ABSTRACT
The activated persulfate (PS) process could produce sulfate radical (SO4·-) and rapidly degrade organic pollutants. The application of Fe3O4 as a promising PS activator was limited due to the rapid conversion of Fe2+ to Fe3+ on its surface. Mo4+ on MoS2 surface could be used as a reducing site to convert Fe3+ to Fe2+, but the separation and recovery of MoS2 was complex. In this study, MoS2/Fe3O4 was prepared to accelerate the Fe3+/Fe2+ cycle on Fe3O4 surface and achieved efficient separation of MoS2. The results showed that MoS2/Fe3O4 was more effective for PS activation compared to Fe3O4 or MoS2, with a removal efficiency of 91.8% for 20 mg·L−1 tetracycline (TC) solution under the optimal conditions. Fe2+ and Mo4+ on MoS2/Fe3O4 surface acted as active sites for PS activation with the generation of SO4•−, •OH, •O2−, and 1O2. Mo4+ acted as an electron donor to promote the Fe3+/Fe2+ cycling and thus improved the PS activation capability of MoS2/Fe3O4. The degradation pathways of TC were inferred as hydroxylation, ketylation of dimethylamino group and C-N bond breaking. This study provided a promising activated persulfate-based advanced oxidation process for the efficient degradation of TC by employing MoS2/Fe3O4 as an effective activator.
HIGHLIGHTS
MoS2/Fe3O4 was an effective, recoverable PS activator for tetracycline degradation.
TC was removed by free radical and non-free radical degradation pathways.
Mo4+ promoted the regeneration of Fe2+ on the Fe3O4 surface.
INTRODUCTION
Tetracycline (TC), as a broad-spectrum drug, has been widely prescribed for the treatment of bacterial infections in humans and animals (Peng et al. 2022). However, only a small portion of TC is absorbed and digested by organisms, while the residue is discharged into the water ecosystem, leading to the occurrence of drug-resistant bacteria and posing a great threat to human health (Cheng & Ji 2022; Liu et al. 2022b). Consequently, it is necessary to remove TC from the wastewater (Neolaka et al. 2023; Zhang et al. 2023d).
Persulfate (PS)-based advanced oxidation processes (AOPs) are considered as an effective method for decomposing pollutants due to their ability to produce sulfate radical () (Wang et al. 2023). Compared with hydroxyl (
),
produced by the activated PS has a higher redox potential (2.5–3.1 V) (Zhou et al. 2023) and a longer lifetime (30–40 μs) (Wang et al. 2022; Shabanloo et al. 2023), which is conducive to the degradation and mineralization of the pollutants due to sufficient time provided for
diffusion and contact with organic pollutants.
can be produced by activating PS using various methods, including ultraviolet (UV), heat, ultrasound, transition metal ions (Diao et al. 2020; Du et al. 2020; Deng et al. 2023; Cabrera-Reina et al. 2023) and heterogeneous transition metal catalysts (Cai et al. 2022; Zhang et al. 2023a). The activation of PS by heterogeneous transition metals is independent of additional energy requirements. Additionally, these metals can be easily recovered after the reaction and thus secondary pollution is avoided, which is more favorable in practical applications (Li et al. 2022b, Dai et al. 2023).
Recently, Fe3O4 was frequently applied to PS activation for the removal of organic pollutants due to its excellent magnetism and environmental friendliness (Peng et al. 2018). Fe2+ had the ability to activate PS and generate more , which was efficient for the degradation of organic pollutants (Zhu et al. 2022). Nonetheless, Fe2+ on the surface of Fe3O4 tended to be oxidized to Fe3+ during the activation of PS, and agglomeration was likely to occur among Fe3O4 nanoparticles due to their inherent magnetic properties, leading to the reduction of their specific surface area and the efficiency of activating PS (Zhang et al. 2022a).
Molybdenum disulfide (MoS2) was widely utilized in the wastewater treatment due to its low toxicity, abundant active center and high electron mobility (Jlidi et al. 2021). It was proved that MoS2 could be used to activate PS to enhance the degradation of organic pollutants (Wang et al. 2021a). Mo4+ exposed due to S defect in MoS2 could be used as an active site to reduce Fe3+ to Fe2+, thus promoting the Fe3+/Fe2+ recycling (Lu et al. 2021). Song et al. found that MoS2 could be used as a cocatalyst of Fe2+ to activate PS for the degradation of sulfisoxazole (SIX). The regeneration of Fe2+ was accelerated because Mo4+ in MoS2 could convert Fe3+ to Fe2+, and the removal efficiency of SIX was as high as 97.1% within 40 min (Song et al. 2020a). Kuang et al. added MoS2 into the Fe3+/PS system to degrade p-chloroaniline (PCA), and the electron-rich Mo4+ could transform Fe3+ into Fe2+ through electron transfer, and thus Fe2+ could be continuously produced and continuous degradation of PCA was achieved (Kuang et al. 2021). However, the separation and recovery from the solution of MoS2 could only be performed by repeated filtration or centrifugation, which was a complicated and expensive process.
One possible strategy to solve the above problems is to combine MoS2 and Fe3O4 to prepare a magnetic MoS2/Fe3O4 composite, which can realize rapid separation and recovery of MoS2 under a magnetic field. Mo4+ in MoS2 can promote the Fe3+/Fe2+ cycle on the surface of Fe3O4 and improve its catalytic activity. However, there are few reports on MoS2/Fe3O4 as a PS activator, and the mechanisms are still unclear.
In this study, MoS2/Fe3O4 was prepared as a PS activator by solvothermal hydrothermal method for the effective removal of TC. The structure, morphology and chemical composition of MoS2/Fe3O4 were explored and its catalytic performance for PS was evaluated. Meanwhile, the effects of MoS2/Fe3O4 dosage, PS concentration, initial pH, temperature and common inorganic anions on TC degradation were investigated. The mechanisms of MoS2/Fe3O4 as a PS activator were proposed, the possible degradation pathways of TC in the MoS2/Fe3O4/PS system were revealed, and the risk of TC and its degradation intermediates to aquatic organism were predicted. It is anticipated that MoS2/Fe3O4 can serve as an efficient PS activator for the enhanced degradation of TC and simultaneously be separated and recovered by simple operations.
MATERIALS AND METHODS
Chemicals
Detailed information on chemicals and reagents was shown in the Supporting Materials (Text S1).
Synthesis of catalysts
Synthesis of Fe3O4
Fe3O4 was prepared using a solvothermal method (Dolatabadi et al. 2023). Specifically, 1.2 g FeCl3·6H2O and 0.5 g Na3C6H5O7·2H2O were dissolved in 30 mL ethylene glycol (EG) solution at 35 °C by magnetic stirring. Then 3.04 g CH3COONa was added into the above solution and stirred for 1 h to obtain the precursor. The precursor was poured into a 50 mL Teflon-lined stainless steel autoclave and placed in a constant temperature oven at 200 °C for 12 h. The resulting solid products were collected, washed alternately with deionized water and anhydrous ethanol and then dried in a vacuum drying oven at 55 °C for 10 h to obtain Fe3O4 nanoparticles.
Synthesis of MoS2/Fe3O4
First, 1 mmoL of (NH4)6Mo7O24·4H2O was added to 35 mL deionized water by stirring. Different doses of Fe3O4 (0.1, 0.15 and 0.2 g) were added and stirred for 30 min. Then, 2.66 g thiourea was added to the above mixture and stirred again for 30 min. The mixture was then transferred into a 50 mL Teflon-lined stainless steel autoclave and placed in a constant temperature oven at 180 °C for 12 h. Then the solid products were collected and washed alternately with deionized water and anhydrous ethanol. After drying at 55 °C for 10 h in a vacuum drying oven, MoS2/Fe3O4 was obtained, which was labeled as FeM-0.1, FeM-0.15, and FeM-0.2 based on the dosage of Fe3O4, respectively.
Characterization
Detailed information on characterization was given in the Supporting Materials (Text S2).
Experimental procedure
400 mL TC (20 mg·L−1) was added into a 500 mL beaker and placed in a thermostatic water bath (25 °C) for magnetic agitation. The initial pH of the solution was adjusted using 0.1 mol·L−1 HCl or 0.1 mol·L−1 NaOH, and then 4 mmol·L−1 PS and 0.4 g·L−1 MoS2/Fe3O4 were added to initiate the reaction. At an interval of 10 min, a 5 mL aliquot sample was collected and filtered by a 0.22 μm filter. At the same time, the reaction was immediately terminated with 0.5 mL Na2S2O3 (0.1 mol·L−1). After the reaction, MoS2/Fe3O4 was recovered and washed alternately with deionized water and anhydrous ethanol three times. Finally, it was dried under vacuum conditions at 60 °C for the recycling experiment.
Analysis methods
TC concentration was detected using a UV-visible spectrophotometer (UV-1800, AUCY Scientific, China) at 357 nm (Fatimah et al. 2023). The pH meter (Five Easy Plus, Mettler-Toledo, China) was used to determine the pH of the solution. The inductively coupled plasma spectrometry (ICP, Agilent 700 ICP-OES, USA) was used to detect the concentrations of the leaching metal ions in the liquid in the reaction system. Electron paramagnetic resonance spectroscopy (EPR, Bruker A300, Germany) was used to analyze the active species for TC degradation. The degradation intermediates of TC were identified using a liquid chromatography-mass spectrometer (LC-MS, Waters 2695, USA) and the details are shown in Text S3. The calculation method and data for the Fukui function are presented in Text S4. Details of the degradation intermediates and their toxicity assessment are summarized in Text S5 and Table S3.
Effect of loading dose of Fe3O4 in MoS2/Fe3O4 on TC degradation
As presented in Fig. S1, the degradation efficiency of TC was 91.8% in the MoS2/Fe3O4 system, while it decreased to 85.5 and 81.1% in the FeM-0.1 and FeM-0.2 systems, respectively. It was reported that the specific surface area and the number of active sites increased with the rising addition of Fe3O4, consequently improving the TC degradation efficiency (Li et al. 2023e; Sun et al. 2020). However, excessive Fe3O4 might cover the reactive active site on the MoS2 surface and influence the electron transfer (Song et al. 2023). As a result, FeM-0.15 was chosen for further investigation, and MoS2/Fe3O4 was referred to FeM-0.15 in the following discussion.
RESULTS AND DISCUSSION
Characterization of MoS2/Fe3O4
(a) SEM image of Fe3O4; (b) SEM image of MoS2; (c) SEM image of MoS2/Fe3O4; (d) TEM image of MoS2/Fe3O4; (e) HRTEM image of MoS2/Fe3O4.
(a) SEM image of Fe3O4; (b) SEM image of MoS2; (c) SEM image of MoS2/Fe3O4; (d) TEM image of MoS2/Fe3O4; (e) HRTEM image of MoS2/Fe3O4.
(a) EDS pattern of MoS2/Fe3O4; (b–f) EDS element mapping of Mo, Fe, O and S of MoS2/Fe3O4.
(a) EDS pattern of MoS2/Fe3O4; (b–f) EDS element mapping of Mo, Fe, O and S of MoS2/Fe3O4.
XPS spectra of the MoS2/Fe3O4: (a) Fe 2p, (b) Mo 3d, (c) O 1s and (d) S 2p.
TC degradation
PS activation performance of MoS2/Fe3O4
Comparison of this work with previous reports on the use of different catalysts to degrade TC
Catalyst . | Preparation method . | Structure . | Dosage (g·L−1) . | TC (mg·L−1) . | Reaction time (min) . | Removal efficiency (%) . | Reference . |
---|---|---|---|---|---|---|---|
Fe3O4 | Commercial purchase | Nanoparticles | 1 | 100 | 90 | 89.0 | Hou et al. (2012)) |
Fe3O4@JDC | Soaking and calcination method | Nanoparticles anchored on the surface of tubular JDC | 0.1 | 10 | 60 | 90.2 | Zhang et al. (2023c) |
BC300-MoS2-1 | Hydrothermal method | Flower-like cluster framework | 0.05 | 20 | 120 | 78 | Su et al. (2022) |
FeOOH@MoS2 | Hydrothermal-chemical deposition method | Nanosheet structure | 0.4 | 50 | 30 | 85 | Yi et al. (2021) |
Fe3O4/CoS2 | Hydrothermal method | Regular polyhedron | 0.1 | 20 | 20 | 88.3 | Qiu et al. (2024) |
C@Fe3O4 | One-pot hydrothermal method | Hollow sphere structure | 0.5 | 100 | 90 | 78.5 | Peng et al. (2018) |
CoFe2O4@MoS2 | One-pot hydrothermal method | Nanoparticles distributed in the MoS2 flakes | 0.2 | 10 | 30 | 80.4 | Peng et al. (2022) |
Fe3O4/MoS2 | Hydrothermal method | Nanoparticles attached on the surface of MoS2 | 0.4 | 20 | 60 | 91.8 | This work |
Catalyst . | Preparation method . | Structure . | Dosage (g·L−1) . | TC (mg·L−1) . | Reaction time (min) . | Removal efficiency (%) . | Reference . |
---|---|---|---|---|---|---|---|
Fe3O4 | Commercial purchase | Nanoparticles | 1 | 100 | 90 | 89.0 | Hou et al. (2012)) |
Fe3O4@JDC | Soaking and calcination method | Nanoparticles anchored on the surface of tubular JDC | 0.1 | 10 | 60 | 90.2 | Zhang et al. (2023c) |
BC300-MoS2-1 | Hydrothermal method | Flower-like cluster framework | 0.05 | 20 | 120 | 78 | Su et al. (2022) |
FeOOH@MoS2 | Hydrothermal-chemical deposition method | Nanosheet structure | 0.4 | 50 | 30 | 85 | Yi et al. (2021) |
Fe3O4/CoS2 | Hydrothermal method | Regular polyhedron | 0.1 | 20 | 20 | 88.3 | Qiu et al. (2024) |
C@Fe3O4 | One-pot hydrothermal method | Hollow sphere structure | 0.5 | 100 | 90 | 78.5 | Peng et al. (2018) |
CoFe2O4@MoS2 | One-pot hydrothermal method | Nanoparticles distributed in the MoS2 flakes | 0.2 | 10 | 30 | 80.4 | Peng et al. (2022) |
Fe3O4/MoS2 | Hydrothermal method | Nanoparticles attached on the surface of MoS2 | 0.4 | 20 | 60 | 91.8 | This work |
(a) Removal efficiency of TC in various systems. (b) Kinetics of TC degradation. Reaction conditions: [MoS2/Fe3O4] = 0.4 g·L−1, [PS] = 4 mmoL, [TC] = 20 mg·L−1, initial pH = 7, T = 25 °C and reaction time = 60 min.
(a) Removal efficiency of TC in various systems. (b) Kinetics of TC degradation. Reaction conditions: [MoS2/Fe3O4] = 0.4 g·L−1, [PS] = 4 mmoL, [TC] = 20 mg·L−1, initial pH = 7, T = 25 °C and reaction time = 60 min.
Effects of MoS2/Fe3O4 dosage and PS concentration
Effects of reaction parameters on the TC degradation: (a) the dosage of MoS2/Fe3O4; (b) PS concentration; (c) initial pH; (d) Zeta potential of MoS2/Fe3O4 at different pH values; (e) temperature; (f) inorganic anions (Cl−, ,
) and HA. Reaction conditions: [MoS2/Fe3O4] = 0.4 g·L−1, [PS] = 4 mmoL, [TC] = 20 mg·L−1, initial pH = 7, T = 25 °C and reaction time = 60 min.
Effects of reaction parameters on the TC degradation: (a) the dosage of MoS2/Fe3O4; (b) PS concentration; (c) initial pH; (d) Zeta potential of MoS2/Fe3O4 at different pH values; (e) temperature; (f) inorganic anions (Cl−, ,
) and HA. Reaction conditions: [MoS2/Fe3O4] = 0.4 g·L−1, [PS] = 4 mmoL, [TC] = 20 mg·L−1, initial pH = 7, T = 25 °C and reaction time = 60 min.

Effects of pH and temperature
At initial pH values of 3, 5 and 7, the removal efficiencies of TC were 94.8, 93.5 and 91.8%, respectively. However, at initial pH values of 9 and 10, the removal efficiencies of TC decreased to 84.2 and 42.9%, respectively (Figure 7(c)). It was thus inferred that the favorable TC degradation in the MoS2/Fe3O4/PS system could be achieved within the initial pH range of 3–7. The pH in the MoS2/Fe3O4/PS system was monitored during the reaction, as shown in Figure S2. When initial pH values were 3, 5 and 7, the pH dropped below 3.30 after 10 min and continued to decrease in subsequent reactions. When initial pH values were 9 and 10, the pH remained above 3.89 within 60 min. The zero charge point (pHPZC) of the MoS2/Fe3O4 surface was approximately 3.66 (Figure 7(d)). Therefore, the MoS2/Fe3O4 surface carried positive charges after 10 min at initial pH values lower than 7, and with negative charges was easily adsorbed on the surface of MoS2/Fe3O4, which promoted the activation of PS and improved the removal efficiency of TC (Zhu et al. 2022). However, the MoS2/Fe3O4 surface carried negative charges within 60 min at initial pH values higher than 9, and the electrostatic repulsion prevented the contact between MoS2/Fe3O4 and PS, inhibiting the degradation of TC (Luo et al. 2021b).
The effect of temperature on TC degradation in the MoS2/Fe3O4/PS system is shown in Figure 7(e). When the temperature was increased from 25 to 35 °C, the removal efficiencies of TC were higher than 90% after 40 min. When the temperature raised to 45 and 55 °C, the removal efficiency reached 90% after 30 min. It indicated that the elevated temperature could accelerate TC degradation in the MoS2/Fe3O4/PS system. Elevated temperature was conducive to the thermal decomposition of PS, promoting the generation of (Guo et al. 2016; Zhang et al. 2022b).
Effects of inorganic anions and humic acid





















Mechanisms
EPR testing and quenching experiment






















(a–c) EPR spectrum in the MoS2/Fe3O4/PS system; (d) Effects of radical scavengers on the degradation of TC. Reaction conditions: [MoS2/Fe3O4] = 0.4 g·L−1, [PS] = 4 mmoL, [TC] = 20 mg·L−1, initial pH = 7, T = 25 °C and reaction time = 60 min.
(a–c) EPR spectrum in the MoS2/Fe3O4/PS system; (d) Effects of radical scavengers on the degradation of TC. Reaction conditions: [MoS2/Fe3O4] = 0.4 g·L−1, [PS] = 4 mmoL, [TC] = 20 mg·L−1, initial pH = 7, T = 25 °C and reaction time = 60 min.


















XPS analysis

XPS spectra of MoS2/Fe3O4 before and after use: (a) Mo 3d; (b) Fe 2p; (c) O 1s; (d) S 2p.
XPS spectra of MoS2/Fe3O4 before and after use: (a) Mo 3d; (b) Fe 2p; (c) O 1s; (d) S 2p.
Mechanisms of MoS2/Fe3O4 activating PS








TC degradation pathway and toxicity assessment
Fukui index was employed to analyze the main sites for reactive species on TC molecule on the basis of the Density functional theory (DFT) calculation (Cheng & Ji 2022; Wu et al. 2022). In this study, ,
,
and
were the attack species participated in TC degradation in the MoS2/Fe3O4/PS system. Therefore,
and
were taken into account to predict the active site of TC in the degradation process (Chen et al. 2023). Fukui index (
,
and
) and the Natural Population Analysis (NPA) charges distribution are shown in Table S1, and the optimized structure of TC is listed in Fig. S3. Generally, the Fukui exponent values with larger
and
were more likely to be attacked by
and free radicals (
, ·OH,
), respectively (Cheng & Ji 2022). Hence, N14 (0.0581) and C7 (0.0751) with higher
was susceptible to
attack. C7 (0.0455) and C12 (0.0401) with larger
values were vulnerable reaction sites for
,
,
attack, which could undergo oxidation, hydroxylation, and double bond breaking (Zhang et al. 2023b). Although C12 had a larger
value, the attack of free radicals was difficult because of the saturated sites and steric hindrance (Wu et al. 2023).





Acute toxicity of (a) fish LC50 (96 h); (b) daphnid LC50 (48 h); (c) green algae EC50 (48 h); chronic toxicity of (d) fish; (e) daphnid; (f) green algae.
Acute toxicity of (a) fish LC50 (96 h); (b) daphnid LC50 (48 h); (c) green algae EC50 (48 h); chronic toxicity of (d) fish; (e) daphnid; (f) green algae.
Recyclability and stability of MoS2/Fe3O4
(a) Recyclability experiments of MoS2/Fe3O4 and (b) leaching concentration of metal ions.
(a) Recyclability experiments of MoS2/Fe3O4 and (b) leaching concentration of metal ions.
CONCLUSIONS
In this work, MoS2/Fe3O4 was successfully prepared as PS activator to improve the removal of TC and the separation of MoS2. Compared to single Fe3O4 and MoS2, MoS2/Fe3O4 had higher activation activity for PS activation, achieving a removal efficiency of 91.8% for 20 mg·L−1 TC after 60 min under the conditions of 0.4 g·L−1 MoS2/Fe3O4, 4 mmoL PS and pH 7. MoS2/Fe3O4 also demonstrated favorable stability during recycling experiments. The degradation process of TC in the MoS2/Fe3O4 system could be divided into free radical decomposition pathway (,
and
) and non-free radical decomposition pathway (
), with
,
and
identified as the predominant active species. The three major degradation pathways of TC in the MoS2/Fe3O4 system were proposed as hydroxylation, ketylation of dimethylamino group and C-N bond breaking. The toxicity of TC and its intermediates could be effectively reduced by using MoS2/Fe3O4 to activate PS. In conclusion, MoS2/Fe3O4 prepared was an efficient PS activator to promote the decomposition of refractory organic pollutants.
ACKNOWLEDGEMENTS
This work was supported by the Science and Technology Development Program of Jilin Province, China (Nos. 20230203168SF).
AUTHOR CONTRIBUTIONS
L. Z. arranged the resources, wrote the review and edited the article, and rendered support in funding acquisition. Q. Z. investigated the work, rendered support in data curation, and wrote the original draft. T. C. investigated the work and rendered support in data curation. C. W. investigated the work and rendered support in data curation. C. X. investigated the work and edited the article. J. G. supervised the work, arranged the resources, and wrote the review. X. P. investigated the work and edited the article. S. L. validated the article and visualized the data.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.
REFERENCES
Author notes
Lanhe Zhang and Qi Zhang are co-first authors.