Cu–Co–O-Codoped Graphite Carbon Nitride as an Efficient Peroxymonosulfate Activator for Sulfamethoxazole Degradation: Characterization, Performance, and Mechanism

Abstract

This study presents the development of a novel Cu–Co–O-codoped graphitic carbon nitride (g-C₃N₄) catalyst for efficient peroxymonosulfate (PMS) activation to degrade sulfamethoxazole (SMX) in aqueous environments. The synthesized Cu–Co–O-g-C₃N₄ catalyst demonstrated exceptional catalytic performance, achieving 90% SMX removal within 10 min—significantly outperforming pristine g-C₃N₄ (14%) and O-doped g-C₃N₄ (22%)—with a reaction rate constant of 0.63 min⁻¹. The superior activity was attributed to the synergistic effects of Cu-Co bimetallic doping and oxygen incorporation, which enhanced the active sites, stabilized metal ions, and minimized leaching. Mechanistic studies revealed a dual-pathway degradation process: (1) a radical pathway dominated by sulfate radicals (SO₄•⁻) and (2) a non-radical pathway driven by singlet oxygen (¹O₂), with the latter identified as the dominant species through quenching experiments. The catalyst exhibited broad pH adaptability and optimal performance at neutral to alkaline conditions. Characterization techniques (XRD, FTIR, XPS) confirmed successful doping and revealed that oxygen incorporation modified the electronic structure of g-C₃N₄, improving charge carrier separation. This work provides a sustainable strategy for antibiotic removal, addressing key challenges in advanced oxidation processes (AOPs), and highlights the potential of multi-heteroatom-doped carbon nitride catalysts for water purification.

1. Introduction

As a typical antibiotic, sulfamethoxazole (SMX), which is widely used to treat infectious diseases in humans and animals, is frequently detected in aquatic environments. The presence of SMX in a water environment (even at a low concentration of 0.5 μg/L) can cause severe effects on the human body and ecosystems. In addition, it is difficult to degrade SMX using conventional technologies such as biodegradation, adsorption, and coagulation. Recently, the efficient removal of SMX was obtained using advanced oxidation technology. Yu et al. [1] demonstrated that 30 μM SMX could be completely degraded by activated peroxymonosulfate at pH 3–5.

Advanced oxidation technologies (AOTs), which are based on the highly oxidizing sulfate radical (SO₄•^–) and hydroxyl radical (•OH), have a high ability to degrade refractory organic contaminants. Compared with •OH, SO₄•^– has a higher redox potential, a longer half-life, and higher selectivity, attracting more attention to sulfate radical-based advanced oxidation technology in water treatment [2]. Generally, transition metals, ultraviolet light, and carbonaceous materials are employed to activate peroxymonosulfate (PMS) to generate SO₄•^– radicals [3]. Among them, as effective, controllable, and recycling activators, transition metals (such as Co, Cu, and Fe [4]) are widely used to activate PMS. However, metal ions are leached during the PMS activation process, which impedes their practical application in water plants [5]. Thus, it is important to seek efficient and eco-friendly PMS activators to improve the removal of refractory organics.

Graphitic carbon nitride (g-C₃N₄) is a nonmetallic polymeric semiconductor with good chemical stability, environmental friendliness, and a low cost, which has been extensively used as a catalyst in the field of photocatalysis. Recently, some studies indicated that PMS could also be activated by g-C₃N₄ in the absence of light. Thus, the potential and performance of g-C₃N₄ as an activator in the field of PMS activation have been widely investigated. However, research results showed that the PMS activation performance of g-C₃N₄ was inefficient and time-consuming. To improve the PMS activation efficiency, g-C₃N₄ is usually modified by doping metal oxides or transition metals. The abundant electron cavities in g-C₃N₄ provide ideal points to bind with transition metals, which can stabilize metal ions, prevent their leaching, and improve the activation efficiency of g-C₃N₄.

Generally, a single metal or a single nonmetal is doped into g-C₃N₄ to modify the catalytic performance of g-C₃N₄ [6]. Co²⁺ is the best PMS activator among transition metals. Obvious improvements in the PMS activation efficiency after doping Co²⁺ into g-C₃N₄ have been reported, determining that the monochlorophenols’ degradation performance was improved in a Co-doped g-C₃N₄/PMS system compared to a g-C₃N₄/PMS system [7]. In addition, Cu²⁺ is a common catalyst for SO₄•^– production. The PMS activation performance was also improved by doping Cu²⁺ into g-C₃N₄. Compared with single-metal doping, more active sites could be obtained by bimetal doping. The synergistic function between doped bimetals could further improve the PMS activation performance. At the same time, g-C₃N₄ as a carrier can reduce the leaching of Co²⁺ and Cu²⁺, promote the conversion of Cu (II) to Cu (I) and Co (III) to Co (II), and improve the catalytic activity of Co-Cu-doped g-C₃N₄. In addition, O-g-C₃N₄, with O atom doping, had a superior activated performance and a wider pH application range than g-C₃N₄. Thus, it was speculated that bimetal (Co and Cu) and nonmetal (O) codoped g-C₃N₄ could have better catalytic performance and a broader pH application range.

Here, Cu–Co–O-g-C₃N₄ was innovatively prepared by doping bimetal (Co, Cu) and O into g-C₃N₄ for PMS activation for the removal of SMX.

2. Materials and Methods

2.1. Materials and Reagents

Anhydrous sodium sulfite (97% purity) was purchased from Tianjin Benchmark Chemical Reagents Co., Ltd. (Tianjin, China). Copper chloride hexahydrate (analytical grade), anhydrous oxalic acid (99% purity), methanoic acid (chromatographic purity), and acetonitrile (chromatographic purity) were provided by Aladdin Reagent Co., Ltd. (Shanghai, China). Peroxymonosulfate (PMS) and sulfamethoxazole (SMX, 98% purity) were obtained from Sigma Reagent Co, Ltd. (Shanghai, China). Cobalt chloride hexahydrate (99% purity) was supplied by Tianjin Yongsheng Fine Chemicals Co., Ltd. (Tianjin, China). Urea (99% purity) was purchased from Tianjin Tianli Chemical Reagent Co., Ltd. (Tianjin, China). Sodium hydroxide (NaOH, 96% purity) was obtained from Tianjin Continental Chemical Reagent Factory (Tianjin, China).

2.2. Synthesis of CN, O-CN and Cu–Co–O-g-C₃N₄

CN, O-CN and Cu–Co–O-g-C₃N₄ were prepared via one-step pyrolysis. Urea (20 g), oxalic acid anhydrous (8 g), and different amounts of copper chloride hexahydrate and cobalt chloride hexahydrate were mixed and ground in a mortar and pestle with the same molar concentrations of copper and cobalt of 0.16, 0.24, 0.32, 0.40, 0.48, 0.56, and 0.64 mM, respectively, and designated as 1-MOCN to 7-MOCN. Then, the mixtures were heated to 500 °C for 2 h, at a rate of 5 °C min⁻¹ under an air atmosphere. To avoid any possible impurity, the completely ground samples were washed three times using deionized water. Then, the powder was dried in an 80 °C oven for 12 h and collected for subsequent application.

2.3. Characterization of Catalysts

The morphological features of the catalysts were examined using scanning electron microscopy (SEM, ZEISS SIGMA-500, Carl Zeiss, Oberkochen, Germany) and transmission electron microscopy (TEM, JEOL JEM-2100, JEOL, Tokyo, Japan). Fourier-transform infrared (FTIR) spectra were acquired using a Shimadzu FTIR-8400S spectrometer (Shimadzu, Kyoto, Japan), while crystallographic analysis was carried out via X-ray diffraction (XRD) using a Bruker D8 Advance diffractometer (Bruker, Berlin, Germany). Surface properties, including the specific surface area and pore size distribution, were determined from nitrogen adsorption–desorption isotherms (Quantachrome SI), with Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods applied for data analysis. Chemical states were probed via X-ray photoelectron spectroscopy (XPS, Thermo Fisher ESCALAB 250xi, Thermo Fisher Scientific, Waltham, MA, USA). PMS concentration was analyzed by a UV–Vis spectrophotometer (Jasco V-530, JASCO, Tokyo, Japan) equipped with quartz cuvettes of 1 cm light path, the test wavelength was 352 nm. SMX was analyzed by high performance liquid chromatography (HPLC, Agilent, Agilent Technologies, Santa Clara, CA, USA).

2.4. SMX Degradation Experiments

The degradation experiments were carried out in a 250 mL glass beaker with 100 mL of SMX solution (10 mg/L). A certain amount of catalyst was added into the SMX solution, and the pH of the SMX solution was adjusted to the desired values using NaOH (0.5 M). Then, the degradation reaction was triggered by adding 0.3 mL of PMS solution (50 mg/L). At regular intervals, the samples were extracted from the mixed solution and immediately filtered with a 0.22 μm filter. After filtering, saturated sodium thiosulfate was rapidly added to quench the reaction. Then, the SMX concentration and residual PMS concentration of the extracted samples were monitored, respectively. Each experiment was conducted in triplicate.

3. Results and Discussion

3.1. Characterization Results of Catalysts

3.1.1. Morphological and Structural Analysis

SEM and TEM were employed to investigate the microstructural properties of CN and Cu–Co–O-g-C₃N₄. Figure 1a–c show that both catalysts exhibited a similar platelet-like layered morphology, suggesting that oxygen incorporation had minimal impact on their overall structure. TEM observations (Figure 1d) further supported this finding, revealing well-ordered stacked layers at the catalyst edges.

The crystallinity of Cu–Co–O-g-C₃N₄ was assessed using selected area electron diffraction (Figure 1e), which confirmed its partially crystalline nature. High-resolution TEM (Figure 1f) revealed lattice fringes with a measured spacing of 0.337 nm, consistent with the (002) crystallographic plane. Additionally, HAADF-STEM combined with EDX elemental mapping (Figure 1g–l) demonstrated the uniform distribution of C, N, and O across the catalyst matrix. The EDX spectra further confirmed the successful integration of oxygen into the carbon nitride framework.

The crystalline phase and structural integrity of the catalysts were examined using X-ray diffraction (Figure 2a). All samples exhibited characteristic g-C₃N₄ diffraction patterns, with the (002) peak at 27.5° (d-spacing = 0.337 nm) corresponding to interlayer stacking and the (100) peak at 12.8° (d-spacing = 0.675 nm) reflecting in-plane periodicity. These findings confirm that the fundamental g-C₃N₄ framework remained intact following metal doping [8].

Notably, the progressive attenuation of the peak intensities with the increasing dopant concentration suggests that metal incorporation significantly reduced the material’s crystallinity. This inverse relationship between the doping level and the diffraction peak intensity indicates that the introduced metal species disrupt the long-range order of the carbon nitride matrix while preserving its basic structural motifs.

3.1.2. Surface Functional Groups and Pore Size Distribution

FTIR spectroscopy was employed to characterize the surface functional groups of the synthesized samples (Figure 2b) [9]. The distinct peak at 809 cm⁻¹ was attributed to the breathing vibration of triazine rings, confirming the integrity of the carbon nitride framework. Multiple absorption bands in the 1200–1700 cm⁻¹ range corresponded to aromatic C-N heterocycle stretching vibrations, with characteristic peaks at 1638, 1575, 1464, and 1406 cm⁻¹ originating from the structural vibrations of heptazine units [10]. The peaks observed at 1320 and 1238 cm⁻¹ were assigned to N-(C)₃ group vibrations, while the broad absorption band spanning 3000–3500 cm⁻¹ indicated the presence of uncondensed terminal amino groups. All the catalysts exhibited highly similar FTIR spectra, demonstrating the preservation of the fundamental carbon nitride structure during synthesis. The absence of detectable oxygen-related peaks was likely due to the doping concentration being below the instrument’s detection limit [11].

Nitrogen adsorption–desorption measurements (Figure 2c) revealed that all the samples displayed typical type IV isotherms with H3-type hysteresis loops, characteristic of materials containing slit-shaped mesopores [12]. The pore size distribution calculated using the BJH method (Figure 2d) indicated the coexistence of mesopores and macropores in the samples. With increasing oxygen content, Cu–Co–O-g-C₃N₄ showed a decreasing trend in specific surface area and pore volume, which could be attributed to the pore structure contraction induced by phosphate incorporation [13]. Notably, the pore size distributions exhibited minimal variations among different samples, suggesting that the doping treatment did not significantly alter the overall pore structure characteristics of the materials [14].

3.2. Degradation of SMX

3.2.1. Catalytic Activity of Cu–Co–O-g-C₃N₄

The influence of the doped bimetal mass ratio of Cu–Co–O-g-C₃N₄ on the SXM degradation process was investigated and is shown in Figure 3. With the increase in the mass of the doped bimetal, an evident increase in the SMX removal efficiency was obtained (Figure 3a) [15]. This is attributed to the large number of activated sites reacting with PMS, which were introduced into the g-C₃N₄ by doping the Cu-Co bimetal. In addition, the rate constant (k) markedly rose with the increase in the mass of the doped bimetal. When the doped bimetal mass ratio rose to 0.229 mg, the rate constant of Cu–Co–O-g-C₃N₄/PMS systems reached 0.669 min⁻¹. Notably, from 4-MOCN to 6-MOCN, the degradation curves almost overlapped, suggesting that the improvement in the reaction rate became slight [16]. Thus, taking economics into account, G5 was selected as the optimal activator for the subsequent experiments [17]. We examined the concentration of metal ions in the solution after the reaction and found that the level of dissolved metal ions was very low (0.01 mg/L), making it environmentally friendly.

The SMX degradation via activated PMS with different activators is shown in Figure 4a. The removal efficiency was 10%, 14%, 22%, and 90%, respectively, by PMS alone, g-C₃N₄/PMS, O-g-C₃N₄/PMS, and the 5-MOCN/PMS system. SMX was removed by the 5-MOCN/PMS system within 10 min compared with the PMS alone [18], g-C3N4/PMS, and O-g-C₃N₄/PMS systems [16]. In addition, as shown in Figure 3b, the correlation coefficients were all higher than 0.98, indicating that the SMX degradation process followed pseudo-first-order kinetics. The reaction rate of the 5-MOCN /PMS systems (0.63 min⁻¹) increased compared with O-g-C₃N₄ (0.034 min⁻¹) and g-C3N4 (0.014 min⁻¹), illustrating that the catalytic activity of g-C₃N₄ was promoted by the codoped Cu–Co bimetal and oxygen, which further suggested that the 5-MOCN had superior activation activity in SMX elimination [19]. We also did surface adsorption experiments and found almost no adsorption effect, which was corroborated by the smaller specific surface area in the BET test.

3.2.2. Reactive Oxygen Species in the 5-MOCN/PMS System

To identify the key reactive species involved in the degradation process, quenching experiments were conducted. Previous studies indicate that tert-butanol (TBA) [20] efficiently scavenges hydroxyl radicals (HO•), with a rate constant ranging from 3.8 to 7.6 × 10⁸ M⁻¹ s⁻¹, while methanol (MeOH) is known to trap both HO• and sulfate radicals (SO₄•⁻), with rate constants of 9.7 × 10⁸ M⁻¹ s⁻¹ and 2.5 × 10⁷ M⁻¹ s⁻¹, respectively.

In contrast, para-benzoquinone (p-BQ) [20] is capable of quenching HO•, SO₄•⁻, and singlet oxygen (¹O₂), exhibiting rate constants of 1.2 × 10¹⁰ M⁻¹ s⁻¹, 2.51 × 10⁹ M⁻¹ s⁻¹, and 2 × 10⁹ M⁻¹ s⁻¹, respectively [21]. L-histidine is used to quench ·O₂⁻ [22]. However, concerns have been raised regarding p-BQ’s reliability as an ¹O₂ quencher due to its potential direct reaction with PMS [23]. To assess this possibility under experimental conditions, PMS decomposition was monitored in various systems.

As shown in Figure 5, we found •O₂⁻, ¹O₂ to be the main active species in the reaction. Of note, a small portion of HO• SO₄•⁻ could be generated from the reaction.

3.2.3. Influences of Various Reaction Conditions

The PMS concentration has an important influence on the SMX degradation process. An inappropriate PMS concentration could result in an increase in the application cost and excessive salt concentration being introduced into the water environment. Thus, the influence of PMS concentration on SMX degradation was investigated and is depicted in Figure 6. The SMX degradation efficiency increased from 0.216 to 0.48, and the reaction rate constant increased as the PMS concentration changed from 0.4 mM to 1.1 mM [24]. However, a slight decrease in the reaction rate constant was obtained when the PMS concentration further increased to 1.2 mM (Figure 6b). This phenomenon might be caused by the reaction between extra PMS and SO₄⁻, which could generate low reactive SO₅^-, eventually leading to the reduction in SMX degradation efficiency. Hence, the optimal PMS concentration was selected as 1.1 mM. Figure 6c shows the effect of different catalyst concentrations on SMX degradation. Figure 6e shows the effect of different pHs on SMX degradation. We have found that the best performance is achieved under neutral to alkaline conditions, it may be that the alkaline conditions have some activating effect on persulfate, but too much alkali leads to short-term decomposition of large amounts of persulfate, generating free radicals too late to react with SMX. Figure 6g shows the effect of different humic acids on SMX degradation.

The degradation efficiency of the catalyst reuse is provided in Figure S1 in the Supporting Information.

3.3. Reaction Mechanism

XPS characterization was conducted to investigate the valence state evolution of Co and Cu species during PMS activation over fresh and spent Cu–Co–O-g-C3N4 catalysts. Post-reaction analysis indicated relative stability in the bulk elemental composition [25], although noticeable depletion of the Co and Cu content (particularly Co) was observed, implying possible metal leaching during the catalytic process. As evidenced by high-resolution XPS spectra (Figure 7b), substantial changes in oxidation states occurred: the relative abundance of Cu²⁺ species rose from 56.8% to 63.3%, accompanied by a proportional decrease in the Cu⁺ content (16.2% to 7.1%) [26], clearly demonstrating the active participation of Cu⁺/Cu²⁺ redox pairs in PMS activation. Parallel transformations were detected in the Co²⁺/Co³⁺ system, further confirming the dual-metal catalytic mechanism [27].

The catalytic process initiates through direct PMS activation by Co(II) and Cu(I) centers, yielding sulfate radicals (SO₄⁻) while oxidizing the metal sites to higher valence states [24]. These activated metal species subsequently participate in secondary reactions with PMS to generate peroxymonosulfate radicals (SO₅⁻). The synergistic interplay between Cu and Co redox cycles significantly enhances the overall catalytic efficiency. Meanwhile, a fraction of SO₄⁻ undergoes conversion to SO₅⁻ through interaction with excess PMS. The resultant SO₅⁻ species reacts with water molecules to produce singlet oxygen (¹O₂) [9]. These reactive oxygen species collectively mediate the oxidative degradation of SMX, ultimately achieving complete mineralization to carbon dioxide and water [28].

Co(II) + HSO₅⁻ → Co(III) + SO₄•⁻ + OH−.

(1)

Cu(I) + HSO₅⁻ → Cu(II) + SO₄•⁻ + OH−.

(2)

Co(III) + HSO₅⁻ → Co(II) + SO₅•⁻ + H+.

(3)

Cu(II) + HSO₅⁻ → Cu(I) + SO₅^•− + H+.

(4)

Co(III) + Cu(I) → Co(II) + Cu(II).

(5)

SO₄•⁻ + HSO₅⁻ →SO₅•⁻ + HSO₄⁻.

(6)

2SO₅•⁻ + H₂O → 1.5¹O₂ + 2HSO₄⁻.

(7)

SO₄•⁻/¹O₂ + SMX → degradation products + CO₂ + H₂O.

(8)

The mass spectra of the catalyzed oxidative decomposition of SMX are provided in the Supporting Information. LC-MS was used to identify the reaction intermediates present in the Cu–Co–O-g-C3N4/PMS/SMX system (Figure S2). As illustrated in Figure 2, five possible pathways in the degradation of SMX are presented. In general, the degradation pathways can be classified into cleavage between nitrogen and sulfur bonds, hydroxylation of SMX, and oxidation of the amine group.

4. Conclusions

This study successfully developed a Cu–Co–O-codoped g-C₃N₄ (Cu–Co–O-g-C₃N₄) catalyst through a simple one-step pyrolysis method, demonstrating exceptional efficiency in activating peroxymonosulfate (PMS) for the degradation of sulfamethoxazole (SMX). The key findings are summarized as follows:

• Superior Catalytic Performance: The Cu–Co–O-g-C₃N₄/PMS system achieved 90% SMX removal within 10 min, outperforming pristine g-C₃N₄ (14%) and O-doped g-C₃N₄ (22%), with a reaction rate constant (0.63 min⁻¹) 45-fold higher than that of g-C₃N₄ alone. The synergistic effect of Cu-Co bimetallic doping and oxygen incorporation enhanced the active sites, stabilized metal ions, and minimized leaching, ensuring sustained catalytic activity. The Cu⁺/Cu²⁺ and Co²⁺/Co³⁺ redox cycles facilitated continuous PMS activation while promoting ROS generation.
• Optimized Reaction Conditions: The system exhibited broad pH adaptability, with optimal performance at neutral to alkaline conditions. Excessive PMS (>1.1 mM) led to self-scavenging of SO₄•⁻, highlighting the importance of balanced oxidant dosing.
• Mechanistic Insights: Characterization (XRD, FTIR, XPS) confirmed successful doping and revealed that O incorporation modified the electronic structure of g-C₃N₄, reducing its bandgap and enhancing charge carrier separation. The dual radical/non-radical mechanism was elucidated through quenching experiments and PMS decomposition studies, with ¹O₂ identified as the dominant species.
• Environmental Implications: The Cu–Co–O-g-C₃N₄ catalyst offers a sustainable solution for antibiotic removal, addressing challenges such as metal leaching and pH sensitivity in conventional advanced oxidation processes (AOPs). This work provides a design strategy for multi-heteroatom-doped carbon nitride catalysts, emphasizing the synergy between bimetal doping and nonmetal modification for water purification.
• Future Perspectives: Further research should focus on (1) long-term stability in real wastewater matrices, (2) scalability of synthesis, and (3) applications for other emerging contaminants.
• Significance: This study advances the development of efficient eco-friendly PMS activators for wastewater treatment, contributing to safer water resources and sustainable environmental remediation.