Photo-transformation of nitrate and fulvic acid driven by guest iron minerals

Highlights

	● Guest α-Fe2O3 facilitated the reduction of NO3– to NH4+ via e– and CO2•– pathways.
	● Fulvic acid, acting as a hole scavenger, enhanced the reduction of NO3– by α-Fe2O3.
	● h +, •OH, and RNS are significant reactive species in the oxidation of fulvic acid.
	● α -Fe2O3 inhibited the photo-transformation of CHON and CHONS of fulvic acid.

Abstract

The photochemical interactions between nitrate (NO₃^–) and natural organic matter (NOM) are vital for aquatic chemistry. However, the effects of guest iron minerals, which may enter the aquatic environments due to both human and natural activities, on those interactions are widely ignored. This work evaluated the effects of hematite (α-Fe₂O₃) on the photochemical conversion products and pathways of NO₃^–, fulvic acid (FA) under 12 h of ultraviolet irradiation. The addition of 0.4 g/L of guest α-Fe₂O₃ accelerated the reduction of NO₃^– by 24.3%, with NH₄⁺ as the primary reduction product, and hampered the mineralization of FA. These effects were dependent on the dosage amount of α-Fe₂O₃ and FA concentrations. The studies on the molecule-level changes of FA revealed that the complete oxidation to CO₂ and the partial oxidation pathways that alter the molecular composition of FA were suppressed, and the mineralization rate decreased by 27.8%. Particularly, the conversion rates of CHON and CHONS were reduced by 21.0% and 20.3%, respectively, increasing the unsaturated products. The scavenging experiments and quantitative measurements of hydroxyl radicals (•OH) proposed that the photogenerated electrons and holes from α-Fe₂O₃ were the key for the altered transformation of NO₃^– and FA. This work revealed the guest effects of iron mineral particles on the photochemical interactions between NO₃^– and NOM in the natural surface waters.

Graphical abstract

Keywords

Guest iron minerals / Photochemistry / Nitrate / Fulvic acid / Conversion

1 Introduction

Iron minerals are abundant on Earth’s surface and characterized by their high specific surface areas, reactivities, and photocatalytic capabilities (Pan et al., 2023; Yu et al., 2024). Numerous studies have revealed the vital roles of iron mineral particles in affecting aquatic environments where they are formed (Zhao et al., 2022; Pan et al., 2023; Yu et al., 2024). It is noted that iron minerals may also enter aquatic environments from external sources due to events including flood (Mishra et al., 2021), mining (Galvão et al., 2020), and the corrosion of ship hulls (Furlan et al., 2023). A recent study reported that shallow hydrothermal eruptions can release iron mineral particles into the ocean, significantly improving diazotroph activity and enhancing carbon export fluxes (Bonnet et al., 2023). This work revealed that the iron mineral particles entering aquatic ecosystems may cause guest effects. Given that the guest iron mineral particles may be photo-responsive and their amount can be considerable (up to 0.17 wt%–4.93 wt%) (Poulton and Raiswell, 2002; Zheng et al., 2023), they may interfere with the photochemistry of substances in aquatic environments.

Nitrate (NO₃^–) and natural organic matter (NOM) are widely present photosensitive substances in aquatic environments, playing a key role in the carbon and nitrogen cycles of aquatic ecosystems (Guo et al., 2022; Hou et al., 2023; Yu et al., 2023). Iron minerals, due to their unique photoreactive properties, mediate the abiotic reduction of NO₃^– on particle surfaces under sunlight irradiation through photochemical reactions (Kleber et al., 2021; Fan et al., 2023). The rapid recombination of photogenerated holes and electrons (h⁺, e^–) is the rate-limiting step for NO₃^– reduction (Yang et al., 2022a).

NOM, which is ubiquitous in surface water environments, acts as an effective free radical scavenger, and iron mineral particles can react with NOM through various mechanisms (Wu et al., 2021; Wu et al., 2023a), such as adsorption (Kleber et al., 2021), co-precipitation (Chen et al., 2020), metal complexation (Wang et al., 2023), and ligand-metal charge transfer (LMCT) (Chen et al., 2023). However, the photocatalytic effect of iron minerals on the photo-transformation of NO₃^– and NOM is not a simple coupling. Furthermore, the impact of altered transformation pathways on NO₃^– and NOM products is currently lacking. Considering the potential input of guest iron minerals, investigating the photochemical mechanism of iron minerals on the abiotic transformation of coexisting NO₃^– and NOM is crucial for a comprehensive assessment of carbon and nitrogen cycles in the aquatic environment.

In this work, the guest effects of iron mineral particles on the photochemical transformation of NO₃^– and NOM were evaluated. Ferric oxide (α-Fe₂O₃) is widely distributed in nature with high abundance and is often used as a model compound for iron minerals (Zheng et al., 2023). Fulvic acid (FA) is an important component of NOM, representing a class of organic substances within NOM that possess high biological activity and significant environmental importance (Hu et al., 2018). The influence of α-Fe₂O₃ on the transformation products and pathways of NO₃^– was analyzed under different experimental conditions. Fourier transform ion cyclotron resonance mass spectrometry coupled with electrospray ionization (ESI FT-ICR-MS) was employed to investigate the impacts of α-Fe₂O₃ introduction on the transformation of FA at the molecular level. Additionally, a combination of scavenging experiments and quantitative measurements of hydroxyl radicals (•OH) confirmed the involvement of predominant reactive species in the transformation of NO₃^– and FA, and a mechanism for the photo-mediated mineral-driven conversion of NO₃^– and FA was proposed. This work elucidates the NO₃^– and FA photoconversion process in response to α-Fe₂O₃, and conduces to improve our understanding of the environmental effects of guest iron mineral input into aquatic environments.

2 Materials and methods

2.1 Materials and reagents

The stock solution of FA was prepared according to a previous study (Yang et al., 2021a). Nano-ferric oxide (α-Fe₂O₃, 99.9%), potassium iodide (KI, 99%), sodium chromate (Na₂CrO₄, 98%), and methylviologen (C₁₂H₁₄Cl₂N₂∙xH₂O, 98%) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. Fulvic acid (FA, 90%) was obtained from Shanghai Yuanye Biotechnology Co., Ltd., China. Other chemical reagents such as potassium nitrate (KNO₃, 99%), terephthalic acid (TPA, 98%), and 2-hydroxyterephthalic acid (hTPA, 98%) were procured from Sinopharm Chemical Reagent Co., Ltd. All chemicals were used as received without further purification.

2.2 Photochemical experiments

The photochemical experiments were conducted in a beaker with circulating cooling water. Photochemical reactions were conducted with the irradiation of either an ultraviolet lamp (UV₂₅₄, TUV PL-S, 9 W, Philips, China) or a xenon lamp (CEL-PF-300-T6, 300 W, Beijing Zhongjiao Jinyuan Technology Co., Ltd., China). The wavelength of the xenon lamp was ranged from 300 to 2500 nm. Our experiments were divided into six groups, (1) NO₃^– + UV₂₅₄ (N + L); (2) NO₃^– + α-Fe₂O₃ + UV₂₅₄ (N + Fe + L); (3) FA + UV₂₅₄ (F + L); (4) FA + α-Fe₂O₃ + UV₂₅₄ (F + Fe + L); (5) FA + NO₃^– + UV₂₅₄ (F + N + L); (6) FA + NO₃^– + α-Fe₂O₃ + UV₂₅₄ (F + N + Fe + L). For all the experiments, the initial concentrations of NO₃^–, FA, and α-Fe₂O₃ were 20 mg-N/L, 100 mg-C/L, and 0.4 g/L, respectively, unless otherwise mentioned. Such relatively high concentrations were chosen to amplify the reaction rates. Magnetic stirring was applied for all the reactions, and the stirring speed was 200 r/min. Liquid samples were collected using syringes and subsequently filtered through 0.45 μm filters (Nanjing Ronghua Scientific Equipment Co., Ltd., China) after 12 h for further analysis of NO₃^–, nitrite (NO₂^–), ammonium (NH₄⁺), total nitrogen (TN), and total organic carbon (TOC) of products (Text S1). More detailed descriptions of the measurement methods, including characterization of α-Fe₂O₃ (Text S2), and quantification of •OH (Text S3, Fig. S1) were presented in the Supporting Information.

2.3 Identification and molecular level analysis of FA

The fluorescence of products was measured in a 1 cm quartz cuvette by three-dimensional excitation-emission matrix fluorescence spectroscopy (3D-EEM, Hitachi F7000, Japan). Deionized water was used in the reference cell as the reference sample. The data analysis was performed using the DOMFluor toolbox for Matlab (R2022a) (Du et al., 2023). The detailed analysis methods are described in Text S4.

The molecular level compositional changes of FA were further analyzed in the ESI FT-ICR-MS (Bruker SolariX, USA) (Du et al., 2023; Hu et al., 2023). The products were solid-phase extracted (SPE) using Agilent Bond Elut PPL (Agilent, 100 mg, 3 mL, USA) to exclude the impact of inorganic ions and concentrate the samples. Elution procedures were similar to the previous study (Du et al., 2023). The negative ion mode was utilized to analyze the molecular level components. The mass-to-charge ratio (m/z) analysis software was applied to assign formula values within 100−800 Da automatically. Weighted averages of elemental ratios of H/C_w and O/C_w, molecular weight (MW_w), modified aromaticity index (AI_modw), double-bond equivalents (DBE_w), nominal oxidation states of carbon (NOSC_w), and the weighted average of each indicator were calculated based on the identified formulas (Du et al., 2023; Li et al., 2023).

2.4 Mass difference network analysis

The linkage analysis was conducted to investigate the reaction pathways between FA and its products in various reaction systems for photochemical transformation. We calculated the accurate molecular masses of all molecules based on the precise atomic masses provided by the International Union of Pure and Applied Chemistry (IUPAC). Based on the 27 photochemical reactions that FA may undergo as reported in the literature, we calculated the mass difference between precursors and products, and simultaneously matched the calculated mass differences with the precise values of the relevant NOM photochemical transformation reactions (Table S1) (Li et al., 2023; He et al., 2024). A program code in Matlab (R2022a) was developed to match the mass differences between precursors and products with those of the 27 reactions and to conduct the linkage analysis. IF the mass difference between two molecules is consistent with the mass difference of the matched reaction, it suggests a potential connection between the two molecules.

2.5 Natural water sample

The natural water sample was collected from the Xuanwu Lake (32° 3′ 51″ N, 118° 48′ 34″ E), in Nanjing, China. The sample was filtered through a 0.45 μm filter membrane and then stored at 4 °C. Detailed water quality parameters were summarized in Table S2.

3 Results and discussion

3.1 Characterization of α-Fe₂O₃

The X-ray diffraction results in Fig. S2(a) demonstrated that the diffraction peaks were precisely corresponding to the hematite standard card (JCPDS PDF#33-0664). High-resolution transmission electron microscopy images showed that nano-ferric oxide owned needle-like morphology with diameters of 30–50 nm (Fig. S2(b)). Additionally, as shown by the UV-vis diffuse reflectance spectroscopy (DRS, Fig. S2c), α-Fe₂O₃ clearly exhibits excellent light absorption. The band gaps (E_g) of α-Fe₂O₃ obtained from the Tauc plots using the Kubelka−Munk function was 1.85 eV (Fig. S2(d)) (Zheng et al., 2023). These results revealed the photochemical properties of α-Fe₂O₃.

3.2 Transformation of nitrogen species induced by α-Fe₂O₃

A dark control absorption experiment was first conducted, which confirmed that the adsorption of NO₃^– over α-Fe₂O₃ was negligible (Fig. S3). Meanwhile, this control experiment also revealed that NH₄⁺ with a concentration of ~6 mg-N/L was released from 100 mg-C/L of FA even under dark conditions (Wu et al., 2020a). After 12 h of irradiation by UV₂₅₄, c(NO₃^–) decreased by 3.4% (N + L in Fig.1(a)), and other products including NO₂^– and NH₄⁺ were also identified (Fig.1(b) and 1(c)). This change can be ascribed to the photochemical transformation of NO₃^–, as described in Eq. (1) (Tugaoen et al., 2017; Yang et al., 2021b). When α-Fe₂O₃ was introduced, c(NO₃^–) decreased by 9.2% (N + Fe + L in Fig.1(a)). This improved conversion percentage of NO₃^– can be ascribed to e^– (Eq. (2)), which mediated the reduction of NO₃^– (Eq. (3)).

Fig.1 Changes in nitrogen species concentration (a) NO3–, (b) NO2–, and (c) NH4+. (d) The removal rates of TN and TOC under UV254 irradiation. Experimental conditions: c0(NO3–) = 20 mg-N/L, c0(FA) = 100 mg-C/L, α-Fe2O3 dosage = 0.4 g/L (N, F, Fe, and L are the abbreviations of NO3–, FA, α-Fe2O3, and UV254 irradiation, respectively).

Full size|PPT slide

(1)${\mathrm{N}\mathrm{O}}_3^{-}+\mathit{h}\mathit{v}{\to [\mathrm{N}\mathrm{O}}_3^{-}{]}^{\ast }\to {\mathrm{R}\mathrm{N}\mathrm{S}+\mathrm{O}}^{\text{•}–}{\to \mathrm{N}\mathrm{O}}_2^{-}{\to \mathrm{N}\mathrm{H}}_4^{+}$

(2)$\alpha \text{-}{\mathrm{F}\mathrm{e}}_2{\mathrm{O}}_3+\mathit{h}\mathit{v}\to {\mathit{h}}^{+}+{\mathit{e}}^{-}$

(3)${\mathrm{N}\mathrm{O}}_3^{-}+{\mathit{e}}^{-}{+\mathrm{H}}^{+}\to {\mathrm{N}\mathrm{O}}_2^{-}{/\mathrm{N}\mathrm{H}}_4^{+}{+\mathrm{H}}_2\mathrm{O}$

when FA was exposed to UV₂₅₄-irradiation, 1.0 mg-N/L of NO₃^– was generated from the transformation of NH₄⁺, whose concentration dropped from 6.8 to 4.1 mg-N/L (F + L in Fig.1(a) and 1(c)). The oxidation from NH₄⁺ to NO₃^– may be ascribed to •OH that was generated from the photobleaching of FA (Eqs. (4)–(5)) (Huang et al., 2008; Kong et al., 2023). The introduction of α-Fe₂O₃ resulted in an inhibition of the FA photolysis due to its light-screening effect (Hou et al., 2023; Wu et al., 2023b). This effect was also evidenced by a decrease in the NH₄⁺ and TOC removal rate (F + Fe + L in Fig.1(c) and 1(d)). It should be noted that little NO₂^– was generated in these two systems (Fig.1(b)). In addition, a higher removal rate of TN (21.2% for F + Fe + L) was observed, as compared with the case without α-Fe₂O₃ (18.3% for F + L). Previous studies have reported that the oxidation of N-containing aromatic pollutants by •OH radicals resulted in the release of NO_x (Garcia-Segura et al., 2017). This difference indicated that the presence of α-Fe₂O₃ is beneficial for the production of gaseous nitrogen products. The detailed transformation mechanism was discussed in the FA conversion section.

(4)$\begin{array}{rl}& \mathrm{F}\mathrm{A}+\mathit{h}\mathit{v}\to {[}^3\mathrm{F}\mathrm{A}{]}^{\ast }\to \mathrm{B}\mathrm{y}\text{-}\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{s}\\ & +{\mathrm{C}\mathrm{O}}_2+{\mathrm{N}\mathrm{O}}_{\mathit{x}}+\text{•}\mathrm{O}\mathrm{H}\end{array}$

(5)${\mathrm{N}\mathrm{H}}_4^{+}+\mathit{h}\mathit{v}\mathrm{o}\mathrm{r}\text{•}\mathrm{O}\mathrm{H}\to {\mathrm{N}\mathrm{O}}_2^{-}{(\mathrm{o}\mathrm{r}\mathrm{N}\mathrm{O}}_{\mathit{x}})\to {\mathrm{N}\mathrm{O}}_3^{-}$

when NO₃^– and FA were both presented, the transformation of NO₃^– was accelerated and a significantly higher concentration of NO₂^– was generated (F + N + L in Fig.1 and b), as compared to the systems with either NO₃^– or FA in the presence. The presence of FA is beneficial for the transformation from NO₃^– to NO₂^– since FA can scavenge the •OH, which prevents its reactions with NO₂^– and reactive nitrogen species (RNS) (Sharpless and Linden, 2001). Meanwhile, the TOC removal rate in the F + N + L system was also higher than that in the F + L and N + L systems (Fig.1(d)), which was ascribed to the FA mineralization over RNS (NO• or NO₂•, Eq. (6)) (Gong et al., 2022; Li et al., 2022). It was noted that when α-Fe₂O₃, NO₃^–, and FA were all presented under the irradiation of UV₂₅₄, the conversion efficiency of NO₃^– increased to 37.5%, higher than that of N + F + L (30.2%). Simultaneously, less NO₂^– was produced and more NH₄⁺ was preserved, indicating that the improved charge separation due to the consumption of h⁺ over FA facilitated the reduction of NO₃^– to NH₄⁺. The promotion of NO₃^– reduction by α-Fe₂O₃ was also observed in the systems containing 2 mg-N/L of NO₃^– and 2 or 4 mg-C/L of FA (Fig. S4(a)), and the concentration of NO₂^– decreased in the system of F + N + Fe + L (Fig. S4(b)). As the initial concentration of FA decreased, the concentration of NH₄⁺ exhibited a more pronounced increase after the reaction (Fig. S4(c)). At low concentrations, FA may primarily undergo oxidation of h⁺, resulting in a limited oxidation of NH₄⁺. Additionally, NO₃^– was also continuously reduced by e^– to generate NH₄⁺, which led to an increase in the accumulated concentration of NH₄⁺.

(6)${\mathrm{R}\mathrm{N}\mathrm{S}+\mathrm{F}\mathrm{A}\to \mathrm{B}\mathrm{y}\text{-}\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{s}\to \mathrm{C}\mathrm{O}}_2+{\mathrm{N}\mathrm{O}}_{\mathit{x}}$

Different from UV₂₅₄, xenon lamps did not show any significant effects on the above-mentioned processes (Fig. S5). This difference indicated that the interactions among NO₃^–, FA, and α-Fe₂O₃ were wavelength-dependent and more sensitive to the UV₂₅₄ irradiation (Bu et al., 2020; Wu et al., 2020b; Yang et al., 2022b; Zheng et al., 2023). In addition, the NO₃^– transformation rate was also affected by the concentrations of α-Fe₂O_3, and 0.4 g/L yielded the highest removal rate of NO₃^– among all the studied concentrations (Fig.2(a)). This concentration also produced the largest amount of NO₂^– and preserved the largest amount of NH₄⁺ (Figs. S6a and S6b. The effects of α-Fe₂O₃ concentrations are reasonable since the lower concentrations of α-Fe₂O₃ produce a smaller amount of active species while the higher concentrations can shield the light, which both inhibit the reaction efficiency (Wang et al., 2021).

Fig.2 The effects of different parameters on the NO3– reduction, including the concentration of (a) α-Fe2O3 and (b) FA. Experimental conditions: c0(NO3–) = 20 mg-N/L, c0(FA) = 100 mg-C/L for (a), α-Fe2O3 dosage = 0.4 g/L for (b).

Full size|PPT slide

Fig.2(b) demonstrated that the concentration of NO₃^– gradually decreased with increasing initial concentration of FA in the range of 10 to 200 mg-C/L. However, NO₂^– and NH₄⁺ concentrations changed in more complicated ways with the variations of FA concentrations (Figs. S6c and 6d). Specifically, the concentrations of the generated NO₂^– increased in the range from 10 to 100 mg-C/L of FA but decreased at 200 mg-C/L of FA. For NH₄⁺, its concentrations increased after the reactions when FA concentrations were 10, 20, and 50 mg-C/L, but decreased when FA concentrations were 100 and 200 mg-C/L. This result suggested that FA played some vital roles in nitrogen transformation, and its roles are relevant to its concentrations. At low concentrations, FA primarily acted as an h⁺ scavenger, facilitating the reduction of NO₃^–. The low-molecular-weight organic compounds generated from the oxidation of FA served as better h⁺ scavengers, promoting the photo-reduction of NO₃^– to NH₄⁺ (Kumar et al., 2021). As the concentration of FA increases, the photolysis of FA becomes more pronounced, resulting in an increased oxidation of NH₄⁺ (Eqs. (4)–(5)).

3.3 Transformation of FA optical properties and molecular level

The UV absorbance in both systems showed a decreasing trend over the reaction time (Fig. S7a and b). The UV₂₅₄ value of the F + N + Fe + L system was lower than that of the F + N + L system, indicating a higher degree of aromaticity in the products of the F + N + Fe + L system after a 12 h reaction (Fig. S7c). The results in Fig. S7d demonstrated that the E₂/E₃ ratio in the F + N + Fe + L system continuously increased in the first 6 h and then reached a plateau. The increase in the beginning is attributed to the cleavage of chemical bonds in FA during photolysis, while the plateau may be due to the dynamic equilibrium of oxidation and demethylation processes occurring during the reaction (Du et al., 2023; He et al., 2024). These results collectively suggest that α-Fe₂O₃ altered the transformation pathways of FA, inhibiting further reactions of larger molecules and resulting in products with higher molecular weights, which is consistent with the observed decrease in the mineralization rate of TOC.

Furthermore, we conducted infrared spectroscopy (FTIR) analysis on the products for the F + N + L and F + N + Fe + L systems (Fig. S8). The FTIR spectra revealed the presence of −OH (3480 cm⁻¹), C−H (2950 cm⁻¹), O−C=O (1740 cm⁻¹), C=O (1610 cm⁻¹), and −C−OH (1380 cm⁻¹) (Gong et al., 2020; Yang et al., 2024). The peaks at 700−800 cm⁻¹ are attributed to di-substituted aromatics (Gong et al., 2020). The infrared spectra of the FA after reactions exhibited a slight reduction in peak intensity, particularly for C−OH, which may indicate the oxidation of phenolic substances within the FA.

3D-EEM analysis was applied to characterize the chromophore and fluorescent of FA. The regional fluorescence integration method was used to quantify the intensity of fluorescence, and fluorescence spectra were divided into five regions according to a previous method (Chen et al., 2003). The strongest fluorescence intensity of the pristine FA was located in region III, which was ascribed to fulvic acid-like substances (Fig. S9) (Wells et al., 2022; Chen et al., 2003). This type of FA is rich in lignins/CRAM-like structure, aromatic protein, and other substances with fluorescent properties. As the reaction time prolonged in the F + L system, the fluorescence intensity of the samples gradually diminished (Fig. S10a). Additionally, after a 2 h reaction, new peaks appeared in region V of the EEM spectra for the products, and the same outcome was observed in the F + N + L and F + N + Fe + L treatment groups. However, after a 12 h reaction, there were significant differences in the position and intensity of the fluorescence peaks of the products across different regions. In the F + N + L system, the degradation of fluorescent substances in FA was further amplified, as compared with the F + L system (Fig. S10b). The presence of α-Fe₂O₃ did not promote the diminishing of fluorescent substances, as a discernible fluorescence could still be observed even after 12 h of UV₂₅₄ irradiation (F + N + Fe + L in Fig. S10c). We also conducted a quantitative analysis and the results (Fig. S11) revealed that the fluorescence intensities of regions III, IV, and V in the F + N + Fe + L group were significantly higher than those in the F + L and F + N + L groups. As a result, the introduction of α-Fe₂O₃ hampered the profound degradation of fulvic acid-like, soluble microbial by-product-like, and humic acid-like substances (Chen et al., 2003). This result confirms that the aforementioned photocatalytic reactions involving α-Fe₂O₃ altered the oxidation pathways of FA, leading to the observed differences in fluorescence characteristics of the products.

The molecular-level changes of FA in the different systems were identified using ESI FT-ICR-MS (Du et al., 2023; Hu et al., 2023), and the results are summarized in Tab.1. After FA underwent different photochemical reactions, the MW_w of products decreased. The F + Fe + L and F + N + Fe + L systems exhibited higher molecular quantities and MW_w values, but lower O/C_w and H/C_w ratios as compared to the F + L and F + N + L groups. Meanwhile, the introduction of α-Fe₂O₃ inhibited the extent of oxidation of aromatic and unsaturated molecules, resulting in high AI_mod,_w, and DBE_w values of the products (Qiao et al., 2020; Du et al., 2023).

Tab.1 Abundance weighted average molecular properties of different samples

	Pristine FA	F + L	F + Fe + L	F + N + L	F + N + Fe + L
Number	5506	5133	5798	5326	5655
MWw	387.422	356.948	365.439	364.829	367.071
O/Cw	0.535	0.545	0.529	0.539	0.530
H/Cw	1.192	1.293	1.271	1.286	1.243
AImod,w	0.198	0.109	0.150	0.123	0.175
DBEw	9.403	7.798	8.171	7.919	8.347

The molecules were further divided into three categories. The molecules present in both pristine FA and products are referred as “Conserved”, while the molecules that present exclusively in the pristine FA are termed as “Removed” and the molecules present only in the treated samples are considered as “Produced” (Zhang et al., 2021; Hu et al., 2023). After reactions, 2003, 1968, 2388, and 1927 molecules were removed, respectively from F + L, F + Fe + L, F + N + L, and F + N + Fe + L, while 1996, 2260, 2204, and 2075molecules were respectively produced. FA photolysis primarily removed molecules with MW values ranging from 400 to 600 Da (Fig.3(a)). Both NO₃^– and α-Fe₂O₃ facilitated the removal of relatively large molecules (600 Da < MW < 800 Da) in FA. More smaller molecules (MW < 200 Da) were removed in the F + N + L and F + N + Fe + L systems, as compared to the F + L and F + Fe + L systems. Additionally, the F + Fe + L group exhibited the highest abundance of products < 200 Da (Fig.3(b)). Studies have documented that low molecular weight (MW < 200 Da) substances are more likely produced from the oxidation of FA over •OH (Zeng et al., 2020; Hu et al., 2023), which is discussed in the later section.

Fig.3 The MW distributions of (a) removed and (b) produced molecules. (c) Size proportional Venn diagrams. The Van Krevelen diagrams of composition variation in (d) pristine FA, (e) F + N + L, and (f) F + N + Fe + L. (1: Lipids; 2: Aliphatic/Proteins; 3: Carbohydrates; 4: Unsaturated hydrocarbons; 5: Lignins/CRAM-like structures; 6: Tannin; 7: Aromatic structures.) The number of molecules that are removed and produced by (g) lipids, (h) aromatic structures, and (i) carbohydrate compounds.

Full size|PPT slide

The products from FA transformations varied among different systems according to the Venn diagram (Fig.3(c)). Furthermore, all molecules detected were classified into seven regions based on the ratio of H/C and O/C (Text S5) (Xu et al., 2018). For the pristine FA, molecules were mainly located in regions 2, 5, and 6, which corresponded to aliphatic/proteins (9.4%), lignins/CRAM-like structures (68.6%), and tannin (14.9%), respectively (Fig.3(d)). After subject to the reactions in different systems, the molecules were more scattered regarding the H/C and O/C values (Fig.3(e), 3(f), and S12). For example, in the system of F + N + L system, the removed molecules shared a similar distribution pattern with the pristine state but produced molecules with lower O/C values and higher H/C values preferably (Fig.3(e)). As a comparison, in the F + N + Fe + L system, molecules with higher O/C values were preferably removed (Fig.3(f)). The difference between the two systems implied that oxygen addition was more significant in the later system.

To differentiate the transformation of different molecules, we performed a zone-based counting of the removed and produced molecules (Fig.3 and S13). α-Fe₂O₃ notably influenced the transformation of tannin, lignin/CRAM-like structures, and aliphatic protein molecules within FA, whereas the variance among the remaining four classes of substances was insignificant. The introduction of α-Fe₂O₃ did not result in any regular changes in tannin, unsaturated hydrocarbons, lignins/CRAM-like compounds, or aliphatic proteins (Fig. S13). However, its introduction facilitated the production of lipids (21.2%−28.3%, Fig.3(g)) and aromatic structured compounds (30.6%−47.7%, Fig.3(h)), and hampered the production of carbohydrates (35.1%−39.7%, Fig.3(i)), regardless of whether NO₃^– was present or not. The findings revealed that α-Fe₂O₃ hindered the complete oxidation of FA, resulting in predominantly partial oxidation products. This result is consistent with the mineralization rate and 3D-EEM of the products.

Mass difference network analysis was further conducted to elucidate the reaction pathways of FA based on the 27 transformation reactions proposed in the DOM photo-transformation research (Table S1). As demonstrated in Fig.4(a) and Fig.4(b), oxygen addition was the most predominant type of reaction, indicating that oxidation was the major reaction in all the systems. Consistently, the F + N + Fe + L system experienced more oxygen addition reactions than the F + N + L system. It was worth noting that the addition of α-Fe₂O₃ in F + L slightly inhibited amine reactions, such as hydrolysis of amide (−CONH + H₂O), deamination (−NH), and oxidative displacement of amine (−NH₂ + OH), which explained the observed increase in NH₄⁺ concentration in the F + Fe + L products. However, the increase in these reaction quantities is not significant in F + Fe + L, and it is speculated that h⁺ played a role in facilitating these reactions (Fig.4(b)). Except for this inhibition, the introduction of α-Fe₂O₃ accelerated all the other reactions, suggesting that α-Fe₂O₃ activated the photochemical transformation of FA. However, these transformations were maintained at a lower mineralization rate (Fig.1(d)).

Fig.4 (a)–(b) The counts of potential reactions classed by mass difference network analysis in the photochemistry of FA. The related transformation reactions and numbers are summarized in Table S3. (c)–(d) Classification and counting of FA photochemical precursors and products according to elements.

Full size|PPT slide

The photochemical reactions of FA are closely related to the molecular composition of the elements (Hou et al., 2023; He et al., 2024). As shown in Fig.4(c), the F + N + L system removed the highest number of precursors. The introduction of α-Fe₂O₃ inhibited the removal of all the analyzed precursors (F + N + Fe + L vs. F + N + L). The inhibited removal of nitrogen-containing molecules (CHON and CHONS) was ascribed to the response of nitrogen transformation to α-Fe₂O₃ photocatalysis (Liu and Wang, 2019). Furthermore, among the photochemical products of FA, CHON was the most dominant, which was accounted for 45.8%−50.6% of the final products. In addition, CHON was also the substance with the most pronounced differences among different reaction systems (Fig.4(d)). Owing to the formation of NO_X during the photolysis of FA, the quantity of CHON molecules in the F + L group was lower than the other groups. Meanwhile, the photolysis of NO₃^– generated RNS, which facilitated the oxidation of FA and thus the formation of CHON molecules (Gong et al., 2022). α-Fe₂O₃ weakened the photolysis of FA by light screening effect or the photo-transformation of NO₃^–, inhibiting the further reactions of nitrogen-containing compounds. Consequently, it can be speculated that the photochemical transformation and generation of nitrogen-containing compounds of FA are highly sensitive to α-Fe₂O₃.

3.4 Identification of the main active species for NO₃^– and FA conversion

To identify the responsible reactive species, excessive amounts of KI (0.1 mmol/L), Na₂CrO₄ (0.25 mmol/L), and methyl viologen (0.05 mmol/L) were added into the system of F + N + Fe + L as the scavengers for h⁺, e^–, and carbon dioxide radicals (CO₂^•–), respectively (Jiang et al., 2022). We observed that NO₃^– reduction was enhanced by adding KI (Fig.5(a)). Specifically, the concentration of the produced NO₂^– was 1.92 times higher than that of the control group after 12 h of UV₂₅₄ irradiation (Fig.5(b)). The TN removal rate in the KI group increased to 8.9% (Fig.5(c)). This result suggested that the presence of h⁺ inhibited NO₃^– reduction to NO₂^– and the gaseous products (Shi et al., 2022; Silveira et al., 2023). However, the TOC removal rate remained almost unchanged when h⁺ was scavenged, indicating that the h⁺ was not the primary species for the TOC removal. On the contrary, e^– and CO₂^•– were both vital to the transformation of nitrogen and their scavenging inhibited the NO₃^– reduction, and e^– made a more prominent contribution to the reduction of NO₃^– (Fig.5(e)).

Fig.5 Nitrogen substance concentration: (a) NO3– and (b) NO2– in reactive species scavenging experiments. (c) The changes of TOC and TN in reactive species scavenging experiments. (d) Quantification of •OH concentration in different systems. (e) Photo-transformation of NO3– and FA driven by guest α-Fe2O3.

Full size|PPT slide

To investigate the interface reaction of α-Fe₂O₃ and the role of FA, we performed X-ray photoelectron spectroscopy (XPS) on the α-Fe₂O₃ products after the reaction. The results indicated that the surface Fe(II) content of α-Fe₂O₃ increased in both the N + Fe + L and F + N + Fe + L groups, with a higher Fe(II) content observed in the F + N + Fe + L group (Fig. S14 and Table S4). UV₂₅₄ excited the α-Fe₂O₃ surface to generate h⁺ and e^–, and e^– reduced Fe(III) to Fe(II) (Liu and Wang, 2019). The presence of FA consumed h⁺, thereby promoting the production of e^– and facilitating the reduction of Fe(III) (Shu et al., 2022; Yu et al., 2024).

We performed ESR analysis for the NO• and •OH by diethyldithiocarbamate (DETC)₂−Fe²⁺ and dimethyl-pyrroline N-oxide (DMPO), successfully identifying signals for both species (Fig. S15). The ESR signals of NO• were detected in the N + L, N + Fe + L, F + N + L, and F + N + Fe + L systems. Notably, the ESR signal of NO• in the F + N + Fe + L system was diminished, indicating that NO₃^– preferentially reduced by e^–, thereby validating our earlier hypothesis (Fig.5(e)). We measured the accumulation of •OH in the different systems (Fig.5(d)). When FA was absent, only a negligible amount of •OH was generated after 12 h (13.10 μmol/L for N + L and 19.75 μmol/L for N + Fe + L). On the contrary, even the photolysis of FA under UV₂₅₄ irradiation resulted in a continuous production of •OH, whose concentration reached 225.6 μmol/L (F + L). The F + N + L system accumulated more •OH than the F + L system, which may explain the higher TOC removal rate in the former one (Fig.1(d)). We noted that when α-Fe₂O₃ was absent (N + L, F + L, and F + N + L), a higher •OH accumulation amount also resulted in a higher TOC removal rate. This consistency indicates the vital roles of •OH on TOC removal when α-Fe₂O₃ was absent.

On the other hand, the F + Fe + L system also accumulated more •OH than the F + L system, due to the benefits of the structural Fe in α-Fe₂O₃ or electron transfer at the interface between α-Fe₂O₃ and FA on the production of •OH (Shu et al., 2022; Wang et al., 2024; Yu et al., 2024). Interestingly, the •OH concentration of the F + N + Fe + L group reached a plateau after 4 h of reaction. However, the fluorescence intensity of the system continued to diminish after 4 h (Fig. S10c), indicating that the degradation of FA after 4 h was predominantly attributed to other oxidation pathways, possibly h⁺ and LMCT.

The above-proposed effects of α-Fe₂O₃ were also verified in the natural surface water. The results indicated that the introduction of α-Fe₂O₃ brought insignificant difference regarding NO₃^– concentrations for the natural surface water (Fig.6(a)). This small difference was possibly due to the low concentration of NOM since the TOC of the surface water was only 6.1 mg-C/L (Table S2). However, the introduction of guest iron mineral particles inhibited the formation of NO₂^– and the removal of TOC, while preserving NH₄⁺ (Fig.6(b)−Fig.6(d)). All these trends were consistent with those in the spiked water.

Fig.6 Changes in nitrogen species concentration under UV254 conditions. (a) NO3–, (b) NO2–, and (c) NH4+. (d) The removal rate of TN and TOC under UV254 conditions.

Full size|PPT slide

4 Conclusions

This study sheds light on the ubiquitous but underappreciated phenomenon that guest iron minerals may affect the biogeochemical processes in aquatic environments. We provided evidence that e^– tends to reduce NO₃^– to NH₄⁺, and the oxidation of FA can be mainly attributed to h⁺, and free radicals (•OH and RNS), when NO₃^–, FA, and α-Fe₂O₃ are irradiated by light simultaneously. These reactions collectively inhibited the photolysis of both FA and NO₃^–. This inhibition also results in weakened oxidation of lignin/CRAM-like structures, aliphatic/proteins, and tannin in FA, especially the limited transformation of nitrogen-containing molecules. In addition, the introduction of α-Fe₂O₃ increased the reduction of NO₃^– and promoted the demethylation, carboxylic acid, and sulfate reaction pathways of FA.

The results in natural surface water also verified that the introduction of guest iron minerals has consequential impacts on the photochemical fates of NO₃^– and FA. Understanding these alterations at the molecular level may help identify the potential environmental impacts, particularly the abiotic transformations of N and C in aquatic environments driven by guest iron mineral particles. Overall, it is imperative for future surface water environmental management and assessment to consider the conjoint effects of the photochemical process of guest minerals on the migration and transformation of photosensitive substances.

References

[1] Bonnet S, Guieu C, Taillandier V, Boulart C, Bouruet-Aubertot P, Gazeau F, Scalabrin C, Bressac M, Knapp A N, Cuypers Y. . (2023). Natural iron fertilization by shallow hydrothermal sources fuels diazotroph blooms in the ocean. Science, 380(6647): 812–817

[2] Bu L, Zhu N, Li C, Huang Y, Kong M, Duan X, Dionysiou D D. (2020). Susceptibility of atrazine photo-degradation in the presence of nitrate: impact of wavelengths and significant role of reactive nitrogen species. Journal of Hazardous Materials, 388: 121760

[3] Chen C, Dong Y, Thompson A. (2023). Electron transfer, atom exchange, and transformation of iron minerals in soils: the influence of soil organic matter. Environmental Science & Technology, 57(29): 10696–10707

[4] Chen C, Hall S J, Coward E, Thompson A. (2020). Iron-mediated organic matter decomposition in humid soils can counteract protection. Nature Communications, 11(1): 2255

[5] Chen W, Westerhoff P, Leenheer J A, Booksh K. (2003). Fluorescence excitation−emission matrix regional integration to quantify spectra for dissolved organic matter. Environmental Science & Technology, 37(24): 5701–5710

[6] Du P, Liu W, Zhang Q, Zhang P, He C, Shi Q, Huang C H, Wang J. (2023). Transformation of dissolved organic matter during UV/peracetic acid treatment. Water Research, 232: 119676

[7] Fan Q, Wang L, Fu Y, Li Q, Liu Y, Wang Z, Zhu H. (2023). Iron redox cycling in layered clay minerals and its impact on contaminant dynamics: a review. Science of the Total Environment, 855: 159003

[8] Furlan P Y, Jaravata E J, Furlan A Y, Kahl P. (2023). Will it rust? A set of simple demonstrations illustrating iron corrosion prevention strategies at sea. Journal of Chemical Education, 100(2): 1081–1088

[9] Galvão E S, Sant A, Cavichini A, Rangel C V G T, Orlando C G P, Grilo C F, Soares J, Oliveira K S S, Sá F, Junior A C. (2020). Tracing iron ore tailings in the marine environment: an investigation of the Fundão dam failure. Chemosphere, 257: 127184

[10] Garcia-Segura S, Mostafa E, Baltruschat H. (2017). Could NO_x be released during mineralization of pollutants containing nitrogen by hydroxyl radical? Ascertaining the release of N-volatile species. Applied Catalysis B: Environmental, 207: 376–384

[11] Gong G, Xu L, Zhang Y, Liu W, Wang M, Zhao Y, Yuan X, Li Y. (2020). Extraction of fulvic acid from lignite and characterization of its functional groups. ACS Omega, 5(43): 27953–27961

[12] Gong S, Ding C, Liu J, Fu K, Pan Y, Shi J, Deng H. (2022). Degradation of Naproxen by UV-irradiation in the presence of nitrate: efficiency, mechanism, products, and toxicity change. Chemical Engineering Journal, 430: 133016

[13] Guo Y, Guo Z, Wang J, Ye Z, Zhang L, Niu J. (2022). Photodegradation of three antidepressants in natural waters: important roles of dissolved organic matter and nitrate. Science of the Total Environment, 802: 149825

[14] He H, Sun N, Li L, Zhou H, Hu A, Yang X, Ai J, Jiao R, Yang X, Wang D, Zhang W. (2024). Photochemical transformation of dissolved organic matter in surface water augmented the formation of disinfection byproducts. Environmental Science & Technology, 58(7): 3399–3411

[15] Hou J, Zhang R, Ge J, Ma C, Yi Y, Qi Y, Li S L. (2023). Molecular and optical signatures of photochemical transformation of dissolved organic matter: nonnegligible role of suspended particulate matter in urban river. Science of the Total Environment, 903: 166842

[16] Hu D, Zeng Q, Zhu J, He C, Shi Q, Dong H. (2023). Promotion of humic acid transformation by abiotic and biotic Fe redox cycling in nontronite. Environmental Science & Technology, 57(48): 19760–19771

[17] Hu S, Lu Y, Peng L, Wang P, Zhu M, Dohnalkova A C, Chen H, Lin Z, Dang Z, Shi Z. (2018). Coupled kinetics of ferrihydrite transformation and As(V) sequestration under the effect of humic acids: a mechanistic and quantitative study. Environmental Science & Technology, 52(20): 11632–11641

[18] Huang L, Li L, Dong W, Liu Y, Hou H. (2008). Removal of ammonia by OH radical in aqueous phase. Environmental Science & Technology, 42(21): 8070–8075

[19] Jiang C, Zhang M, Dong G, Wei T, Feng J, Ren Y, Luan T. (2022). Photocatalytic nitrate reduction by a non-metal catalyst h-BN: performance and mechanism. Chemical Engineering Journal, 429: 132216

[20] Kleber M, Bourg I C, Coward E K, Hansel C M, Myneni S C B, Nunan N. (2021). Dynamic interactions at the mineral–organic matter interface. Nature Reviews. Earth & Environment, 2(6): 402–421

[21] Kong S, Liu X, Jiang H, Hong W, Zhang J, Song W, Yan S. (2023). Photobleaching-induced changes in the optical and photochemical properties of algal organic matter. Water Research, 243: 120395

[22] Kumar A, Rana A, Guo C, Sharma G M, Katubi K M, Alzahrani F M, Naushad M, Sillanpää M, Dhiman P, Stadler F J. (2021). Acceleration of photo-reduction and oxidation capabilities of Bi₄O₅I₂/SPION@calcium alginate by metallic Ag: wide spectral removal of nitrate and azithromycin. Chemical Engineering Journal, 423: 130173

[23] Li, J.L., Zhai, X. and Du, L (2022). Effect of nitrate on the photochemical production of carbonyl sulfide from surface seawater. geophysical research letters, 49, (13): e2021GL097051.

[24] Li X, Shen J, Cao H, Zhang W, Sun Z, Ma F, Gu Q. (2023). Molecular transformation of dissolved organic matter during persulfate-based advanced oxidation: response of reaction pathways to structure. Chemical Engineering Journal, 474: 146256

[25] Liu Y, Wang J. (2019). Reduction of nitrate by zero valent iron (ZVI)-based materials: a review. Science of the Total Environment, 671: 388–403

[26] Mishra A, Alnahit A, Campbell B. (2021). Impact of land uses, drought, flood, wildfire, and cascading events on water quality and microbial communities: a review and analysis. Journal of Hydrology, 596: 125707

[27] Pan Y, Zheng X, Zhao G, Rao Z, Yu W, Chen B, Chu C. (2023). Water vapor condensation on iron minerals spontaneously produces hydroxyl radical. Environmental Science & Technology, 57(23): 8610–8616

[28] Poulton S, Raiswell R. (2002). The low-temperature geochemical cycle of iron: from continental fluxes to marine sediment deposition. American Journal of Science, 302(9): 774–805

[29] Qiao W, Guo H, He C, Shi Q, Xiu W, Zhao B. (2020). Molecular evidence of arsenic mobility linked to biodegradable organic matter. Environmental Science & Technology, 54(12): 7280–7290

[30] Sharpless C M, Linden K G. (2001). UV photolysis of nitrate: effects of natural organic matter and dissolved inorganic carbon and implications for UV water disinfection. Environmental Science & Technology, 35(14): 2949–2955

[31] Shi H, Li C, Wang L, Wang W, Bian J, Meng X. (2022). Photocatalytic reduction of nitrate pollutants by novel Z-scheme ZnSe/BiVO₄ heterostructures with high N₂ selectivity. Separation and Purification Technology, 300: 121854

[32] Shu Z, Pan Z, Wang X, He H, Yan S, Zhu X, Song W, Wang Z. (2022). Sunlight-induced interfacial electron transfer of ferrihydrite under oxic conditions: mineral transformation and redox active species production. Environmental Science & Technology, 56(19): 14188–14197

[33] Silveira J E, de Souza A S, Pansini F N N, Ribeiro A R, Scopel W L, Zazo J A, Casas J A, Paz W S. (2023). A comprehensive study of the reduction of nitrate on natural FeTiO₃: photocatalysis and DFT calculations. Separation and Purification Technology, 306: 122570

[34] TugaoenH O N, Garcia-SeguraS, Hristovski K, WesterhoffP (2017). Challenges in photocatalytic reduction of nitrate as a water treatment technology. Science of the Total Environment, 599–599: 1524–1551

[35] Wang D, Cai L, Song S, Giannakis S, Ma J, Vione D, Pulgarin C. (2024). Bacterial inactivation in sunlit surface waters is dominated by reactive species that emanate from the synergy between light, iron, and natural organic matter. Applied Catalysis B: Environmental, 343: 123573

[36] Wang D, Mueses M A, Márquez J A C, Machuca-Martínez F, Grčić I, Moreira R P M, Puma G L. (2021). Engineering and modeling perspectives on photocatalytic reactors for water treatment. Water Research, 202: 117421

[37] Wang S, Wen J, Mu L, Hu X, Feng R, Jia Y. (2023). Highly active complexes of pyrite and organic matter regulate arsenic fate. Journal of Hazardous Materials, 458: 131967

[38] Wells M J, Hooper J, Mullins G A, Bell K Y. (2022). Development of a fluorescence EEM-PARAFAC model for potable water reuse monitoring: implications for inter-component protein–fulvic–humic interactions. Science of the Total Environment, 820: 153070

[39] Wu B, Berg S M, Remucal C K, Strathmann T J. (2020a). Evolution of N-containing compounds during hydrothermal liquefaction of sewage sludge. ACS Sustainable Chemistry & Engineering, 8(49): 18303–18313

[40] Wu X, Liu P, Gong Z, Wang H, Huang H, Shi Y, Zhao X, Gao S. (2021). Humic acid and fulvic acid hinder long-term weathering of microplastics in lake water. Environmental Science & Technology, 55(23): 15810–15820

[41] Wu Y, Bu L, Duan X, Zhu S, Kong M, Zhu N, Zhou S. (2020b). Mini review on the roles of nitrate/nitrite in advanced oxidation processes: radicals transformation and products formation. Journal of Cleaner Production, 273: 123065

[42] Wu Y, Huang X, Xu J, Huang W, Li J, Mailhot G, Wu F. (2023a). Insight into the effect of natural organic matter on the photooxidation of arsenite induced by colloidal ferric hydroxides in water. Water Research, 232: 119683

[43] Wu Y, Huang X, Xu J, Huang W, Li J, Mailhot G, Wu F. (2023b). Insight into the effect of natural organic matter on the photooxidation of arsenite induced by colloidal ferric hydroxides in water. Water Research, 232: 119683

[44] Xu H, Li Y, Ding M, Chen W, Wang K, Lu C. (2018). Simultaneous removal of dissolved organic matter and nitrate from sewage treatment plant effluents using photocatalytic membranes. Water Research, 143: 250–259

[45] Yang B, Wang C, Cheng X, Zhang Y, Li W, Wang J, Tian Z, Chu W, Korshin G V, Guo H. (2021a). Interactions between the antibiotic tetracycline and humic acid: examination of the binding sites, and effects of complexation on the oxidation of tetracycline. Water Research, 202: 117379

[46] Yang L, Zhang Z, Chen Z. (2021b). Formation of nitrite and ammonium during the irradiation of nitrate-containing water by VUV/UV. Journal of Water Process Engineering, 40: 101801

[47] Yang S, Wang K, Yu X, Xu Y, Ye H, Bai M, Zhao L, Sun Y, Li X, Li Y. (2024). Fulvic acid more facilitated the soil electron transfer than humic acid. Journal of Hazardous Materials, 469: 134080

[48] Yang W, Wang J, Chen R, Xiao L, Shen S, Li J, Dong F. (2022a). Reaction mechanism and selectivity regulation of photocatalytic nitrate reduction for wastewater purification: progress and challenges. Journal of Materials Chemistry. A, Materials for Energy and Sustainability, 10(34): 17357–17376

[49] Yang X, Rosario-Ortiz F L, Lei Y, Pan Y, Lei X, Westerhoff P. (2022b). Multiple roles of dissolved organic matter in advanced oxidation processes. Environmental Science & Technology, 56(16): 11111–11131

[50] Yu C, Qian A, Lu Y, Liao W, Zhang P, Tong M, Dong H, Zeng Q, Yuan S. (2024). Electron transfer processes associated with structural Fe in clay minerals. Critical Reviews in Environmental Science and Technology, 54(1): 13–38

[51] YuHXiB ShiLTanW (2023). Chemodiversity of soil dissolved organic matter affected by contrasting microplastics from different types of polymers. Frontiers of Environmental Science & Engineering, 17(12), 153

[52] Zeng Q, Wang X, Liu X, Huang L, Hu J, Chu R, Tolic N, Dong H. (2020). Mutual interactions between reduced Fe-bearing clay minerals and humic acids under dark, oxygenated conditions: hydroxyl radical generation and humic acid transformation. Environmental Science & Technology, 54(23): 15013–15023

[53] Zhang B, Shan C, Wang S, Fang Z, Pan B. (2021). Unveiling the transformation of dissolved organic matter during ozonation of municipal secondary effluent based on FT-ICR-MS and spectral analysis. Water Research, 188: 116484

[54] Zhao Z, Peng S, Ma C, Yu C, Wu D. (2022). Redox behavior of secondary solid iron species and the corresponding effects on hydroxyl radical generation during the pyrite oxidation process. Environmental Science & Technology, 56(17): 12635–12644

[55] Zheng X, Wu B, Zhou C, Liu T, Wang Y, Zhao G, Chen B, Chu C. (2023). Sunlight-driven production of reactive oxygen species from natural iron minerals: quantum yield and wavelength dependence. Environmental Science & Technology, 57(2): 1177–1185

Acknowledgements

The authors would like to thank the financial support from the National Natural Science Foundation of China (Nos. 52100178, 52370072, and 42361144873) and the Natural Science Foundation of Jiangsu Province (BK20210933).

Conflict of Interests

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Electronic Supplementary Material

Supplementary material is available in the online version of this article at https://doi.org/10.1007/s11783-025-1927-5 and is accessible for authorized users.

RIGHTS & PERMISSIONS

2025 Higher Education Press 2025