Selective photocatalytic mineralization of organic pollutants using eggshell- and seashell-based hydroxyapatite–TiO₂ composites

Abstract

Biogenic hydroxyapatite (HAp)–TiO₂ composites were synthesized from waste eggshells and seashells as calcium precursors and evaluated as sustainable photocatalysts for organic pollutant degradation. The resulting HAp-derived seashell–TiO₂ (HSST) and eggshell–TiO₂ (HEST) catalysts exhibited comparable crystalline phases and porous structures, indicating minimal structural differences. Under UVA irradiation, both catalysts achieved high decomposition efficiencies (> 90%) for BPA, BTA, and MB; however, CR degradation with HEST was comparatively lower, at approximately 80%. Distinct differences were observed in mineralization behavior: HEST showed superior total organic carbon removal for BPA, BTA, and MB, whereas HSST exhibited enhanced mineralization of CR, primarily due to its higher dye adsorption capacity. Reactive oxygen species (ROS) scavenging experiments identified ¹O₂ and O₂^•⁻ as the dominant oxidizing species, with OH• playing a secondary role. Both composites maintained stable degradation performance across a wide pH range, while mineralization was strongly influenced by solution pH and pollutant type through surface charge effects, adsorption–degradation coupling, and reactive species dynamics. Overall, HEST favored mineralization of non-dye and selected dye contaminants, whereas HSST more effectively removed anionic dyes via enhanced adsorption. This study demonstrates a scalable approach for converting shell waste into tailored photocatalysts for sustainable water purification.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-56175-4.

Introduction

Hydroxyapatite (HAp) has attracted increasing attention in environmental engineering and materials science due to its outstanding physicochemical properties, including a strong affinity for organic compounds and inorganic ions, a high ion-exchange capacity, and excellent structural stability¹. These characteristics originate from its unique crystal structure, abundant surface functional groups, and high surface reactivity, allowing it to interact with a wide range of pollutants in aqueous environments². Consequently, HAp has been widely employed in environmental purification processes such as adsorption, ion exchange, chemical precipitation, nuclear waste immobilization, and advanced oxidation and reduction reactions³. In particular, its hydrophilic nature and surface hydroxyl and phosphate groups promote the formation of stable surface complexes with contaminants, enhancing pollutant capture and retention⁴. Due to these multifunctional properties, HAp is recognized as a versatile and promising material for water purification and environmental protection.

In addition to its conventional role as an adsorbent, HAp has increasingly been used as a functional component of photocatalytic systems via the formation of metal-supported structures or composites with semiconductor photocatalysts⁵. Of these systems, HAp–TiO₂ composites have received considerable attention due to their superior photocatalytic performance compared with single-component catalysts. In these composites, TiO₂ nanoparticles are uniformly dispersed and immobilized on the surface of HAp, increasing the number of accessible photoactive sites and improving the light utilization efficiency⁶. The strong interfacial interactions between HAp and TiO₂ also facilitate charge transfer and suppress the recombination of photogenerated electron–hole pairs⁷. This prolongs the lifetime of reactive species and promotes the generation of reactive oxygen species (ROS), thus accelerating pollutant degradation. As a result, HAp–TiO₂ composites exhibit improved photocatalytic activity, stability, and reusability, making them highly attractive for use in water treatment.

Despite these advantages, conventional synthesis routes for HAp-based photocatalysts often involve multiple chemical treatment steps, high-temperature processes, and the extensive use of high-purity reagents⁸. These procedures increase production costs and generate secondary environmental concerns, thus limiting large-scale practical implementation. These complex synthesis pathways also suffer from poor reproducibility and are difficult to translate to an industrial scale⁹. To overcome these limitations, the use of waste biomass materials, such as eggshells and seashells, has been pursued as an environmentally friendly and economically viable alternative for HAp production¹⁰. These forms of waste, which are generated in high volumes by the food and marine industries, are rich in calcium carbonate. Through simple pretreatment, calcination, and chemical conversion processes, high-purity HAp can be efficiently synthesized from these resources. This reduces material costs and energy consumption while promoting waste valorization and resource recycling. Accordingly, biomass-derived HAp represents a sustainable pathway for large-scale catalyst production that is consistent with circular economy principles¹¹.

The performance of HAp–TiO₂ composites synthesized from biomass-derived precursors in the photocatalytic degradation of organic pollutants in aqueous systems has been promising^12,13. However, most existing studies have investigated the removal efficiency of these composites under limited experimental conditions, with relatively few systematic evaluations of pollutant-dependent behavior¹⁴. In particular, comprehensive comparative analysis of various organic contaminants, including ionic dyes, neutral micropollutants, and nitrogen-containing compounds, remains scarce. Furthermore, the influence of the biomass origin on the surface properties, adsorption behavior, ROS generation, and mineralization efficiency of the resulting composite has not been sufficiently clarified, particularly for eggshell- and seashell-derived HAp-based materials¹⁰. This information is required to elucidate structure–performance relationships and establish rational design strategies for advanced photocatalysts. Therefore, integrated studies that simultaneously consider the pollutant type, precursor origin, and reaction mechanisms are essential to improve the efficiency, versatility, and practical applicability of biomass-derived HAp–TiO₂ systems¹⁵.

In response to these research gaps, this study synthesized TiO₂-based photocatalysts incorporating HAp derived from waste eggshells and seashells and systematically evaluated the degradation of representative organic pollutants in water. Shell-based biomass waste was used as a functional catalyst precursor to reduce material production costs while improving resource efficiency and waste minimization. Consequently, the proposed strategy aligns with circular economy principles by converting biogenic waste into value-added materials for environmental remediation. This work also contributes directly to the United Nations Sustainable Development Goals, particularly SDG 6 (Clean Water and Sanitation) and SDG 12 (Responsible Consumption and Production)¹⁶. Overall, the integration of waste-derived HAp with TiO₂ represents a promising and environmentally responsible platform for advanced water-treatment applications.

Results and discussion

Characterization of HEST and HSST

X-ray diffraction (XRD) and Brunauer–Emmett–Teller (BET) analyses were conducted to examine the crystalline phases and surface textural properties of HAp-derived eggshell–TiO₂ (HEST) and HAp-derived seashell–TiO₂ (HSST). The XRD patterns revealed that both materials had similar crystalline compositions, with characteristic diffraction peaks corresponding to HAp and TiO₂, along with minor contributions from calcite, CaO, and Ca(OH)₂ (Fig. 1)^17–21. These findings indicate that both the HEST and HSST catalysts were successfully synthesized as HAp/TiO₂ composite materials. According to the N₂ adsorption–desorption isotherm analysis (Supplementary Fig. 1), both HEST and HSST exhibited H3 hysteresis loops²², which is characteristic of mesoporous and macroporous materials composed of non-rigid aggregates of plate-like particles, indicating the presence of slit-shaped pores within the layered structure. The absence of a distinct adsorption plateau at high relative pressures supported the dominant mesoporous and macroporous architecture of both samples²³. BET analysis (Table 1) confirmed that both HEST and HSST had significantly higher contributions from mesopores and macropores than from micropores. Although HSST exhibited slightly higher specific surface area and total pore volume than HEST, the analytical coefficient of variation was below 4.52%, indicating that these differences were not statistically significant. Overall, the N₂ adsorption–desorption and XRD results demonstrated that HEST and HSST had nearly identical textural and crystalline properties, indicating that the biomass source did not significantly influence their fundamental physicochemical characteristics. This similarity allowed their adsorption and photocatalytic performance to be compared without structural bias.

Fig. 1.

XRD spectra for the biogenic HAp–TiO₂ composites (a) HEST and (b) HSST synthesized from different shell precursors.

Table 1.

Surface area and pore volume of HEST and HSST.

Adsorbent	Surface area (m2/g)		Total pore volume (cm3/g)			Average pore diameter (nm)
	BET	External	Micropore	Mesopore	Total
HEST	26.572	11.746	0.005	0.132	0.137	20.652
HSST	27.536	14.743	0.005	0.145	0.150	21.804

Photocatalytic degradation and mineralization of organic pollutants

The photocatalytic removal performance of HEST and HSST was evaluated using bisphenol A (BPA) and benzotriazole (BTA) as representative non-dye contaminants and congo red (CR) and methylene blue (MB) as dye contaminants. As shown in Fig. 2, both catalysts exhibited comparable removal efficiencies for BPA, BTA, and MB, achieving degradation rates of more than 92%. Overall, a higher removal efficiency was observed for non-dye contaminants than for dye contaminants. Notably, HSST had a higher removal rate for the dyes than HEST, particularly CR. The enhanced affinity of HSST for dye molecules was supported by adsorption experiments, which revealed higher adsorption capacities for CR and MB compared with HEST (Supplementary Fig. 2). This behavior can be attributed to the strong interactions between the functional groups of CR and MB and Ca²⁺ species present on the catalyst surface²⁴. Consequently, HSST exhibited more favorable degradation kinetics for dye contaminants, indicating an advantage in its pollutant decomposition efficiency (Supplementary Table 1).

Fig. 2.

Photocatalytic removal performance of HEST and HSST for organic pollutants (60 µM organic pollutant, 1 g∙L^− 1 catalyst, time = 120 min).

Despite the similarly high degradation efficiencies of both composites, clear differences were observed in their mineralization performance, as reflected by total organic carbon (TOC) removal (Fig. 3). For BPA, HEST and HSST achieved comparable TOC removal of approximately 20%, indicating similar mineralization behavior. In contrast, HEST exhibited substantially higher TOC removal for BTA and MB, reaching about 47% and 37%, respectively, suggesting more effective oxidation of degradation intermediates. Conversely, HSST showed superior mineralization for CR, achieving a TOC removal of approximately 45%, compared with less than 10% for HEST, which is consistent with its higher adsorption capacity that enhances contact between dye molecules and reactive sites²⁵. However, for MB, despite stronger adsorption onto HSST, HEST achieved higher TOC removal, indicating that adsorption alone does not ensure effective mineralization and that surface reactivity and reactive oxygen species dynamics also play critical roles²⁶. Overall, these results demonstrate that while both HEST and HSST effectively degrade target pollutants, mineralization efficiency strongly depends on pollutant type and catalyst properties, highlighting the complementary roles of enhanced dye adsorption by HSST and superior mineralization capability of HEST in photocatalytic water treatment.

Fig. 3.

Photocatalytic TOC removal performance of HEST and HSST for organic pollutants (60 µM organic pollutant, 1 g∙L^− 1 catalyst, time = 120 min).

Role of ROS species in HEST and HSST photocatalysis

To identify the dominant ROS involved in pollutant degradation over HEST and HSST, scavenging experiments were conducted using TBA, L-his, and BQ to selectively quench hydroxyl radicals (OH•), singlet oxygen (¹O₂), and superoxide radicals (O₂^•⁻), respectively (Fig. 4). For BPA, the addition of all scavengers reduced removal efficiency, although TBA caused only minor inhibition, indicating a limited contribution of OH•, whereas L-his and BQ strongly suppressed degradation, reducing removal to below 20% and confirming the dominant roles of ¹O₂ and O₂^•⁻. A similar trend was observed for BTA, with ¹ O₂ and O₂^•⁻ governing degradation and OH• playing a negligible role. In contrast, the degradation of the dyes CR and MB was strongly affected by both scavenging and adsorption effects: BQ and L-his markedly suppressed CR and MB removal, respectively, while simultaneously increasing their adsorption onto HEST and HSST, suggesting that the scavengers acted as molecular bridges enhancing surface accumulation (Supplementary Fig. 3)²⁷. Notably, BQ addition to the CR system reduced removal efficiency below that achieved by adsorption alone, indicating competitive adsorption of BQ that displaced CR from reactive sites and suppressed photocatalytic degradation. Overall, both HEST and HSST exhibited similar inhibition patterns, identifying ¹ O₂ and O₂^•⁻ as the dominant oxidizing species and OH• as secondary, implying that differences in mineralization efficiency arise primarily from variations in surface physicochemical properties rather than ROS generation²⁸.

Fig. 4.

Photocatalytic removal performance of HEST and HSST in the presence of the scavengers (a) BPA, (b) BTA, (c) CR, and (d) MB (60 µM organic pollutant, 1 g∙L^− 1 catalyst, 10 mM scavenger, time = 120 min).

Scavenger experiments demonstrated that O₂^•⁻ and ¹ O₂ rather than OH• are the major ROSs causing pollutant decomposition in HEST and HSST photocatalytic systems. To further confirm the actual production of these ROSs during photocatalytic reactions, further probe experiments were conducted using selective ROS probe compounds. Nitro blue tetrazolium (NBT) and phenyl methyl sulfoxide (PMSO) were used as probe molecules of O₂^•⁻, and ¹ O₂, respectively (Fig. 5, Supplementary Fig. 4). During photocatalytic reactions catalyzed by HEST and HSST, all probe compounds showed a sharp decrease in concentration over time. In particular, the significant consumption of NBT and PMSO strongly indicated the active production of O₂^•⁻ and ¹ O₂ within the photocatalytic system. These results are consistent with the results of scavenger experiments, further confirming that O₂^•⁻ and ¹ O₂ are the major ROSs contributing to pollutant decomposition in HEST and HSST mediated photocatalytic processes.

Fig. 5.

Probe test of HEST and HSST (50 µM BA, NBT, PMSO, 1 g∙L^− 1 catalyst, time = 120 min).

Effect of pH on performance of HEST and HSST

To verify the influence of surface properties on photocatalytic performance, the effects of solution pH on the degradation and mineralization of BTA, BPA, CR, and MB were systematically examined under pH 3, pH 6, and pH 9 conditions. As shown in Figs. 6a and c and 7a and c, both HEST and HSST maintained consistently high degradation efficiencies across the tested pH range, indicating strong operational stability. For BPA and BTA, the removal efficiency was consistently maintained at about 80–95% under all test conditions. A slight improvement in the removal performance was observed at pH 3, which can be attributed to the favorable electrostatic interaction and efficient ROS generation under acidic conditions. On the other hand, under alkaline conditions, it was observed that the removal efficiency was slightly reduced due to the electrostatic repulsion between the catalyst surface and the contaminant molecule and the partial removal of ROS by hydroxide ions. This surface charge-dependent behavior can be explained based on the zero-charge point (pzc) values of HEST and HSST and the pKa values of the target contaminants. The pH_pzc values for HEST and HSST were about 8–10, whereas the pKa values for BPA and BTA were about 9.8–10.4 and 8.2, respectively (Supplementary Fig. 6)^29,30. Under acidic and neutral pH conditions, BPA and BTA are mainly present in the neutral form, minimizing the electrostatic repulsion and promoting photocatalytic decomposition. However, under alkaline conditions, the two contaminants are more easily deprotonated and are mainly present as negatively charged species. Therefore, the electrostatic repulsion between the negatively charged catalyst surface and the anionic contaminant is greater, resulting in a reduced removal performance at higher pH values. For CR degradation, the decomposition efficiency showed no significant difference between pH 6 and pH 9; however, the efficiency increased under pH 3 conditions. This result is consistent with the electrostatic attraction behavior described above, where CR was rapidly removed through strong adsorption onto the Ca²⁺-rich HAp surface (Supplementary Fig. 5)²⁴. In contrast, MB exhibited a slight decrease in decomposition efficiency at pH 3. In contrast, TOC removal exhibited a stronger pH dependence (Figs. 6b and d and 7b and d). Alkaline conditions promoted BPA mineralization with HSST, acidic conditions promoted CR mineralization, especially with HEST, and BTA mineralization was strongly inhibited at low pH for both catalysts^31–33. MB mineralization was only weakly affected by pH, with improved TOC removal by HEST at pH 3 and a slight decline for HSST³⁴. Overall, these results demonstrate that pH-dependent mineralization is governed by the combined effects of pollutant properties, catalyst surface affinity, and ROS dynamics specific to HEST and HSST.

Fig. 6.

Effects of solution pH on the photocatalytic removal and mineralization performance of HEST and HSST: (a) BPA degradation, (b) BPA TOC removal, (c) BTA degradation, and (d) BTA TOC removal (60 µM organic pollutant, 1 g∙L^− 1 catalyst, time = 120 min).

Fig. 7.

Effects of solution pH on the photocatalytic removal and mineralization performance of HEST and HSST: (a) CR degradation, (b) CR TOC removal, (c) MB degradation, and (d) MB TOC removal (60 µM organic pollutant, 1 g∙L^− 1 catalyst, time = 120 min).

Catalyst stability of HEST and HSST

HEST and HSST synthesized from waste resources showed effective photocatalytic activity and excellent removal performance against various pollutants. Nevertheless, evaluation of catalyst stability is essential to evaluate its practicality as a photocatalyst. Accordingly, a reuse experiment was conducted to check whether the photocatalytic performance of HEST and HSST can be maintained in repeated cycles. In addition, the stability of both catalysts was further evaluated by comparing the physicochemical characterization results obtained before and after the reuse experiment. First, the reuse experiment showed that both HEST and HSST maintain stable photocatalytic performance in three consecutive cycles (Fig. 8, Supplementary Fig. 7). For the organic pollutant (BTA), both catalysts were completely removed in all reuse cycles. In the case of dye pollutant (MB), the removal efficiency was found to be slightly reduced after repeated use, but the decomposition efficiency was maintained at more than 80%, indicating that both catalysts maintained overall photocatalytic activity and stability during repeated operation³⁵.

Fig. 8.

Reusability study of HEST and HSST: (a) HEST-BTA, (b) HEST-MB, (c) HSST-BTA, and (d) HSST-MB (60 µM organic pollutant, 1 g∙L^− 1 catalyst, time = 120 min for 3 cycles).

To further investigate the stability of the catalyst after reuse, SEM, XRD, and XPS analyses were performed to evaluate the potential changes in surface morphology, crystal structure, and surface chemical state. SEM-EDS analysis revealed that both fresh HEST and HSST showed irregular and non-uniform surface morphology with similar particle sizes (Supplementary Fig. 8a–d). After the photocatalytic experiment, no significant morphological changes or particle aggregation was observed in either catalyst (Supplementary Fig. 8e–l). In addition, no noticeable difference was observed in the surface morphology depending on the type of contaminant treated. These observations indicate that the two catalysts maintained structural stability during repeated photocatalytic reactions. Similar results were derived in XRD analysis. After the reuse experiment, the diffraction peaks characteristic of HEST and HSST hardly changed, indicating that the crystal structure of both catalysts was well preserved during photocatalytic operation (Supplementary Fig. 9). However, the EDS results showed that there was a slight change in elemental composition after the reaction (Supplementary Table 2), suggesting that the surface element distribution and chemical bonding state may have partially changed as the catalyst participated in the photocatalytic process.

In addition, XPS analysis revealed a distinct surface interaction mechanism during the photocatalytic decomposition of BTA and MB in both HEST and HSST (Supplementary Figs. 10–15). In the C 1s spectrum, the intensity of the C–H peak decreased after photocatalytic decomposition, whereas the intensities of the C–O–C and C = O peaks (286 eV and 287–288 eV) increased for both catalysts³⁶. The O 1s spectrum showed an increase in the intensity of the hydroxyl-related peak after the photocatalytic reaction, suggesting that additional surface hydroxyl groups were formed³⁷. In the Ti 2p spectrum, a peak of about 457.5 eV decreased, whereas the intensity of other Ti-related peaks increased after the reaction. This change is related to the formation of Ti³⁺ species induced by the oxygen-vacancy production during the photocatalyst³⁸. Partial reduction of Ti⁴⁺ to Ti³⁺ under light irradiation results in oxygen deletion, which can easily adsorb and dissociate water molecules to form surface hydroxyl groups³⁹. These hydroxyl groups subsequently contribute to the production of ROS, enhancing photocatalytic activity⁴⁰. Under light irradiation, photogenerated electrons are transferred to dissolved oxygen molecules adsorbed on oxygen-vacancy and Ti³⁺ defect sites, resulting in the formation of O₂^•⁻⁴¹. Surface hydroxyl groups further facilitate interfacial charge transfer and promote ROS generation. Subsequently, the generated O₂^•⁻ species participate in secondary ROS formation pathways, leading to the production of ¹O₂. In the N 1s spectrum, no apparent nitrogen-related peaks were observed after either fresh catalyst or BTA decomposition. However, after MB decomposition, a new N 1s peak appeared, indicating that residual MB molecules and/or intermediate products were adsorbed on the catalyst surface.

Overall, these results suggest that photocatalytic degradation proceeded through interfacial electron-transfer interactions between the catalyst surface and pollutant molecules, leading to the generation of reactive oxygen species (ROS) and subsequent pollutant decomposition. Furthermore, the HEST and HSST catalysts exhibited both high pollutant removal efficiencies and excellent catalytic stability during repeated use⁴². These findings demonstrate that the developed catalysts possess competitive photocatalytic performance compared with previously reported photocatalysts (Supplementary Table 3). In addition, the TOC removal efficiency increased with increasing catalyst dosage, suggesting that further optimization of the process conditions could enhance the mineralization performance of pollutants (Supplementary Fig. 16). These results indicate the potential for complete pollutant decomposition and the eventual reduction of toxicity through the removal of intermediate byproducts³⁵.

Conclusion

In this study, biogenic HAp–TiO₂ composites were successfully synthesized using waste eggshells and seashells as calcium precursors and systematically evaluated for the photocatalytic removal of representative organic pollutants. Structural characterization confirmed that HEST and HSST had similar crystalline phases and porous structures, enabling a reliable comparison of their photocatalytic performance without structural bias. Both catalysts exhibited high degradation efficiencies for BPA, BTA, CR, and MB, with removal rates exceeding 90% in most cases. However, clear differences were observed in their mineralization behavior. HEST achieved higher TOC removal rates for BTA and MB, whereas HSST exhibited enhanced mineralization of CR, mainly due to its higher dye adsorption capacity. These results indicate that mineralization efficiency is governed not only by adsorption behavior but also by surface reactivity and ROS activity. HEST and HSST exhibited similar inhibition patterns, indicating comparable photocatalytic pathways and ROS generation mechanisms. Differences in mineralization performance were attributed primarily to variation in their surface physicochemical properties. Both catalysts exhibited stable degradation over a wide pH range, while mineralization was more sensitive to the pH level and pollutant type. HEST demonstrated superior mineralization of non-dye and selected dye contaminants, whereas HSST achieved enhanced removal of anionic dyes due to its higher adsorption capacity. These complementary properties highlight the potential for the development of target-specific biogenic photocatalysts and the use of shell waste valorization for sustainable water treatment.

Method

Chemicals and analysis methods

All used chemicals and analysis methods in this study are described in the Supplementary Information (Supplementary Text 1).

Synthesis of HEST and HSST

Biogenic HAp–TiO₂ composites were synthesized from waste eggshells and compressed seashells following the procedure shown in Fig. 9. For HEST preparation, waste eggshells were washed with deionized water, dried at 80 °C for 24 h, and calcined at 1000 °C for 1 h to obtain CaO. This calcination condition was selected to ensure the thermal decomposition of CaCO₃-rich shell precursors into reactive CaO, while the 1 h holding time was sufficient to achieve precursor conversion without excessive energy consumption and particle coarsening. The CaO was dispersed in deionized water and reacted with H₃PO₄ at 90 °C under continuous stirring until the pH reached approximately 7, followed by aging for 7 d to promote hydroxyapatite formation. The resulting precipitate was collected, dried at 105 °C for 1 h, and calcined at 1000 °C for 1 h to obtain HAp. This additional calcination step was applied to improve the phase development and crystallinity of HAp after wet precipitation. The obtained HAp was then mixed with TiO₂ at a 1:1 weight ratio in 50 mL of ethanol and stirred for 1 h. After drying at 60 °C for 24 h, the mixture was calcined at 500 °C for 3 h to produce HEST. The final calcination at 500 °C for 3 h was chosen as a moderate thermal treatment to remove residual solvent and moisture, enhance the interfacial contact between HAp and TiO₂, and stabilize the composite structure while avoiding severe sintering or undesirable phase transformation. The HSST composite was synthesized using compressed seashells as the calcium precursor following the same procedure, yielding HAp–TiO₂ composites with identical processing conditions.

Fig. 9.

Schematic illustration of biogenic HAp synthesis from eggshells and compressed seashells via CaO formation and wet precipitation.

Photocatalytic degradation experiments

The adsorption behavior and photocatalytic degradation performance of HEST and HSST were evaluated using BPA and BTA as non-dye contaminants and CR and MB as dye contaminants. In each experiment, 50 mL of a 60 µM aqueous pollutant solution was placed in a quartz beaker, and the catalyst was added at a loading of 1.0 g L^− 1. The suspension was magnetically stirred in the dark for 30 min to establish adsorption–desorption equilibrium, after which photocatalytic reactions were initiated under UVA irradiation (λ = 365 nm) using a photoreactor equipped with six 4 W lamps housed in a black acrylic chamber. Aliquots (1.0 mL) were collected at − 30, 0, 30, 60, and 120 min, where 0 min corresponded to the onset of irradiation, immediately filtered through a 0.20 μm PTFE membrane filter, and analyzed; adsorption was quantified during the dark period (− 30 to 0 min). To examine pH effects, the initial solution pH was adjusted to 3, 6, or 9 using H₂SO₄ or NaOH prior to catalyst addition, while blank experiments were conducted without pH adjustment. Reactive oxygen species scavenging experiments were performed by adding TBA, L-his, or BQ (10 mM) before pollutant addition, followed by adjustment of the pollutant concentration to 60 µM and application of the same catalyst addition, dark adsorption, and UVA irradiation procedures. All experiments were conducted at least duplicate, and the results are reported as mean and standard deviation. Statistical significance was evaluated using a t-test, with significance confirmed at p < 0.05.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

H.Y.: Methodology, Validation, Investigation, Writing—original draft. M.K.: Resources, Methodology, Investigation. S-H.H.: Validation, Investigation, Writing. L.S.: Resources, Methodology, Investigation. I.S.: Resources, Methodology, Investigation. S.A.: Resources, Supervision, Funding acquisition, Writing—Review & Editing. C.-G.L.: Conceptualization, Resources, Supervision, Funding acquisition, Writing—Review & Editing. J.L.: Conceptualization, Resources, Supervision, Writing—Review & Editing.

Funding

This work was supported by the Science Committee of the Ministry of Science and Higher Education of the Republic of Kazakhstan (Grant No. BR27199301), the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No. RS-2025-25412299), with additional funding support from the Ministry of Science and ICT and the Korea Research Foundation (No. RS-2024-00455709).

Data availability

Data supporting the findings of this study are available upon request.

Declarations

Competing interests

The authors declare no competing interests.