1. Introduction Volatile organic compounds (VOCs) pose a significant threat to the environment and human health due to their carcinogenic, teratogenic and mutagenic properties [ 1]. The steady rise in annual toluene emissions has aroused widespread concern due to its ubiquitous presence in emissions from sectors such as plastics manufacturing, printing, and transportation [ 2]. Consequently, controlling VOC emissions has become a growing priority in air pollution research. Conventional catalytic combustion of VOCs in gas–solid systems often suffers from catalyst deactivation and the generation of highly toxic byproducts such as dioxins [ 3]. In this context, the development of novel approaches for VOC treatment is of paramount importance, with a specific focus on achieving both high removal efficiency and enhanced environmental sustainability [ 4]. Recently, advanced oxidation processes (AOPs) have been widely used for VOC elimination in a wet scrubber due to the generation of highly reactive oxygen species (ROS) [ 5], such as hydroxyl radical (•OH), sulfate radical (SO 4•−) [ 6], superoxide radical (O 2•−) [ 7], and singlet oxygen ( 1O 2) [ 8]. Unlike catalytic combustion, AOPs enable VOC removal with higher efficiency, simpler operational requirements, lower costs, and reduced energy consumption [ 9]. Toluene is continuously transferred into the liquid phase via bubbling, necessitating the use of heterogeneous catalysts with both high activity and stability to activate oxidants such as hydrogen peroxide (H 2O 2) and peroxymonosulfate (PMS) [ 10]. However, the application of these catalysts is hindered by several barriers: low availability of active sites, a narrow pH operating range, and high metal leaching under real-world conditions [ 11]. A major limitation of AOPs is that most reactive oxygen species possess extremely short lifetimes (10 −6–10 −9 s) and undergo self-quenching, which strongly prevents them from effectively contacting and degrading pollutants [ 12]. Thus, the design of highly active and durable catalysts for peroxide activation continues to face major obstacles, especially in boosting ROS utilization within heterogeneous AOP systems—a critical step toward the complete oxidation of VOC-laden off-gases. Over the past few years, transition metal sulfides have attracted great attention [ 13]. Transition metal sulfides based AOPs usually rely on the metal active site on the catalyst surface to activate persulfate including PMS (peroxymonosulfate) and PDS (peroxydisulfate) and promote the degradation of organic pollutants, which is similar to the activation mechanism of homogeneous transition metals [ 14]. The role of sulfur on the catalyst surface can be summarized into two aspects: one is the reducing property of catalyst surface S 2− improving electron transfer efficiency [ 15]; the other is that the unsaturated S on the surface of the metal sulfide traps the protons in the solution, producing H 2S and exposing metal active sites, thereby accelerating the reaction [ 16]. Although transition metal sulfides can efficiently activate PMS to generate ROS, their ability to adsorb VOCs is very weak [ 17]. Consequently, the problem of insufficient mass transfer between gas and liquid has not been fundamentally solved, leading to inadequate contact between VOCs and ROS [ 18]. In this study, we designed a novel heterojunction catalyst (M50C50) by ball milling MIL-100(Fe) (a classical MOF) and CoS (a typical transition metal sulfide). The key novelty of this work is threefold. First, unlike previous studies that used MOFs or metal sulfides separately, our M50C50 composite integrates the high adsorption capacity of MIL-100(Fe) for gaseous toluene with the efficient PMS activation ability of CoS, overcoming the long-standing issue of poor gas–liquid mass transfer in AOP wet scrubbers. Second, the heterojunction interface between MIL-100(Fe) and CoS facilitates interfacial electron transfer from the MOF to CoS (as evidenced by XPS binding energy shifts), which lowers the energy barrier for PMS activation and promotes the generation of reactive radicals. Third, the system achieves over 90% toluene degradation within 90 min over a wide pH range (3–11) and reaches ~75% mineralization (CO 2 evolution), with significantly reduced toxic byproducts compared to conventional catalytic combustion. Through identification of reactive oxygen species and intermediates, we elucidate the synergistic mechanism involving interfacial electron transfer and heterojunction-driven PMS activation. To the best of our knowledge, this is the first report combining MOF adsorption and metal-sulfide-driven PMS activation in a wet scrubber for deep oxidation of gaseous VOCs. This work provides not only an efficient catalyst but also a mechanistic insight, laying a foundation for the application of AOP wet scrubbers in industrial waste gas treatment. 2.1. Characterizations of M50C50 The crystal structures of MIL-100(Fe), CoS, and MxCy composites (M70C30, M50C50, and M30C70) were characterized by PXRD in Figure 1a. The diffraction pattern of MIL-100(Fe) is consistent with the simulated pattern reported in the literature [ 22]. The four typical peaks at 2θ = 30.6°, 35.4°, 47.0° and 54.5° correspond to the (100), (101), (102) and (110) crystal planes of standard CoS respectively [ 23]. Confirming the successful synthesis of hexagonal CoS. The PXRD patterns of the MxCy composites exhibit characteristic peaks of both MIL-100(Fe) and CoS, with no detectable impurity peaks. This indicates that the crystal structures of both components are preserved through the ball milling process. Thus, the successful preparation of the MxCy composites is confirmed [ 24]. The composition of MxCy composite materials was determined by X-ray diffraction patterns (XRD). As shown in Figure 1b, the broad peak of pure MIL-100(Fe) near 3400 cm −1 corresponds to the O-H stretching vibration, the peak near 1600 cm −1 corresponds to the C=O stretching vibration of carboxylic groups. Furthermore, the peaks in the range of 1300 cm −1–1000 cm −1 correspond to the C-O stretching vibration [ 25]. The peak at 481 cm −1 is ascribed to the Fe-O stretching vibration [ 26]. The intensity of this band fell with decreasing MIL-100(Fe) proportion in the composites, while another peak at 1176 cm −1—assigned to the Co=S stretching vibration—was also observed. The presence of this peak, along with the characteristic peaks of MIL-100(Fe) confirmed the presence of CoS [ 27]. Although no obvious FT-IR peak shift was detected, the XPS and electrochemical results suggest an electronic interaction at the heterojunction interface. This interaction provides evidence for the formation of an interfacial bond between the two components. The surface chemical composition of MIL-100(Fe), CoS, and the M50C50 was investigated by XPS in Figure 1c. The survey spectrum of pure MIL-100(Fe) shows the main characteristic peaks of C 1s, O 1s and Fe 2p, while that of pure CoS shows peaks for Co 2p, C 1s, O 1s, and S 2p. The survey spectrum of the composite material M50C50 simultaneously contains all the characteristic peaks of C 1s, O 1s, Fe 2p, Co 2p and S 2p, and the relative peak intensities are consistent with the nominal composition. These results directly confirm the coexistence of MIL-100(Fe) and CoS in the composite and the successful formation of the hybrid material. The high-resolution XPS spectrum of Co 2p for pure CoS are shown in Figure 1d. The Co 2p of pure CoS can be deconvoluted into the characteristic peaks of Co 2+ (781.83 eV, 798.14 eV) and Co 3+ (793.51 eV, 778.34 eV), along with two satellite peaks (802.99 eV, 785.47 eV), which is consistent with the mixed valence state of Co in CoS. The Co 2p spectrum of the composite material M50C50 can be fitted as the characteristic peaks of Co 2+ (782.05 eV, 797.84 eV) and Co 3+ (793.93 eV, 778.82 eV), accompanied by two satellite peaks (802.47 eV, 786.68 eV) [ 28]. Notably, the binding energy of Co 2+ exhibits a slight negative shift (0.2–0.3 eV), while that of Co 3+ shows a positive shift (0.4–0.5 eV) compared to pure CoS. These shifts suggest an electron transfer from MIL-100(Fe) to CoS, leading to a slight reduction in the valence state of Co. As expected, pure CoS does not show any Fe 2p peak, confirming that Fe in the composite originates solely from MIL-100(Fe).The high-resolution XPS spectrum of Fe 2p for MIL-100(Fe) and M50C50 as single and composite materials are shown in Figure 1e. The Fe 2p spectrum of pure MIL-100(Fe) can be fitted as the characteristic peaks of 711.60 eV and 724.88 eV, along with two satellite peaks (716.31 eV, 729.44 eV), which is consistent with the mixed valence state of Fe in MIL-100(Fe). The Fe 2p spectrum of the composite material M50C50 can be fitted as the characteristic peaks of 711.54 eV and 724.40 eV, along with two satellite peaks (730.13 eV, 716.33 eV) [ 29]. Pure MIL-100(Fe) has no Co 2p peak, further verifying that Co in the composite material only comes from CoS. The high-resolution XPS spectrum of S 2p for CoS and M50C50 as single and composite materials are shown in Figure 1f. The S 2p spectrum of pure CoS can be fitted as S 2− (163.92 eV, 162.62 eV), SO 32− (169.77 eV), and SO 42− (168.59 eV) [ 30]. The S 2p spectrum of the composite material M50C50 can be fitted as S 2− (163.23 eV, 162.03 eV), S n2− (164.39 eV), SO 32− (169.77 eV), and SO 42− (168.59 eV). The peak area ratio of S n2− slightly increases, and the binding energy of S 2− shows a positive shift of 0.2 eV. These changes indicate an interaction between the Fe atoms in MIL-100(Fe) and the S atoms in CoS. The opposite directional shifts in the binding energies of Co 2p and Fe 2p further confirm electron transfer across the interface from MIL-100(Fe) to CoS. This interfacial charge redistribution is characteristic of heterojunction formation, which is expected to enhance the catalytic performance by facilitating charge separation and transfer. The porous structure of the as-prepared samples was characterized by nitrogen adsorption–desorption measurements. The MIL-100(Fe) sample exhibits a typical type I isotherm with a steep nitrogen uptake at low relative pressures (P/P 0 < 0.1), indicating the presence of abundant micropores ( Figure S1). A slight hysteresis loop can be observed at medium relative pressures, suggesting the existence of mesopores, which may originate from interparticle voids. In contrast, both CoS and the M50C50 display relatively low nitrogen adsorption capacities, and their isotherms show no obvious micropore filling at low pressure, implying a non-porous or macroporous structure. MIL-100(Fe) possesses a high specific surface area of 1150.82 m 2·g −1 and a large pore volume of 0.595 cm 3·g −1, with an average pore diameter of 2.07 nm, further confirming its microporous nature. In comparison, pristine CoS exhibits a very low surface area of only 2.83 m 2·g −1 and a negligible pore volume of 0.025 cm 3·g −1, consistent with its non-porous characteristics. After compositing with MIL-100(Fe), the M50C50 sample shows a slightly increased surface area of 7.27 m 2·g −1 and pore volume of 0.039 cm 3·g −1, indicating that the incorporation of MIL-100(Fe) slightly improves the porosity of the composite, although the overall surface area remains limited due to the dominant contribution of CoS ( Table S1). MIL-100(Fe) exhibits a very high specific surface area (1150.82 m 2/g) but only moderate degradation efficiency (ca. 20%), whereas CoS and M50C50 possess much lower surface areas (2.83 and 7.27 m 2/g) but achieve much higher efficiencies (80% and 92%, respectively). This clearly indicates that the catalytic performance is not dominated by physical adsorption; instead, the chemical activation ability of CoS and the synergistic effect in the M50C50 composite play the decisive role ( Figure S8). The SEM image of MIL-100(Fe) is shown in Figure 2a. MIL-100(Fe) mainly has an octahedral morphology with a particle size of 1 to 3 μm. The SEM image of CoS is shown in Figure 2b. The morphology of CoS is mainly spherical. The SEM image of M50C50 is shown in Figure 2c. After ball milling MIL-100(Fe) and CoS to form M50C50, part of the octahedral morphology of MIL-100(Fe) is retained. The morphology of CoS after ball milling is no longer spherical but irregular and uniformly distributed on the surface of MIL-100(Fe). The TEM image of M50C50 is shown in Figure 2d. The TEM image of M50C50 material can also show that CoS with an irregular shape is distributed around MIL-100(Fe). Figure 2e is the SEM image of material M50C50. Figure 2f–i show that Co, O, S and Fe are uniformly distributed throughout the M50C50, both on the surface and in the interior. Which is consistent with the results of the above SEM and TEM images. These results collectively confirm the successful preparation of the M50C50. The electrochemical properties of the as-prepared samples were further investigated by electrochemical impedance spectroscopy (EIS). The diameter of the semicircle in the high-frequency region reflects the charge transfer resistance at the electrode/electrolyte interface. Among all samples, MIL-100(Fe) exhibits the smaller semicircle diameter, indicating the lowest charge transfer resistance and the most efficient interfacial charge transport. In contrast, CoS shows a significantly larger semicircle, suggesting poor conductivity and limited charge transfer kinetics. The M50C50 displays an intermediate semicircle diameter, implying that the incorporation of MIL-100(Fe) into CoS effectively enhances the charge transfer capability compared to pristine CoS. These results demonstrate that the introduction of MIL-100(Fe) improves the electrical conductivity and facilitates interfacial charge migration compared to pristine CoS. The M50C50 is therefore beneficial for catalytic or electrochemical performance ( Figure S2). 2.2. Catalytic Removal of Toluene in AOP Wet Scrubber As shown in Figure 3a, the removal efficiency of toluene by MIL-100(Fe), CoS and different proportion composite materials (M30C70, M50C50, M70C30) was tested in a control experiment. The experimental results showed that under a reaction time of 90 min, the aqueous solution system alone had only about 7% removal rate for toluene, indicating that toluene is insoluble in water. To improve the removal effect of toluene, after adding PMS, within 90 min, the removal rate of toluene increased from 7% to about 15%, suggesting that the H 2O/PMS system had a slight improvement in the removal performance of toluene. This might be because PMS dissolves in water and can produce a small amount of SO 4•−1O 2 reactive oxygen species [ 31], which exerted a weak oxidation effect on toluene. In the H 2O/CoS system, the removal rate of toluene was around 10%, indicating that the adsorption of CoS on toluene was very weak, almost not adsorbing toluene. While in the H 2O/MIL-100(Fe) system, the removal rate of toluene increased to about 20%, which might be because the dispersed MIL-100(Fe) in the water solution exerted a certain adsorption effect on toluene, increasing the removal rate of toluene [ 32]. It is worth noting that in the H 2O/PMS/CoS system, the removal rate of toluene increased to about 80%, which might be because CoS activated PMS to produce a large amount of reactive oxygen species, which exerted a strong oxidation effect on toluene, resulting in a significant increase in the removal rate of toluene. In the H 2O/PMS/M50C50 system, all three composites with different mass ratios achieved toluene removal efficiencies of approximately 90%. Among which the H 2O/PMS/M50C50 system had the highest removal rate of 92%, compared with the H 2O/PMS/CoS system, the toluene molecules adsorbed on the surface of MIL-100(Fe) were more likely to undergo oxidation or electron charge transfer with the reactive oxygen species, and at the same time, the impedance of the liquid film decreased, promoting the mass transfer between the gas and liquid phases, thereby enhancing the removal rate of toluene, which was significantly higher than that of a single material. The uniform composite material increased the exposure of active sites, jointly improving the adsorption and degradation efficiency of toluene. As shown in Figure 3b, in the H 2O/PMS/CoS system, the production of carbon dioxide was around 65 ppmv, and the mineralization rate of toluene was about 32%. While in the H 2O/PMS/M50C50 system, the production of carbon dioxide was around 160 ppmv, indicating that toluene was mineralized into carbon dioxide and water to a greater extent, resulting in an increase in the production of carbon dioxide in the reactor, and the mineralization rate of toluene increased to about 75%, further proving that toluene can be effectively removed, and the mineralization degree of the composite material was greater than that of the single CoS material [ 33]. It should be noted that the mineralization efficiency in this study was evaluated based on carbon dioxide evolution (approximately 75% within 90 min), whereas total organic carbon (TOC) analysis of the liquid phase was not performed. Therefore, the possibility of residual organic intermediates in the liquid phase cannot be completely excluded. Future work will combine TOC measurement with intermediate identification (LC-MS or GC-MS) to establish a more comprehensive carbon mass balance and further validate the complete degradation of VOCs. To evaluate the synergistic effect between CoS and MIL-100(Fe) in the M50C50, control experiments were conducted using a physical mixture of CoS and MIL-100(Fe) with the same mass ratio under identical reaction conditions. The catalytic activity of both materials was monitored over a continuous operation of 90 min. In contrast, the physical mixture of CoS and MIL-100(Fe) showed notably lower catalytic activity. The initial removal efficiency was 82% at 0 min, and it gradually declined to 80% by the end of the 90 min reaction. The physical mixture exhibited a mean toluene removal of roughly 80.1%, lagging nearly 10% behind the M50C50. Moreover, a slight deactivation trend was observed for the physical mixture, suggesting weaker stability compared to the composite ( Figure S3). The intimate heterojunction interface in the M50C50 facilitates more efficient charge transfer and promotes the activation of PMS, whereas the physical mixture lacks such synergistic coupling, resulting in inferior performance. These results clearly demonstrate that the composite formation, rather than a simple combination of components, is essential for achieving optimal catalytic efficiency. From the above control experiments and the data of carbon dioxide production and mineralization rate, it is easy to see that the M50C50 material has a better removal effect on toluene. Therefore, this material was selected for the subsequent exploration of reaction conditions. The optimization of reaction parameters is of great significance for the practical application of PMS activation. Therefore, in the following research, various reaction parameters such as pH value, dosage of M50C50, and concentration of PMS were mainly explored to investigate the influence on the removal rate of toluene. M50C50 is an important activating substance for generating sulfate free radicals, so adding an appropriate dose of M50C50 is one of the prerequisites for generating sufficient reactive oxygen species [ 34]. Therefore, the influence of different M50C50 dosages (0.01 g/L, 0.02 g/L, 0.05 g/L, 0.1 g/L, and 0.2 g/L) on the removal efficiency of toluene was studied. As shown in Figure 3c, when the dosage of M50C50 increased from 0.01 g/L to 0.1 g/L, the removal rate of toluene rose from 63% to 92.7%, and when the dosage was increased to 0.2 g/L, the removal rate of toluene reached up to 89.3%, even slightly decreased. This might be because the increase in the dosage of M50C50 led to more active sites in the solution, generating more reactive oxygen species, which further improved the removal efficiency of toluene. However, when the dosage of M50C50 continued to increase, the utilization rate of PMS reached a saturation point, and, at this time, the metal ions in the solution would compete with the generated reactive oxygen species, consuming some of the active substances, resulting in a slight decrease in the removal rate of toluene. It should be noted that, in the optimized system, the concentrations of PMS (1.0 g/L) and catalyst (0.1 g/L) were in slight excess relative to toluene (30 ppmv). This was intentionally designed to evaluate the intrinsic catalytic activity without being limited by the availability of reactive species. Under these conditions, the toluene conversion remained above 90% within 90 min. However, as the reaction proceeds, the conversion would eventually tend to zero if the PMS concentration were lowered or the reaction time were extended, because the remaining toluene concentration becomes too low to drive further degradation. Importantly, all comparative experiments (e.g., different catalysts or PMS dosages) were conducted under the same excess conditions, so the conclusions drawn from the relative differences are robust. The catalytic performance under optimized, rather than starved, conditions. Therefore, the subsequent experiments selected a M50C50 dosage of 0.1 g/L. Considering that the production of reactive oxygen species (ROSs, such as SO 4•−, •OH, O 2•−, and 1O 2) in the optimal reaction system is closely related to the concentration of PMS [ 35], the influence of different PMS dosages (0.01 g/L, 0.05 g/L, 0.5 g/L, 1.0 g/L, and 5.0 g/L) on the removal efficiency of toluene was studied. As shown in Figure 3d, when the concentration of PMS increased from 0.01 g/L to 1.0 g/L, the removal rate of toluene rose from around 35% to 90.1%, which might be due to the increase in the concentration of PMS, resulting in more ROSs being produced to degrade toluene in the system [ 36]. However, when the concentration of PMS increased from 1.0 g/L to 5.0 g/L, the removal rate of toluene slightly decreased from 90.1% to 89.3%, possibly because excessive SO 4•− underwent self-quenching reactions, causing a downward trend in the removal rate of toluene [ 37]. Therefore, the subsequent experiments selected a PMS concentration of 1.0 g/L. The pH of the solution strongly influences multiple aspects of the catalytic process, including the interaction of the catalyst with organic substrates and the radical-mediated degradation of pollutants [ 38]. To explore the influence of the initial pH value (3, 5, 5.6, 7, 9, 11) on the activation of PMS by M50C50 for the degradation of toluene, 0.1 mM H 2SO 4 and NaOH were used to control the pH value of the aqueous solution. As shown in Figure 3e, the experimental results indicated that the removal of toluene was effective within a wide pH range. Upon increasing the pH, the toluene degradation efficiency showed a mild reduction, falling from 90.1% at pH 5 to 77.3% at pH 11, and the reasons for this phenomenon are as follows: under acidic conditions, more SO 4•− can be produced, while in alkaline conditions, M50C50 may precipitate and PMS may decompose, resulting in a lower utilization rate and a reduction in the removal rate of toluene. The initial solution had a pH of 5.6, which was considered economically feasible and had a good removal effect within a wide pH range. The optimal pH for the subsequent experiments was selected as 5.6. As shown in Figure 3f, it is the continuous 5 cycle operation test of the M50C50 material. The experimental results show that the toluene removal efficiency of the material has remained at around 90% throughout the process, without any significant decline. This provides a reliable basis for the practical application of the material. The stability and reusability of the catalyst were evaluated through a long-term continuous flow experiment, during which toluene (30 ppmv) was used as the target pollutant. To maintain the oxidative capacity of the system, PMS was replenished every 1.5 h to maintain an initial concentration of 1.0 g/L. The degradation performance was continuously tracked for a total duration of 540 min ( Figure S4). The phase stability of the material was investigated by comparing the X-ray diffraction (XRD) patterns of the fresh and used M50C50 samples. Both samples exhibit diffraction peaks with comparable relative intensities. The absence of peak shifts or the emergence of new reflections confirms that no phase transformation or decomposition occurred during the reaction, indicating excellent structural stability ( Figure S5). 2.3. Identification of the Reactive Oxygen Species To determine the main types of free radicals in the oxidation and degradation process of toluene, the effects of different types of free radical scavengers on the degradation of toluene were investigated. According to reported rate constants, tert-butanol (TBA) selectively quenches •OH (3.8–7.6 × 10 8 M −1s −1) but reacts poorly with SO 4•− (4.0–9.1 × 10 5 M −1s −1). Ethanol, in contrast, effectively scavenges both radicals, with rate constants of 1.2–2.8 × 10 9 M −1s −1 for SO 4•− and 1.6–7.8 × 10 7 M −1s −1 for •OH [ 39]. Based on this difference, ethanol was used as a scavenger for SO 4•− and •OH, while TBA was used to scavenge OH. As shown in Figure 4a,b, after adding ethanol to the H 2O/PMS/M50C50 system, the degradation of toluene was inhibited. The degradation efficiency of toluene decreased from 92% to 12%. Compared with the addition of tertbutanol, where the degradation efficiency decreased from 92% to 25%, the addition of ethanol led to a more significant decrease in the degradation efficiency. This result indicates that both SO 4•− and •OH are important reactive oxygen species in the reaction system [ 40]. According to the report, in the advanced oxidation process based on sulfate radicals, superoxide radicals (O 2•−) and singlet oxygen ( 1O 2) also act as important reactive oxygen species, which have an impact on the reaction [ 41]. Therefore, p-benzoquinone and L-histidine were selected as scavengers to capture O 2•−1O 2 respectively [ 42]. The addition of p-benzoquinone and L-histidine decreased the degradation efficiency to 85% and 77%, respectively, indicating that superoxide radicals and singlet oxygen were generated but played a minor role compared to sulfate and hydroxyl radicals. After adding p-benzoquinone and L-histidine to the reaction system, the degradation efficiency of toluene only decreased slightly, indicating that superoxide radicals and singlet oxygen were generated in the reaction, but their role in toluene degradation was far less significant than that of sulfate radicals and hydroxyl radicals. As shown in Figure 4c, the carbon dioxide production in the H 2O/PMS/M50C50 system after adding ethanol and tertbutanol was analyzed. After adding ethanol, the carbon dioxide production decreased to some extent, from 160 ppmv to 60 ppmv. While after adding tertbutanol, the carbon dioxide production decreased more markedly, from 160 ppmv to 110 ppmv. This indicates that the degradation efficiency of toluene was inhibited and the mineralization ability was weakened, which is consistent with the results of the above scavenging experiments. Further, the reactive oxygen species generated in the H 2O/PMS/M50C50 system were qualitatively detected by EPR technology. As shown in Figure 4d, characteristic signal peaks of DMPO-SO 4•− and DMPO-•OH appeared in the H 2O/PMS/M50C50 system [ 43]. It indicates that SO 4•− and •OH exist in the system. It further proves that the formed SO 4•− and •OH play an important role in the degradation of toluene in the system. 2.4. Generation Pathway of Reactive Oxygen Species in the H 2O/PMS/M50C50 System For the fresh M50C50, the high-resolution XPS spectra of Co 2p, Fe 2p and S 2p were subjected to peak deconvolution to determine the valence states of the constituent elements. The corresponding peak fitting results are presented in the main text, while the raw survey and unprocessed high-resolution spectra are provided ( Figure S6). To gain further insight into the surface chemical composition and elemental valence states of the M50C50 before and after the reaction, high-resolution XPS spectra were deconvoluted. As shown in Figure 5a, both the materials before and after the reaction can display obvious characteristic peaks of Co 2p, Fe 2p, O 1s, N 1s, C 1s, and S 2p. Moreover, the intensity of the peaks after the reaction has weakened, and some characteristic peaks have shifted. As shown in Figure 5b, the analysis of the fine spectra of Fe 2p before and after the reaction reveals that the M50C50 used before has two characteristic peaks, and its high-resolution spectrum can be fitted into 4 peaks [ 44]. After the use, the M50C50 used all have two characteristic peaks, and its high-resolution spectrum can be fitted into 6 peaks. When the M50C50 catalyst undergoes a chemical reaction, the relative content of Fe 3+ decreases from 100% to 46.15%, indicating that Fe 3+ participates in the reaction in the H 2O/PMS/M50C50 system and is reduced to Fe 2+ by PMS. As shown in Figure 5c, the analysis of the fine spectra of Co 2p before and after the reaction reveals that the M50C50 used before has two characteristic peaks, and its high-resolution spectrum can be fitted into6 peaks [ 45]. After the use, the M50C50 used has two characteristic peaks, and its high-resolution spectrum can be fitted into 6 peaks. After the M50C50 catalyst was used, the relative content of Co 2+ decreased from 89.13% to 68.91%, while the relative content of Co 3+ increased from 10.87% to 31.09%. This indicates that Co 3+ participated in the reaction in the H 2O/PMS/M50C50 system and was reduced by PMS to Co 2+. At the same time, the Co 2+ in the reaction was oxidized to Co 3+. This redox cycle of Co 2+/ Co 3+ is facilitated by the interfacial electron transfer from MIL-100(Fe) to CoS. As shown in Figure 5a, the analysis of the fine spectra of S 2p before and after the reaction revealed that the M50C50 before use could be fitted with 5 main characteristic peaks [ 46]. After use, the M50C50 could be fitted with 5 main characteristic peaks. The relative content of S 2− decreased from 15.33% to 12.65%, the relative content of S n2− increased from 10.44% to 14.60%, and the relative content of SO 42− increased from 40.69% to 45.83%. The peak of SO 32− disappeared in the used catalyst, and the characteristic peak of S 0 appeared and the content of S 0 increased. This indicates that some S 2− was oxidized to elemental sulfur, which might be the reason for the oxidation of S 2− by Fe 3+/Co 3+. The content of SO 42− increased, and another part of S 2− participated in the redox cycle regeneration of Fe 3+/Fe 2+ and Co 3+/Co 2+ during the reaction, and also consumed electrons through oxidation to form high-valent sulfur species. To investigate the contribution of dissolved oxygen to the pollutant degradation in the H 2O/PMS/M50C50 system, control experiments were conducted under N 2-bubbling conditions to exclude the influence of oxygen. The degradation efficiency was monitored over 90 min under both air-bubbling and N 2-bubbling conditions. The experimental results showed that under air-bubbling conditions, the removal efficiency remained consistently around 90%, exhibiting stable and efficient degradation performance. Under N 2-bubbling conditions, the removal efficiency also maintained between 89% and 94%, which was nearly identical to that observed under air-bubbling conditions. Throughout the reaction, the two efficiency curves almost overlapped, showing no significant difference. These results indicate that molecular oxygen does not play a crucial role in this system ( Figure S7). The efficient degradation of pollutants is primarily attributed to the direct activation of PMS. The redox cycles of Co 2+/Co 3+, Fe 2+/Fe 3+, and the stepwise oxidation of sulfur species (S 2−→S 0→SO 32−→SO 42−) are schematically illustrated in ( Figure S9). Briefly, Co 2+ activates PMS to generate SO 4•− and is oxidized to Co 3+; Co 3+ is then reduced back to Co 2+ by electrons supplied from S 2− (or Fe 2+). Simultaneously, Fe 3+ is reduced to Fe 2+ by electrons from S 2−, and Fe 2+ further activates PMS to produce SO 4•−. The oxidation of S 2− releases electrons that sustain these metal redox cycles. Based on the above data results, the possible mechanism of M50C50 activating PMS for the degradation of toluene was proposed in Figure 6. Firstly, the activation of PMS usually occurred on the surface of M50C50, where the Co 2+ reacts with PMS to generate Co 3+, SO 4•− and •OH. Subsequently, SO 4•− reacted with H 2O to generate •OH, and SO 5•− could also continue to participate in the reaction to generate SO 4•− and •OH. At the same time, Fe 3+ cannot directly activate PMS by donating electrons. Instead, it is reduced to Fe 2+ by electrons donated from two sources: the oxidation of Co 2+ to Co 3+ by PMS, and the oxidation of S 2− to higher-valent sulfur species (S 0, SO 42−). The generated Co 2+ and Fe 2+ further activate PMS, sustaining the production of reactive radicals. Particularly, S 2− promotes the continuous regeneration of Co 2+ and Fe 2+, allowing the system to continuously decompose PMS [ 47]. Moreover, the high specific surface area of the M50C50 provides abundant anchoring sites for CoS and MIL-100(Fe), reducing particle agglomeration and increasing the surface active sites. The coordinatively unsaturated S atoms on the CoS surface can combine with protons to form H 2S, which is then stripped from the surface, exposing more Co 2+ sites to participate in the catalytic Fenton process [ 48]. Therefore, the efficient degradation of toluene in the H 2O/PMS/M50C50 system can be attributed to the synergistic effect of two factors: first, the heterojunction interface between MIL-100(Fe) and CoS, which facilitates interfacial electron transfer and lowers the energy barrier for PMS activation; second, the Fe 2+/Fe 3+ and Co 3+/Co 2+ redox cycles, which sustain the continuous generation of reactive radicals. 3.2. Catalyst Preparation The solvothermal synthesis of octahedral MIL-100(Fe) was carried out following the method previously reported [ 49] in Figure 7. For the synthesis of octahedral MIL-100(Fe), iron powder (Fe 0, 0.11 g) was first mixed with concentrated HNO 3 (0.15 mL) and deionized water (10.0 mL). The mixture was ultrasonicated for 15 min to ensure complete dissolution. Subsequently, 1,3,5-benzenetricarboxylic acid (H 3BTC, 0.28 g) and hydrofluoric acid (HF, 0.18 mL) were introduced into the solution. The resulting mixture was transferred into a Teflon-lined stainless-steel autoclave and maintained at 150 °C for 24 h. After the reaction, the orange-yellow precipitate was collected by centrifugation, washed alternately with ethanol and deionized water three times, and finally dried under vacuum at 60 °C for 10 h. CoS was prepared by dissolving Co(NO 3) 2·6H 2O (0.1 g) and thioacetamide (TAA, 0.4 g) in ethanol (20.0 mL). The resulting solution was transferred into an autoclave and kept at 120 °C for 8 h. After the reaction, black spherical CoS particles were isolated by centrifugation, rinsed with deionized water, and dried in a vacuum oven overnight at 60 °C. To obtain the MIL-100(Fe)/CoS composites, the as-prepared MIL-100(Fe) and CoS were combined via ball milling at 30 Hz for 20 min. Composites with various mass ratios were prepared and labeled as MxCy, where M and C represent MIL-100(Fe) and CoS, respectively, and x and y indicate their corresponding mass proportions. 3.3. VOC Catalytic Oxidation Test The degradation of toluene gas was carried out in an AOP wet scrubber, as illustrated in Figure 8. A toluene-laden gas stream was produced by passing air through liquid toluene, generating a vapor that was subsequently dispersed into the aqueous phase as fine microbubbles. The gas flow rate and toluene concentration were regulated using mass flow controllers. Prior to entering the reactor, the concentrated toluene vapor was combined with an air stream in a mixing chamber to achieve the desired dilution. Once inside the reactor, the catalyst and PMS were added, initiating a gas–liquid–solid reaction. To monitor the process, a gas chromatograph (GC) fitted with two flame ionization detectors (FIDs) continuously measured the toluene and CO 2 concentrations at both the inlet and outlet ports. In the optimized system, the concentrations of PMS (1.0 g/L) and catalyst (0.1 g/L) were chosen to be in slight excess relative to toluene (30 ppmv) to ensure that the degradation was not limited by the availability of reactive species, allowing a reliable evaluation of the catalytic activity under mass-transfer-sufficient conditions. The reaction was run for 90 min, which was found to be sufficient for the reagents to decrease their concentration significantly based on preliminary tests. 4. Conclusions Toluene is an air pollutant that can cause damage to the environment and human health. It has attracted much attention due to its high toxicity, stubbornness and widespread emissions. To address this challenge, we have designed a highly active M50C50 catalyst featuring a heterojunction interface between MIL-100(Fe) and CoS. This heterojunction facilitates interfacial electron transfer from MIL-100(Fe) to CoS, lowering the energy barrier for PMS activation and enabling deep oxidation of toluene in an AOP wet scrubber. Experimental results show that M50C50 efficiently and stably degrades toluene over a wide pH range. SO 4•− and •OH radicals are generated as the primary reactive oxygen species, responsible for the mineralization of toluene. The redox cycle regeneration of Fe 3+/Fe 2+ and Co 3+/Co 2+ further sustain the continuous generation of reactive radicals. This work provides insight into the deep oxidation and mechanism of VOCs and lays a foundation for the application of AOP wet scrubbers in industrial waste gas treatment. Supplementary Materials The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/catal16060534/s1, Figure S1: The N 2 adsorption–desorption isotherm; Figure S2: The electrochemical impedance spectroscopy; Figure S3: Effect of M50C50 preparation methods on toluene degradation in the AOP wet scrubber; Figure S4: Continuous operation on toluene removal; Figure S5: XRD spectra of the materials before and after the reaction; Figure S6: XPS spectra of M50C50 before and used after reaction; Figure S7: Removal efficiency of toluene in the H2O/PMS/M50C50 system under different bubbling atmospheres; Figure S8: Correlation between specific surface area and toluene degradation efficiency of MIL-100(Fe), CoS, and M50C50; Figure S9: Schematic illustration of the Co 2+/Co 3+ and Fe 2+/Fe 3+ redox cycles, along with the stepwise oxidation of sulfur species (S 2−→ S 0→ SO 32−→ SO 42−); Table S1: BET surface area, pore volume and pore diameter of MIL-100(Fe), CoS and M50C50. Figure 1. ( a) The PXRD pictures and ( b) FT-IR spectra of the MIL-100(Fe), CoS and MxCy composites. ( c) The XPS and the high-resolution XPS spectra of ( d) Co 2p, ( e) Fe 2p and ( f) S 2p for MIL-100(Fe), CoS and M50C50. Figure 1. ( a) The PXRD pictures and ( b) FT-IR spectra of the MIL-100(Fe), CoS and MxCy composites. ( c) The XPS and the high-resolution XPS spectra of ( d) Co 2p, ( e) Fe 2p and ( f) S 2p for MIL-100(Fe), CoS and M50C50. Figure 2. The SEM pictures of ( a) MIL-100(Fe), ( b) CoS and ( c, e) M50C50. The TEM image of ( d) M50C50. ( f– i) The EDS elemental mappings of the M50C50. Figure 2. The SEM pictures of ( a) MIL-100(Fe), ( b) CoS and ( c, e) M50C50. The TEM image of ( d) M50C50. ( f– i) The EDS elemental mappings of the M50C50. Figure 3. The removal efficiency of toluene ( a) and corresponding outlet CO 2 concentration ( b) in various catalytic systems. The influences of ( c) catalyst dosage, ( d) PMS concentration and ( e) initial pH on toluene degradation over M50C50. ( f) The cycle experiments of toluene degradation over M50C50. (Experimental conditions: [Toluene] = 30 ppmv, [PMS] = 1.0 g/L, [Catalyst] = 0.1 g/L, [pH] = 5.6, T = 25 °C). Figure 3. The removal efficiency of toluene ( a) and corresponding outlet CO 2 concentration ( b) in various catalytic systems. The influences of ( c) catalyst dosage, ( d) PMS concentration and ( e) initial pH on toluene degradation over M50C50. ( f) The cycle experiments of toluene degradation over M50C50. (Experimental conditions: [Toluene] = 30 ppmv, [PMS] = 1.0 g/L, [Catalyst] = 0.1 g/L, [pH] = 5.6, T = 25 °C). Figure 4. The effect of various radical scavengers on the toluene removal efficiency ( a), the final toluene removal efficiency ( b) and outlet CO 2 concentration ( c) in the H 2O/PMS/M50C50 system. EPR spectra of •OH and SO 4•− in various catalytic systems ( d). (Experimental conditions: [Toluene] = 30 ppmv, [Catalyst] = 0.1 g/L, [PMS] = 1.0 g/L, [TBA] = 100 mM, [EtOH] = 100 mM, [L-histidine] = 1.0 mM, [p-BQ] = 1.0 mM, T = 25 °C.) Figure 4. The effect of various radical scavengers on the toluene removal efficiency ( a), the final toluene removal efficiency ( b) and outlet CO 2 concentration ( c) in the H 2O/PMS/M50C50 system. EPR spectra of •OH and SO 4•− in various catalytic systems ( d). (Experimental conditions: [Toluene] = 30 ppmv, [Catalyst] = 0.1 g/L, [PMS] = 1.0 g/L, [TBA] = 100 mM, [EtOH] = 100 mM, [L-histidine] = 1.0 mM, [p-BQ] = 1.0 mM, T = 25 °C.) Figure 5. ( a) Survey and ( b) Co 2p, ( c) Fe 2p and ( d) S 2p XPS spectra of fresh and used M50C50 in catalytic degradation of toluene. (Experimental conditions: [Toluene] = 30 ppmv, [PMS] = 1.0 g/L, [Catalyst] = 0.1 g/L, [pH] = 5.6, [T] = 25 °C). Figure 5. ( a) Survey and ( b) Co 2p, ( c) Fe 2p and ( d) S 2p XPS spectra of fresh and used M50C50 in catalytic degradation of toluene. (Experimental conditions: [Toluene] = 30 ppmv, [PMS] = 1.0 g/L, [Catalyst] = 0.1 g/L, [pH] = 5.6, [T] = 25 °C). Figure 6. The possible mechanism of toluene degradation in the H 2O/PMS/M50C50 system. Figure 6. The possible mechanism of toluene degradation in the H 2O/PMS/M50C50 system. Figure 7. Image of synthetic pathways of the catalyst. Figure 7. Image of synthetic pathways of the catalyst. Figure 8. Schematic diagram of the experimental system. Figure 8. Schematic diagram of the experimental system. Zhang, Z.; Ruan, Y. Deep Oxidation of Atmospheric VOCs by MOFs/Metal Sulfide Composites via Fenton-like Reaction: Performance and Mechanism. Catalysts 2026, 16, 534. https://doi.org/10.3390/catal16060534 Zhang Z, Ruan Y. Deep Oxidation of Atmospheric VOCs by MOFs/Metal Sulfide Composites via Fenton-like Reaction: Performance and Mechanism. Catalysts. 2026; 16(6):534. https://doi.org/10.3390/catal16060534 Zhang, Zishi, and Yang Ruan. 2026. "Deep Oxidation of Atmospheric VOCs by MOFs/Metal Sulfide Composites via Fenton-like Reaction: Performance and Mechanism" Catalysts 16, no. 6: 534. https://doi.org/10.3390/catal16060534 Zhang, Z., & Ruan, Y. (2026). Deep Oxidation of Atmospheric VOCs by MOFs/Metal Sulfide Composites via Fenton-like Reaction: Performance and Mechanism. Catalysts, 16(6), 534. https://doi.org/10.3390/catal16060534
