄²⁻ + 4H⁺ (1)

Fig. 5 Effect of FeS₂@GO loading (a), initial pH (b) and H₂O₂ concentration (c) on 4-chlorophenol (4-CP) removal at 25 ± 1 °C in pH 5.0 reaction solutions initially containing 128.6 mg L⁻¹ (1.0 mM) 4-CP and 50 mM H₂O₂ and 0.8 g L⁻¹ FeS₂@GO. Data for FeS₂@GO-only treatment are from single measurements, whereas the other data points are given as means ± standard deviations (n = 3).

The removal of 4-CP in the FeS₂@GO-based Fenton system was obviously pH-dependent. Overall, 4-CP removal rates decreased with increasing pH_i (Fig. 5b), suggesting that acidic conditions facilitated 4-CP removal. For example, 4-CP removal rate at 30 min decreased from 86% for the reaction solution with pH_i 5.0% to 59% for that with pH_i 9.0. When pH_i was increased to 11.0, 4-CP removal rate was inhibited to a relatively low value of 21%, which is attribute to the decreasing formation of ˙OH caused by low Fe²⁺ concentration at high pH.¹ Similarly, pH was a major factor that affected the removal of other organics (e.g. trichloroethylene and 2,4,6-trinitrotoluene) in pyrite Fenton systems.^7,8 However, it is well known that in a classic Fenton system aqueous Fe(II) leading to decomposition of H₂O₂ to hydroxyl radicals was only efficiently occurred at the pH below 4.5.³⁰ In the present work, the FeS₂@GO-mediated removal of 4-CP stayed in a high rate in a slight acidic to alkaline conditions (pH 5.0–9.0), even in a strong alkaline solution (pH 11.0), 49% of 4-CP was removed in 60 min yet. As reported by Bonnissel-Gissinger,¹ pyrite suspension pH declined and reached an acidic equilibrium point during the oxidation by releasing iron and hydrogen, providing an appropriate pH for the subsequent Fenton reaction. Additionally, the presence of H₂O₂ extremely exacerbated pH decreasing, since pyrite dissolution kinetics by H₂O₂ is much faster than that by molecular oxygen in suspensions,^31,32 which leaded to a relatively high removal efficiency for 4-CP under weak acidic to weak alkaline conditions (pH 5.0 to pH 9.0). The similar result was observed by Che et al.⁷ and Bae et al.⁹ in the removal of trichloroethylene and diclofenac, respectively, by pyrite-based Fenton's reaction. Furthermore, the numerous carboxyl groups (–COOH) and hydroxyl groups (–OH) on GO surface also provided an acidic interface for the occurrence of Fenton's reaction.³³

It is well know that in a Fenton system H₂O₂ simultaneously plays as the roles of both of OH˙ producer (eqn (2)) and OH˙ scavenger (eqn (3)).^34,35 Therefore, the concentration of H₂O₂ is a crucial parameter for 4-CP removal in FeS₂@GO Fenton system.

H₂O₂ + Fe²⁺ → Fe³⁺ + OH⁻ + HO˙ (2)

H₂O₂ + OH˙ → OH₂˙ + H₂O (3)

As Fig. 5c showed that the removal of 4-CP was increased with the increasing concentration of H₂O₂ in the FeS₂@GO Fenton system. The 4-CP removal rates declined as the concentration of H₂O₂ varies from 12–50 mM. For example, 4-CP removal rate at 30 min decreased from 17% at the H₂O₂ concentration of 12 mM to 56% and 86% at that of 25 mM and 50 mM, respectively. With the reaction proceeding to 60 min, 4-CP removal rate was respectively increased to 39%, 93% and 100% for H₂O₂ concentrations of 12 mM, 25 mM and 50 mM. As McKibben and Barnes³¹ reported, the general mechanism of Fe²⁺ dissolution from the pyrite surface in an aerobic H₂O₂ solution includes two different pathways (eqn (1), (4) and (5)) and the second pathway with two consecutive reactions is considered as the dominant way for Fe²⁺ generation.

2FeS₂ + 15H₂O₂ → 2Fe³⁺ + 4SO₄²⁻ +2H⁺ + 14H₂O (4)

FeS₂ + 14Fe³⁺ + 8H₂O → 15Fe²⁺ + 2SO₄²⁻ + 16H⁺ (5)

Therefore, the increasing of H₂O₂ concentration accelerates the 4-CP removal in the FeS₂@GO Fenton system. However, the self scavenging of HO˙ caused by H₂O₂ concentration increasing seemed to be more significant during the reaction. As a result, further addition of H₂O₂ to 200 mM did not significantly affect 4-CP removal rates (data was not shown) and H₂O₂ concentration was fixed at an optimal level in the following experiments.

5. The reuse of FeS₂@GO

The reuse and stability were two major concerns for a catalyst to be used in practical applications.^36–38 In the durability test, the same FeS₂@GO samples were repeatedly employed for 4-CP removal for consecutive five times. It showed that there were no obvious attenuation in the extent of 4-CP removal was observed (Fig. 6). However, a slight decline inevitably occurred for the loss of catalyst during the reaction and recovery procedures. For example, the removal rate of 4-CP was over 99% for the initial reaction, but it slightly dropped to about 95% for the fifth cycling. It is worth noticing that the recovery of FeS₂@GO is difficult for its high dispersibility in the reaction solutions. Therefore, further research would still need to be conducted to generate enhanced separation properties for FeS₂@GO.

Fig. 6 Reuse of FeS₂@GO at 25 ± 1 °C in pH 5.0 reaction solutions initially containing 128.6 mg L⁻¹ (1.0 mM) 4-CP, 50 mM H₂O₂ and 0.8 g L⁻¹ FeS₂@GO. (a–e) show 4-CP residue in five degradation reactions. Data points are from single measurement.

6. Identified products

It is well known that organic compounds were expected to oxidize to short-chain carboxylic acids (e.g., formic acid, acetic acid, oxalic acid and maleic acid) and then finally minerals, CO₂ and H₂O.³⁹ In this study, three carboxylic acids including oxalic acid, acetic acid and chloroacetic acid were identified as the primary intermediates through HPLC-DAD analysis and the time profile of the intermediates were given in Fig. 7. It is shown that oxalic acid and acetic acid accumulated in the early stage, and then attenuated after reaching a peak value. For example, oxalic acid accumulated to a peak concentration of 0.35 mM at 5 min and then declined, no oxalic acid was detected after 30 min. Similarly, acetic acid accumulated to a peak concentration of 0.89 mM at 30 min and was not detected after 60 min. However, the same course was not observed on chloroacetic acid. It kept on a relative low level compared with oxalic acid and acetic acid. To investigate the mineralization of 4-CP, concentrations of TOC were measured during the oxidative degradation of 4-CP by FeS₂@GO Fenton system. It is shown in Fig. 8 that after 60 min reaction 91%, 70% and 12% of TOC was removed for FeS₂@GO, FeS₂ and pyrite, respectively.

Fig. 7 Formation of intermediates oxalic acid, acetic acid and chloroacetic acid.

Fig. 8 Removal of TOC at 25 ± 1 °C in pH 5.0 reaction solutions initially containing 128.6 mg L⁻¹ (1.0 mM) 4-CP, 50 mM H₂O₂ and 0.8 g L⁻¹ FeS₂@GO (pyrite or FeS₂). Data points are from single measurement.

7. Proposed mechanism

Previous study has proved that FeS₂ has a higher optical absorption coefficient than silicon but has a comparable band gap energy (E_g = 0.95 eV).⁴⁰ Additionally, single layered GO have a high optical transmittance, especially for visible light.⁴¹ In our experiments, the removal process of 4-CP was exposed to indoor natural light, so we conducted a control experiment in the dark. As a result, no obvious diversity on 4-CP removal was observed under natural light and dark condition, indicating that FeS₂@GO–H₂O₂ system also works well in dark. Furthermore, for the photo-oxidation efficiency of photocatalysts was found to generally depend on light intensity,⁴² the intense light source xenon lamp with a UV-light filter was employed in the experiment for 4-CP removal. Similarly, the concentration of 4-CP declined the same as in dark (Fig. 9). Therefore, it is considered that the FeS₂@GO–H₂O₂ system without the photo-catalytic process would avoid the drawbacks found in light sensitive system.

Fig. 9 Effect of visible light on 4-chlorophenol (4-CP) at 25 ± 1 °C in pH 5.0 reaction solutions initially containing 128.6 mg L⁻¹ (1.0 mM) 4-CP, 50 mM H₂O₂, and 0.8 g L⁻¹ FeS₂@GO. Data for visible light treatment are from single measurement, while the other data points are given as means ± standard deviations (n = 3).

According to the previous studies,^4,7,8,10 the Fenton reaction stimulated by the dissolved Fe²⁺ from pyrite surface is the dominant mechanism for 4-CP removal. Overall, H₂O₂ consumption rates in different reaction systems followed the order of pyrite > FeS₂@GO > FeS₂ (Fig. 10), because GO highly facilitated dispersion of FeS₂ and provided a huge interface for the transformation of H₂O₂, enhancing the 4-CP removal.

Fig. 10 Hydrogen peroxide consumption by pyrite, FeS₂ and FeS₂@GO at 25 ± 1 °C in pH 5.0 reaction solutions initially containing 50 mM H₂O₂ and 0.8 g L⁻¹ FeS₂@GO (pyrite or FeS₂). Data points are given as means ± standard deviations (n = 3).

The role of H₂O₂ in the removal of 4-CP was precursor of HO˙. Because H₂O₂ itself is not an efficient oxidizer of 4-CP, HO˙ generated from H₂O₂ are the actual oxidant in the FeS₂ and FeS₂@GO Fenton systems. It is showed in Fig. 11 that addition of methanol as a HO˙ scavenger substantially retarded the removal of 4-CP both in FeS₂ and FeS₂@GO Fenton systems. Besides, it is worth to notice that in an anaerobic reaction system 4-CP removal was also inhibited to a certain degree (Fig. 11). Thus, it is believed oxygen played an indispensable role in the FeS₂@GO Fenton system for 4-CP removal. First, it is proved that the transformation of halogenated compounds in pyrite suspension was consistent with the finding that pyrite oxidation by O₂ follows a Fenton-like mechanism, in which the reduction of O₂ on the pyrite surface can induce HO˙ formation.^43–48 On the other hand, the dissolved O₂ in reaction solutions oxide the FeS₂@GO surface, releasing Fe²⁺ for the further Fenton reaction (eqn (3)). Therefore, dissolved O₂ is an essential factor in the FeS₂@GO Fenton system. In addition, it is considered that the surface disulfide groups have been the proposed electron donor in reductive dehalogenation.⁴⁹

Fig. 11 Removal of 4-chlorophenol (4-CP) at 25 ± 1 °C in pH 5.0 reaction solutions in the presence of 2.5% wt. methanol (˙OH radical scavenger) and in the absent of O₂ initially containing 128.6 mg L⁻¹ 4-CP, 50 mM H₂O₂ and 0.8 g L⁻¹ FeS₂@GO (or FeS₂). Data for anaerobic treatment are from single measurement, while the other data points are given as means ± standard deviations (n = 3).

Conclusions

This study demonstrated that high-dispersive FeS₂ on GO possesses a relatively high oxidative capacity for removing aqueous 4-CP in H₂O₂ solution, offering a good alternative for the treatment of aqueous organic contaminants. The continuous dissolution of Fe²⁺ from FeS₂@GO surface is considered as the key factor to achieve the oxidation of 4-CP by continuously producing the ˙OH. And oxygen played an indispensable role in the FeS₂@GO Fenton system for 4-CP removal by facilitating the dissolution of Fe²⁺. Besides, no distinguishable diversity on removal of 4-CP was observed in presence and absence of visible light. The results obtained from this study can provide basic knowledge to understand the oxidative degradation mechanism of 4-CP by high-dispersive FeS₂ in H₂O₂ solution and to design a high-dispersive and efficient heterogeneous Fenton catalyst for treatment of organic-contaminated water.

Acknowledgements

The authors acknowledge financial support from “The Opening Foundation of the Environmental Engineering Key Discipline”, the Fundamental Research Funds for the Central Universities and the Zhejiang Provincial Natural Science Foundation of China (no. LY12B05001).

Notes and references

1. P. Bonnissel-Gissinger, M. Alnot, J. J. Ehrhardt and P. Behra, Environ. Sci. Technol., 1998, 32, 2839–2845.
2. H. T. Pham, M. Kitsuneduka, J. Hara, K. Suto and C. Inoue, Environ. Sci. Technol., 2008, 42, 7470–7475.
3. H. T. Pham, K. Suto and C. Inoue, Environ. Sci. Technol., 2009, 43, 6744–6749.
4. R. Weerasooriya and B. Dharmasena, Chemosphere, 2001, 42, 389–396.
5. C. J. Liang, Y. Y. Guo, Y. C. Chien and Y. J. Wu, Ind. Eng. Chem. Res., 2010, 49, 8858–8864.
6. M. Arienzo, Chemosphere, 1999, 39, 1629–1638.
7. H. Che, S. Bae and W. Lee, J. Hazard. Mater., 2010, 185, 1355–1361.
8. R. Matta, K. Hanna and S. Chiron, Sci. Total Environ., 2007, 385, 242–251.
9. S. Bae, D. Kim and W. Lee, Appl. Catal., B, 2013, 134, 93–102.
10. H. Che and W. Lee, Chemosphere, 2011, 82, 1103–1108.
11. L. L. Xu, X. F. Li, J. Q. Ma, Y. Z. Wen and W. P. Liu, Appl. Catal., A, 2014, 485, 91–98.
12. S. Kar and S. Chaudhuri, Chem. Phys. Lett., 2004, 398, 22–26.
13. C. Wadia, Y. Wu, S. Gul, S. K. Volkman, J. H. Guo and A. P. Alivisatos, Chem. Mater., 2009, 21, 2568–2570.
14. D. W. Wang, Q. H. Wang and T. M. Wang, CrystEngComm, 2010, 12, 755–761.
15. D. W. Wang, Q. H. Wang and T. M. Wang, CrystEngComm, 2010, 12, 3797–3805.
16. S. Q. Song, R. C. Rao, H. X. Yang, H. D. Liu and A. M. Zhang, Nanotechnology, 2010, 21, 185602–185607.
17. D. N. Thi, H. P. Ngoc, H. D. Manh and T. N. Kim, J. Hazard. Mater., 2011, 185, 653–661.
18. A. Prakash, S. Chandra and D. Bahadur, Carbon, 2012, 50, 4209–4219.
19. H. X. Wu, G. Gao, X. J. Zhou, Y. Zhang and S. W. Guo, CrystEngComm, 2012, 14, 499–504.
20. W. Fan, W. Gao, C. Zhang, W. W. Tjiu, J. S. Pan and T. X. Liu, J. Mater. Chem., 2012, 22, 25108–25115.
21. Y. Choi, H. S. Bae, E. Seo, S. Jang, K. H. Park and B. S. Kim, J. Mater. Chem., 2011, 21, 15431–15436.
22. T. Zeng, X. L. Zhang, Y. R. Ma, H. Y. Niu and Y. Q. Cai, J. Mater. Chem., 2012, 22, 18658–18663.
23. Y. H. Song, Z. F. He, H. Q. Hou, X. L. Wang and L. Wang, Electrochim. Acta, 2012, 71, 58–65.
24. Y. L. Dong, H. G. Zhang, Z. U. Rahman, L. Su, X. J. Chen, J. Hu and X. G. Chen, Nanoscale, 2012, 4, 3969–3976.
25. G. Y. He, W. F. Liu, X. Q. Sun, Q. Chen, X. Wang and H. Q. Chen, Mater. Res. Bull., 2013, 48, 1885–1890.
26. H. Bader, V. Sturzenegger and J. Hoigné, Water Res., 1988, 22, 1109–1115.
27. K. Kasaka, H. Yamada, S. Matsui, S. Echigo and K. Shishida, Environ. Sci. Technol., 1998, 32, 3821–3824.
28. D. Li, M. B. Muller, S. Gilje, R. B. Kaner and G. G. Wallace, Nat. Nanotechnol., 2008, 3, 101–105.
29. N. A. Zubir, C. Yacou, J. Motuzas, X. W. Zhang and J. C. D. da Costa, Sci. Rep., 2014, 4, 1–8.
30. S. Navalon, M. Alvaro and H. Garcia, Appl. Catal., B, 2010, 99, 1–26.
31. M. A. McKibben and H. L. Barnes, Geochim. Cosmochim. Acta, 1986, 50, 1509–1520.
32. P. C. Singer and W. Stumm, Science, 1970, 167, 1121–1123.
33. X. M. Ren, J. X. Li, X. L. Tan, W. Q. Shi, C. L. Chen, D. D. Shao, T. Wen, L. F. Wang, G. X. Zhao, G. P. Sheng and X. K. Wang, Environ. Sci. Technol., 2014, 48, 5493–5500.
34. C. K. Duesterberg and T. D. Waite, Environ. Sci. Technol., 2006, 40, 4189–4195.
35. X. F. Li, X. Liu, L. L. Xu, Y. Z. Wen, J. Q. Ma and Z. C. Wu, Appl. Catal., B, 2015, 165, 79–86.
36. C. S. Shen, Y. Shen, Y. Z. Wen, H. Y. Wang and W. P. Liu, Water Res., 2011, 45, 5200–5210.
37. C. S. Shen, Y. Z. Wen, X. D. Kang and W. P. Liu, Chem. Eng. J., 2011, 166, 474–482.
38. C. S. Shen, H. Chen, S. S. Wu, Y. Z. Wen, L. N. Li, Z. Jiang, M. C. Li and W. P. Liu, J. Hazard. Mater., 2013, 244–245, 689–697.
39. C. S. Shen, Y. Z. Wen, Z. L. Shen, J. Wu and W. P. Liu, J. Hazard. Mater., 2011, 193, 209–215.
40. S. Ogawa, J. Appl. Phys., 1979, 50, 2308–2311.
41. V. Singh, D. Joung, L. Zhai, S. Das, S. I. Khondaker and S. Seal, Prog. Mater. Sci., 2011, 56, 1178–1271.
42. C. S. Shen, S. F. Song, L. L. Zang, X. D. Kang, Y. Z. Wen, W. P. Liu and L. S. Fu, J. Hazard. Mater., 2010, 177, 560–566.
43. M. Berger, M. Hazen, A. Nejjari, J. Fournier, J. Guignard, H. Pezerat and J. Cadet, Carcinogenesis, 1993, 14, 41–46.
44. C. A. Cohn, M. J. Borda and M. A. A. Schoonen, Earth Planet. Sci. Lett., 2004, 225, 271–278.
45. C. A. Cohn, R. Laffers and M. A. A. Schoonen, Environ. Sci. Technol., 2006, 40, 2838–2843.
46. C. A. Cohn, S. Mueller, E. Wimmer, N. Leifer, S. Greenbaum, D. R. Strongin and M. A. A. Schoonen, Geochem. Trans., 2006, 7, 3.
47. R. T. Lowson, Chem. Rev., 1982, 82, 461–497.
48. J. D. Rimstidt and D. J. Vaughan, Geochim. Cosmochim. Acta, 2003, 67, 873–880.
49. G. W. Luther, Geochim. Cosmochim. Acta, 1987, 51, 3193–3199.

---