Degradation of methyl orange by pyrite activated persulfate oxidation: mechanism, pathway and influences of water substrates

While the textile printing and dyeing industry brings convenience to people's lives, environmental pollution caused by the random discharge of printing and dyeing wastewater has attracted widespread attention (Ahamed Fazil et al. 2021; Rafique et al. 2021). Printing and dyeing wastewater is hard to treat due to its complex composition, high pollutant concentration, difficult degradation, and complicated chemical structure (Zhang et al. 2019). Due to the large volumes of water required by the textile printing and dyeing industry, river or tap water is often used for processing, further increasing the difficulty of treatment. Moreover, commonly used physical, chemical, and biological treatment technologies are not suitable for achieving the desired results (Berkessa et al. 2020; Qian et al. 2020). At present, finding the treatment method for printing and dyeing wastewater has become a hot issue in the field of environmental protection.

Advanced oxidation processes (AOPs) can react rapidly under mild conditions, this has attracted much research attention (Shao et al. 2018; Pham et al. 2021). This technology aims to degrade organic substances through the strong oxidation of some free radicals such as hydroxyl radicals (HO^•) and sulfate radicals (⁠⁠) (Abd Manan et al. 2019; Lojo-López et al. 2021). Currently, sulfate radical-based advanced oxidation processes (SR-AOPs) can effectively degrade a variety of organic contaminants into less complex byproducts and finally mineralize them into carbon dioxide and water (Du et al. 2019; Zhou et al. 2019). What's more, compared with HO^•, has a higher redox potential (E₀ = 2.5−3.1 V), a longer lifetime (t_1/2 = 30−40 μs), and a wider range of pH (Peng et al. 2018; Chen et al. 2020). Therefore, more and more researchers have begun to study AOPs based on ⁠.

In recent years, the produced by activating persulfate (PS) has become a novel direction for printing and dyeing wastewater treatment (Crincoli et al. 2020; Song et al. 2020). For activation of persulfate (PS) or peroxy sulfate (PMS) to produce ⁠, commonly used methods include photoactivation (Tian et al. 2018), ultrasonic activation (Ma et al. 2018), heat activation (Shi et al. 2019), alkaline activation (Dominguez et al. 2019), metal activation (Xie et al. 2020), etc. However, UV activation requires sophisticated equipment and relatively high cost (Jung et al. 2016; Dhaka et al. 2017; Zou et al. 2021). Heat activation requires strict control of temperature and has limitations (Miao et al. 2018). Metal activation could be carried out at normal temperature and pressure easily and conveniently (Li et al. 2021). Among these, iron is a transition metal element, its toxicity is relatively low compared to copper and manganese elements. Fe⁰ and Fe²⁺-activated persulfate have the disadvantages of pH limitation and a large amounts of iron sludge generation after activation (Matzek & Carter 2016), so some iron-containing compounds such as iron ore and steel slag have received attention and research due to their low cost and ease of availability (Wang et al. 2021).

There are many common iron ores on Earth. Natural vanadium–titanium magnetite-activated persulfate is able to decolourize methyl orange (MO) (Zhang et al. 2020a). Hematite nanoparticle-activated persulfate has been used for sonochemical degradation of bisphenol A (Dehvari et al. 2021). Pyrite-activated persulfate could oxidatively degrade Orange G in aqueous solution (Zhang et al. 2020b). Additionally, Pyrite (FeS₂) is one of the lowest cost and easily available natural minerals that mainly exist in water, lakes, sediments and groundwater (Chen et al. 2018). The Fe²⁺ in solution from FeS₂ immediately reacts with PS to produce ⁠, reducing the pollution of iron sludge production. It was found that FeS₂ could reduce Fe(III) to Fe(II) (Liu et al. 2015). Therefore, the formed Fe(III)/Fe(II) cycle could continuously activate PS to produce ⁠, further reducing the consumption of the activator, so pyrite could be used as a Fenton-like activator to effectively degrade organic contaminants (Diao & Chu 2021). Although there are many mixed dyes in actual industrial wastewater, most of them are azo dyes. Methyl orange is a typical azo dye with a special (-N = N-) bond that makes its molecular structure relatively stable (Jiang et al. 2015; Lei et al. 2018). Hence, in this study, MO was selected as the target substance and FeS₂ was used as the activator to activate PS to produce degraded MO to explore the treatment method for printing and dyeing wastewater. The effect and mechanism of pyrite-activated persulfate to degrade MO were explored. Due to the large volumes of water required by the textile printing and dyeing industry, river water or tap water is usually used to process and treat textiles. Therefore, river water and tap water were used to simulate treatment of wastewater contaminated with MO, in order to provide a reference for the economic treatment of printing and dyeing wastewater.

Natural pyrite (FeS₂) was purchased from a mineral product processing plant in Shijiazhuang, Hebei, China. Methyl orange (C₁₄H₁₄N₃SO₃Na) was obtained from Tianjin Tianli Chemical Reagent Co., Ltd, China. Sodium persulfate (PS, Na₂S₂O₈), methanol (MeOH, CH₃OH), and potassium iodide (KI) were produced by Tianjin Kaitong Chemical Reagent Co., Ltd, China. Sodium hydroxide (NaOH) and concentrated sulfuric acid (H₂SO₄) were produced by Sinopharm Chemical Reagent Co., Ltd, China. Tert butanol (TBA, C₄H₉OH) and sodium carbonate (NaHCO₃) were produced by the Tianjin Beichen Founder Reagent Factory, and the above reagents were all analytically pure.

Pyrite after grinding through a 100 mesh sieve, was placed in a beaker and soaked in 0.5 M H₂SO₄ solution for 1 hour and then filtered. The pyrite was washed repeatedly with distilled water until the pH of the filtrate was 7, and then filtered. The filtered sample was dried in an oven at about 40 °C for future use.

After the reaction was completed under the optimal conditions, the pyrite in the reaction solution was filtered, and the pyrite was washed repeatedly with distilled water until the pH of the filtrate was 7, and then filtered. The filtered samples were dried in an oven at about 40°C for the next cycle test.

The initial concentration of MO solution was 0.1 mM (Figure S1 shows the standard curve for MO), after analyzing the experimental results of the adsorption of MO with different concentrations of FeS₂ (Figure S2) and the degradation of MO in FeS₂/PS with different concentrations of PS (Figure S3). The response surface methodology in JMP 13 software was used to optimize the FeS₂/PS dosage. Subsequently, the obtained experimental data were analyzed for the optimal dosage of FeS₂ and PS for subsequent experiments.

Here, 0.1 mM MO was prepared by adding the desired FeS₂ and PS, which were placed on a magnetic stirrer and stirred for 5, 10, 20, 30, 60, 90, 120, and 150 min. Next, 3 mL of the sample were taken and passed through a 0.45 μm filter membrane and 0.5 mL methanol was added to release the free radicals. After mixing evenly, the samples underwent colorimetric analysis at 470 nm (iso-absorption point wavelength, which is the wavelength at the intersection of the acidic and alkaline absorption curves) using a 721 visible photometer. After the reaction had finished, the pH and temperature of the solution were measured. Two sets of experiments were performed in parallel for each group of experiments.

In the pH experiment, the pH of 0.1 mM MO was adjusted to 2, 4, 6, 8, and 10 with 0.5 M H₂SO₄ and 0.5 M NaOH. In the free radical experiment, 3 mL MeOH, 3 mL TBA and no free radical quencher were added to the solution before the start of the experiment. In the experiment with different anions, different concentrations of Cl⁻, ⁠, and were added to the solution. In the experiment with different water substrates, river water from the Uma River in Taigu, tap water, and distilled were used to prepare MO solutions for subsequent experiment.

The content of Fe(II) and Fe(III) was determined using the o-phenanthroline colorimetric method. The content of PS was determined using the colorimetric method. Biological toxicity was determined using an OD₆₀₀ colorimetric method. TOC was determined using a total organic carbon analyzer (Muti N/C-3100 mode1, Shimadzu, Japan). Field emission scanning electron microscopy (SEM; JSM-7001F, Tokyo, Japan), X-ray diffraction (XRD; MiniFlex II, Tokyo, Japan), and X-ray photoelectron spectroscopy (XPS; AXIS ULTRA DLD, Kratos, UK) were utilized to characterize FeS₂. The automatic specific surface and pore analyzer (TriStar II 3020 model, USA) was used to measure the specific surface area, total pore volume, micropore area, micropore volume and mesopore distribution of minerals. The degradation intermediates from MO were analyzed using liquid chromatography-mass spectrometry (LC-MS) (LC-MS-8045, Shimadzu, Japan).

The dose for FeS₂/PS was optimized by response surface methodology, and it was found that the removal rate for MO could reach 92.1–100% after 150 min when [MO] = 0.1 mM, [PS] = 1 mM, [FeS₂] = 1.6 g/L (Figure 1). When [MO] = 0.1 mM, [FeS₂] = 1.6 g/L, the removal rate of MO increased with the increase in PS concentration. On the one hand, the PS itself could degrade MO (Wacławek et al. 2017). On the other hand, the presence of Fe(II) could also activate PS to produce (Marchesi et al. 2012; De Luca et al. 2017), which had a good degradation effect on MO.

To verify the optimization results for FeS₂-activated PS degradation of MO, the removal of MO for different systems under optimal conditions was investigated (Figure 2), and the pH before and after the reaction are shown in Table S1. It can be seen from Figure 2 that when FeS₂ was added alone, the FeS₂ only had a small adsorption effect on MO, fluctuating up and down in the range 0–2.2%. The specific surface area measured by the mesoporous physical analyzer was 0.1500 m²/g, its cumulative pore area and cumulative pore volume (Figure S4) also showed its weak adsorption effect.

When PS was added alone, the degradation rate of MO reached 16.9% in the first 5 min (Figure 2), which was about one-third of the subsequent degradation rate. The rapid reaction may be due to the oxidizing nature of (Ahmad et al. 2013; Yun et al. 2017). Within 5–150 min, PS continued to degrade 31.8% of MO, but the degradation effect was reduced compared with the rapid stage, which might be due to competition between MO degradation products and undegraded MO, and with MO degradation more and more byproducts, competition with PS was increasingly intense, resulting in a progressively slower trend of MO degradation.

When [MO] = 0.1 mM, [PS] = 1.0 mM and [FeS₂] = 1.6 g/L, the degradation of MO showed a fast and then a slow trend. The degradation of MO at 0–120 min was probably due to the oxidation of MO by itself and the increasing dissolution of Fe(II) over time to react with PS and produce more to degrade MO. Furthermore, UV spectra (Figure S5) also showed rapid degradation of MO in the first 120 min. However, the degradation rate of MO slowed down at 120–150 min, probably because the intermediates of MO and MO co-competed for ⁠, resulting in a decrease in the degradation rate of MO.

Under optimal conditions, the efficiency of different systems for MO removal ranged from high to low: FeS₂/PS > PS > FeS₂. Table 1 shows the removal efficiencies of other activated PS systems for MO. The UV/PS system could also degrade MO, although MO was degraded over a short time, the PS dose was 42 times higher than the target substance (Frontistis et al. 2015). Copper tailings could activate PS to degrade MO, but the PS dose was also much greater than the target substance, which was 50 times that of MO (Wang et al. 2020a). It was found that Zn⁰ could activate PS to degrade MO. The Zn⁰/PS system showed that 90% MO was removed after 180 min, but the removal of MO by PS alone was 28% and the removal of MO by Zn⁰ alone was 51%. Therefore, the degradation of MO by Zn⁰-activated PS producing was only 11% (Li et al. 2014). In summary, compared to other published reports in the literature, in this paper, the PS dose was low relative to the target material, and the produced by FeS₂ activation of PS could degrade 43.3% MO. Moreover, FeS₂ is a low cost and easily available natural mineral. Therefore, the FeS₂/PS system has good prospects for application for the removal of MO.

Before the reaction started, a free radical scavenger was added to the solution. MeOH containing α-hydrogen could quench and HO^• with rate constants of 1.6 × 10⁷ M⁻¹ s⁻¹and 1.9 × 10⁹ M⁻¹ s⁻¹, respectively (Xiao et al. 2020). Tert butanol (TBA) is usually used as a scavenger for HO^• because the rate constant of (4.0 × 10⁵ M⁻¹ s⁻¹) is much lower than that of HO^• (6.0 × 10⁸ M⁻¹ s⁻¹) (Ding et al. 2020). Therefore, the respective contributions of HO^• and could be explored through the addition of TBA and MeOH. In addition, to effectively quench HO^• and within the system, we added methanol (1.5 M) at 15,000 times the dose of MO (0.1 mM) and tert butanol (0.6 M) at 6,000 times the dose of MO (0.1 mM). The pH before and after the reaction is shown in Table S2.

After the addition of MeOH, the removal of MO in FeS₂/PS system was 73.0% after 150 min of reaction. As and HO^• were not present in this system (Figure 3), the MO removal rate was mainly through the adsorption of FeS₂ and the degradation of PS itself (Figure 2). The addition of TBA only hindered the reaction of the system with HO^• (Chen et al. 2019; Shao et al. 2020). At 150 min, the removal effect of MO after the addition of TBA was similar to that without any inhibitor (Figure 3), indicating that the contribution of hydroxyl radicals in the system was relatively low. The removal rate after the addition of methanol was subtracted from that of TBA, and the result was 20.5%, which was the removal rate of MO by ⁠. Therefore, it could be observed that the contribution rate of to the degradation of MO was about 22.4%.

In the FeS₂/PS system, the content of PS gradually decreased with the increase in time, which fully demonstrated the important role of PS in FeS₂/PS degradation of MO. The residual rates of PS at 30, 60, 90, 120, and 150 min were 95.6%, 78.8%, 61.1%, 53.4%, and 33.6% (Figure 4(a)). Fe(II) could activate PS to produce for the effective degradation of MO. Therefore, the concentrations of Fe(II) and Fe(III) in FeS₂/PS system were determined by the o-phenanthroline method. It could be found that the concentrations of Fe(II) and Fe(III) were very low (Figure 4(b)), but their concentrations increased with time, probably because when Fe(II) was in solution from FeS₂ in the pre-reaction stage, it immediately reacted with PS to produce ⁠. After a period of time, the PS concentration gradually decreased, the dissolved Fe(II) could not react with PS immediately causing the detected Fe(II) concentration in the solution to increase.

When the initial pH of the solution was 4, 6 or 8, the MO removal rates after 150 min were 95.0%, 91.8% and 91.4% (Figure 5). There was no significant difference in the removal rate of MO at 150 min when pH = 4–8, but pH = 4 had a promoting effect before 120 min. The degradation in the first 5 min might be due to the oxidative nature of PS itself, and at 5–150 min might be due to Fe(II) dissolution from FeS₂, and Fe(II) activated PS produced SO₄^•− to degrade MO.

When the initial pH ≥ 10, the removal rate of MO was relatively poor, and the removal rate was only 40.4% after 150 min (Figure 5). As the Fe(II) for FeS₂ was difficult to dissolve when the pH was high, this stage might be the degradation of the PS itself. In 5–150 min, the removal rate of MO tended to be flat. At this stage, the removal rate only increased by 17.1%. When the initial pH ≤ 2, the removal rate of MO, was only 33.6% after 150 min, and the reason might be that excessive Fe(II) could also react with ⁠.

By measuring the pH at different reaction times in the degradation of MO in the FeS₂/PS system (Figure S7), it was found that the pH of MO gradually decreased in the degradation of the FeS₂/PS system and it decreased to 2.87 after 150 min. In addition to the pH of 2 and 10 system reactions after the solution pH were 1.97 and 4.70. The remaining pH systems were all in the range of 2.81–2.89 (Table S3). The pH of the solution after the reaction might be reduced because the reaction of Fe(II) and PS produced H⁺, Fe(III) and ⁠, which made the pH of the solution less.

Sewage or natural waters usually contain a certain amount of anions. These anions will have an impact on the degradation of pollutants by free radicals. This paper explored the effect of different concentrations of Cl⁻, and in the FeS₂/PS system. The pH values before and after the reaction are shown in Table S4, and the k_obs and t_1/2 values were determined according to the quasi-first-order reaction equation ln(C/C₀) = −k_obs to evaluate the effect of various concentrations of anions (Table S5).

In exploring Cl⁻ tests, it was found that a low concentration (0.01–0.1 mM) of Cl⁻ had a slight inhibitory effect on the degradation of MO by the FeS₂/PS system (Figure 6(a)). After 150 min, 0.01 and 0.1 mM Cl⁻ were inhibited by 6.4% and 3.5%, respectively, but the overall degradation trend was the same as when Cl⁻ was not added. This might be attributed to the occurrence of Cl⁻ which might convert into Cl^• and ⁠, low-active free radicals (Equations (4)–(6)) (Liu et al. 2016; Luo et al. 2019; Zhang et al. 2020b), the reduction of made the degradation rate of FeS₂/PS system to MO lower. High concentrations (1–10 mM) of Cl⁻ had a slight promotion effect within 0–30 min and a slight inhibitory effect after 30–150 min.

In the experiment of exploring the influence of ⁠, it was found that the degradation trend of the system with the addition of low concentration (0.01–0.1 mM) of was consistent with that without (Figure 6(b)), which only produced a slight inhibitory effect. It might be that the reaction of and generated HCO₃^•, a low reactive free radical (Equation (7)), which reduced the reaction rate (Fan et al. 2015; Gao et al. 2018; Lian et al. 2019). The inhibitory effect of high concentrations (1–10 mM) of was more obvious after 60 min. On the one hand, as the concentration of increased, more low-active free radicals were produced, which made the inhibitory effect more noticeable. On the other hand, it might result from the introduction of a large amount of ⁠, which made the pH of the solution weakly alkaline. In the aforementioned study, the degradation of the system would be inhibited in an alkaline environment.

Through previous studies on pH and different concentrations of anions and other factors, it was found that complex environmental factors could reduce the degradation efficiency of MO in the FeS₂/PS system. Large numbers of water sources of printing and dyeing plants come from river water. Therefore, the river water from the Uma River in Taigu, tap water, and distilled water were used to simulate printing and dyeing wastewater. The effect of different water substrates on the degradation MO in FeS₂/PS system were investigated, and Table S6 shows the quality indicators of river water and tap water.

The research results showed that the degradation of MO in river water and tap water was far less than that in distilled water without pH adjustment in the FeS₂/PS system (Figure 7(a)). This was found by measuring some physical and chemical parameters of river water and tap water. The pH of river water and tap water was weakly alkaline and they both contained many anions. These factors would affect the degradation of MO in the FeS₂/PS system to a greater or lesser extent.

To further effectively degrade the MO in different water substrates, the pH of the MO solution of different water substrates was adjusted to 6 for the degradation test. The study found that the degradation effect was similar to the result of no pH adjustment (Figure 7(b)), indicating that when the pH of wastewater from diverse water substrates was adjusted to 6, the MO could not be effectively degraded. This might be due to the fact that at pH = 6, the anions in solution cannot be effectively removed.

In order to effectively degrade MO in different water substrates, the pH of different water substrates was adjusted to 4 for degradation tests (Figure 7(c)). It was found that for the degradation effect of distilled water > tap water > river water, after 150 min of reaction, the removal rate was 95.0%, 89.8% and 87.3%, which reached the expected treatment effect. According to the carbonation binding state distribution diagram, it was found that there was no in the system with pH = 4 (Figure S8). Therefore, acidification (pH = 4) could promote the degradation of MO in river water in the FeS₂/PS system.

In the actual treatment of printing and dyeing wastewater, the reusability and stability of the catalyst were demonstrated by the reuse test of FeS₂ with economic and environmental factors. As shown in Figure 8(a), FeS₂ could still effectively activate persulfate to degrade MO after five cycles, and the degradation rate of MO could still reach 92.2% at 150 min. This indicated that FeS₂ had good recycling value.

To further explore the elemental composition and surface electronic states before and after the FeS₂ reaction, fresh and used FeS₂ were analyzed by XPS spectroscopy. Figure S9 shows that fresh and used FeS₂ mainly consisted of elements such as Fe, S, C, O, and Na, which was consistent with the energy-dispersive X-ray spectroscopy (EDS) spectrum (Figure S10). Fresh FeS₂ detection elements such as C and O might be surface oxides because FeS₂ was in contact with some non-metals and metals in the air. Fe and S on the surface of the used FeS₂ were significantly reduced, the reason might be that these elements were dissolved and released into the water.

Figure 8(c) shows the high-resolution Fe 2p spectrum of fresh and used FeS₂. After fitting, it was found that the Fe 2p_3/2 orbit had three peaks of 707.3, 709.6 and 710.8 eV, assigned to Fe(II)-S, Fe(II)-O, and Fe(III)-O (Ye et al. 2021). The proportions before reaction were 52.9%, 14.0% and 33.1%, respectively. After the reaction, the Fe(II)-S peak significantly decreased to 23.2% and the Fe(III)-O peak increased to 55.9%.

The high-resolution S 2p spectra of fresh and used FeS₂ are shown in Figure 8(d). For the 162.5 eV and 163.6 eV peaks, the S 2p_3/2 and S 2p_1/2 orbitals of were, respectively, accounting for 89.4% before the reaction, but decreased to 65.5% after the reaction. Peaks at 169 and 170.2 eV were S 2p_3/2 and S 2p_1/2 orbitals of sulfate (⁠⁠) (Diao et al. 2018). The total proportion was 10.6% before the reaction and increased to 34.5% after the reaction. This implied that in FeS₂ was oxidized to in the oxidation process. Corresponding slight changes in the phase and strength of fresh and used FeS₂ could also be observed from XRD (Figure 8(b)) and Fourier transform infrared (FTIR) (Figure S11). These results indicated that Fe(III)/Fe(II) cycling was accelerated in FeS₂, but was non-renewable in the FeS₂/PS system and, once was depleted, Fe(II) activation PS will be the main reason for the degradation of MO.

To explore the changes in biological toxicity before and after different treatments, OD₆₀₀ was used to characterize the number of microorganisms, observe the growth of microorganisms, and characterize the changes in toxicity. As can be seen from Figure S12, the OD₆₀₀ values between the MO, H₂O/PS, H₂O/FeS₂, MO/PS, MO/FeS₂, H₂O/PS/FeS₂, MO/PS/FeS₂, and H₂O treatments were 1.30, 1.31, 1.26, 1.28, 1.39, 1.17, 1.00, and 1.22 at 24 h, respectively. Their growth trends all steadily increased in the previous period, indicating that there were no significant difference in the number of microorganisms between the treatments, which proved that the toxicity change was not obvious. Although the OD₆₀₀ value of MO/PS/FeS₂ treatment decreased and then increased during 24–72 h, other treatments also fluctuated within the range of 1.2–1.6, probably due to excessive nutrient consumption in the early stage of the reaction, resulting in insufficient nutrients after 24 h and competition between microorganisms, which caused the OD₆₀₀ values to fluctuate up and down. In addition, the biological toxicity of the FeS₂/PS system at 72 h was not significantly different from that of MO, which has good application prospects.

The degree of mineralization of MO in FeS₂/PS system was determined by measuring the TOC concentration in the solution after the reaction. As shown in Figure 9, the removal rate of TOC increased with the increase in the reaction time, 14.1% TOC was removed after 2.5 h of reaction in the FeS₂/PS system. The removal rate of TOC reached 30.5% after 5 h of reaction in this system.

Through LC-MS further analysis of the degradation intermediates, it could be seen from Figure S13 and Table S7 that the AOPs could open the -N = N- double bond of the azo dye and carry out effective degradation. Its degradation pathways are illustrated in Figure 10(a). In the wastewater contaminated with MO, Na⁺ exists in the form of ions, so it was speculated that m/z = 304 was P1. Then two pathways were deduced. In the first pathway, P2, P3, and P4 were demethylation, demethylation, and deaminylation products in turn. P5 was the -N = N- double bond-breaking product, and P6 was the deaminylation product. Path II was the process of removing the SO³⁻ group into P7 and P8, P9 was the product of double bond-breaking, and these eventually became other intermediates, CO₂ and H₂O. Based on the above results, a speculative mechanism for MO degradation in the FeS₂/PS system is shown in Figure 10(b).

• (1)

  The optimal conditions for degradation of MO in the FeS₂/PS system obtained by response surface methodology were: [MO] = 0.1 mM, [PS] = 1.0 mM, [FeS₂] = 1.6 g/L, the removal effect of MO could reach 92.9% after 150 min, meanwhile the removal rate of TOC reached 14.1%. Furthermore, the degradation products had no biological toxicity. In addition, the played a major role in this system, and OH• had little effect.

• (2)

  In the FeS₂/PS system, when the initial pH was 4–8, the removal rate of MO remained unchanged at 150 min. Cl⁻ had little effect on the degradation of MO in the FeS₂/PS system. However, the presence of would have an inhibitory effect. In addition, at low concentration (0.01–0.1 mM) had little effect, but at high concentration (1–10 mM) had a certain inhibitory effect.

• (3)

  The river water and tap water could greatly inhibit the degradation of MO in the FeS₂/PS system, but acidification (pH = 4) could promote the degradation of MO in river water and tap water.

• (4)

  The mechanism for FeS₂/PS was Fe(II) dissolved from FeS₂ that could activate PS to produce Fe(III) and which could degrade the target substance, meanwhile, dissolved by FeS₂ could reduce Fe(III) to Fe(II). Therefore, the formed Fe(III)/Fe(II) cycle could continuously activate PS to produce ⁠.

This work was supported by the Natural Science Foundation of Shanxi Province, China (No. 202103021224130, No. 202103021224139); the Fund for Shanxi ‘1331 Project’ (No. 20211331-15); the Youth Foundation of Shanxi Province, China (No. 201901D211353, No. 201901D211385); Key R&D Program Projects in Shanxi Province, China (No. 201903D221015); the Scientific and Technological Innovation Planning of Higher Education Institutions of Shanxi Province, China (No. 2019L0383, No. 2020L0155). Thanks to the staff of the College of Resources and Environment, Shanxi Agricultural University, the staff of the Shanxi Institute of coal chemistry, and we are grateful to the anonymous reviewers and the editors for their valuable suggestions and comments on revising and improving this paper.

All authors declare no conflict of interest.

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