Synergistic oxidation and adsorption of arsenic(III) by epigallocatechin gallate activated iron(III) peroxydisulfate system Authors: Dun Fu, Yongqing Wang, Tong Wu, Sheng He, Shuxin Han, Mengting Zhu Published: 2026 ABSTRACT Groundwater contamination in mining areas has become more severe, particularly due to elevated arsenic (As) pollution risks. Advanced oxidation processes by peroxydisulfate (PDS-AOPs) have demonstrated excellent efficiency in stubborn contaminants treatment. In this study, we propose a synergistic strategy involving polyphenol species and Fe(III) to enhance PDS activation, significantly enhancing arsenite (As(III)) oxidation and adsorption. The findings demonstrated that As(III) removal efficiency could achieve 98.2% within 10 min under the experimental condition (0.15 mM Fe(III), 0.2 mM PDS, and 0.02 mM epigallocatechin gallate (EGCG)), representing an increase of 84.4% compared to the Fe(III)/PDS without EGCG. This phenomenon is primarily due to the capacity of EGCG to chelate Fe(III) and facilitate its reduction, promoting the interconversion of Fe(III) and Fe(II) through enhanced redox cycling. The process in turn boosts the activation of PDS, leading to the production of reactive oxygen species such as SO4•⁻,·OH and O2•⁻, all of which contribute synergistically to the oxidation of As(III).This study also adopted density functional theory (DFT) calculation, transmission electron microscopy (TEM) and X-ray photoelectron spectroscopy (XPS) analysis to revealing the mechanism of As(III) removal. Besides, the As(V) adsorbed onto iron-based aggregate through the reaction is in the pentavalent state, indicating that the reaction system strengthens the oxidation of As(III). This study achieved efficient removal of As(III) and met the Chinese national standard of 0.01 mg/L, which provide the theoretical foundation for utilizing iron-based catalysts in PDS activation technology.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-57863-x. Introduction Arsenic (As) levels in groundwater can be elevated by the weathering of As-bearing minerals, as well as by mining activities that use As-containing reagents or disturb As-rich ores and rocks1. Extended contact with high concentrations of As in the environment can lead to As poisoning, cancer and other health problems2. In Northern China, where the coal industry is relatively developed, the area affected by high-risk As pollution in groundwater has exceeded 3,000 km23. As(III), highly toxic than As(V), is commonly found in groundwater. As(III) predominantly is present as H3AsO3 in groundwater, making it difficult to remove with conventional techniques like adsorption and coagulation4. To decrease the harmful effects of water with As(III) contamination, formulating effective methods to convert As(III) to As(V) is of critical importance. Over recent years, advanced oxidation processes by peroxydisulfate (PDS-AOPs) have demonstrated excellent efficiency in removing complex and refractory pollutants from wastewater5. This technology degrades and mineralizes stubborn pollutants by creating reactive species via disrupte of O-O bond in PDS or peroxymonosulfate (PMS), which have been widely used as oxidants in PDS-AOPs. In particular, PDS could be activated using heterogeneous catalysts (e.g., MnO2, nZVI, CuFe2O4 spinel) or homogeneous metal ions (e.g., Fe(II), Co(II), Cu(II)), demonstrating great potential for the removal of obstinate contaminant6,7. Among these, the development of Fe(II)-activated PDS has achieved a high degree of maturity. Nevertheless, the stringent requirement for acidic pH conditions (pH ~ 3) and the generation of substantial iron sludge limit its broader application8. In contrast, Fe(III) activation offers distinct advantages in terms of broader pH adaptability, the capacity for sustained and gradual radical release, making it particularly suitable for long-term applications such as groundwater contamination remediation9. Nevertheless, this system still encounters critical challenges, such as a slow initial reaction rate, dependence on naturally occurring reducing agents in the environment for efficient iron cycling (Fe(III)/Fe(II)), and an unstable cycling mechanism, which impede further enhancement and precise control of its treatment efficiency. Natural polyphenols (NPs) have garnered significant attention due to their distinct advantages in green sustainability, environmental friendliness, and multifunctional synergistic effects10. These compounds not only efficiently facilitate the reduction of Fe(III) to Fe(II), but also exhibit low or negligible toxicity in both themselves and their degradation products, minimizing the risk of secondary pollution. Furthermore, NPs possess inherent complexation abilities, enabling them to form stable complexes with iron ions11. This characteristic substantially expands the effective pH range of the system, allowing for efficient operation even under near-neutral environmental conditions. In addition, certain NPs can function as sacrificial electron donors by competing with pollutants for free radical12. This mechanism steers the radical chain process to enhance the generation of highly reactive intermediate species, while also directly activating PDS through quinone intermediates, thereby creating multiple synergistic activation pathways. As a result, NPs deliver higher efficiency, longer durability and better environmental friendliness for pollutant degradation than single chemical reducing agents. For example, zero-valent iron nanoparticles green-synthesized from leaf extracts possess excellent As(III) oxidation capacity, and those derived from guava leaves present optimal arsenic conversion performance across different pH environments13. Furthermore, acting as analogs of humic substances, NPs can directly oxidize As(III) to As(V) through semiquinone radicals and reactive oxygen species generated under neutral and alkaline conditions14. At this point, this research seeks to explore the enhancement of As(III) oxidation and adsorption using NPs-promoted Fe(III)/PDS system. The work systematically evaluates the collaborative effects of NPs and Fe(III) on PDS activation and elucidates the mechanisms underlying As(III) oxidation and immobilization. Key operational parameters are optimized to maximize arsenic removal efficiency. Furthermore, density functional theory (DFT) calculations, transmission electron microscopy (TEM), and X-ray photoelectron spectroscopy (XPS) were used to examine the mechanisms underlying As(III) removal. Materials and methods Materials Gallic acid (GA), tea polyphenols (TP), procyanidin (PC), luteolin (LT), and epigallocatechin gallate (EGCG) were procured from Shanghai Titan Technology Co., Ltd. (China). PDS, FeCl3, KBH4 and NaAsO2 were obtained from Sinopharm Chemical Reagent Co., Ltd. (China). All experiments for As(III) removal were conducted using ultra-pure water. Methods Experiments Batch experiments were carried out using a set of 500 mL glass vials. Five types of polyphenols (GA, TP, PC, LT, and EGCG) were selected for comparative experimental analysis. Fe(III)-polyphenol complexes work solution were prepared as follows: At room temperature, 1 mg/L As(III) and 0.15 mM Fe(III) solutions were thoroughly mixed in 300 mL reaction solution. Subsequently, the reaction was initiated by adding different types of polyphenols (0.02 mM) and PDS (0.2 mM) to the mixed solution, respectively. Additionally, under the same reaction system, no polyphenols were added, which served as the blank control. Each treatment is conducted in triplicate. The experiments were conducted over a 10 min period, with samples periodically collected and filtered for analysis. The study systematically investigated the impact of key parameters on As(III) removal. The influence of common coexisting ions (Cl−, HCO3−, SO42−, H2PO4−) was also evaluated to simulate real mine drainage conditions. Additionally, the role of dissolved oxygen (DO) was examined by comparing reactions performed under aerobic conditions with those carried out under an N2-purged anaerobic environment. ROS identification To identify the ROS generated in reaction system, quenching experiment was carried out using tert-butanol (TBA) and methanol (MeOH). Prior to the addition of EGCG and PDS, 500 mM of either TBA or MeOH was introduced into the reaction mixture under consistent conditions. Furthermore, electron paramagnetic resonance (EPR) spectroscopy was utilized to detect the presence of radical species, using a Bruker EMXplus-6/1 spectrometer (Germany). Analytical techniques The As(III) concentration was determined by hydride generation atomic fluorescence spectrometry (HG-AFS, SA-20, China). Fe(II) and Fe(III) species were tested via the 1,10-phenanthroline method. PDS concentration was measured by UV-Vis spectroscopy at 352 nm based on its complexation with iodide in the presence of NaHCO3. The crystallinity of iron-containing aggregates was characterized by XRD (Rigaku SmartLab SE, Japan). The morphological characteristics were examined via SEM (ZEISS Sigma 300, Germany) and TEM (FEI Talos F200X, USA), and the surface elemental composition was characterized using XPS (Thermo Scientific K-Alpha, USA). Computational analysis First-principles DFT calculation was performed using the Vienna Ab Initio Simulation Package, employing plane-wave basis sets and the projector-augmented wave method. Exchange-correlation interactions were treated with the Perdew-Burke-Ernzerhof generalized gradient approximation, while van der Waals forces were accounted for using Grimme’s DFT-D3 model. Lattice parameters were a = 16.2 Å, b = 16.2 Å, c = 21.7 Å with a 450 eV energy cutoff and a 2 × 2 × 1 Gamma k-point mesh. Structural optimization was carried out until the forces on atoms fell below 0.02 eV/Å, using an energy convergence threshold of 10− 5 eV. Results and discussion Evaluation of As(III) removal in NPs/Fe(III)/PDS system As(III) removal by various NPs GA, TP, PC, LT and EGCG were separately incorporated to Fe(III)/PDS system to assess their effects on As(III) removal efficiency. The addition of NPs could significantly enhance the As(III) removal, as illustrated in Fig. 1a. More specifically, EGCG exhibits best performance achieving a removal rate of As(III) of 98.2% after 10 min, which represents an increase of 84.4% and 92.7%, compared to Fe(III)/PDS and EGCG/PDS system. The data imply that neither Fe(III) nor EGCG are capable of effectively activating PDS. Thus, the observed phenomenon could be explained by the potent reducing capability of EGCG, thereby enabling the reduction of Fe(III) to Fe(II) and consequently promotes PDS activation15. At the same time, the structures of NPs contain at least two ortho- or para-hydroxyl functional groups, which could enable strong chelating reduction interactions with metal ions16. Meanwhile, EGCG contains six ortho-phenolic hydroxy groups, which significantly enhance its redox activity. Bu et al.17 reported that these ortho- and para-dihydroxy benzene functionalities can facilitate Fe(III)/Fe(II) cycle via electron shuttling, thereby enhancing PDS activation and radical generation. PDS concentration on As(III) removal The influence of various PDS concentration were investigated on As(III) removal in EGCG/Fe(III)/PDS reaction system under the concentrations of EGCG and Fe(III) held constant at 20 µM and 0.15 mM, as shown in Fig. 1b. Under the influence of 0.1 mM to 0.2 mM PDS concentration, the removal rate of As(III) rose from 80.2% to 89.2%. This is because higher concentrations of PDS could produce more active species, which is conducive to the oxidative removal of As(III)18. However, as the PDS increased from 0.25 mM to higher levels (0.5 mM and 1 mM), the As(III) removal was not enhanced as PDS concentration increases. It may be due to the enhancement of the side reaction between excessive PDS and ·OH19, decreasing the availability of ·OH for As(III) oxidation. Therefore, 0.2 mM was identified as the optimal concentration of PDS in EGCG/Fe(III)/PDS system. In the optimum reaction system, the remaining of SO42−, the primary by-product of PDS, revealed that SO42− concentration rapidly increased to 32 mg/L as the reaction proceeded (Fig. S1). SO42− generation exhibited a positive correlation with As(III) removal efficiency, indicating that PDS was quickly activated and decomposed in the EGCG-mediated Fe(III) system. Notably, SO42− are ubiquitous in natural water bodies and do not exert adverse effects on water quality. Fe(III) concentration As illustrated in Fig. 1c, the removal rate of As(III) rose markedly from 18.3% to 91.8% as the Fe(III) increased within the range of 0.05 to 0.15 mM. It’s been studied that the chelation of EGCG with Fe(III) can efficiently inhibit the precipitation of Fe(III) and rapidly facilitate its reduction to Fe(II)20. Therefore, with the increase of Fe(III) content, more EGCG-Fe(III) chelates will be produced, which will enhance molecular collision rates and promote PDS decomposition within the reaction system. Nevertheless, upon increasing the Fe(III) concentration to 0.2 mM, the As(III) removal showed no significant improvement. It is possible that following the reduction of surplus Fe(III) while the generated Fe(II) could potentially interact with the active compound, leading to an ineffective utilization of the substance21. To ensure that the system meets the low economic cost and stable removal effect, 0.15 mM Fe(III) was identified as the optimal parameter for the subsequent research. EGCG concentration EGCG plays a key role for enhancing Fe(II)/Fe(III) redox cycling while its concentration requires careful optimization. As illustrated in Fig. 1d, as the EGCG concentration rose from 10 µM to 20 µM, the efficiency of As(III) removal surpassed 90.0%. And the further increase of EGCG dosage to 100 µM showed a negative effect on As(III) removal. This is because EGCG will also be oxidized by some active substances, for example, ·OH and SO4• -22, which may impair As(III) removal, particularly with excessive EGCG dosing. EGCG can be degraded into dozens of complex by-products, mainly including gallic acid, degalloylated flavanols, and their corresponding ortho-quinone derivatives23. The first two groups pose negligible risks to environmental safety, while ortho-quinone derivatives have potential high risks. Additionally, this study focuses on the efficient removal of As(III) within a short reaction time of 10 min, and it is believed that extending the reaction time will significantly reduce the potential risks of ortho-quinone by-products. Fig. 1(a) As(III) removal efficiency by various NPs; the effect of (b) PDS, (c) Fe(III), and (d) EGCG concentration on As(III) removal. (Experimental conditions: [polyphenols]0=0.02 mM, [Fe(III)]0=0.15 mM, [PDS]0=0.2 mM, [As(III)]0=1 mg/L.) (a) As(III) removal efficiency by various NPs; the effect of (b) PDS, (c) Fe(III), and (d) EGCG concentration on As(III) removal. (Experimental conditions: [polyphenols]0=0.02 mM, [Fe(III)]0=0.15 mM, [PDS]0=0.2 mM, [As(III)]0=1 mg/L.) Effects of pH, DO and co-existing anions pH Given that As(III) predominantly exists in the form of soluble H3AsO3 across a broad range of pH, this study systematically examined the influence of pH on its removal. In the preliminary pH interval ranging from 3 to 7 (Fig. 2a), the removal efficiencies of As(III) were 22.7%, 82.1%, 84.3%, 60.5% and 22.1%, respectively. In comparison with neutral and alkaline conditions, the As(III) removal is better with acidic condition, in agreement with previous study of Li et al.22. The EGCG/Fe(III)/PDS has shown inhibitory effects on As(III) removal in neutral environment because Fe(III) rapidly precipitates in the form of iron hydroxide precipitation led to reduction of iron activated PDS efficiency24. While under extreme acidic conditions, the generation of (Fe(II)(H2O))2 is leads to a reduction in the activity and effectiveness of Fe(II) in the system25, which reduces PDS activation efficiency and fewer active species, impairing As(III) oxidation removal. From Fig. 2b, FeOH2+ is the predominant Fe(III) species identified, accounting for approximately 75% at pH 3. Furthermore, as shown in Fig. 2c, Fe2+ is the dominant Fe(II) species within the pH range of 1–7, indicating that EGCG participates in reducing of Fe(III) to Fe(II) across reactions (1) and (2). 1 2 DO condition O2 plays an important role in the operation of PDS-activated system. Compared with anaerobic conditions, the existence of O2 promotes the production of highly reactive oxidative species, thereby increasing the overall oxidation capacity and As(III) oxidation efficiency17. Under air-saturated conditions, the removal rate of As(III) reached 89.9% within 10 min, as illustrated in Fig. 2d. Under hypoxic conditions, the removal rates dropped to 74.5%, indicating the participation of O2 in PDS activation, which promotes the oxidative elimination of As(III). The evolution of superoxide radicals (O2•-) could be the primary factor enhancing removal efficiency (Eqs. (3) and (4))26.3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{Fe(II)}} + {\mathrm{O}}_{{\mathrm{2}}} \to {\mathrm{O}}_{{\mathrm{2}}}^{{ \cdot - }} + {\mathrm{Fe(III)}}$$\end{document}4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{O}}_{{\mathrm{2}}} ^{{\cdot - }} + {\mathrm{S}}_{{\mathrm{2}}} {\mathrm{O}}_{{\mathrm{8}}} ^{{{\mathrm{2}} - }} \to {\mathrm{SO}}_{{\mathrm{4}}} ^{{\cdot - }} + {\mathrm{SO}}_{{\mathrm{4}}} ^{{{\mathrm{2}} - }} + {\mathrm{O}}_{{\mathrm{2}}}$$\end{document} Co-existing anions Figure 2e presents an investigation into the impact of co-existing anions on As(III) removal in the EGCG/Fe(III)/PDS reaction system. The existence of Cl- has a negligible effect on As(III) removal. Within the concentration range of 0–1 mM, the addition of SO42-, HCO3− and H2PO4- resulted in a decrease in As(III) removal rates by 28.4%, 86.5%, and 63.8%, respectively. Especially the presence of HCO3− and H2PO4−, further elevation of concentration led to nearly complete inhibition of As(III) removal. This can mainly be explained by the consumption of PDS by inorganic anions, which directly limits the generation of reactive species27. Moreover, complexes formed between inorganic anions and iron ions also decrease the effective concentration of Fe(II), which hinders PDS activation and subsequent production of radicals, ultimately reducing As(III) oxidation7. Fig. 2Effects of (a) pH; (b) Fe(III) and (c) Fe(II) species distribution at various pH; Effect of (d) DO, (e) co-existing anions on As(III) removal in EGCG/Fe(III)/PDS system; (f) The concentration changes of EGCG and Fe(II) in the EGCG/Fe(III)/PDS reaction system. (Experimental conditions: [EGCG]0=0.02 mM, [Fe(III)]0=0.15 mM, [PDS]0=0.2 mM, [As(III)]0=1 mg/L.) Effects of (a) pH; (b) Fe(III) and (c) Fe(II) species distribution at various pH; Effect of (d) DO, (e) co-existing anions on As(III) removal in EGCG/Fe(III)/PDS system; (f) The concentration changes of EGCG and Fe(II) in the EGCG/Fe(III)/PDS reaction system. (Experimental conditions: [EGCG]0=0.02 mM, [Fe(III)]0=0.15 mM, [PDS]0=0.2 mM, [As(III)]0=1 mg/L.) Fe(II) content The levels of Fe(II) as well as residual EGCG in the EGCG/Fe(III)/PDS system were analyzed (Fig. 2f). Fe(II), the reductive by-product of Fe(III), was quantified, showing that its concentration rose from approximately 1 µM to 3 µM over the reaction period. This confirms that Fe(III) was reduced to Fe(II) through multiple pathways, and the produced Fe(II) further promoted PDS activation and As(III) oxidation. Meanwhile, EGCG concentration decreased from 20 µM to 2 µM within 10 min, indicating that EGCG acted as an effective electron donor to drive Fe(III) reduction and sustain Fe(III)/Fe(II) redox cycling. This cycling enhanced the system’s oxidative capacity, reduced reliance on external iron, and inhibited iron sludge formation, thus improving environmental compatibility and process sustainability28. Detection of ROS Quenching experiment By conducting quenching experiments, MeOH was selected to quench SO4• - and ·OH radicals while TBA was used as a scavenger specifically for ·OH29. As illustrated in Fig. 3, MeOH and TBA have different degrees of inhibition on As(III) removal. However, the inhibitory capacity of MeOH is notably more potent than that of TBA. When the concentration of TBA increases from 0 to 100 mM, the As(III) removal efficiency decreases from 94.3% to 86.4% (Fig. 3a). As the MeOH concentration increased from 0 to 100 mM, the removal rate of As(III) significantly declined from 94.3% to 71.5% (Fig. 3b). This observation suggests that both SO4• - and ·OH contribute to promoting As(III) removal within the EGCG/Fe(III)/PDS reaction system. Fig. 3Effects of TBA (a) and MeOH (b) on As(III) removal in the EGCG/Fe(III)/PDS reaction system. (Experimental conditions: [EGCG]0=0.02 mM, [Fe(III)]0=0.15 mM, [PDS]0=0.2 mM, [As(III)]0=1 mg/L.) Effects of TBA (a) and MeOH (b) on As(III) removal in the EGCG/Fe(III)/PDS reaction system. (Experimental conditions: [EGCG]0=0.02 mM, [Fe(III)]0=0.15 mM, [PDS]0=0.2 mM, [As(III)]0=1 mg/L.) Notably, even at a high concentration of 100 mM MeOH, only partial inhibition was observed, indicating the existence of additional reaction pathways beyond free radical oxidation. MeOH primarily scavenges free SO4• - and ·OH in the bulk solution, but it has limited access to reactive species adsorbed on the surface of iron-based aggregates. In this system, EGCG-promoted Fe(III)/Fe(II) cycling continuously generates surface-bound radicals30, which are less affected by aqueous radical scavengers. Moreover, direct interfacial electron transfer between As(III) and iron-based aggregates also contributes to As(III) oxidation. These non-radical pathways are insensitive to MeOH, which explains the residual removal efficiency under high MeOH concentrations. Combined with EPR and XPS results, both free radical oxidation and surface-mediated non-radical pathways jointly govern As(III) removal in the EGCG/Fe(III)/PDS system. EPR analyses As shown in Fig. 4a, the characteristic spectrum of DMPO-·OH adduct exhibits the features of a 1:2:2:1 quartet31; meanwhile, the DMPO- SO4• - adduct’ s spectrum displays comparatively low intensity, which can be attributed to the rapid attenuation of its signal over time32. In the EGCG/Fe(III)/PDS reaction system, the signal intensities of DMPO-·OH and DMPO-SO4• - were higher than those in the Fe(III)/PDS reaction system. This result indicates that EGCG facilitates the generation of SO4• - and ·OH within the EGCG/Fe(III)/PDS reaction system. The weak signals of DMPO-·OH and DMPO-SO4• - were detected with single PDS33. Fig. 4EPR spectra of (a) DMPO-·OH, DMPO-SO4•− and (b) DMPO-O2•− in EGCG/Fe(III)/PDS, Fe(III)/PDS and PDS reaction system. EPR spectra of (a) DMPO-·OH, DMPO-SO4•− and (b) DMPO-O2•− in EGCG/Fe(III)/PDS, Fe(III)/PDS and PDS reaction system. In addition, the EPR spectra revealed the presence of a characteristic peak corresponding to O2• - (Fig. 4b). This could be attributed to the reaction between Fe(II), produced when EGCG reduces Fe(III), and dissolved oxygen, resulting in the generation of O2• -. As a result, O2• - efficiently promotes the activation of S2O82− through electron transfer, leading to the production of SO4• -34,35. Fig. 5Hypothesized pathway for ROS formation in the EGCG-enhanced Fe(III)/PDS reaction system. Hypothesized pathway for ROS formation in the EGCG-enhanced Fe(III)/PDS reaction system. Furthermore, semiquinone radicals (SQ• -) derived from EGCG can transfer electrons to O2, generating H2O234. The H2O2 produced may be further activated by SQ• -, contributing to sustained radical generation and enhanced oxidative capacity of the system14. On the basis of the EPR analysis, it can be concluded that SO4• -, ·OH and O2• - are the main reactive oxygen species responsible for As(III) oxidation in the EGCG/Fe(III)/PDS reaction system, as illustrated in Fig. 5. Characterization of iron-based aggregate XRD and TEM To confirm contribution of iron-based aggregate to the removal of As(III), the iron-based aggregate produced in EGCG/Fe(III)/PDS reaction system was collected. Physicochemical properties of iron-based aggregate were analyzed by a range of characterization techniques. As displayed in Fig. 6a, comparing the peak position and peak intensity of the obtained XRD pattern with the standard spectrum, the iron-based aggregate produced has higher diffraction peak intensity, indicating that they have better crystallinity. Meanwhile, obvious lattice stripes were also observed in the TEM image (Fig. 6b), which further confirmed the crystal composition of the ferrite. Through phase analysis, the aggregate was mainly composed of Fe3O436. Fig. 6(a) XRD spectra, (b) TEM, (c)SEM and (d) EDS images of iron-based aggregate. (a) XRD spectra, (b) TEM, (c)SEM and (d) EDS images of iron-based aggregate. SEM Figure 6(c-d) depict the surface morphology and energy spectrum of iron-based aggregate. The appearance of the iron-based aggregate shows a massive three-dimensional structure. There are both smooth and rough areas on the surface of the iron-based aggregate. Many particles are stacked on the surface, which increases the specific surface area and exhibits effective adsorption performance, bridging and trapping of heavy metals15. The composition of the main elements on the surface of the aggregate was preliminarily determined by energy spectrum scanning. It was found that the peak intensity of O atom was the strongest, followed by Fe atom, indicating that the content of oxygen and iron in the aggregate was relatively high. It is worth noting that the proportion of As atoms on the aggregate surface, indicating that the aggregate had significant adsorption effect on As. XPS As shown in Fig. 7, the surface elements of iron-based aggregate was analyzed qualitatively and quantitatively by XPS. The main peaks of O 1s were located at 531.65, 530.07 and 533.34 eV, respectively, referring to C = O, O2− and C-O species37. In the Fe 2p spectrum, the peaks observed at 711.51 eV and 724.30 eV suggest the existence of Fe(III)-O species, while the peaks at 715.72 eV and 727.67 eV were attributed to Fe(II)-O species38. This suggests that the aggregate releases Fe(II), and the Fe(III) present on its surface likely originates from PDS oxidation by Fe(II)39. The main peaks in S 2p spectra were 168.68, 170.07, 166.75, 162.80 and 164.27 eV, respectively, corresponding to SO42−, SO3− and S22− species40. Notably, the peaks at 45.71 eV and 44.05 eV are associated with As(V) and As(III), respectively, with the surface proportion of As(V) significantly exceeding that of As(III), indicating that oxidative adsorption has become the dominant mechanism for As(III) removal in the system. Fig. 7XPS spectra of aggregate O 1s (a), Fe 2p (b), S 2p (c), and As 3d (d). XPS spectra of aggregate O 1s (a), Fe 2p (b), S 2p (c), and As 3d (d). DFT study on As(III)/As(V) oxidation adsorption of iron-based aggregate The As complexes formed on the surface of metal oxides exhibit unique and diverse forms of existence when they adsorb As. DFT calculation was employed to investigate the oxidation and adsorption characteristics of As(III) and As(V) on iron-based aggregates. Using Fe as the main adsorption site, DFT was used to study the stable adsorption configuration and surface adsorption energy on the surface of Fe3O4. As shown in Fig. 8, Fe3O4 is more conducive to the adsorption of As(III) and As(V) with bidentate binuclear coordination, and the adsorption energy are − 1.221 eV and − 0.443 eV respectively. The adsorption energy of Fe3O4 for As(III) and As(V) is negative, indicating that the adsorption process releases energy, making the system reach a more stable state and spontaneous adsorption process33. According to the proportion of As(V) on the surface of aggregate is higher than that of As(III), it indicates that bidentate binuclear complexation is more significant in the adsorption of As(V). And it is obvious that bidentate binuclear complexation could efficiently improve the utilization of adsorption sites of aggregate41. The Gibbs free energy change (ΔG) optimized by calculating the energy barrier difference of the adsorption energy reaching the equilibrium structure after the adsorption of As(III) and As(V) reveals the spontaneous oxidation of As(III) by Fe3O4. Results show that the energy barrier difference ΔG of the adsorption energy reaching the equilibrium structure is negative25, indicating that surface reactions involve both As(III) adsorption and its oxidative transformation. Fig. 8Schematic diagram of oxidation and adsorption capacity of Fe3O4 on As(III). Schematic diagram of oxidation and adsorption capacity of Fe3O4 on As(III). Possible As(III) removal mechanism in the EGCG/Fe(III)/PDS reaction system A plausible oxidative removal mechanism for the EGCG/Fe(III)/PDS reaction system is suggested through the analysis and interpretation before, as illustrated in Fig. 9. Under the relative challenge of Fe(III) activating PDS, EGCG enhances the rate of conversion from Fe(III) to Fe(II), through chelating and reducing Fe(III), which boosts the activity and effectiveness of Fe(II) and intensifies PDS activation to yield ·OH, SO4• -, and O2• -17. In EGCG/Fe(III)/PDS reaction system, these active radicals with high oxidation potential oxidize As(III) into As(V), promoting the reduction of As(III) toxicity while being more conducive to fixation41, The key to As(III) removal is the Fe3O4-based aggregate produced by the complexation reaction between EGCG and Fe(III), which could induce heterogeneous Fenton process to adsorb and precipitate As without further alkaline regulation42. The DFT result shows that the iron-based aggregate can spontaneously adsorb oxidized As(III) and combine with As in a bidentate binuclear complex structure, which ultimately greatly reduces the toxicity of As(III) in solution and adsorbs As. Fig. 9Possible oxidative removal mechanism. Possible oxidative removal mechanism. Conclusion This study has simulated the iron source environment of closed pit mine water and constructed a natural polyphenol EGCG to participate in the removal of As(III) through the Fe(III)/PDS reaction system. The findings indicated that As(III) was removed with an efficiency of 98.2%. Due to the close relationship between the form of iron and environmental factors, acidic environments can promote As(III) removal. The presence of co-existing ions could disrupt the generation of reactive species while O2 participation is more conducive to the As(III) removal. It has been found that the coexistence of active substances ·OH, SO4•−, and O2•− can rapidly convert As(III) into As(V) through oxidation, effectively reducing the toxicity of As(III) and facilitating further removal. Thus, efficient removal of As(III) relies critically on the swift oxidation of As(III) by reactive species present in the system. Iron-based aggregate was generated by the complexation reaction between EGCG and Fe(III), serving as excellent adsorption and precipitation carriers of As(III) and As(V), further achieving high efficiency in As(III) removal. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1