Toward sulfur retention and H₂S mitigation in composting: elucidating and validating the molecular pathway of Fe(II)/Fe(III)-driven sulfur incorporation into humic acid

Abstract

This study elucidated the molecular mechanisms by which iron of different valence states (Fe(II)/Fe(III)) regulates the migration and stabilization of sulfur in humic acid (HA) during composting, as well as its key role in efficiently suppressing hydrogen sulfide (H₂S) emissions. The results demonstrated that the iron valence state dictates the final pathway and stability of sulfur fixation: Fe(III) acted as an oxidative catalytic center, driving sulfur through an oxidation pathway, ultimately leading to its covalent fixation within the complex HA framework as stable functional groups such as sulfonyl and sulfonamide (e.g., C₁₉H₃₀N₄O₄S), achieving the “long-term sequestration” of sulfur. This pathway resulted in a 58.04% increase in sulfur-containing humic acid content and synergistically achieved a 94.68% reduction in H₂S emissions. In contrast, Fe(II) directed sulfur through a reductive pathway, forming unstable intermediates like thioethers (e.g., C₁₀H₁₆N₄S₂), which accumulated rapidly in the early stages but proved difficult to maintain over the long term. Abiotic synthesis experiments further confirmed the direct catalytic role of Fe(III) in the formation of stable organic sulfur. The study also found that iron additives significantly suppressed the expression of functional genes (e.g., dsrA) in sulfate-reducing bacteria, thereby reducing H₂S generation at the source. Based on these findings, this study ultimately establishes a dual-safeguard mechanism for iron-mediated H₂S emission reduction, namely “source inhibition of microbial reduction” and “end-of-pipe catalytic oxidative fixation”, providing a theoretical basis and technical pathway for the precise control and resource recovery of sulfur pollution during composting.

---

Green foundation

1. This work advances green chemistry by introducing a non-microbial, iron-mediated strategy for composting that simultaneously mitigates toxic H₂S emissions and valorizes sulfur by incorporating it into humic acid as a slow-release nutrient, shifting waste management from pollution control to resource recovery.

2. Quantitatively, Fe(III) addition reduced H₂S emissions by 94.68% and increased the content of stable, sulfur-containing humic acid molecules by 58.04%, demonstrating a qualitative shift towards forming highly stable sulfonyl/sulfonamide groups for long-term sulfur retention.

3. Future research could make this process greener by exploring the use of iron-rich industrial wastes (e.g., red mud) as Fe(III) sources and optimizing process parameters to minimize energy and resource input, further enhancing its sustainability and economic viability.

1. Introduction

Humic acid (HA), as a key component of compost organic matter, directly determines the agricultural value and soil remediation potential of the product based on its content level.^1,2 Studies have shown that the environmental functions of HA depend not only on its carbon skeleton structure but also on the various heteroatoms it contains.³ In addition to essential elements such as carbon, hydrogen, oxygen, and nitrogen, sulfur also plays a particularly critical role.⁴ Once organic fertilizers are incorporated into the soil, the stably bound sulfur in HA becomes a core carrier for realizing its environmental functions: on the one hand, it can serve as a slow-release sulfur source for plants; on the other hand, characteristic functional groups such as sulfonyl, thioether, and thiophene within its molecular structure can effectively chelate heavy metal ions (e.g., Cd²⁺, Hg⁺), participate in redox reactions, and promote the degradation of organic pollutants, thereby significantly enhancing the environmental benefits of compost products.^5–7 However, during the composting of sulfur-rich organic waste, the biogeochemical cycle of sulfur is highly prone to imbalance.⁸ Excessive activity of sulfate-reducing bacteria can lead to the release of large amounts of hydrogen sulfide (H₂S) gas, which not only results in the loss of sulfur nutrients but also causes serious environmental pollution.^9–11 Existing biological control technologies often exhibit delayed responses and unstable effectiveness under frequently fluctuating composting conditions.¹²

To control sulfur loss, the introduction of iron-based additives has become a widely explored strategy. The strong chemical affinity between iron and sulfur enables rapid binding through redox reactions, forming inorganic sulfides such as FeS.^13,14 Simultaneously, iron can suppress the metabolic activity of sulfate-reducing bacteria, thereby reducing H₂S generation at the source.¹⁵ However, such inorganic sulfides often exhibit chemical instability in the complex physicochemical environment of composting and are prone to morphological re-transformation. As a result, the sulfur fixation effect tends to be unsustainable, failing to achieve long-term stable sequestration of sulfur at a fundamental level.

It is noteworthy that although research and practice have confirmed that the addition of iron (Fe(II)/Fe(III)) can enhance the overall content of HA and effectively suppress H₂S emissions.¹⁶ A deeper and more critical process has long been overlooked: how does the intervention of iron affect the migration and fixation of sulfur into the molecular skeleton of HA? As the most abundant and structurally stable organic component in compost, HA should theoretically serve as an ideal long-term “sulfur sink”.^17,18 Its extensive aromatic framework and abundant functional groups provide a molecular environment for sulfur to be stably fixed via covalent bonds, offering a sulfur retention potential far superior to that of inorganic sulfides.^19,20

Therefore, this study employed laboratory-scale simulated composting combined with high-resolution liquid chromatography-mass spectrometry (HR-LC/MS) to analyze the evolution of sulfur species in HA under the addition of iron (Fe(II)/Fe(III)). Simulated reactions between purified HA and sulfur-containing compounds were further conducted to validate the proposed sulfur fixation pathways. The core innovation of this research lies in treating iron valence states as a “molecular switch” that directly regulates the chemical speciation and fixation routes of sulfur within HA, thereby laying a foundation for achieving highly efficient and stable sulfur fixation during composting. The significance of this work resides in its pioneering coupling of molecular-level sulfur speciation analysis in HA with pollutant reduction and compost quality improvement. It provides new insights and technical support for the synergistic regulation of organic matter and sulfur during composting, thereby promoting a paradigm shift in organic waste treatment from singular pollution control toward integrated resource recycling and functional enhancement. Current research bridging waste resource recovery and soil environmental chemistry remains scarce. Therefore, this work contributes to filling this critical knowledge gap.

2. Materials and methods

2.1. Composting experimental design

This study utilized fresh chicken manure as the primary raw material for aerobic composting experiments. The chicken manure was collected from a small-scale poultry farm in Harbin, Heilongjiang Province, China. Due to its high content of sulfur-containing proteins, it served as an ideal material for investigating sulfur transformation during composting.²¹ The raw materials were air-dried naturally and sieved to remove impurities before being mixed with rice straw to adjust the initial carbon-to-nitrogen ratio (C/N) to 25 : 1 and the moisture content to 60.00 ± 5.00%.²² The experiment included three treatment groups: CK group: control group, without any iron addition; DF group: supplemented with ferrous chloride (FeCl₂) at a rate of 1.00% of the raw material dry weight; TF group: supplemented with ferric oxide (Fe₂O₃) at a rate of 1.00% of the raw material dry weight. FeCl₂ and Fe₂O₃ were selected as the sources of Fe(II) and Fe(III), respectively. This selection was designed to ensure that the added iron would stably maintain its initial valence state during the composting process, thereby clearly elucidating the functional differences between ferrous and ferric iron (Text S1 for detailed selection criteria).

The experiment was conducted using lab-scale intelligent composting reactors with an effective volume of 10 L, equipped with a temperature gradient control system. The operating parameters were controlled according to a dynamic temperature curve derived from multiple optimized preliminary trials (Fig. S1).¹³ The composting reactors were operated under strictly controlled conditions to ensure reproducibility. A continuous aeration system was employed, with the air flow rate maintained at 0.4 L min⁻¹ kg⁻¹ volatile solids (VS) throughout the process using calibrated air pumps and flowmeters. The moisture content was monitored every 48 hours by measuring the weight loss of the reactor, and deionized water was added as needed to maintain the moisture content within the target range of 55%–65%. To ensure homogeneity and prevent anaerobic conditions, the compost mixtures were manually turned every 5 days for the first 30 days (active thermophilic and cooling phases), and then every 10 days during the maturation phase (last 30 days) (To ensure the stable existence of Fe(II) and Fe(III), the composting process parameters are adjusted within ±5% of their average values, so as to facilitate ions of different valence states to better exert their functions). The composting period lasted 60 days, with sampling performed on days 0, 3, 13, 38, and 60. Each treatment group was set up with three biological replicates. During each sampling event, five sub-samples were collected from each replicate using the five-point sampling method and thoroughly mixed to form a composite sample for subsequent HA extraction and analysis. Changes in pH and organic matter content during composting are shown in Fig. S2.

2.2. Extraction and quantification of humic acid

HA extraction was performed following the standardized method of the International Humic Substances Society (IHSS).²³ Briefly, 2 g of the dried compost sample was mixed with 0.1 M Na₄P₂O₇ and NaOH at a ratio of 1 : 10 (w/v). The mixture was shaken at 25 °C for 24 hours. Suspended solids were then separated by centrifugation at 12 000 rpm for 10 minutes, and the supernatant was filtered through a 0.45 μm micropore membrane. The resulting clear liquid constituted the HA solution. The total organic carbon (TOC) content of the HA was determined using a Shimadzu TOC-VCPH analyzer.

2.3. Structural determination of humic acid

2.3.1. Humic acid sample preparation. A 10 mL aliquot of the HA sample was filtered through a 0.45 μm membrane to obtain a clarified solution. Desalting was performed using an HLB solid-phase extraction (SPE) cartridge.²⁴ The cartridge was first activated and equilibrated by sequentially rinsing it three times with 5 mL of methanol and three times with 5 mL of ultrapure water under gravity flow. The sample solution was then loaded onto the cartridge bed. The breakthrough solution was discarded to remove non-volatile salt ions. Subsequently, the cartridge was rinsed with 5 mL of ultrapure water to eliminate residual salts. Finally, the target components were eluted using methanol. The collected eluate was divided into two portions. One portion was subjected to rotary evaporation to remove the solvent and subsequently reconstituted with ultrapure water, yielding a desalted HA solution for subsequent experiments. The other portion was used for high-resolution liquid chromatography-mass spectrometry (LC-MS) analysis. This method, optimized for the SPE procedure, effectively utilized the specific adsorption capacity of the HLB sorbent for polar compounds, achieving efficient separation of salt ions from the HA matrix.

2.3.2. Molecular structure identification of humic acid. The molecular structure of HA was characterized using an ultra-high-performance liquid chromatography (HPLC) system (Thermo Scientific UltiMate 3000) coupled to a Thermo Scientific Q Exactive Orbitrap mass spectrometer (MS).²⁵ Chromatographic separation was performed on an Accucore C18 column (2.1 mm × 100 mm, 2.6 μm particle size) at a flow rate of 0.3 mL min⁻¹. The mobile phase consisted of solvent A (water containing 1% formic acid) and solvent B (methanol containing 1% formic acid). A gradient elution program was employed as follows: 0–3 min, 25% B; 3–15 min, linear increase to 95% B; 15–18 min, 95% B; 18.01–21.00 min, re-equilibration to 25% B. The column temperature was maintained at 40 °C, and the injection volume was 2 μL. Prior to each injection, a 3 min equilibration time was allowed. Detection of analytes was accomplished using mass spectrometry. Mass spectrometric detection was conducted in both positive and negative electrospray ionization (ESI) modes. The mass spectrometer scanned over a mass range of m/z 85–1500 to identify target compounds. The optimal instrument working parameters were established as follows: sheath gas flow rate, 45 mL min⁻¹; auxiliary gas flow rate, 10 mL min⁻¹; spray voltage, 3.80 kV; capillary temperature, 320 °C; auxiliary gas heater temperature, 320 °C. Detected compounds were matched against databases (mzCloud, ChemSpider) using a mass tolerance of 5 ppm for precursor ions and 0.5 Da for fragment ions. A peak alignment time tolerance of 0.2 min and a signal-to-noise (S/N) ratio threshold of 3 were applied. Quantitative assessment of each HA molecule was performed by calculating its relative peak area in the chromatogram.

2.4. Composting process monitoring and analytical methods

The sensor probe was connected to the gas sampling port of the sealed composting reactor, ensuring that the detection element was in direct contact with gases released from the compost matrix. Instrument readings were continuously monitored throughout the process, and the hydrogen sulfide concentration was recorded once values stabilized. The methods for determining ferrous ion, free iron oxide, and amorphous iron oxide contents during composting are detailed in Text S2. The analysis of abundance changes of genes (Sat, AprA, and DsrA) associated with the dissimilatory sulfate reduction pathway is described in Text S3. The detection methods for inorganic sulfur and total sulfur are provided in Text S4, and their variation trends are shown in Fig. S7 and S9.

2.5. Organic synthesis verification experiments

To clarify the mechanistic roles of Fe(II) and Fe(III) in the conversion of inorganic sulfur into HA-bound organic sulfur, a series of organic synthesis experiments were designed.

2.5.1. Design of simulated reaction system and selection of sulfur source. Pyrogallolwas used as a canonical molecular proxy for the aromatic backbone of HA, and sodium thiosulfate (Na₂S₂O₃) was employed as the sulfur source, with their molar ratio set at 2 : 1. Fe₂O₃ was added as a catalyst at 1% of the total mass to construct the reaction system. The reaction was carried out at 45 °C in a carbonate buffer at pH 8.0 for 24 hours to simulate average composting conditions. The selection of pyrogallol is based on its well-established value in humification research. As a polyhydroxybenzene derivative, it possesses the high nucleophilic reactivity and redox activity characteristic of the polyphenolic and quinoid subunits that form the core structure of natural HA.²⁶ Under the mild oxidative conditions mediated by Fe(III) in this system, pyrogallol readily undergoes oxidative coupling and polymerization, forming complex aromatic networks analogous to those found in natural HA. This property makes it an ideal and widely recognized model compound for studying the abiotic incorporation of sulfur atoms into HA.^27,28

Regarding the sulfur source, thiosulfate (S₂O₃²⁻) was selected due to its central role in the biogeochemical cycle of sulfur and its relevance as a key reactive intermediate in composting environments. On one hand, it can be generated through the oxidation of H₂S; on the other hand, it can be further transformed via abiotic or biological pathways. More importantly, under the catalysis of metal oxides such as Fe(III), S₂O₃²⁻ undergoes homolytic cleavage, generating highly reactive sulfur-containing radicals. These radicals readily undergo electrophilic addition or radical coupling reactions with the electron-rich aromatic rings of HA precursors like pyrogallol, thereby directly simulating the fixation of reactive sulfur species into the HA framework.^29,30

2.5.2. Validation experiments on the promoting effect of iron valence states. Based on the feasibility of the iron-mediated sulfur transformation pathway verified in the aforementioned model reaction, another set of experiments was conducted to perform validation within a system more closely resembling authentic HA. In these validation experiments, pyrogallol was replaced with the desalted HA from the 60-day compost of the CK group. The catalysts were set as FeCl₂ (Fe(II) group) and Fe₂O₃ (Fe(III) group), respectively, while a control group (CKS group) containing only HA and sodium thiosulfate without iron addition was also established. These three validation reaction groups were likewise conducted at 45 °C in a pH 8.0 buffer for 24 hours.³¹

All experimental groups described above (the initial model reaction and the three validation reaction groups) were set up with three biological replicates. After the reaction, samples were collected and desalted following the pretreatment method for HA samples described in section 2.3.1, so as to remove adsorbed sulfur structures and facilitate the identification of covalently bound organic sulfur species. The molecular structures of the products were finally analyzed using the LC-MS method outlined in section 2.3.2, in order to verify the specific catalytic role of Fe(II) and Fe(III) in the formation of organic sulfur.

2.6. Statistical analysis

Statistical graphs depicting HA molecules and H₂S content were generated using Origin 2021 software. Chemical formulas for HA molecules were drawn using ChemDraw Professional 22.0.0. Correlation analysis and statistical significance testing were performed using R software (version 4.4.2). Raw LC-MS data were processed using Compound Discoverer 3.2 software (Thermo Fisher Scientific, USA). Structural formulas of HA molecules were drawn using ChemDraw Professional 22.0.0.

3. Results

3.1. Iron-driven sulfur immobilization as SHAM and its role in H₂S mitigation

3.1.1. Construction and validation of valence-specific iron environments. A foundational premise of this study was the successful construction of valence-specific iron environments. Crucially, the chemical speciation of iron was effectively managed by the addition of distinct iron sources (Fig. S3–S5). As substantiated by iron speciation monitoring data, the DF group maintained significantly higher levels of Fe(II) during the critical early thermophilic phase, establishing a predominant reducing milieu. Conversely, in the TF group, the added ferric oxide (Fe₂O₃) persisted predominantly as stable crystalline free iron oxides, thereby sustaining an oxidative catalytic interface throughout the process. The comparable background levels of amorphous iron oxides across treatments further underscore that the observed differential effects originated from the distinct nature of the added iron sources. This successful construction of valence-controlled iron environments provides the foundational context for interpreting the subsequent divergent behaviors of sulfur transformation. Against this backdrop, our analysis of sulfur incorporation began with rigorous methodological validation. The SPE desalting procedure, a critical step prior to LC-MS analysis, was shown to have a high recovery efficiency of 92.52% ± 4.61% for the HA matrix itself, as determined by TOC analysis. Furthermore, the analytical reproducibility among biological replicates was excellent, with the relative standard deviation (RSD) of SHAM content averaging 5.12% (Table S1). Through rigorous quality control, the observed dynamic changes in SHAM content and composition are verified to be free from artifacts arising from sample pretreatment or analytical processes.

3.1.2. Iron-driven SHAM formation and compositional shift linked to H₂S mitigation. Based on the dynamic changes in total HA concentration (Fig. S6) and LC-MS molecular characterization results, this study calculated the dynamic changes in sulfur-containing humic acid molecule (SHAM) content during the composting process (Fig. 1a). The experimental results showed that in the control group (CK), the SHAM content increased slowly from an initial 0.68 mg g⁻¹ to 1.12 mg g⁻¹ (a 64.71% increase) over the 60-day composting period (Table S2 displayed the molecular structures of SHAM identified in the CK group.), while the cumulative H₂S release reached 847.43 mg m⁻³ (Fig. 1b). This indicated a risk of secondary pollution due to sulfur volatilization in unregulated chicken manure composting.³² Following iron ion addition, the DF treatment group not only achieved an 85.33% reduction (P < 0.01) in H₂S emissions compared to CK, but also exhibited an SHAM content at the end of composting that was 10.72% higher (P < 0.05) than that of the CK group. In comparison, the TF treatment group demonstrated superior H₂S emission reduction efficacy (inhibition rate of 94.68%). SHAM accumulation commenced gradually from the start of composting, and by the end, the cumulative SHAM content in TF was 58.04% higher (P < 0.05) than in CK.

Fig. 1 Evolution rules of sulfur-containing HA promoted by Fe(II)/Fe(III) at different composting stages and their contributions to inhibiting H₂S release. (a) Variations in the content of sulfur-containing HA molecules (SHAM) during composting in CK, DF (Fe(II)), and TF (Fe(III)). (b) Cumulative amount of hydrogen sulfide gas released during composting. (c) Diversity analysis of sulfur-containing HA molecules. (d) Results of correlation analysis among ΔSHAM, NMDS1, and ΔH₂S. ΔSHAM: difference in SHAM content between each treatment group and the CK group (mg g⁻¹), representing the sulfur increment resulting from iron addition. NMDS1: score of each sample on the first principal coordinate of NMDS, representing the characteristic value of its SHAM molecular composition structure. ΔH₂S: difference in cumulative H₂S release between the CK group and each treatment group (mg m⁻³), representing the net H₂S emission reduction achieved by iron addition.

Furthermore, to resolve the overall differences in SHAM molecular composition among different treatment groups, this study employed non-metric multidimensional scaling (NMDS). Based on the relative abundances of all identified SHAM molecules, a similarity distance matrix between samples was constructed using the Bray–Curtis distance algorithm (Fig. 1c), with 95% confidence ellipses added. In this analysis, NMDS1 represented the first ordination axis, which explained the greatest proportion (72.97%) of variation among samples, and its scores intuitively reflected the relative position of each sample within the overall molecular composition profile; NMDS2 explained 19.05% of the variation. The results showed that the addition of iron ions, particularly Fe(III), significantly altered the NMDS1 scores of the samples, indicating a profound reconfiguration of the molecular composition of SHAM by iron. Pearson correlation analysis further revealed that both the shift in SHAM composition (NMDS1) and the increase in sulfur fixation (ΔSHAM) were significantly positively correlated with the reduction in H₂S emission (ΔH₂S) (P < 0.05, Fig. 1d). This finding bridged the statistical variation along the ordination axis with ecological function, suggesting that the efficient suppression of H₂S not only depended on the quantity of sulfur fixed but was also closely associated with the formation of specific SHAM structures induced by iron.

In the NMDS analysis, distinct clustering of samples driven by different iron valences was observed in the ordination space (Fig. 1c), indicating that the compositional variation represented by NMDS1 was not random but systematically regulated by iron valence states. This raised a key scientific question: did this separation originate from the preferential catalysis by Fe(II) and Fe(III) of sulfur-containing molecules with distinct chemical stability and oxidation states? A higher NMDS1 score might signify a molecular composition reconfigured toward more stable, less degradable, or less volatile sulfur species. To test this hypothesis and identify the specific molecular entities driving the divergence, subsequent research would systematically identify and structurally characterize key SHAM molecules that exhibited significant differences under iron mediation.

3.2. Identification of key sulfur-containing humic acid mediated by Fe(II)/Fe(III)

As shown in Fig. 2a and b, the volcano plots reveal the differential impact of iron valence states on sulfur speciation transformation, based on the log₂ fold-change in abundance (X-axis) of SHAMs between the treatment and control groups, and the statistical significance (−log₁₀p-value, Y-axis). Here, a log₂ fold-change > 0 indicates significant upregulation of the SHAM in the treatment group, while a log₂ fold-change < 0 indicates significant downregulation. A total of 14 and 12 significantly enriched SHAMs (upregulated in abundance, P < 0.01) were identified in the DF and TF groups, respectively, preliminarily pinpointing key molecular carriers for sulfur stabilization (Tables S3 and S4). Further Pearson correlation analysis revealed significant positive correlations between differential molecules and hydrogen sulfide reduction (ΔH₂S) (Fig. 2c and d). In the DF group, seven differential molecules including C₂₀H₃₂N₄S and others correlated with ΔH₂S-a (P < 0.05), while six molecules including C₁₈H₃₂O₅S and others showed correlation with ΔH₂S-b in the TF group (P < 0.05). These correlation characteristics suggest that the addition of Fe(II) and Fe(III) achieves directional sulfur fixation by specifically inducing the covalent binding of sulfur-containing molecules with HA. To elucidate the molecular mechanism of iron-mediated sulfur stabilization, subsequent research will systematically investigate the functional group configurations, sulfur bonding patterns, and structure–activity relationships with iron valence states of the differential SHAMs, based on their molecular formula characteristics. This will enable the establishment of a molecular regulatory network model for iron-enhanced sulfur fixation during composting.

Fig. 2 Identification of SHAM with significant differences mediated by iron. (a) SHAM showing significant differences in the TF group compared to the CK group; (b) SHAM showing significant differences in the DF group compared to the CK group. (c) Correlation heatmap between SHAM with significantly increased abundance in the DF group and the H₂S emission reduction amount. (d) Correlation heatmap between SHAM with significantly increased abundance in the TF group and the H₂S emission reduction amount. *P < 0.05, **P < 0.01, ***P < 0.001. Red and blue color scales represent positive and negative correlations, respectively.

3.3. Iron valence-dependent regulation of sulfur speciation in humic acid revealed by molecular structure recognition

3.3.1. Fe(III) promoted formation and stable accumulation of sulfonyl-containing sulfones. Fig. 3 illustrated the structures of SHAM in the TF group that showed a significant positive correlation with H₂S emission reduction (P < 0.05). The results demonstrated that during the mid-composting phase (day 13), Fe(III) catalysis promoted the binding of sulfur atoms to alkyl-substituted carbons on aromatic rings, forming thioether structures (TF-1).³³ The abundance of TF-1 increased sharply by 193.67% (P < 0.05) but subsequently declined to initial levels by the maturity phase, revealing the dynamic instability of thioether linkages. Concurrently, sulfur formed sulfonyl bridges (–SO₂–) that covalently linked aromatic molecules, generating sulfone compounds (TF-2, TF-3, and TF-4).³⁴ These compounds showed a significant increase in abundance (P < 0.05) from day 13 to day 60, indicating that the sulfonyl group serves as a stable form for sulfur retention in compost. Fe(III) addition also enabled sulfur incorporation into HA as thioketones (TF-5). Thioketone (TF-5), acting as a highly reactive intermediate, reached a transient peak concentration (24.68 μg g⁻¹) on day 38 before rapidly declining. Furthermore, thiazole heterocyclic structures (TF-6) became significantly enriched (P < 0.05) in the later stages, evidencing the dual benefits of synergistic sulfur–nitrogen fixation for nutrient retention.^35,36 Therefore, while Fe(III) diversified sulfur speciation, sulfonyl-dominated sulfone compounds emerged as the core carriers for sulfur fixation due to their chemical inertness. This constitutes the primary reason for the high contribution of HA sulfur fixation in the TF group to H₂S emission suppression. However, SHAM containing thioether bonds, thioketones, and other unsaturated linkages demonstrated suboptimal accumulation efficiency during the composting process.

Fig. 3 Structural formulas of SHAM in the TF group that exhibited a significant positive correlation with H₂S emission reduction. HA molecules were designated as TF-1 to TF-6.

3.3.2. Fe(II) induced formation and dynamic changes of thioether linkages and polysulfur sites. In DF, the study identified seven SHAMs that showed a significant positive correlation with H₂S reduction (P < 0.05). Fig. 4 showed that the addition of Fe(II) favored the participation of sulfur in HA formation via thioether bonds (DF-6 and DF-7). Day 13–38 of composting was the period of significant thioether accumulation (P < 0.05). The DF-6 structure consisted of a purine ring and a long-chain fatty acid connected by a thioether bond; the content of this compound reached 439.61 μg g⁻¹ on day 38 of composting (P < 0.05). The DF-7 structure contained a terpenoid derivative and cysteine, and its content reached 233.74 μg g⁻¹ on day 13 of composting. However, the content of both decreased significantly in the later stages of composting (P < 0.05). Furthermore, the addition of Fe(II) also promoted the formation of various sulfur-containing heterocyclic structures, such as the thiophene ring in DF-1 and the thiazole ring in DF-4. Their content dynamics also exhibited a trend of first increasing and then decreasing (P < 0.05). Notably, DF-2 and DF-5, which possessed multiple sulfur sites, and DF-3, which contained a sulfonyl group, were able to stably accumulate during the composting process. Therefore, Fe(II) and Fe(III) induced the formation of different types of SHAMs. This suggested that these iron species facilitated the directional accumulation of sulfur in HA through distinct pathways, thereby reducing H₂S release.

Fig. 4 Structural formulas of SHAM in the DF group that exhibited a significant positive correlation with H₂S emission reduction. HA molecules were designated as DF-1 to DF-7.

3.4. Molecular pathways of Fe(II)/Fe(III)-mediated sulfur incorporation into humic acid formation

Based on the structural identification of key differential SHAM in section 3.3, this study reveals the specificity of iron valence states in selecting sulfur species. To integrate these findings and elucidate their underlying connections, we constructed a comprehensive reaction pathway diagram (Fig. 5), which systematically illustrates the molecular mechanisms by which Fe(II) and Fe(III) drive the transformation of sulfur from reactive inorganic forms into stable organic structures fixed within HA.

Fig. 5 Molecular mechanism of Fe(II)/Fe(III)-mediated incorporation of various sulfur species into HA formation.

In composting environments, the interconversion between sulfate and sulfide forms the foundation of the sulfur cycle. The redox cycling of Fe(II)/Fe(III) facilitates this process, driving sulfur speciation transformations (Fig. 5a). The strong oxidizing capacity of Fe(III) dominates the oxidative pathway of sulfur. As demonstrated by Yin et al., Fe(III) can oxidize H₂S to intermediates such as thiosulfate, which subsequently generates sulfur radicals (e.g., ˙SO₃⁻).¹⁸ Corroborating our findings, sulfone compounds with sulfonyl bridges (TF-2, TF-3, TF-4) identified in this study represent the terminal stable products formed via radical addition reactions between such sulfur radicals and the aromatic backbone of HA, followed by further oxidation. Furthermore, studies indicate that Fe(III) acts as a Lewis acid, enhancing the electrophilicity of sulfur intermediates and thereby facilitating the formation of structures such as sulfonamides.^37,38 This offers a mechanistic explanation for the observed enrichment of such stable structures in the TF group.

On the other hand, Fe(II) is known to maintain a higher concentration of reduced sulfur species in the system, facilitating their nucleophilic reactions with ketones or alkenes in HA precursors to form thioethers or thioketones.³⁹ The results of this study confirm this, with significant formation of thioether linkages (DF-6, DF-7) observed in the DF group. Furthermore, these reduced sulfur structures were found not to be endpoints; they can be further oxidized within the composting environment. Hoque et al. demonstrated that thioethers can ultimately be oxidized to form sulfonyl compounds.^40,41 This mechanism links the initial fixation role of Fe(II) with the ultimate stabilization by Fe(III), collectively explaining the predominance of sulfonyl structures in the later stages of composting. Regarding the formation of sulfur heterocycles, existing studies report that H₂S can react with compounds containing amino and alkene groups to form thiazoles.⁴² Thiazole and thiophene structures were identified in both the DF and TF groups in this study. Therefore, using the SHAM structural identification as core evidence and integrating fundamental iron–sulfur chemistry, this study, for the first time, constructs a valence-dependent molecular pathway network for sulfur fixation applicable to the complex composting system. This network clearly elucidates how Fe(II) and Fe(III) achieve efficient and stable sequestration of sulfur within HA by regulating its redox state.

3.5. Experimental verification of the proposed molecular pathways

This study designed a two-step organic synthesis strategy. This approach aimed to validate whether the molecular pathways proposed in Fig. 5, which were inferred from the complex composting environment, represent the iron-valency-dependent chemical processes for SHAM formation.

3.5.1. Verification of pathway basic feasibility using the core structure of humic acid. In this study, pyrogallol was selected as a structural analog of the aromatic core of HA. A model reaction system was constructed with pyrogallol and sodium thiosulfate under Fe₂O₃ catalysis,²⁸ aiming to investigate whether stable functional groups such as sulfonyl groups could participate in the formation of HA via an abiotic pathway under Fe(III) catalysis. LC-MS analysis of the reaction products successfully identified phenyl sulfate esters (RPs-7, RPs-16) and sulfones (RPs-22), confirming the presence of sulfonyl structures. As shown in Fig. 6a, in the presence of both Fe₂O₃ and oxygen, the system triggered a benzene ring cleavage reaction,⁴³ generating olefinic products, including fumaric acid and (2Z,4E)-2-hydroxymuconic acid, which accounted for 1.75% of the total products. Simultaneously, some phenolic hydroxyl groups were oxidized to quinone structures, such as 5,8-dihydroxy-1,4-naphthoquinone and hydroxy-1,4-benzoquinone. These intermediates further reacted with sulfur radicals to form thiophene derivatives (RPs-1, RPs-4, RPs-5, RPs-20) and thioethers (RPs-6, RPs-21) (Table S5).

Fig. 6 Validation of sulfur incorporation pathways in simulated systems. (a) Fe₂O₃ was added to the pyrogallol–sodium thiosulfate system to demonstrate the molecular pathway of inorganic sulfur incorporation into HA. (b) Molecular structures and relative abundance of sulfur-containing humic acids across the CKS, Fe(II), and Fe(III) groups. Red and gray colors represent molecules with sulfonyl groups and other sulfur-containing species, respectively. (c) and (d) Fe(III) and Fe(II) were added to the HA–sodium thiosulfate system to verify their enhancement effects on HA-bound sulfur formation: (c) content of SHAM; (d) content of sulfonyl-containing compounds.

3.5.2. Validation of valency-specific effects within an authentic humic acid matrix. Although the model compound experiments confirmed the fundamental chemical principles, a key question remained unresolved: when the reactants are structurally complex natural HA, do these transformation pathways dependent on iron speciation still hold and dominate? To address this question and bridge the gap between the simplified model and the actual composting system, this study further conducted validation experiments using desalted HA extracted from mature CK compost as the reaction substrate. The results from this real reaction system showed clear quantitative consistency with observations from the composting process (Fig. 6b): the addition of Fe(III) not only significantly increased the total SHAM content (+128.40%), but more importantly, it selectively and markedly enhanced the relative abundance of stable sulfonyl-containing compounds by 29.59% (Fig. 6c and d). This effect highly aligned with the molecular characteristics of the SHAM components in the TF compost group, confirming that even within the macromolecular framework of complex natural HA, Fe(III) can still preferentially and efficiently direct sulfur towards stable, oxidized end products. This final validation step thus definitively established a causal link between the simulated chemical reactions and the actual composting phenomena, demonstrating that the differential outcomes of sulfur fixation in HA are essentially a chemical process selectively governed by iron speciation.

4. Discussion

In composting systems, achieving long-term stabilization of sulfur is crucial for resource recovery and pollution control. However, merely immobilizing sulfur in soluble organic matter or inorganic iron–sulfur minerals (e.g., FeS) still carries the risk of re-release due to degradation or oxidation. Therefore, directing sulfur into the chemically inert and structurally complex framework of HA is regarded as key to achieving long-term sulfur recycling. As the main insoluble organic component in compost, the three-dimensional aromatic structure of HA can provide covalent binding sites, forming organic sulfur that is resistant to rapid microbial utilization, thereby facilitating a transition from “temporary storage” to “long-term sequestration”. The core of this study lies in elucidating the regulatory roles of Fe(II) and Fe(III) in this process. The results revealed that the valence state of iron not only influences the “quantity” of sulfur fixation but also determines the “quality” and “final pathway” of its immobilization.

4.1. Iron valence-dominated divergence in molecular pathways determines the stability of sulfur fixation

The most direct evidence from this study indicates that Fe(III) acts as an oxidative catalytic center, driving sulfur transformation along an oxidative pathway, with sulfonyl and sulfonamide groups (e.g., C₁₉H₃₀N₄O₄S) as the terminal stable products. This enabled the TF group to achieve a 94.68% reduction in H₂S emissions over the 60-day composting period, along with a 58.04% increase in SHAM content, reflecting a synergistic enhancement in both “quantity and quality”. The exceptional stability of the sulfonyl group, which is the cornerstone of these long-term sequestration products, is rooted in its distinct electronic structure and consistently demonstrated across scientific disciplines: firstly, the very high bond dissociation energy of the S=O bond (∼535 kJ mol⁻¹) and the strong, resonance-stabilized nature of the sulfonyl bridge (–SO₂–) form a rigid and thermally robust molecular scaffold, which is exploited in high-performance engineering plastics like polysulfones that retain integrity under extreme conditions.^44,45 Secondly, the sulfur atom in a sulfonyl group is in its highest oxidation state (+6), is tetrahedrally surrounded by four highly electronegative atoms, creating a symmetric, low-energy state that is kinetically inert to nucleophilic attack and oxidative degradation, accounting for the metabolic stability of sulfonamide drugs and herbicides.⁴⁶ Thirdly, its identification as a dominant and persistent form of organic sulfur in ancient sediments and coals through S K-edge XANES spectroscopy provides direct evidence for its geochemical stability over geological timescales.⁴⁷ In contrast, Fe(II) tended to stabilize reduced sulfur species, directing a reductive pathway that preferentially formed intermediates such as thioethers (e.g., C₁₀H₁₆N₄S₂).⁴⁸ These structures accumulated rapidly in the early to middle stages of composting but exhibited poor stability in the later phase. Although the total SHAM content in the DF group increased by 53.80% compared to the CK group, its molecular composition lacked fundamental structural renovation, resulting in lower product stability (Fig. 1c). Furthermore, abiotic organic synthesis experiments, conducted under conditions excluding microbial interference, reproduced this iron valence-guided pattern of SHAM formation, providing the most direct causal evidence for the proposed chemical fixation pathways.

4.2. Dynamic transformation of iron species and pathway contributions

Based on the aforementioned differences in molecular transformation pathways, a key question arises: in the dynamic and complex composting system, is iron-mediated sulfur transformation strictly dependent on the initial valence state of iron? Dynamic monitoring of iron speciation revealed that the DF and TF groups successfully established iron-dominated chemical environments centered on Fe(II) and Fe(III), respectively (Fig. S3–S5). This indicates that by regulating the oxidation–reduction microenvironment of the compost pile, a sustained predominance of the targeted iron valence state can be achieved, although not as an absolutely static equilibrium. Due to the aerobic conditions of composting, localized dynamic transformations of iron species inevitably occur. For instance, in the DF group, the total Fe(II) content decreased by 5.75%, and the content of free iron oxides increased by 5.56% in the later stage. It is noteworthy that such transformations are directly related to sulfur conversion products, as exemplified by the appearance of minor sulfonyl structures (e.g., DF-3) in the later phase of the DF group.

Nevertheless, these localized transformations did not alter the dominant pathway patterns in the two systems, a conclusion further supported by the analysis of inorganic sulfur species (Fig. S7). The DF group exhibited a higher accumulation of inorganic sulfur, which aligns with the findings reported by Chen et al., indicating that Fe(II) tends to maintain sulfur in its reduced state and promotes the formation of iron–sulfur minerals or polysulfides, thereby retaining sulfur in unstable inorganic phases.⁴⁹ In contrast, Fe(III), as a strong oxidant, can deeply oxidize sulfur into highly reactive intermediates such as sulfonyl groups (–SO₂–) through radical-initiated reactions.⁵⁰ These intermediates readily undergo covalent binding with the aromatic structures of HA, thereby channeling sulfur into the humification pathway and achieving a qualitative shift from “inorganic temporary storage” to “chemical stabilization”. Furthermore, studies by Li et al. have indicated that the addition of exogenous iron can activate the reactivity of iron oxides, which directly suppresses the metabolic activity of sulfate-reducing bacteria (SRB) through mechanisms such as competitive inhibition, thereby reducing H₂S generation at the source.^51,52 Consequently, the high efficiency of H₂S emission reduction achieved in the DF group (85.33%) and TF group (94.68%) is not only a result of sulfur chemical fixation but also reflects the net effect of synergistic interactions between microbial inhibition and chemical processes.

4.3. Dual safeguard mechanism for H₂S emission reduction: from source inhibition to end immobilization

Overall, iron-mediated sulfur transformation during composting is not unidirectionally determined solely by the initial iron valence state, but rather constitutes a dynamic, multi-pathway coupled process. Therefore, this study quantitatively analyzed key sulfur redox functional genes (dsrA, Sat, aprA; Text S5, Fig. S8). Compared to the CK group, both DF and TF treatments significantly reduced the relative abundance of sulfur cycle genes. This confirms that iron addition effectively suppresses the metabolic potential of sulfate-reducing bacteria, achieving “source reduction” of H₂S. Based on the mechanisms described above, this study further proposes a theoretical framework. In composting systems amended with iron-based materials, H₂S emission reduction is achieved through the synergistic action of dual pathways: “source inhibition” and “terminal fixation”. On one hand, iron-based additives can inhibit the activity of microorganisms such as sulfate-reducing bacteria, thereby reducing H₂S generation at the source. On the other hand, iron (particularly when Fe(III) is dominant) promotes the conversion of reactive sulfur species into intermediates like sulfonyl groups via abiotic catalytic oxidation. These intermediates are then stably incorporated into HA skeleton through covalent bonding, forming persistent organic sulfur. The superiority of Fe(III) lies in its ability to efficiently drive both pathways simultaneously. It not only effectively inhibits microbial sulfur production but also possesses strong oxidative catalytic capability, directing sulfur toward conversion into chemically stable, non-releasable organic forms. This approach enhances emission reduction efficiency while ensuring the long-term stability of sulfur in the final compost product. From a “microbial–chemical” coupling perspective, this theory systematically explains the mechanisms of iron-mediated sulfur transformation and fixation, providing a theoretical foundation for sulfur pollution control and resource recovery during composting.

5. Environment benefits

In this study, industrial ferric oxide (Fe₂O₃) powder was used, with a market price of approximately 282.80 USD per ton. At an addition rate of 1% on a dry weight basis, only 10 kg of Fe₂O₃ is required per ton of compost, resulting in a raw material cost increase of about 2.83 USD. This additional cost can be fully offset by the comprehensive benefits brought by the compost product, indicating good economic feasibility. Moreover, the 1% Fe₂O₃ addition corresponds to a pure iron content of only 0.70%, which is significantly lower than the average iron content in soil (1%–2%) and thus poses no adverse effects on soil structure or ecological safety. Experimental results showed that the total sulfur content in the compost product of the TF treatment group increased by 18.66% compared to the CK group (Fig. S9), with 67.05% of the sulfur attributed to the enhanced HA-mediated sulfur fixation pathway. Based on a demonstration project scale estimation, in the composting of 100 000 tons of sulfur-containing waste, the addition of Fe(III) could fix approximately 167 tons of sulfur, equivalent to reducing about 177 tons of hydrogen sulfide emissions. This approach not only significantly lowers environmental risks but also saves costs associated with hydrogen sulfide treatment.

It should be pointed out that while this study utilized industrial ferric oxide to validate the technical principle and feasibility of iron-enhanced sulfur fixation, a more cost-effective alternative iron source with significant environmental co-benefits, namely red mud, can be adopted in practical engineering. Red mud is a solid waste residue generated from the alumina industry, containing 15% to 30% iron oxides, expressed as Fe₂O₃ and FeO. China has an immense stockpile of red mud, with annual accumulation exceeding 100 million tons. Its long-term storage not only occupies substantial land but also poses ongoing environmental risks. Currently, the comprehensive utilization rate of red mud remains low, and its disposal often requires additional cost. If applied in the composting process, it would eliminate the need to purchase iron additives and could potentially generate economic returns through environmental subsidies or waste disposal fees for utilizing solid waste, thereby achieving a net economic gain from iron source substitution. For practical implementation, red mud also offers advantages in transportation and application costs. Sourced primarily from aluminum industry bases, it is often geographically close to urban waste treatment facilities, which can significantly reduce logistics expenses. Furthermore, the alkaline nature of red mud can partially neutralize compost acidity, reducing the dosage of pH-adjusting agents. By employing a “waste treats waste” model to use red mud in composting sulfur-rich waste, it is possible to substantially reduce hydrogen sulfide emissions, enhance the HA and sulfur nutrient content of the final compost product, and achieve large-scale consumption of red mud while mitigating its environmental impact. This approach enables the simultaneous improvement of both environmental and economic outcomes.

Author contributions

Jin Zhou: conceptualization, writing – original draft, data curation, formal analysis. Yiqi Yin: data curation. Jiaran Chen: resources. Yu Deng: visualization. Rui Sun: methodology. Caihong Song: methodology. Zimin Wei: writing – review & editing, methodology, funding acquisition, project administration.

Conflicts of interest

The authors declare no competing financial interest.

Data availability

The data supporting this article have been provided as part of the supplementary information (SI). Supplementary information: additional information including Texts S1–S5, Fig. S1–S9, and Tables S1–S6. See DOI: https://doi.org/10.1039/d5gc05649g.

Acknowledgements

This work is funded by National Natural Science Foundation of China (grant number: 52170126, 52370148 and 32372820).

References

1. L. Sun, C. Zheng, J. Yang, C. Peng, C. Xu, Y. Wang, J. Feng and J. Shi, Biol. Fertil. Soils, 2016, 52, 31–39.
2. K. M. Malik, K. S. Khan, M. Billah, M. S. Akhtar, S. Rukh, S. Alam, A. Munir, A. M. Aulakh, M. Rahim, M. M. Qaisrani and N. Khan, Agronomy, 2021, 11, 2514.
3. H. Sun, L. Luo, J. Wang, D. Wang, R. Huang, C. Ma, Y.-G. Zhu and Z. Liu, Environ. Sci. Technol., 2022, 56, 6639–6646.
4. Y. Dong, M. Gao, W. Qiu, L. Xiao, Z. Cheng, H. Peng and Z. Song, J. Hazard. Mater., 2024, 464, 132942.
5. Y. Li, C. Lu, N. Zhu, J. Chao, W. Hu, Z. Zhang, Y. Wang, L. Liang, J. Chen, D. Xu, Y. Gao and J. Zhao, J. Hazard. Mater., 2022, 430, 128447.
6. M. Zhao, X. Liu, Z. Li, X. Liang, Z. Wang, C. Zhang, W. Liu, R. Liu and Y. Zhao, J. Hazard. Mater., 2021, 407, 124647.
7. S. Chaudhary, S. S. Sindhu, R. Dhanker and A. Kumari, Microbiol. Res., 2023, 271, 127340.
8. C. V. Preble, S. S. Chen, T. Hotchi, M. D. Sohn, R. L. Maddalena, M. L. Russell, N. J. Brown, C. D. Scown and T. W. Kirchstetter, Environ. Sci. Technol., 2020, 54, 16097–16107.
9. Q. Fu, Q. Xu, Z. Liu, D. Wang, X. Liu, D. He, Y. He, Y. Li, J. Yang and A. Duan, Chem. Eng. J., 2022, 430, 133150.
10. L. Zhu, W. Li, C. Huang, Y. Tian, B. Xi, W. Wu and Y. Yan, J. Environ. Manage., 2024, 353, 120195.
11. Z. Chen, Y. Chen, X. Hao, Q. Wen, X. Wang, A. Liu, D. Gao, K. Hu, N. A. Samak, M. Yang and J. Xing, Chem. Eng. J., 2025, 508, 160632.
12. Z. Zhang, S. Song, W. He, X. Ren, Y. Ren, H. Yuan, H. Pang, J. Yang and J. Lu, J. Hazard. Mater., 2025, 492, 138114.
13. Q. Wu, W. Gao, X. Wang, R. Sun, Y. Jiao, X. Li and Z. Wei, Chem. Eng. J., 2025, 517, 164452.
14. Q. Zeng, X. Wang, X. Liu, L. Huang, J. Hu, R. Chu, N. Tolic and H. Dong, Environ. Sci. Technol., 2020, 54, 15013–15023.
15. N. Paepatung, N. Boonapatcharoen, W. Songkasiri, H. Yasui and C. Phalakornkule, Chem. Eng. J., 2022, 435, 135175.
16. Y. Han, J. Wang, T. Liu, M. Wei, S. Wang, J. Guo, X. Ma, Z. Li, N. Wang and Y. Sang, J. Hazard. Mater., 2025, 486, 137030.
17. M. Tarek, J. S. Santos, V. Márquez, M. Fereidooni, M. Yazdanpanah, S. Praserthdam and P. Praserthdam, J. Energy Chem., 2024, 90, 388–411.
18. R. Yin, C. Fan, J. Sun and C. Shang, Chem. Eng. J., 2018, 336, 587–594.
19. F. Yang and M. Antonietti, Adv. Sci., 2020, 7, 1902992.
20. M. Ren, Q. Zhuang, X. He, W. Liu, C. Guo, H. Ye, J. R. Reinfelder, C. Ma, J. Li and Z. Dang, Environ. Sci. Technol., 2024, 58, 7357–7366.
21. K. Wang, C. He, S. You, W. Liu, W. Wang, R. Zhang, H. Qi and N. Ren, J. Hazard. Mater., 2015, 300, 745–753.
22. L. Wan, X. Wang, C. Cong, J. Li, Y. Xu, X. Li, F. Hou, Y. Wu and L. Wang, Bioresour. Technol., 2020, 301, 122730.
23. X. Ge, X. Xiao, Y. Zhang, X. Zhu, C. Chu and B. Chen, Environ. Sci. Technol., 2025, 59, 6579–6589.
24. P. S. Bäuerlein, J. E. Mansell, T. L. ter Laak and P. de Voogt, Environ. Sci. Technol., 2012, 46, 954–961.
25. S. Papadopoulos Lambidis, T. Schramm, K. Steuer-Lodd, S. Farrell, P. Stincone, R. Schmid, I. Koester, R. Torres, T. Dittmar, L. Aluwihare, C. Simon and D. Petras, Environ. Sci. Technol., 2024, 58, 19289–19304.
26. M. Shin, E. Park and H. Lee, Adv. Funct. Mater., 2019, 29, 1903022.
27. L. Chen, K. Zhang, Y. Shen, Z. Chen and W. Fang, Green Chem., 2025, 27, 8585–8593.
28. H. T. Abdel-Mohsen, J. Conrad and U. Beifuss, Green Chem., 2014, 16, 90–95.
29. Y. Sun, Y. Huang, S. Zheng, H. Chen, J. Pei, Y. Yang, H. Ji and C. Han, Chem. Eng. J., 2025, 511, 161675.
30. Z. Zhou, P. Q. Tran, E. S. Cowley, E. Trembath-Reichert and K. Anantharaman, Nat. Rev. Microbiol., 2025, 23, 122–140.
31. C. Kim, Y.-P. Chin, H. Son and I. Hwang, Water Res., 2022, 212, 118107.
32. A. Feilberg, M. J. Hansen, D. Liu and T. Nyord, Nat. Commun., 2017, 8, 1069.
33. J. Kuzmin, J. Röckl, N. Schwarz, J. Djossou, G. Ahumada, M. Ahlquist and H. Lundberg, Angew. Chem., Int. Ed., 2023, 62, e202304272.
34. C. Morrill, J. E. Gillespie and R. J. Phipps, Angew. Chem., Int. Ed., 2022, 61, e202204025.
35. T. S. Lobana, Coord. Chem. Rev., 2021, 441, 213884.
36. S. Tian, S. Yang, J. Wang, Q. Li, C. Li, X. Cui and P. Ning, J. Hazard. Mater., 2022, 440, 129827.
37. J. R. Ludwig, S. Phan, C. C. McAtee, P. M. Zimmerman, J. J. Devery III and C. S. Schindler, J. Am. Chem. Soc., 2017, 139, 10832–10842.
38. Z. Wu and D. A. Pratt, Nat. Rev. Chem., 2023, 7, 573–589.
39. C. Kim, V. R. Panditi, P. R. Gardinali, R. S. Varma, H. Kim and V. K. Sharma, Chem. Eng. J., 2015, 279, 307–316.
40. M. A. Hoque, T. Jiang, D. L. Poole and S. S. Stahl, J. Am. Chem. Soc., 2024, 146, 21960–21967.
41. Z. Rao, X. Li, Y. G. Fang, J. S. Francisco, C. Zhu and C. Chu, J. Am. Chem. Soc., 2023, 145, 10839–10846.
42. V. A. Palchykov, Coord. Chem. Rev., 2025, 537, 216677.
43. B. Qu, C. Liu, Y. Zhang, Z. Fu, Y. S. Zhang, A. Li and G. Ji, Chem. Eng. J., 2023, 474, 145736.
44. B. An, L. Sun, T. Sun and Y. Li, Angew. Chem., Int. Ed., 2025, 64, e202421906.
45. W. Jiang, K. Zhu, H. Ma, J. Liu, C. Zhang, Y. Dai and H. Jia, Environ. Sci. Technol., 2023, 57, 8680–8690.
46. Z. Wang, Z. Zhang, W. Zhao, P. Sivaguru, G. Zanoni, Y. Wang, E. A. Anderson and X. Bi, Nat. Commun., 2021, 12, 5244.
47. O. Carrión, C.-Y. Li, M. Peng, J. Wang, G. Pohnert, M. Azizah, X.-Y. Zhu, A. R. J. Curson, Q. Wang, K. S. Walsham, X.-H. Zhang, S. Monaco, J. M. Harvey, X.-L. Chen, C. Gao, N. Wang, X.-J. Wang, P. Wang, S. J. Giovanonni, C.-P. Lee, C. P. Suffridge, Y. Zhang, Z. Luo, D. Wang, J. D. Todd and Y.-Z. Zhang, Nat. Microbiol., 2023, 8, 2326–2337.
48. T. Hao, Y. Huang, F. Li, Y. Wu and L. Fang, Water Res., 2022, 219, 118587.
49. S.-C. Chen, X.-M. Li, N. Battisti, G. Guan, M. A. Montoya, J. Osvatic, P. Pjevac, S. Pollak, A. Richter, A. Schintlmeister, W. Wanek, M. Mussmann and A. Loy, Nature, 2025, 646, 925–933.
50. Z. She, J. Wang, X. Pan, D. Ma, Y. Gao, S. Wang, X. Chuai, Y. Mu and Z. Yue, Water Res., 2024, 254, 121404.
51. R. Li, L. Zhang, Y. Yang, W. Zhang, Y. Zhen, J. Sun and F. Jiang, Water Res., 2026, 288, 124706.
52. H. Dai, Z. Huang, C. Wang, Y. Hu, X. Xu, F. Zhao, Y. Pi, J. Qian, F. Hu and X. Peng, Sep. Purif. Technol., 2025, 361, 131225.

---