Oxidative degradation of Orange G in aqueous solution by persulfate activated with pyrite

Large quantities of dye can be lost in water during dye production and their application, mainly in the textile and leather industry. Azo dyes make up almost 60% of the synthetic dyes used industrially, and most of them are recalcitrant, toxic and potentially carcinogenic in nature (Routoula & Patwardhan 2020). A variety of technologies, such as adsorption, biodegradation, ozonation, electrochemical oxidation and chemical reduction, are tried to remove azo dye from water, but these processes have high costs and limited applicability (Pavithra et al. 2019). Advanced oxidation processes (AOPs) were proposed as alternative technologies for treatment of refractory organic wastewater due to the high oxidizing ability of involved reactive species (Wang & Wang 2018).

In recent years, AOPs based on persulfate (PS) and permonosulfate (PMS) have found an increasingly wide utilization in environmental pollution control, especially for the treatment of biorefractory wastewater (Wang & Wang 2018). Sulfate radical (SO₄^•−), a strong oxidizing agent (E° = 2.5–3.1 V vs. SHE), can be formed by PS and PMS with several activators, such as UV-light, transition metal ions (like Co²⁺, Cu²⁺, Ag⁺, Fe²⁺ and Mn²⁺), heat, carbon, and electrode (Wang & Wang 2018). Iron is the most widely used PS activator because it is effective, relatively non-toxic, environmentally friendly and more cost-effective than other transition metals. However, Fe²⁺/PS system has the problems of rapid consumption of ferrous ion, narrow pH range and iron sludge production. To solve these problems, heterogeneous catalysts like zero-valent iron (Cao et al. 2019), goethite (Su et al. 2019), Fe₃O₄-sepiolite (Xu et al. 2019) and sulfide minerals (Chen et al. 2017; Lian et al. 2019) were developed. Among these heterogeneous catalysts, sulfide minerals hold great application prospects, because they could not only continually release Fe²⁺ ions to activate PS and PMS to produce SO₄^•− and hydroxyl radicals (•OH), but also promote the cycling of Fe³⁺/Fe²⁺ couple with reducible sulfur in these minerals (Chen et al. 2017; Lian et al. 2019).

Orange G (OG), a typical azo dye in textile wastewaters, has been the subject of intense investigations. Several catalysts, such as ferrous ion (Xu & Li 2010), sewage sludge-derived magnetic nanocomposites (Li et al. 2017b) and iron-immobilized resin chars (Shi et al. 2015), have been used to activate PS for OG degradation. However, information about OG degradation by PS activated with sulfide minerals could not be found in the literature. In this study, pyrite (FeS₂), the most widely distributed sulfide in the earth's crust, was applied to degrade OG in aqueous solution in the presence of PS. Effects of various factors, initial solution pH, pyrite particle size, pyrite dosage, initial PS concentration and co-existing anions, were investigated. Reactive species responding for OG degradation in the pyrite/PS system were identified by quenching experiments and electron paramagnetic resonance (EPR) tests. The changes of Fe, O and S speciation during pyrite oxidation were characterized by X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD). The degradation intermediates of OG were analyzed with high resolution liquid chromatography quadropole time-of-flight mass spectrometry (LC-QTOF-MS), and its degradation pathway was proposed. Finally, the mineralization of OG in the pyrite/PS system was examined.

Orange G (OG) with HPLC grade (˃96%) was purchased from Shanghai Macklin Biochemical Co. Ltd, China. Potassium persulfate (PS, Na₂S₂O₈), methanol (MeOH, CH₃OH), tert-butyl alcohol (TBA, C₄H₉OH) and p-benzoquinone (BQ) were purchased from Sinopharm Group Chemical Reagent co. Ltd, China. All the above chemicals and reagents were of analytical grade and employed without further purification. Raw pyrite ore with high purity was purchased from Gaowantong Fossil Specimen Museum, Guilin, Guangxi Province, China. The raw ore was first milled with a ceramic mortar and sieved, and the resulting powder was then washed with 95% ethanol, 1 M HNO₃ and deionized water, successively. The as-prepared powder was finally dried at 30 °C and stored before use.

All experiment reactions were operated on a magnetic stirrer with constant temperature water bath at a stirring rate of 150 rpm. Batch experiments were conducted in a 300 mL cylindrical glass container at 30 ± 0.5 °C. The container was covered with aluminium-foil paper to prevent the effect of light. OG aqueous solution was prepared by adding a certain amount of OG into deionized water. The experiments were started immediately after PS and pyrite were added into 200 mL OG aqueous solution successively. The pH was adjusted to desired values by 1 M H₂SO₄ or NaOH before adding pyrite into the solution. Samples (1.0 mL) were withdrawn at preselected time points, immediately quenched by 0.5 mL of methanol and filtrated with 0.45 μm membrane before analysis. Scavenging tests were conducted by adding a certain dose of MeOH, TBA or BQ to the reaction solution before the reaction started. Unless otherwise stated, the experimental conditions were set as [PS]₀ = 5 mM, [pyrite]₀ = 1.0 g/L, pH₀ = 6.0, [OG]₀ = 100 μM. Each experiment was performed in duplicate at least and the mean values were used for analysis. The error bars in the figures of the following contents represent standard deviation.

The concentration of OG in aqueous solutions was determined by UV–vis spectrophotometry (752N, Shanghai INESA) at a maximum absorption wavelength of 478 nm. Solution pH was measured by a pH meter (PB-10, Sartorius). The concentration of Fe²⁺ was determined by spectrophotometric method with 1, 10-phenanthroline, and total dissolved Fe was also determined in the same way using hydroxylamine as the reducing agent. Total organic carbon (TOC) was measured by a TOC analyzer (Vario TOC Select, Germany). The degradation intermediates of OG were analyzed using LC-QTOF-MS. The mobile phase was 1 mM oxalic acid and acetonitrile aqueous solution, and the mass spectra were recorded across the range of 50–1,000 m/z with the negative scan mode.

The crystalline phase of the pyrite powder was determined by an X-ray powder diffractometer (XRD; X'Pert3 Powder, Netherlands Panaco) system equipped with Cu-Kα radiation source at 45 kV and 200 mA. The surface elements of the pyrite powder was analyzed using X-ray photoelectron spectroscopy (XPS; ESCALAB250Xi, ThermoFisher) with monochromatized Al-Kα radiation. EPR tests were conducted on an EPR spectrometer (Bruker EMX 10/12, Bermen, Germany) to detect the generated active species, and 5,5-dimethy-1-pyrroline-N-oxide (DMPO) was chosen as spin-trapping reagent in the pyrite/PS system.

As can be seen from Figure 1, 6.4% of OG was degraded in aqueous solution in 60 min by PS alone, suggesting that the oxidation of OG by PS alone was weak. It was reported that pyrite could degrade organic pollutants in the presence of O₂ (Zhang et al. 2015). However, the removal rate of OG was only 3.0% with pyrite alone, illustrating that pyrite had no obvious oxidative degradation and adsorption towards OG. On the contrary, the addition of pyrite and PS together to the solution resulted in a more rapid and prominent destruction of OG. The degradation rate of OG sped up from 20 min and then slowed down from 30 min, and the reasons may be the lower solution pH (seen from Figure 2(b)) and residual OG concentration in the later period of the reaction. OG was completely removed within 60 min, which indicated that pyrite could effectively activate PS for the degradation of organic pollutants. The rapid removal of OG may be attributed to the generated reactive species such as SO₄^•− and •OH by pyrite activated PS (seen from Figure 3).

Chemical scavengers were added into the solution before the reaction started to identify the involved reactive species such as SO₄^•−, •OH and O₂^•−. MeOH and TBA were frequently selected as chemical scavengers. MeOH which contains α-hydrogen can quench both SO₄^•− and •OH with the rate constants of 1.6 × 10⁷ and 1.9 × 10⁹ M⁻¹S⁻¹, whereas TBA is often used as a more effective scavenger for •OH, because its rate constant with SO₄^•− (4.0 × 10⁵ M⁻¹S⁻¹) is much lower than that with •OH (6.0 × 10⁸ M⁻¹S⁻¹) (Fu et al. 2019). BQ has been demonstrated to be an efficient scavenger for O₂^•− (9.6 × 10⁸ M⁻¹S⁻¹) (Wang et al. 2018). Effect of MeOH, TBA and BQ on OG degradation in the pyrite/PS system were displayed in Figure 3(a). Significant inhibitory effects on OG degradation were observed after the addition of 1.0 M MeOH and 1.0 M TBA, and the inhibitory effect of MeOH was much more stronger than that of TBA, suggesting that both SO₄^•− and •OH were the dominant reactive species and had played important roles in the degradation process. No obvious inhibition effect was observed with the addition of 2.0 mM BQ, indicating that O₂^•− contributed little to OG degradation. EPR technique was used to further identify the generated reactive species. As shown in Figure 3(b), the signals assigned to DMPO-HO• and DMPO-SO₄^•− adducts were detected in the pyrite/PS system. This verified it was SO₄^•− and •OH that were the dominant reactive species responsible for OG degradation.

Figure 5(a) shows XRD pattern of the fresh and used pyrite at 2-Theta values from 20 to 90°. The sharpness of XRD reflections clearly illustrated that the fresh pyrite was highly crystalline. The XRD pattern of the fresh and used pyrite matched well with the standard XRD data of pyrite (JCPDF: 42-1340), which demonstrated that the main component of pyrite was FeS₂ and its crystal structure was not destructed after used.

To explore the elemental composition and the surface electronic states of pyrite, XPS spectroscopy was further performed. As shown in Figure 5(b), the fresh pyrite was mainly composed of Fe, S, C, O and Al elements. The detection of C and O may be caused by the adsorption of amorphous C and O on the sample in contact with air and some nonmetal and metal (such as Si, Mg and Al) oxide on the surface of pyrite. After used, Fe and S on the surface of pyrite were significantly reduced, and the reason might be that these elements were oxidized and then released into aqueous phase. Figure 5(c) showed the high-resolution Fe 2p spectra of the fresh and used pyrite. There were two peaks at 724.8 eV and 711.2 eV, which corresponded to Fe 2p_1/2 and Fe 2p_3/2, respectively, and were characteristic of Fe(II) and Fe(III) on the surface of pyrite (Xia et al. 2018). Satellites of Fe(II) and Fe(III) at 715.0 and 719.9 eV were also found in the Fe 2p spectra of the fresh and used pyrite. Clearly, the proportion of Fe(II) species on the surface of pyrite declined from 72.4 to 64.6% after used, indicating the oxidation of Fe(II) into Fe(III) species on the surface of pyrite. The high-resolution S 2p spectra of the fresh and used pyrite are shown in Figure 5(d), and the bonding energies at 162.4, 163.6, 164.8 and 168.6 eV originated from the monosulfide (S²⁻), polysulfide (S_n²⁻), elemental sulfur (S⁰) and sulfate (SO₄²⁻), respectively. The percentages of S²⁻, S_n²⁻, S⁰ and SO₄²⁻ species on the surface of the fresh pyrite catalyst were 61.5, 28.6, 1.1 and 8.8%, respectively. After used, the percentages of these sulfur species changed to 56.1, 25.7, 4.5, and 13.7%, respectively. This means that S²⁻ and S_n²⁻ in the pyrite were oxidized into S⁰ and SO₄²⁻ in the oxidation process, which is in good accord with other reports (Zhang et al. 2017; Fan et al. 2018). It was believed that reducible sulfur in sulfide minerals could convert Fe³⁺ back into Fe²⁺ and then improve PS activation (Zhang et al. 2017; Zhou et al. 2018). Our previous study also verified reductive sulfur in sulfide minerals could enhance the regeneration of Fe²⁺ and thus promote the degradation efficiency of organic pollutants by activated PS (Zhang et al. 2020). Therefore, we presumed that reductive sulfur in pyrite in this work accelerated Fe³⁺/Fe²⁺ cycle and thus OG degradation, which could be seen in Figure 2(c), where Fe²⁺ stabilized at a certain concentration instead of depletion in the later reaction period.

The degradation intermediates of OG were analyzed using LC-QTOF-MS. Figure S1 shows the liquid chromatograms and the corresponding mass spectra of the OG solution at the reaction time of 20 min. The degradation intermediates of OG meanly included nine compounds and their potential molecular structures are listed in Table S1. The degradation pathway of OG in the pyrite/PS system is proposed in Figure 6. The azo bond of OG was first attacked by active species via epoxidation, and simultaneously the benzene ring and naphthalene ring of OG were oxidized through hydroxylation, resulting the formation of intermediate P1. The intermediate P1 was then transformed into P2, P3 and P4 through the cleavage of the N – N bond, denitrogenation and desulfonation. P4 was then oxidized into P9 and finally mineralized into CO₂ and H₂O. P5 could be obtained from P2 and P3 via processes like ring-opening, decarbonylation and hydroxylation, and it was further oxidized into P6, P7 and P8, which were subsequently mineralized into SO₄^2–, CO₂ and H₂O.

The concentration of TOC was measured to determine the degree of OG mineralization in the pyrite/PS system. As shown in Figure 7, the removal rate of TOC increased with the reaction time and reached a maximum value of 34.4%. The mineralization rate of OG was much slower than its degradation rate, and the effluent of the pyrite/PS system must be further treated. As we known, azo dyes are xenobiotic compounds, and they are difficult to degrade by aerobic biological method due to the electron withdrawing nature of the azo bond. Combination of AOPs and biological processes are often applied for industrial textile wastewater treatment, and AOPs could improve the biodegradability prior to biological treatment (Paździor et al. 2019). Thus, the pyrite/PS system can be used as pretreatment for azo dye wastewater.

Pyrite could effectively activate PS for the degradation of OG in aqueous solution, and a complete destruction of OG was achieved within 60 min. Lower solution pH, smaller pyrite particles and higher pyrite dosage were beneficial for OG degradation. Higher PS concentration was in favour of OG degradation, but excess PS would decrease the removal efficiency of OG. The addition of HCO₃⁻ and H₂PO₄⁻ but Cl⁻ had inhibitory effects on the destruction of OG. SO₄^•− and •OH were the dominant reactive species responsible for OG degradation in the pyrite/PS system. The azo bond, naphthalene ring and benzene ring of OG were all destroyed by the generated reactive species. The mineralization rate of OG in the pyrite/PS system was only 34.4% after 60 min of reaction. The pyrite/PS system can be used to treat azo dye wastewater or as pretreatment method to improve its biodegradability.

All relevant data are included in the paper or its Supplementary Information.