CoMn2O4 Catalyst Prepared Using the Sol-Gel Method for the Activation of Peroxymonosulfate and Degradation of UV Filter 2-Phenylbenzimidazole-5-sulfonic Acid (PBSA)

In this study, a bimetallic oxide catalyst of cobalt-manganese (CoMn2O4) was synthesized using the sol-gel method, and it was then characterized using a variety of techniques such as scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD) spectroscopy, X-ray photoelectron spectroscopy (XPS), and nitrogen adsorption–desorption isotherms. The obtained novel catalyst, i.e., CoMn2O4, was then used as an activator of peroxymonosulfate (PMS) for the catalytic degradation of a commonly-used UV filter, 2-phenylbenzimidazole-5-sulfonic acid (PBSA) in water. The effects of various factors (e.g., catalyst dosage, PMS concentration, reaction temperature, and pH) in the process were also evaluated. Chemical scavengers and electron paramagnetic resonance (EPR) tests showed that the •OH and SO4•− were the main reactive oxygen species. Furthermore, this study showed that CoMn2O4 is a promising catalyst for activating PMS to degrade the UV filters.


Introduction
In recent years, UV filters have been increasingly used in personal care products (PCPs), such as sunscreen creams, lipsticks, shampoos, and hair gels. Given their dramatically increasing daily use, UV filters are continuously being discharged into the aquatic environment, and have become a class of emerging contaminants (ECs) and also formed "false persistent" pollution [1,2]. A large number of studies have shown that some organic UV filters can mimic the effects of various hormones in living organisms, thereby interfering with the normal endocrine functions of animals and humans, resulting in potentially serious health threats [3][4][5][6][7][8]. Unfortunately, UV filters are relatively stable in nature and are considered difficult to be biodegraded, making it difficult for municipal wastewater treatment plants to completely remove them during treatment [9,10]. In view of the yearly increases in production and emission of these ECs, the environmental pollution caused by UV filters have become a research hotspot in the environmental field worldwide. Therefore, it is necessary to apply some cost-effective techniques such as advanced oxidation processes (AOPs), to ensure the efficient elimination of such ECs from the water environment [11].
Over the past few decades, traditional AOPs have mainly focused on hydroxyl radicals ( • OH) as the main reactive oxidative species (ROS) to oxidize organic pollutants [12]. As is well known, hydrogen peroxide (H 2 O 2 ) is commonly used as an oxidant to decompose organic pollutants during the production of • OH under certain conditions [13][14][15]. However, as a liquid, H 2 O 2 is inconvenient to transport and easily self-decomposes under normal temperature conditions, which results in a low utilization rate. Meanwhile, an increasing number of studies have shown that activated persulfate (PS), including peroxydisulfate (PDS) and peroxymonosulfate (PMS), can produce sulfate radicals (SO 4 •− ), which have high oxidizing power for the selective degradation of organic pollutants [16][17][18].
The key to the application of the SR-AOPs is to increase the yield of SO 4 •− . Usually, the methods of activating PS have included heat [25], UV [26], and transition metal ions [27]. However, some of these methods can be expensive, whilst others would cause secondary pollution. Heterogeneous catalysts for PS activation such as Fe 3 O 4 have become the focus of current research because they do not pose a secondary pollution problem, have a fast activation rate at normal temperature and pressure, and the process does not require additional energy consumption [28,29]. These types of catalysts have the advantages of environmental friendliness, easy magnetic separation, and low cost. However, their catalytic effects are not satisfactory, and drawbacks also exist, such as low catalytic activity, low oxidant utilization rate, and incomplete degradation of the organic intermediates. In order to improve the catalytic performance, different transition metals, such as Cu, Mn, Cr, Co, etc., have been added to the Fe 3 O 4 [30,31]. Alternatively, humic acid, EDTA, polyhydroquinone, etc., have been coated on the surface of the Fe 3 O 4 [32][33][34][35][36][37][38]. Manganese, a multi-use metal with many stable oxides, has also been used in bimetallic oxides together with Co where it exhibits higher catalytic ability than Co 3 O 4 , Mn 2 O 3 , and their mixtures for PMS activation to degrade Rhodamine B [39]. However, very few systematic studies have been reported on the heterogeneous catalysts containing Mn, as well as their catalytic activities on PMS. Therefore, these composite materials may be promising candidates for the SR-AOPs [40]. This work was aimed at studying the performance of CoMn 2 O 4 in activating PMS for the degradation properties of a common UV filter, 2-phenylbenzimidazole-5-sulfonic acid (PBSA) [41,42]. Generally, metal oxides could be prepared using several methods, i.e., the coprecipitation, hydrothermal, and sol-gel methods. Amongst them, the sol-gel method is considered an attractive synthetic method in which the prepared metal oxides have a higher degree of structural and compositional uniformity [43,44]. Therefore, in this study, CoMn 2 O 4 was prepared using the sol-gel method and then it was characterized using the following techniques: SEM, TEM, XRD, BET, and XPS [45]. Moreover, the effects of various factors on the degradation of PBSA were assessed, and the main reactive oxygen species (ROS) in the PMS/CoMn 2 O 4 system were confirmed as stimulating the activation mechanisms.

Preparation of the CoMn 2 O 4 Catalyst
The CoMn 2 O 4 catalyst was prepared using the sol-gel method as described in Reference [44]. Typically, Co(NO 3 ) 2 ·6H 2 O (0.015 mol) and Mn(NO 3 ) 2 ·4H 2 O (0.030 mol) were dissolved in 90 mL of water, and then glycine (0.045 mol) was added. The obtained solution was then stirred and placed in a water bath at 80 • C until it was completely dissolved. Afterwards, citric acid (0.045 mol) was slowly added to the solution, which was stirred until a gel was formed. The obtained wet gel was then placed in an oven and dried at 250 • C for 1 h. Subsequently, the obtained dry gel was ground and placed in a muffle furnace, and then calcined at 500 • C for 2 h at 10 • C/min to prepare the CoMn 2 O 4 nanopowders.

Characterization Methods
The surface elemental composition of the sample was analyzed by the ESCALAB 250XI X-ray photoelectron spectroscopy (XPS) (Thermo Fisher Scientific, Waltham, MA, USA) using Al K-alpha radiation under conditions optimized for the maximum signal (spot size, 500 µm; lens mode, standard; analyzer mode, CAE; pass energy 30.0 eV; energy step size, 0.050 eV). Wide scans were recorded for the CoMn 2 O 4 , whilst the core level peaks that were recorded in detail were: C 1s and O 1s, Co 2p and Mn 2p.
The morphologies were determined using the Quanta400FEG scanning electron microscope (SEM, FEI, Hillsboro, OR, USA) at 20 kV and the JEM-2100F transmission electron microscope (TEM, JEOL, Tokyo, Japan). The elemental composition was determined using a Horiba EX-250 energy-dispersive X-ray (EDX, Kyoto, Japan) at 20 kV.
The crystal structure of the synthesized sample was confirmed through the X-ray diffraction spectra recorded in the 2θ range of 5-80 • (scan rate of 0.06 • s −1 ), using a Cu-Kα (λ = 0.154 nm) wavelength D8-advanced X-ray diffractometer (XRD, Bruker, Karlsruhe, Germany) at 40 kV and 30 mA.
The specific surface area and the pore size distribution were determined using the TriStar II 3020 surface area and porosity analyzer (Micromeritics, Atlanta, GA, USA) at the liquid nitrogen temperature (−196 • C).

Catalytic Test Procedure
Thereafter, 100 mL PBSA solution (5 mg/L) was added in a conical flask. A certain amount of PMS was then added to the reaction solution, and the mixture was shaken in a water bath at 25 • C. The catalyst CoMn 2 O 4 was then added to initiate the reaction. Then, a 0.8 mL solution was sampled with a pipette at defined time intervals, filtered through a 0.45 µm microporous membrane, and injected into a vial of high-performance liquid chromatography (HPLC). The vial was filled with 0.2 mL of methanol as a quencher. The solution concentration was measured using HPLC.

Analytical Methods
The concentration of PBSA in the sample was analyzed by a Shimadzu LC-20A HPLC with a diode array detector (DAD). The specific conditions: an Agilent Zorbax SB-C18 column (4.6 × 250 mm, 5 µm, Santa Clara, CA, USA) was used; the mobile phase was a MeOH and formic acid solution (0.30%); the flow rate was 1.0 mL/min; the injection volume was 20 µL; and the quantitative wavelength was 303 nm.

Characterization of the Catalyst
The morphology and structure of the CoMn 2 O 4 were revealed by the SEM and TEM images. As shown in Figure 1a, it can be seen that the sample was in the form of irregular flakes with a uniform distribution, where fine particles were distributed on the edge of the block. There was a certain agglomeration and fluffy accumulation, as well as many pores between the particles. As presented in Figure 1b, it can be seen that the grains of the CoMn 2 O 4 powder had an irregular polyhedral structure. Furthermore, the energy-dispersive X-ray (EDX) elemental analysis spectrum of the CoMn 2 O 4 in Figure 1c indicated that the catalyst contained C, O, Co, and Mn elements.

Characterization of the Catalyst
The morphology and structure of the CoMn2O4 were revealed by the SEM and TEM images. As shown in Figure 1a, it can be seen that the sample was in the form of irregular flakes with a uniform distribution, where fine particles were distributed on the edge of the block. There was a certain agglomeration and fluffy accumulation, as well as many pores between the particles. As presented in Figure 1b, it can be seen that the grains of the CoMn2O4 powder had an irregular polyhedral structure. Furthermore, the energy-dispersive X-ray (EDX) elemental analysis spectrum of the CoMn2O4 in Figure 1c Table 1.    The crystal structure of the CoMn 2 O 4 nanoparticle was examined using XRD, as shown in Figure 2. The diffraction peaks appeared at 18 [30]. Moreover, no other peaks were observed on the XRD pattern of the sample, indicating the high purity of the CoMn 2 O 4 obtained using the sol-gel method. The crystal structure of the CoMn2O4 nanoparticle was examined using XRD, as shown in Figure  2. The diffraction peaks appeared at 18.2°, 29.3°, 31.2°, 32.8°, 36.3°, 38.7°, 44.7°, 51.8°, 54.3°, 56.5°, 59.0°, 60.6°, 65.1°, and 74.9°, respectively, which was consistent with the CoMn2O4 crystalline structure (JCPDS 77-0471) as in Reference [30]. Moreover, no other peaks were observed on the XRD pattern of the sample, indicating the high purity of the CoMn2O4 obtained using the sol-gel method. As shown in Figure 3a, the nitrogen adsorption-desorption isotherm displays a type IV isotherm with a wide H3 hysteresis area, indicating the existence of mesopores. This can be further revealed through the corresponding pore-size distribution plots, which was calculated using the BJH (Barrett-Joyner-Halenda) method from the desorption branch isotherm as shown in Figure 3b. These pores may be formed by the agglomeration and fluffy accumulation of the CoMn2O4 particles, as shown in the SEM and TEM images (Figure 1a,b). The specific surface area (SSA) of CoMn2O4 was approximately 24.23 m 2 g −1 , which was calculated using the multipoint BET (Brunauer-Emmett-Teller) method. Figure 3c,d displays XPS surveys of the elemental compositions and chemical states of the CoMn2O4, respectively. Two components of Co were found in the CoMn2O4, that is Co 2+ at 780.5 eV and Co 3+ at 782.0 eV ( Figure 3c). Mn existed in three forms, that is, Mn 2+ , Mn 3+ , and Mn 4+ at 641.1, 642.1, and 643.4 eV, respectively ( Figure 3d). This indicated that the Co and Mn species in the CoMn2O4 existed as mixed valences. These results were consistent with the features of spinel-type CoMn2O4. Thus, the above characterizations confirmed the successful preparation of the CoMn2O4 nanomaterial by the sol-gel method. As shown in Figure 3a, the nitrogen adsorption-desorption isotherm displays a type IV isotherm with a wide H3 hysteresis area, indicating the existence of mesopores. This can be further revealed through the corresponding pore-size distribution plots, which was calculated using the BJH (Barrett-Joyner-Halenda) method from the desorption branch isotherm as shown in Figure 3b. These pores may be formed by the agglomeration and fluffy accumulation of the CoMn 2 O 4 particles, as shown in the SEM and TEM images (Figure 1a The crystal structure of the CoMn2O4 nanoparticle was examined using XRD, as shown in Figure  2. The diffraction peaks appeared at 18.2°, 29.3°, 31.2°, 32.8°, 36.3°, 38.7°, 44.7°, 51.8°, 54.3°, 56.5°, 59.0°, 60.6°, 65.1°, and 74.9°, respectively, which was consistent with the CoMn2O4 crystalline structure (JCPDS 77-0471) as in Reference [30]. Moreover, no other peaks were observed on the XRD pattern of the sample, indicating the high purity of the CoMn2O4 obtained using the sol-gel method. As shown in Figure 3a, the nitrogen adsorption-desorption isotherm displays a type IV isotherm with a wide H3 hysteresis area, indicating the existence of mesopores. This can be further revealed through the corresponding pore-size distribution plots, which was calculated using the BJH (Barrett-Joyner-Halenda) method from the desorption branch isotherm as shown in Figure 3b. These pores may be formed by the agglomeration and fluffy accumulation of the CoMn2O4 particles, as shown in the SEM and TEM images (Figure 1a,b). The specific surface area (SSA) of CoMn2O4 was approximately 24.23 m 2 g −1 , which was calculated using the multipoint BET (Brunauer-Emmett-Teller) method. Figure 3c,d displays XPS surveys of the elemental compositions and chemical states of the CoMn2O4, respectively. Two components of Co were found in the CoMn2O4, that is Co 2+ at 780.5 eV and Co 3+ at 782.0 eV (Figure 3c). Mn existed in three forms, that is, Mn 2+ , Mn 3+ , and Mn 4+ at 641.1, 642.1, and 643.4 eV, respectively ( Figure 3d). This indicated that the Co and Mn species in the CoMn2O4 existed as mixed valences. These results were consistent with the features of spinel-type CoMn2O4. Thus, the above characterizations confirmed the successful preparation of the CoMn2O4 nanomaterial by the sol-gel method.  (d) Figure 3. N2 adsorption-desorption isotherms (a) and pore size distributions (b); Co 2p (c) and Mn 2p (d) spectrum of CoMn2O4.

Catalytic Oxidation of the PBSA
To investigate the activation efficiencies of various catalysts/PMS systems, dynamic experiments to remove the PBSA were conducted as shown in Figure 4. From Figure 4a,b, neither the conventional nanocarbons (CNT-COOH and rGO) nor the classical metal catalysts (Co3O4 and Mn3O4) were effective in activating PMS to remove the PBSA under the selected conditions. Comparatively, the CoMn2O4/PMS system not only exhibited considerable removal efficiency of the PBSA but also showed a universality for the degradation of other pollutants (e.g., UV filter benzophenone-4 (BP-4) and phenol).

Catalytic Oxidation of the PBSA
To investigate the activation efficiencies of various catalysts/PMS systems, dynamic experiments to remove the PBSA were conducted as shown in Figure 4. From Figure 4a,b, neither the conventional nanocarbons (CNT-COOH and rGO) nor the classical metal catalysts (Co 3 O 4 and Mn 3 O 4 ) were effective in activating PMS to remove the PBSA under the selected conditions. Comparatively, the CoMn 2 O 4 /PMS system not only exhibited considerable removal efficiency of the PBSA but also showed a universality for the degradation of other pollutants (e.g., UV filter benzophenone-4 (BP-4) and phenol). (d) Figure 3. N2 adsorption-desorption isotherms (a) and pore size distributions (b); Co 2p (c) and Mn 2p (d) spectrum of CoMn2O4.

Catalytic Oxidation of the PBSA
To investigate the activation efficiencies of various catalysts/PMS systems, dynamic experiments to remove the PBSA were conducted as shown in Figure 4. From Figure 4a,b, neither the conventional nanocarbons (CNT-COOH and rGO) nor the classical metal catalysts (Co3O4 and Mn3O4) were effective in activating PMS to remove the PBSA under the selected conditions. Comparatively, the CoMn2O4/PMS system not only exhibited considerable removal efficiency of the PBSA but also showed a universality for the degradation of other pollutants (e.g., UV filter benzophenone-4 (BP-4) and phenol).  The dosage of the catalysts is a key index in the process of oxidative degradation. Figure 4c shows the effect of different amounts of CoMn2O4 on the degradation of the PBSA. In the absence of PMS, the CoMn2O4 could not remove the PBSA effectively even though the concentration of CoMn2O4 was increased to 50 mg/L. This indicated the weak adsorption of CoMn2O4 (at only about 16%). With 250 mg/L PMS, the CoMn2O4 could efficiently activate the PMS to degrade the PBSA, and the degradation reaction conformed to the pseudo-first-order kinetics. The calculated first-order rate constant (k) at 100 mg L −1 was 1.47 × 10 −1 min −1 , which was 2.2 times higher than that at 25 mg L −1 CoMn2O4 (6.67 × 10 −2 min −1 ). The degradation rate of the PBSA was increased together with the amount of the catalyst because the catalytically active sites increased with the amount of CoMn2O4, and thus more active sites could activate PMS to produce more ROS.
As a precursor to SO4 •− , the concentration of PMS has a great influence on the degradation of pollutants. The effect of different concentrations of PMS on the degradation of the PBSA was also estimated and shown in Figure 4d. The PMS alone could not degrade PBSA when there was no catalyst present. With the increasing amount of PMS, the degradation rate of the PBSA also increased. The k value increased from 0.30 × 10 −1 to 1.06 × 10 −1 min −1 when the PMS dosage was increased from 125 to 250 mg L −1 , and this value could be further increased to 1.87 × 10 −1 min −1 with an increased dosage of PMS (500 mg L −1 ). Under certain conditions of the catalyst (50 mg/L) and within a certain range, the PMS amount was increased, leading to the increasing generation of SO4 •− , which consequently accelerated the catalytic degradation rate of the PBSA.
In the CoMn2O4/PMS system, the removal of the PBSA was also affected by the reaction temperature. As shown in Figure 4e, the higher temperature had a positive effect on the removal of the PBSA. When the dosages of CoMn2O4 and PMS were individually set as 10 and 250 mg/L, respectively, the degradation rate of the PBSA reached 65.87% after 40 min at 25 °C, and when the temperature was increased to 35 °C, the degradation rate increased to 93.76%. Thereafter, the temperature was further increased to 45 °C, the complete removal of PBSA was observed after 14 min. This indicated that the oxidation reaction was an endothermic reaction. The elevated reaction temperature could significantly increase the degradation rate of the PBSA. The k values at 25, 35, and 45 °C were 2.73 × 10 −2 , 4.53 × 10 −2 , and 2.30 × 10 − min −1 , respectively, and thus the activation energy of the reaction was calculated to be 93.12 kJ/mol according to the Arrhenius equation.
The addition of PMS would significantly reduce the solution pH value. The pH of the PBSA solution would reduce from approximately an initial value of 7.3 to 3.4 after the addition of PMS. Thus to examine the effects of real pH on the activation efficiency of CoMn2O4, the PBSA solution and a certain amount of PMS were added into a conical flask, followed by a pH adjustment with NaOH (0.01 M) or H2SO4 (0.01 M). As shown in Figure 4f, the PBSA did not degrade well in a more acidic and alkaline environment. The above experimental phenomena could be explained by the following With 250 mg/L PMS, the CoMn 2 O 4 could efficiently activate the PMS to degrade the PBSA, and the degradation reaction conformed to the pseudo-first-order kinetics. The calculated first-order rate constant (k) at 100 mg L −1 was 1.47 × 10 −1 min −1 , which was 2.2 times higher than that at 25 mg L −1 CoMn 2 O 4 (6.67 × 10 −2 min −1 ). The degradation rate of the PBSA was increased together with the amount of the catalyst because the catalytically active sites increased with the amount of CoMn 2 O 4 , and thus more active sites could activate PMS to produce more ROS.
As a precursor to SO 4 •− , the concentration of PMS has a great influence on the degradation of pollutants. The effect of different concentrations of PMS on the degradation of the PBSA was also estimated and shown in Figure 4d. The PMS alone could not degrade PBSA when there was no catalyst present. With the increasing amount of PMS, the degradation rate of the PBSA also increased. The k value increased from 0.30 × 10 −1 to 1.06 × 10 −1 min −1 when the PMS dosage was increased from 125 to 250 mg L −1 , and this value could be further increased to 1.87 × 10 −1 min −1 with an increased dosage of PMS (500 mg L −1 ). Under certain conditions of the catalyst (50 mg/L) and within a certain range, the PMS amount was increased, leading to the increasing generation of SO 4 •− , which consequently accelerated the catalytic degradation rate of the PBSA. In the CoMn 2 O 4 /PMS system, the removal of the PBSA was also affected by the reaction temperature. As shown in Figure 4e, the higher temperature had a positive effect on the removal of the PBSA. When the dosages of CoMn 2 O 4 and PMS were individually set as 10 and 250 mg/L, respectively, the degradation rate of the PBSA reached 65.87% after 40 min at 25 • C, and when the temperature was increased to 35 • C, the degradation rate increased to 93.76%. Thereafter, the temperature was further increased to 45 • C, the complete removal of PBSA was observed after 14 min. This indicated that the oxidation reaction was an endothermic reaction. The elevated reaction temperature could significantly increase the degradation rate of the PBSA. The k values at 25, 35, and 45 • C were 2.73 × 10 −2 , 4.53 × 10 −2 , and 2.30 × 10 −1 min −1 , respectively, and thus the activation energy of the reaction was calculated to be 93.12 kJ/mol according to the Arrhenius equation.
The addition of PMS would significantly reduce the solution pH value. The pH of the PBSA solution would reduce from approximately an initial value of 7.3 to 3.4 after the addition of PMS. Thus to examine the effects of real pH on the activation efficiency of CoMn 2 O 4 , the PBSA solution and a certain amount of PMS were added into a conical flask, followed by a pH adjustment with NaOH (0.01 M) or H 2 SO 4 (0.01 M). As shown in Figure 4f, the PBSA did not degrade well in a more acidic and alkaline environment. The above experimental phenomena could be explained by the following two aspects: on the one hand, the declined performance at an acidic pH may be due to the formation of CoOH + , which limits the formation of SO 4 •− ; whilst the poor performance at an alkaline pH may be associated with the formation of less-reactive Co(OH) 2 precipitates [40]. On the other hand, the combined effect of the pK a of PMS (9.4, second pK a of its parent acid), the point of zero charge value of CoMn 2 O 4 (around 7.2), and the pK a of PBSA (pK a1 = 4.0 and pK a2 = 11.9) might have also contributed to the observed results [46]. The dissolved organic matter (DOM) is an important factor because it might quench the free radicals, and thus affect the degradation of the target contaminants. HA represents a typical organic matter in sewage and surface waters. Therefore, the effect of HA on the catalytic PMS oxidation of PBSA by CoMn 2 O 4 was studied. As shown in Figure 5, the degradation of the PBSA was inhibited when the HA (1-10 mg/L) was added to the system, and this inhibition intensified as the concentration of HA increased. Specifically, 1 mg/L of HA had little effect on the degradation of the PBSA by CoMn 2 O 4 ; whereas, when the HA concentration was increased to 10 mg/L, the removal rate of the PBSA in 30 min had significantly decreased from 91.24% to 67.92%. The corresponding k values were 4.80 × 10 −2 min −1 and 3.53 × 10 −2 min −1 at the HA levels of 1.0 and 10.0 mg L −1 , respectively. The reason may be that HA acts as a degradable organic substance in competition with the PBSA to react with ROS such as free radicals in the system [24]. two aspects: on the one hand, the declined performance at an acidic pH may be due to the formation of CoOH + , which limits the formation of SO4 •− ; whilst the poor performance at an alkaline pH may be associated with the formation of less-reactive Co(OH)2 precipitates [40]. On the other hand, the combined effect of the pKa of PMS (9.4, second pKa of its parent acid), the point of zero charge value of CoMn2O4 (around 7.2), and the pKa of PBSA (pKa1 = 4.0 and pKa2 = 11.9) might have also contributed to the observed results [46]. The dissolved organic matter (DOM) is an important factor because it might quench the free radicals, and thus affect the degradation of the target contaminants. HA represents a typical organic matter in sewage and surface waters. Therefore, the effect of HA on the catalytic PMS oxidation of PBSA by CoMn2O4 was studied. As shown in Figure 5, the degradation of the PBSA was inhibited when the HA (1-10 mg/L) was added to the system, and this inhibition intensified as the concentration of HA increased. Specifically, 1 mg/L of HA had little effect on the degradation of the PBSA by CoMn2O4; whereas, when the HA concentration was increased to 10 mg/L, the removal rate of the PBSA in 30 min had significantly decreased from 91.24% to 67.92%. The corresponding k values were 4.80 × 10 −2 min −1 and 3.53 × 10 −2 min −1 at the HA levels of 1.0 and 10.0 mg L −1 , respectively. The reason may be that HA acts as a degradable organic substance in competition with the PBSA to react with ROS such as free radicals in the system [24]. In this experiment, we measured the TOC removal of the reaction solutions, and the results are shown in Figure 6a. The TOC removal rate of PBSA in the CoMn2O4/PMS system was 29.59% after 60 min, indicating that the CoMn2O4/PMS system could not only rapidly degrade the PBSA, but it could also convert it into CO2, H2O, and other inorganic substances.
In this study, the catalyst was separated and dried after the catalytic degradation reaction. The recyclability of the CoMn2O4 was evaluated through a multi-cycle experiment of the treated catalyst, and the results are shown in Figure 6b. It could be seen that approximately 80% of the PBSA could be eliminated within 60 min after undergoing the reactions for three repetitions, indicating reasonable stability and reusability of the catalyst. In this experiment, we measured the TOC removal of the reaction solutions, and the results are shown in Figure 6a. The TOC removal rate of PBSA in the CoMn 2 O 4 /PMS system was 29.59% after 60 min, indicating that the CoMn 2 O 4 /PMS system could not only rapidly degrade the PBSA, but it could also convert it into CO 2 , H 2 O, and other inorganic substances. Nanomaterials 2019, 9, x FOR PEER REVIEW 8 of 13 two aspects: on the one hand, the declined performance at an acidic pH may be due to the formation of CoOH + , which limits the formation of SO4 •− ; whilst the poor performance at an alkaline pH may be associated with the formation of less-reactive Co(OH)2 precipitates [40]. On the other hand, the combined effect of the pKa of PMS (9.4, second pKa of its parent acid), the point of zero charge value of CoMn2O4 (around 7.2), and the pKa of PBSA (pKa1 = 4.0 and pKa2 = 11.9) might have also contributed to the observed results [46]. The dissolved organic matter (DOM) is an important factor because it might quench the free radicals, and thus affect the degradation of the target contaminants. HA represents a typical organic matter in sewage and surface waters. Therefore, the effect of HA on the catalytic PMS oxidation of PBSA by CoMn2O4 was studied. As shown in Figure 5, the degradation of the PBSA was inhibited when the HA (1-10 mg/L) was added to the system, and this inhibition intensified as the concentration of HA increased. Specifically, 1 mg/L of HA had little effect on the degradation of the PBSA by CoMn2O4; whereas, when the HA concentration was increased to 10 mg/L, the removal rate of the PBSA in 30 min had significantly decreased from 91.24% to 67.92%. The corresponding k values were 4.80 × 10 −2 min −1 and 3.53 × 10 −2 min −1 at the HA levels of 1.0 and 10.0 mg L −1 , respectively. The reason may be that HA acts as a degradable organic substance in competition with the PBSA to react with ROS such as free radicals in the system [24]. In this experiment, we measured the TOC removal of the reaction solutions, and the results are shown in Figure 6a. The TOC removal rate of PBSA in the CoMn2O4/PMS system was 29.59% after 60 min, indicating that the CoMn2O4/PMS system could not only rapidly degrade the PBSA, but it could also convert it into CO2, H2O, and other inorganic substances.
In this study, the catalyst was separated and dried after the catalytic degradation reaction. The recyclability of the CoMn2O4 was evaluated through a multi-cycle experiment of the treated catalyst, and the results are shown in Figure 6b. It could be seen that approximately 80% of the PBSA could be eliminated within 60 min after undergoing the reactions for three repetitions, indicating reasonable stability and reusability of the catalyst. In this study, the catalyst was separated and dried after the catalytic degradation reaction. The recyclability of the CoMn 2 O 4 was evaluated through a multi-cycle experiment of the treated catalyst, and the results are shown in Figure 6b. It could be seen that approximately 80% of the PBSA could be eliminated within 60 min after undergoing the reactions for three repetitions, indicating reasonable stability and reusability of the catalyst.

Activation Mechanism
Classical quenching tests were carried out to distinguish the contribution of different ROS in the CoMn 2 O 4 /PMS system. Typically, a certain amount of ethanol (EtOH) or tert-butanol (TBA) was added to the system. As shown in Figure 7, using a 1000:1 molar ratio of EtOH (or TBA) to PMS, the degradation rate was obviously inhibited, indicating the dominated role of the • OH and SO 4 •− in the PBSA removal [47,48]. Moreover, we could judge that the major ROS in the reaction system was mainly • OH, rather than SO 4 •− . To further confirm this assumption, EPR tests were carried out, as shown in Figure 8. After five min, DMPO-OH and DMPO-SO 4 signals were observed, and these results were consistent with the radical quenching tests.

Activation Mechanism
Classical quenching tests were carried out to distinguish the contribution of different ROS in the CoMn2O4/PMS system. Typically, a certain amount of ethanol (EtOH) or tert-butanol (TBA) was added to the system. As shown in Figure 7, using a 1000:1 molar ratio of EtOH (or TBA) to PMS, the degradation rate was obviously inhibited, indicating the dominated role of the • OH and SO4 •− in the PBSA removal [47,48]. Moreover, we could judge that the major ROS in the reaction system was mainly • OH, rather than SO4 •− . To further confirm this assumption, EPR tests were carried out, as shown in Figure 8. After five min, DMPO-OH and DMPO-SO4 signals were observed, and these results were consistent with the radical quenching tests. Based on the results from the XPS survey, classical quenching, and EPR tests, a possible mechanism for PMS activation by CoMn2O4 was proposed under the selected reaction conditions. The specific catalytic reaction process is as follows: PMS first reacts with Co and Mn ions (≡Co 2+ , ≡Mn 2+ , and ≡Mn 3+ ) on the surface of CoMn2O4, producing SO4 •− (Equations (1)- (3)). Then, it produces SO4 •− which reacts with water to produce • OH (Equation (4)), and finally, the SO4 •− and • OH degrade and mineralize the PBSA, as described in Reference [40].

Activation Mechanism
Classical quenching tests were carried out to distinguish the contribution of different ROS in the CoMn2O4/PMS system. Typically, a certain amount of ethanol (EtOH) or tert-butanol (TBA) was added to the system. As shown in Figure 7, using a 1000:1 molar ratio of EtOH (or TBA) to PMS, the degradation rate was obviously inhibited, indicating the dominated role of the • OH and SO4 •− in the PBSA removal [47,48]. Moreover, we could judge that the major ROS in the reaction system was mainly • OH, rather than SO4 •− . To further confirm this assumption, EPR tests were carried out, as shown in Figure 8. After five min, DMPO-OH and DMPO-SO4 signals were observed, and these results were consistent with the radical quenching tests. Based on the results from the XPS survey, classical quenching, and EPR tests, a possible mechanism for PMS activation by CoMn2O4 was proposed under the selected reaction conditions. The specific catalytic reaction process is as follows: PMS first reacts with Co and Mn ions (≡Co 2+ , ≡Mn 2+ , and ≡Mn 3+ ) on the surface of CoMn2O4, producing SO4 •− (Equations (1)-(3)). Then, it produces SO4 •− which reacts with water to produce • OH (Equation (4)), and finally, the SO4 •− and • OH degrade and mineralize the PBSA, as described in Reference [40]. ≡Co 2+ + HSO5 − → ≡Co 3+ + SO4 •− + OH − (1) ≡Mn 2+ + HSO5 − → ≡Mn 3+ + SO4 •− + OH − (2) Based on the results from the XPS survey, classical quenching, and EPR tests, a possible mechanism for PMS activation by CoMn 2 O 4 was proposed under the selected reaction conditions. The specific catalytic reaction process is as follows: PMS first reacts with Co and Mn ions (≡Co 2+ , ≡Mn 2+ , and ≡Mn 3+ ) on the surface of CoMn 2 O 4 , producing SO 4 •− (Equations (1)- (3)). Then, it produces SO 4 •− which reacts with water to produce • OH (Equation (4)), and finally, the SO 4 •− and • OH degrade and mineralize the PBSA, as described in Reference [40].

Conclusions
In this study, a CoMn 2 O 4 catalyst was synthesized using the sol-gel method, and the obtained catalyst had high catalytic activity for PMS to degrade the PBSA. The degradation of the PBSA was mainly affected by several factors, such as catalyst dosage, PMS concentration, and reaction temperature. In the reaction system, chemical scavenger and electron paramagnetic resonance (EPR) tests proved that • OH and SO 4 •− were the major radicals, playing a dominant role in the PBSA degradation. In addition, the efficient catalytic performance might be attributed to the Co-Mn synergy in the synthesized material. These findings might contribute to the potential application of CoMn 2 O 4 in SR-AOPs.