Carbon Nanotube-Supported FeCo₂O₄ as a Catalyst for an Enhanced PMS Activation of Phenol Removal

Abstract

Peroxymonosulfate (PMS) activation has gained increasing attention for its water remediation. In this work, carbon nanotube-supported FeCo₂O₄ nanoparticles (FeCo₂O₄/CNT) were prepared and showed tremendous potential as a catalyst for PMS activation. The synergistic effect between FeCo₂O₄ and CNT in FeCo₂O₄/CNT promotes its better catalytic performance than individual CNT or FeCo₂O₄. The synthesized FeCo₂O₄/CNT could reach 100% phenol removal with a k value of 0.30 min⁻¹ within 15 min ([PMS] = 0.3 g L⁻¹, [FeCo₂O₄/CNT] = 0.3 g L⁻¹). FeCo₂O₄/CNT can adapt well to a wide pH range (4–9) and a complex water component (with inorganic ions or organic matter). Moreover, the catalytic mechanism investigation suggested that both radical and non-radical pathways are accountable for the efficient removal of phenol.

1. Introduction

Recently, concerns over the presence of persistent organic pollutants in waters have escalated. Traditional water treatment technologies often fall short in adequately removing persistent and toxic compounds, leading to the need for more robust and efficient approaches. Advanced oxidation processes (AOPs), which are a group of powerful chemical treatment methods used to remove pollutants, offering a promising solution to address environmental pollution challenges [1,2,3]. Among various AOPs, the activation of PMS, which involves the generation of highly reactive sulfate radicals (SO₄^•−) that can effectively degrade contaminants, has gained significant attention due to its ability to break down recalcitrant pollutants into less harmful byproducts (e.g., CO₂ and H₂O) [4,5].

Catalysts are critical factors in PMS activation, as they facilitate the conversion of PMS to active species, such as SO₄^•−, hydroxyl radicals (•OH), and singlet oxygen (¹O₂). Transition metals, such as iron, copper, nickel, and platinum, possess unique electronic properties due to the presence of partially filled d orbitals in their valence shells, which give rise to their unique properties (e.g., allow them to form coordination complexes with other molecules or ions), making them important in a broad range of applications, such as catalysis [6], energy storage [7], and electronics [8]. In addition, transition metals often exhibit multiple oxidation states, which allows them to participate in redox reactions and facilitate electron transfer processes during catalysis. To date, catalysts based on various transition metals, such as α-MnO₂, Fe₃O₄, CuO, Co₃O₄, and Co@N-C, have been designed for PMS activation [9,10,11,12]. Among the catalysts, Co-based materials have demonstrated remarkable efficiency for PMS activation in the degradation of diverse contaminants. However, cobalt is a relatively expensive material compared to other transition metals used as catalysts. It is considered a potentially toxic metal, and its release into the environment can have adverse impacts on ecosystems and human health. The high cost of cobalt and environmental concerns can limit its widespread application, especially in large-scale industrial processes.

Strategies, such as catalyst modification, support optimization, and exploring alternative lower-cost materials, could help overcome some of these challenges and enhance the practical application of Co-based materials in advanced oxidation processes [13,14]. Constructing bimetallic oxides is an effective approach for enhancing catalytic performance [15]. Bimetallic oxides are widely used as catalysts for various chemical reactions due to their unique properties and synergistic effects between different metal components. These catalysts consist of two different metal oxides, typically with one metal acting as the active site for catalysis while the other enhances the catalytic performance [16]. Bimetallic oxides often exhibit higher catalytic performance compared to their monometallic counterparts. The presence of two metals in close proximity can promote electron transfer or facilitate the activation of reactants, leading to enhanced reactivity and conversion efficiency. Moreover, the interaction between the metals can create unique active sites and alter the electronic and geometric properties, further enhancing catalytic performance. Ren et al. [17] prepared MFe₂O₄ (M=Co, Cu, Mn, and Zn) for PMS activation, and found CoFe₂O₄ with the highest catalytic performance coupled with PMS. Zhang and co-workers [18] synthesized a single-atom Zr-doped Co₃O₄ catalyst for PMS activation. They demonstrated that the Co-O bonds of Co₃O₄ were elongated and the d-band center of Co was increased by Zr doping, resulting in an enhanced catalytic performance towards PMS activation. Zhou et al. [19] found that the removal of 2,4-dichlorophenol during the FeCo₂O₄/PMS system was better than mono-metallic oxide (Fe₃O₄ or Co₃O₄)-activated PMS. The results proved that the surface -OH groups on FeCo₂O₄ could activate PMS and generate active species for pollutant degradation. Zhang et al. [20] employed a magnetically separable CuFe₂O₄ spinel for the PMS activation of iopromide. Their work revealed that the activity of CuFe₂O₄ was much better than that of CuO, and the Cu(II)-Cu(III)-Cu(II) cycle on CuFe₂O₄ contributed to highly efficient SO₄^•− production for iopromide degradation.

In this study, CNT-supported FeCo₂O₄ is prepared for PMS activation. CNTs possess exceptional properties that make them suitable supports for bimetallic oxide catalysts in catalytic reactions. The high surface area, excellent thermal and chemical stability properties, and strong adsorption capacity provide an ideal platform for catalyst deposition. The tubular structure of CNTs also offers enhanced mass transport properties, facilitating the diffusion and interaction between oxidants and target pollutants. The combination of FeCo₂O₄ with CNT supports can create a highly active and stable catalytic system for PMS activation. The catalytic performance is evaluated by employing phenol as the model pollutant. The optimal reaction conditions (including FeCo₂O₄ dosage, PMS concentration, reaction temperature, and solution pH) for phenol removal are performed. Furthermore, the generated reactive species and catalytic mechanism are studied and explored.

2. Materials and Methods

2.1. Chemicals

In this study, carboxylated multi-walled CNTs ([L] = 15 μm, [d]_{outer surface} = 30–50 nm), iron nitrate nonahydrate (Fe(NO₃)₃·9H₂O), cobalt nitrate hexahydrate (Co(NO₃)₂·6H₂O), ammonia solution (NH₃·H₂O), methanol, tert-butyl alcohol (TBA), PMS (KHSO₅·0.5KHSO₄·0.5K₂SO₄), phenol, and β-carotene were acquired from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Ethanol was offered by Sinopharm Chemical Regent Co., Ltd. (Shanghai, China). Sodium hydrogen carbonate (NaHCO₃), potassium chloride (KCl), and other regents were bought from Yongda Chemical Reagent Co., Ltd. (Tianjin, China). All the chemicals were in an analytical state or of superior purity and employed without any further purification. Ultrapure water was used in all experiments.

2.2. Preparation of FeCo₂O₄/CNT

The FeCo₂O₄/CNT was prepared through a hydrothermal procedure. In brief, 0.2 g of CNT was uniformly distributed in 60 mL of ethanol by sonication treatment. Then, a mixture containing 1 mL of 0.2 M Fe(NO₃)₃·9H₂O and 2 mL of 0.2 M Co(NO₃)₂·6H₂O solution was poured into the CNT suspension with ongoing agitation at an ambient temperature. After 6 h of reaction, 1.25 mL of NH₃·H₂O and 1.75 mL of H₂O were gradually added to the suspension one drop at a time. Then, the reaction temperature was increased to 80 °C and maintained for 10 h. The acquired solution was then moved to a Teflon autoclave and hydrothermal treatment was performed at 150 °C for 3 h. After cooling down, the obtained suspension was cleaned with ethanol and water several times. Ultimately, the filter cake was treated with drying conditions at 60 °C for 8 h to obtain FeCo₂O₄/CNT.

2.3. Characterization

The microstructure and element distribution of prepared FeCo₂O₄/CNT was observed through a transmission electron microscope (TEM, FEI Talos F200S, Columbia, MD, USA) furnished with an energy-dispersive X-ray detector (EDX). The crystal structure was studied by X-ray powder diffraction (XRD, Rigaku SmartLab SE, Tokyo, Japan). The chemical properties of the materials before and after the reaction were characterized using X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha, MA, USA). A fully auto-surface area analyzer (Micromeritics ASAP2020, Norcross, GA, USA) was employed to determine the specific surface areas (SSAs).

2.4. Experimental Section

In this study, phenol was picked as the model organic containment for evaluating the catalytic performance of FeCo₂O₄/CNT. In general, the prepared FeCo₂O₄/CNT (0.03 g) was dispersed in 100 mL of phenol solution (20 mg L⁻¹) by sonification for 5 min. Under continuous stirring, 0.03 g of PMS was poured into the suspension to provoke the reaction. At designated time points, 5 mL of aliquots were gathered and immediately quenched with 1 mL of ethanol to terminate phenol degradation. Finally, the samples were strained utilizing a 0.22 μm filter for subsequent examination.

2.5. Analysis

Phenol concentrations were analyzed using high-performance liquid chromatography (HPLC, Shimadzu LC16, Kyoto, Japan) equipped with a C18 column. The mobile phase was delivered at a flow rate of 0.8 mL min⁻¹ (V_CH3OH/V_H2O = 7/3). The generated radicals were determined using electron paramagnetic resonance spectrometry (EPR, Bruker EMXplus-6/1, Karlsruhe, Germany). 5,5-Dimethyl-1-pyrroline N-oxide (DMPO) was picked as the radical quencher of •OH and SO₄^•−. 2,2,6,6-Tetramethyl-4-piperidone (TEMP) was employed to identify ¹O₂. The contribution of generated reactive species for phenol degradation was measured by quenching tests. Ethanol was applied as the trapping agent for •OH and SO₄^•−, while TBA and β-carotene were implemented as the specific scavengers for •OH and ¹O₂, respectively.

3. Results and Discussion

3.1. Characterization of Catalysts

The TEM technique and EDX mapping were utilized to examine the structure and spatial arrangement of FeCo₂O₄ on the FeCo₂O₄/CNT surface. In the TEM image (Figure 1a), CNT appears as long, tubular structures with a characteristic hollow core. The FeCo₂O₄ nanoparticles are visible as contrasting features attached to or dispersed on the CNT. The EDX mapping shown in Figure 1b–f displays the spatial arrangement of specific elements, which can confirm the presence and spatial arrangement of C, Co, Fe, and O elements within the prepared FeCo₂O₄/CNT.

Determining the Brunauer–Emmett–Teller (BET) surface areas and pore size distribution of FeCo₂O₄/CNT is crucial for understanding its porosity, surface characteristics, and adsorption properties, which have a substantial impact on the catalytic process. The BET surface areas for CNT, FeCo₂O₄, and FeCo₂O₄/CNT were 96.5, 185.1, and 99.8 m² g⁻¹, respectively. The pore size distribution of different materials is displayed in Figure 2b. The average pore sizes for CNT and FeCo₂O₄/CNT were 21.1 and 16.0 nm, respectively, which was in accordance with the structure of CNT. The average pore size of FeCo₂O₄/CNT exhibited a comparably lower value compared to that of CNT alone. This occurrence can be understood in terms of the conglomerating and covering of metal oxide particles, which can occupy more outer space and consequently reduce the pore size of the carrier. The crystal structures of CNT, FeCo₂O₄, and FeCo₂O₄/CNT were analyzed by XRD measurements. As presented in Figure 2c, the characteristic diffraction of the (002) and (100) planes of graphite carbon (PDF# 41-1487) occurs both on CNT and FeCo₂O₄/CNT. For FeCo₂O₄/CNT, the diffraction peaks at 19.0, 31.3, 36.9, 44.9, 59.5, and 65.4 were matched with the (111), (220), (311), (400), (511), and (440) lattice planes, respectively (PDF# 71-0816). There were no other peaks that could be observed on FeCo₂O₄/CNT, indicating the prepared FeCo₂O₄/CNT possessed high purity.

To gain an insight into the chemical structure of prepared FeCo₂O₄/CNT, XPS measurements were analyzed and the corresponding data are illustrated in Figure 3. The survey scan spectrum (Figure 3a) provides visual evidence of the C, O, Co, and Fe species present on the FeCo₂O₄/CNT. The C 1s peak value of 284.8 eV was used for the calibration of the binding energies of Co 2p, Fe 2p, and O 1s peaks. Figure 3b displays the Co 2p spectrum; the binding energies measured at 802.8 and 785.4 eV match those expected for the satellites. The peaks identified at 780.6 and 796.3 eV indicate the presence of ≡Co(II) species, while the peaks at 779.3 and 794.7 eV suggest the existence of ≡Co(III) species. Additionally, the ≡Co(II)/≡Co(III) ratio was calculated to be 1.80. Figure 3c illustrates the XPS spectrum of Fe 2p on FeCo₂O₄/CNT; a satellite peak at 718.0 eV belonging to Fe 2p_3/2 can be observed. The distinctive peaks corresponding to Fe 2p_1/2 and Fe 2p_3/2 were observed at binding energies of 724.0 and 710.9 eV, respectively. The difference in binding energies between the satellite and Fe 2p_3/2 peaks was 7.1 eV. These results demonstrate the presence of ≡Fe(III) species [21]. The O 1s spectrum depicted in Figure 3d can be accurately fitted with three peaks. The O1 peak at 529.7 eV is attributed to the characteristic signature of metal-bonded oxygen (O_lat). The O2 peak observed at 531.0 eV and the O3 peak at 532.3 eV were attributed to the characteristic signals of surface hydroxyl oxygen bonds (-OH) and physically adsorbed oxygen [22].

3.2. Performance of FeCo₂O₄/CNT for Phenol Removal

The activities of catalysts for activating PMS to degrade phenol were conducted through batch tests and are presented in Figure 4. As it can be observed, PMS alone only results in a ~3% phenol degradation efficiency, which can be ascribed to the relatively low redox potential of PMS (E⁰ = 1.82 V). When PMS was activated by CNT, the removal efficiency of phenol displayed little improvement, indicating PMS could not be effectively triggered by individual CNT. The phenol removal efficiency was significantly enhanced when FeCo₂O₄ or FeCo₂O₄/CNT was introduced into the reaction system. After 15 min of reaction, the removal efficiency of phenol increased to 40% in the FeCo₂O₄/PMS system. By contrast, FeCo₂O₄/CNT-activated PMS reached 100% phenol removal within 15 min. The apparent first-order reaction rate constants (k) of the catalysts were obtained by fitting the correlation of the reaction time and −ln(C/C₀). The data are a good fit for the model with an R² of 0.95. The k value for phenol removal in the FeCo₂O₄/CNT/PMS system was 0.30 min⁻¹, which was remarkably better than that of the CNT/PMS (0.001 min⁻¹) and FeCo₂O₄/PMS (0.030 min⁻¹) systems. These results verify that FeCo₂O₄/CNT has a superior performance for activating the PMS of phenol degradation. The prepared FeCo₂O₄/CNT paves the way for more efficient and sustainable approaches to tackle the challenges of environmental pollution.

To evaluate the optimal reaction conditions of phenol removal, a series of PMS activation experiments by the as-synthesized FeCo₂O₄/CNT catalyst under various experimental parameters were performed. The correlation between phenol degradation and the amount of FeCo₂O₄/CNT used are shown in Figure 5a. The dosage of the catalyst played a significant role in catalytic processes for organic pollutant degradation because a more active site could be created with an increased catalyst dosage. It was shown that after 10 min of reaction, 0.5 g L⁻¹ of FeCo₂O₄/CNT showed a greater phenol removal efficiency (~100%) and k value (0.58 min⁻¹) than 0.1 g L⁻¹ of FeCo₂O₄/CNT concentration (η = 85%, k = 0.20 min⁻¹), demonstrating that the increased catalyst facilitated the degradation of phenol during PMS activation. As a main reactant for phenol removal, the effect of PMS concentration on catalytic performance was also investigated. As can be observed from Figure 5b, the removal of phenol also shows positive changes at PMS concentrations varying between 0.1 to 0.3 g L⁻¹. Phenol removal at a PMS concentration of 0.1 g L⁻¹ (η = 51%, k = 0.06 min⁻¹) was much lower than that of 0.3 g L⁻¹ (η = 96%, k = 0.30 min⁻¹), which was due to the fact that the increased substrate could lead to more of the reactant converting into active species for phenol removal. However, with an additional increment in the PMS concentration to 0.4 g L⁻¹, both the phenol removal efficiency and k value experienced a reduction, resulting in values of 86% and 0.17 min⁻¹, respectively. Given the fact that PMS (HSO₅⁻) can react with SO₄^•− and generate SO₅^•− with a low redox potential (E_{SO5•−/SO52−} = 0.81 ± 0.02 V) [23], the inhibition of the catalytic performance may have been due to the consumption of SO₄^•− by adding excessive PMS. It was reported that the solution pH generally had a significant impact on catalytic processes. Herein, the influence of initial solution pH on FeCo₂O₄/CNT activating PMS for phenol removal was evaluated. As shown in Figure 5c, FeCo₂O₄/CNT displays high phenol removal under a wide pH range (4–9) with k values of 0.34–0.80 min⁻¹. The good pH adaptability of the prepared-eCo₂O₄/CNT encourages the potential for practical wastewater treatment.

The reaction temperature was considered to be one of the important factors affecting catalytic activity. As estimated in Figure 6, the performance was enhanced gradually when the reaction temperature was increased from 15 to 45 °C. Compared with the reactions at 15 °C (η_{5 min} = 86%, k = 0.20 min⁻¹), 25 °C (η_{5 min} = 85%, k = 0.30 min⁻¹), or 35 °C (η_{5 min} = 97%, k = 0.69 min⁻¹), phenol removal was increased to 100% within 5 min at 45 °C, accounting for a 1.81–6.25-fold increase in the k value (k = 1.25 min⁻¹). This result indicates the high temperature susceptibility of the FeCo₂O₄/CNT/PMS system. The activation energy (Ea) was calculated to determine the susceptibility. As seen from Figure 6c, the slope value of the line related to ln(k) and 1000/T is analyzed to be −5.85; thus, the Ea for phenol degradation in the FeCo₂O₄/CNT/PMS system is 48.6 kJ mol⁻¹.

It is well known that there are various inorganic ions and natural organic substrates in actual water, which can lead to an increase in the ionic strength and may have an influence on the catalytic capability of FeCo₂O₄/CNT. Thus, the performance of the FeCo₂O₄/CNT/PMS system with inorganic ions (Cl⁻, H₂PO₄⁻, or HCO₃⁻) or natural organic matter (humic acid, HA) to regulate ionic strength was studied to further evaluate its potential in practical applications. As can be observed in Figure 7, the phenol removal efficiencies are little influenced by adding H₂PO₄⁻ (10 mmol L⁻¹) or HA (20 mg L⁻¹). This phenomenon is consistent with the result of Zhang et al. [18]. Studies have shown that ionic strength mainly affects the interactions of electrostatic bonds; yet, it has little effect on covalent or ionic bonds [24]. The abovementioned results reveal that the catalytic reaction in the FeCo₂O₄/CNT/PMS system is accompanied by a strong force between PMS and the active sites on FeCo₂O₄/CNT. However, it is notable that the catalytic performance was enhanced with the presence of 10 mmol L⁻¹ Cl⁻ or 10 mmol L⁻¹ HCO₃⁻. After 2.5 min of the reaction, phenol removal reached 100%. The addition of a high concentration of HCO₃⁻ can make the solution approach alkaline conditions, which can be beneficial for the activation of PMS. The presence of Cl⁻ in the FeCo₂O₄/CNT/PMS system could lead to the production of chloride species (e.g., Cl^• and Cl₂^•−), which could also contribute to the degradation of pollutants. Recently, Wang and co-workers [25] found that ¹O₂ could be effectively produced through the reaction between PMS (HSO₅^•−) and Cl⁻, which thereafter enhanced the degradation of pollutants. Therefore, the enhancement of performance with the presence of Cl⁻ can be ascribed to the generation and participation of chloride species and/or ¹O₂ in the FeCo₂O₄/CNT/PMS system.

3.3. Catalytic Mechanism

According to the previous studies, the process of PMS activation involves both radical and non-radical pathways [26,27]. While radical pathways, such as SO₄^•− and •OH formations, are well-known mechanisms for PMS activation, non-radical pathways can also play a significant role. Understanding both radical and non-radical pathways is important for comprehensively understanding the overall mechanism of PMS activation. To identify and characterize the formation of active species during the reaction, quenching tests were performed by adding TBA to scavenge •OH, ethanol to quench both •OH and SO₄^•−, and β-carotene to quench ¹O₂, respectively. The results in Figure 8 illustrate that, with the addition of 0.2–0.5 mol L⁻¹ of TBA, the phenol degradation is similar to the results of the control experiment; however, the terminal degradation efficiency decreases slightly to 96%. By comparison, the removal was weakened by the presence of ethanol. After 15 min of the reaction, the phenol removal was 87%. These results indicate that the contribution of SO₄^•− to phenol degradation is relatively greater than that of •OH. Non-radical pathways involve reactions that do not directly generate radicals, but still participate in PMS activation. The efficiency of phenol removal was remarkably inhibited to 76% when 0.1 mmol L⁻¹ of β-carotene was introduced to the system, indicating the contribution of ¹O₂ to phenol degradation.

To elucidate the active species of PMS activation by FeCo₂O₄/CNT in depth, EPR measurements were conducted using specific capture agents to detect particular reactive species. In this study, the capture agents used were DMPO for •OH and SO₄^•− and TEMP for ¹O₂. As can be observed in Figure 9a, a seven-fold peak belonging to DMPOX [28,29], which refers to the oxidation products of DMPO (when DMPO reacts with various ROSs, it undergoes oxidation, resulting in the formation of DMPOX), can be found when DMPO is added to the FeCo₂O₄/CNT/PMS system. By analyzing the resulting DMPO adducts of EPR measurements, it strengthens the notion that both •OH and SO₄^•− are implicated in phenol removal. Singlet oxygen, which also plays a significant role in PMS activation, was also detected and characterized. A distinct triplet signal can be discerned in Figure 9b, showing a consistent relative intensity ratio of 1:1:1, which corresponds to the TEMP-¹O₂ adduct. This result confirms the presence of ¹O₂ in the FeCo₂O₄/CNT/PMS system.

The interaction between metal oxide and CNT during the reaction can synergistically enhance the catalytic performance, resulting in a significant overall improvement. As mentioned above, the phenol removal efficiency of FeCo₂O₄/CNT (96%) was significantly better than that of CNT (3%) and FeCo₂O₄ (35%) alone. As shown in Figure 10, when PMS and mechanically mixed CNT and FeCo₂O₄ are employed together, the phenol removal efficiency reaches 56% within 10 min, which is greater than the sum of PMS activated by CNT and FeCo₂O₄ (38%). The results demonstrate that there is a synergistic effect between CNT and FeCo₂O₄ in the FeCo₂O₄/CNT/PMS system. The synergistic effect mainly includes the following two aspects: (1) enhanced electron transfer: CNT possesses unique electronic properties, including high electrical conductivity and electron mobility. The introduction of CNT into the system can facilitate the efficient transfer of electrons between bimetallic oxide nanoparticles and PMS, leading to enhanced reactive species generation. (2) Increased active sites: the large surface area of CNT offers numerous active sites for anchoring bimetallic oxides and provides more sites for the adsorption of PMS and pollutants, thereby promoting higher catalytic activity.

To further elucidate the underlying mechanisms of PMS activation, the chemical structure of FeCo₂O₄/CNT following the reaction was characterized by XPS measurements. The C 1s peak value of 284.8 eV was used for the calibration of the binding energies of the Co 2p spectrum. It was recognized that activating PMS by Co-based oxides to generate active species was accompanied by the conversion of ≡Co(II) to ≡Co(III). The reaction involved the transfer of an electron from ≡Co(II) to PMS, resulting in the formation of ≡Co(III) species, enabling the generation of active species (•OH and SO₄^•−). Meanwhile, the ≡Co(III) species was reduced by PMS, regenerating ≡Co(II) and completing the catalytic cycle. This continuous electron transfer mechanism ensured sustained PMS activation and reactive species generation.

$$\equiv {\mathrm{Co}\left(\mathrm{II}\right)+\mathrm{HSO}}_5^{-}\to \equiv {\mathrm{Co}\left(\mathrm{III}\right)+\mathrm{OH}}^{-}{+\mathrm{SO}}_4^{•-}$$

(1)

$$\equiv {\mathrm{Co}\left(\mathrm{III}\right)+\mathrm{HSO}}_5^{-}\to \equiv {\mathrm{Co}\left(\mathrm{II}\right)+\mathrm{H}}^{+}{+\mathrm{SO}}_5^{•-}$$

(2)

$$\equiv {\mathrm{Co}\left(\mathrm{II}\right)+\mathrm{HSO}}_5^{-}\to \equiv {\mathrm{Co}\left(\mathrm{III}\right)+•\mathrm{OH}+\mathrm{SO}}_4^{2-}$$

(3)

$$\equiv \mathrm{Co}\left(\mathrm{III}\right)+{\mathrm{e}}^{-}\to \equiv \mathrm{Co}\left(\mathrm{II}\right)$$

(4)

$${\mathrm{O}}_{\mathrm{lat}}\to {\mathrm{O}}^{*}$$

(5)

$${\mathrm{O}}^{\ast }{+\mathrm{HSO}}_5^{-}\to { \mathrm{HSO}}_4^{-}+{}_{}{}^1\mathrm{O}{}_2$$

(6)

Figure 11a exhibits the Co 2p spectrum of used FeCo₂O₄/CNT; the satellites located at 803.4 and 786.9 eV can be observed. The peaks found at 781.7 and 796.7 eV correspond to the ≡Co(II) species, while the peaks at 780.2 and 795.3 eV belong to the ≡Co(III) species. The ≡Co(II)/≡Co(III) ratio for used FeCo₂O₄/CNT was calculated to be 1.70, which was slightly reduced by 0.1% compared to fresh FeCo₂O₄/CNT. The little difference between the ≡Co(II)/≡Co(III) ratio before and after the reaction suggested the reduction rate of ≡Co(III) to ≡Co(II) was enhanced in the FeCo₂O₄/CNT/PMS system, allowing for the superior removal of phenol. The ≡Co(II)/≡Co(III) ratio for FeCo₂O₄ was calculated to be 1.34, which was obviously lower than that of fresh FeCo₂O₄/CNT. The result can be attributed to the fact that CNT favors the generation of low-valent Co, which is beneficial for improving catalytic activity. The XPS spectrum of Fe 2p on used FeCo₂O₄/CNT also presented a distinct satellite peak and the difference in binding energies between the satellite and Fe 2p_3/2 peaks was slightly decreased to 6.3 eV. The O 1s spectrum of used FeCo₂O₄/CNT is depicted in Figure 11c. The O1, O2, and O3 peaks can be observed at 530.3, 531.7, and 533.3 eV, respectively. Compared with fresh FeCo₂O₄/CNT, the ratio of metal-bonded oxygen significantly decreased from 41.3% to 20.5%. It was reported that some O_lat could be released and subsequently converted into active oxygen (O^∗). The generated O^∗ could react with HSO₅⁻ and transfer into ¹O₂ [30].

4. Conclusions

In conclusion, a CNT-supported FeCo₂O₄ catalyst was successfully prepared in this study. The prepared FeCo₂O₄/CNT showed tremendous potential as a catalyst for the activation of PMS. The k value of FeCo₂O₄/CNT for phenol removal was 0.30 min⁻¹, which was 300- and 10-times greater than that of CNT (0.001 min⁻¹) and FeCo₂O₄ (0.030 min⁻¹), respectively. Both radical (•OH and SO₄^•−) and non-radical (¹O₂) pathways were demonstrated to be responsible for the superior removal of phenol. The enhanced electron transfer and increased active sites caused by the synergistic effect between FeCo₂O₄ and CNT enhanced the overall catalytic performance of the material. This study can provide an insight into the design of highly effective catalysts for PMS activation.