Facile synthesis of CoOOH@MXene to activate peroxymonosulfate for efficient degradation of sulfamethoxazole: performance and mechanism investigation

Using MXene as substrate, CoOOH@MXene with different mass content of CoOOH were prepared and used to active peroxymonosulfate (PMS) for the sulfamethoxazole (SMX) degradation. The sample characterizations demonstrated the successful preparation of CoOOH@MXene. CoOOH@MXene possessed much higher BET surface area (183.82 m2/g) than CoOOH (85.36 m2/g) and MXene (6.89 m2/g) due to the good dispersibility of CoOOH particles on MXene. Due to its large surface area, 1.3CoOOH@MXene displayed the best catalytic performance for the degradation of SMX. With 0.2 g/L of 1.3CoOOH@MXene and 0.5 mM of PMS, 20 μM of SMX was completely eliminated in 10 min. The degradation followed pseudo-first-order kinetic model well, with rate constants of 0.33 min−1 for 1.3CoOOH@MXene and 0.054 min−1 for CoOOH. Influencing factors of initial pH, catalyst dosage, PMS concentration, SMX concentration, and co-existing anions on SMX degradation were assessed systematically. Recycling tests verified the excellent reusability and stability of the catalyst. Quenching experiments and electron paramagnetic resonance analysis substantiated that 1O2 played a leading role. Moreover, the intermediates were identified, and degradation pathways and activation mechanism of CoOOH@MXene for PMS were proposed. This work may highlight the application of MXene with transition metals in PMS activation.


Introduction
Antibiotics are continually detected in the water environments. Due to their high toxicity and possibility of producing resistant bacteria, they pose a huge threat to aquatic organisms and human health (Karkman et al. 2018). As a universally used sulfonamide antibiotics, SMX presents in both the wastewater discharged from pharmaceutical manufacturing factories and conventional wastewater treatment plants (Wu et al. 2022). Unfortunately, due to the chemical stability and poor biodegradability of SMX, the traditional wastewater treatment techniques are unable to thoroughly eliminate SMX from aqueous solution (Du et al. 2021). Hence, there is an urgent need to develop an effective technology to exhaustively remove SMX to minimize its perniciousness.
Advanced oxidation processes (AOPs) have increasingly drawn widespread attention and exhibit great development prospects as this approach can produce powerful reactive species to oxidize stubborn macromolecular organic contaminants into innocuous small compounds (Karim et al. 2022). At present, the widely used AOPs mainly include photocatalysis, wet oxidation, ozone oxidation, and Fenton method (Huang et al. 2017). Compared to photocatalytic degradation, SO 4 •− -based AOPs with stronger oxidation ability, wider pH adaptation range, better recyclability, and stability have showed fascinating performance in water and wastewater treatment (Wen et al. 2016;Zhang et al. 2020a). Generally, SO 4 •− can be obtained through activating PMS or PS by visible-light , electrocoagulation (EC) , alkaline, and transition metals (Du et al. 2016;Yu et al. 2019). Without the requirements of external energy supply or chemical reagent consumption, transition metal activation (Co 2+ , Cu 2+ , Fe 2+ , Fe 3+ , Mn 2+ , and Ni 2+ ) has been regarded as one of the most effective and feasible approach (Madihi-Bidgoli et al. 2021). Yet, the activation of transition metals in a homogeneous system is limited on account of the metal ion loss, discommodious recovery, and inevitable sludge generation (Liang et al. 2012). Heterogeneous systems can overcome these shortcomings. Accordingly, much efforts have been made to construct solid catalysts of transition metals especially Co-based compounds .
As a newly reported two-dimensional (2D) layered material, MXene consists of carbonitrides and transition metal carbides (Naguib et al. 2011). Thanks to its well-defined layered structure, excellent metal conductivity, abundant Lewis acid sites and surface functional groups, and structural stability (Anasori et al. 2017), MXene has been extensively applied in reinforcement for nanocomposites (Ling et al. 2014), water purification (Guo et al. 2015), and catalysis (Zhao et al. 2017). In contrast with LDHs, due to a large number of negatively charged -F and -OH on the surfaces of MXene, it was supposed to provide superior water dispersibility and strong anchoring for the free cations through electrostatic interaction as well (Liu et al. 2017b). These merits promote the production and uniform distribution of transition metal nanoparticles. As a result, MXene is considered to be a promising supporting material for nanocatalysts. Liu et al. found that the unique sandwich-like structure of MXene could prevent the aggregation of Co 3 O 4 and increase the active site amount, causing Co 3 O 4 @MXene achieved higher BPA degradation (0.3984 min −1 ) than Co 3 O 4 (0.1439 min −1 ) . Wang et al. pre-sented that the MXene not only acted as the substrate for Co 2+ aggregation, but also afforded a large amount of -F and -OH groups (Wang et al. 2020). Ding et al. synthesized α-Fe 2 O 3 /MXene nanocomposites through using MXene as the catalyst substrate, and the MXene substrate could promote size reduction and good dispersion of α-Fe 2 O 3 nanoparticles, thereby improving the catalytic performance (Ding et al. 2020). Our group firstly demonstrated that Lewis acid sites and hydroxyl groups on MXene played a non-negligible role in PMS activation by Prussian blue analogues@MXene, achieving higher COU degradation (0.121 min −1 ) than PBA (0.023 min −1 ) (Zeng et al. 2021). Overall, previous literature suggested that employing MXene as a supporting material could enhance the catalytic activity, showing good prospects in organics degradation.
As a naturally occurring mineral, cobalt oxyhydroxide (CoOOH) has received increasingly interests as an alternative catalyst to activate PMS owing to its effective electron transfer rate, high hydrophilicity with plentiful of hydroxyl groups and active sites (Decree et al. 2015). Besides, the resulting CoOH + is regarded as efficient active species and the limiting step of the overall catalytic reaction rate in Co 2+ combined PMS system (Shi et al. 2014). Though promising, the catalytic activity of CoOOH nanoparticles for PMS is required to be improved. Herein, we synthesized a novel CoOOH@MXene by supporting CoOOH nanoparticles onto the multi-layered MXene via co-precipitation method. It is expected that MXene would greatly increase the dispersibility and catalytic activity of CoOOH nanoparticles. Nevertheless, to date, there is few study on examining the catalytic activity of CoOOH@MXene in AOPs. The obtained CoOOH@MXene was characterized and used to activate PMS for SMX degradation. The catalytic performance was systematically assessed from the perspective of initial pH, catalyst dosage, PMS concentration, SMX concentration, and co-existing anions. The active species participated in the catalytic system were identified by radical scavenging experiments and EPR analysis. The activation mechanism and oxidation intermediates of SMX were clarified as well.

Chemicals
Ti 3 AlC 2 (MAX) powder (purity ≥ 98%) was acquired from Forsman Co., Ltd. (Beijing, China). PMS was purchased from Aladdin Biochemical Technology Co., Ltd (Shanghai, China). Other reagents were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). In this study, all chemicals are of analytical grade without purification, and ultrapure water was used throughout the experiments produced by a Millipore system (Bedford, USA).

Synthesis of CoOOH@MXene
The synthesis procedure of CoOOH@MXene is schematically illustrated in Fig. 1. MXene used in this study was prepared by hydrofluoric acid etching method (Lukatskaya et al. 2013). Typically, 1 g Ti 3 AlC 2 was dispersed in 10 mL HF solution and stirred vigorously at 35 °C for 24 h. Then, the obtained material was thoroughly washed with ultrapure water and dried in a vacuum oven at 60 °C overnight until reaching a constant weight. Afterwards, soaking 0.1 g MXene in 60 mL dimethyl sulfoxide ultrasonically for 6 h, and then washed it repeatedly with ethanol and ultrapure water, and dried it in a vacuum oven at 60 °C overnight for further use.
The CoOOH@MXene was prepared by a modified chemical precipitation method. 0.1 g of MXene and different mass of CoCl 2 ·6H 2 O (0.05, 0.13, 0.21 g) were mixed in 25 mL of ultrapure water and sonicated for 15 min to obtain a homogeneous mixture. Subsequently, 0.1 M of NaOH solution was added dropwise into the above mixture until the pH reached 11.5. After keeping the resulting solution in water 1 3 bath (60 °C) for 2 h, 5 mL H 2 O 2 (30%) was injected into the solution and continued to react for 2 h. Finally, the resultant solids were washed and dried at 60 °C. The products were named as xCoOOH@MXene, where x represented 0.5, 1.3, and 2.1 according to the designed weight ratio of MXene to CoCl 2 ·6H 2 O.
Details on the characterization and experimental methods are described in Text S1-S2.

Characterization
The micro-structure and morphology of MXene and 1.3CoOOH@MXene were analyzed using SEM. From Fig. 2a, it can be clearly seen that the dense layer structure of Ti 3 AlC 2 transforms into accordion-like MXenes after being etched in HF solution. As depicted in Fig. 2b and c, the as-synthesized 1.3CoOOH@MXene still maintains the layered structure of MXene and many tiny particles are obviously loaded onto the surface and interlayers of MXene, which is probably a connection with the loading of CoOOH nanoparticles. The SEM images imply that the surface of 1.3CoOOH@MXene is much rougher than the pure MXene. Meanwhile, EDS was also recorded to reveal the chemical composition. As seen in Fig. 2d, there are four different elements (Ti, C, O, and Co) on the surface of the 1.3CoOOH@ MXene composites, and all elements were well distributed in the whole area.
The crystal information of MXene and 1.3CoOOH@ MXene were identified by XRD patterns (Fig. 2e). Clearly, the characteristic diffraction peaks located at 20.2°, 36.  . For the XRD patterns of the composite, the characteristic peaks of (004), (103), and (110) belonging to MXene can be distinguished, but the intensity is significantly weakened, confirming that the crystal structure of MXene is well maintained. The disappeared (002) peak is probably ascribed to the loss of titanium from MXene. Further, the characteristic peaks of CoOOH are obviously strengthened with increasing the mass ratio of CoOOH. The above observation indicates the successful synthesis of CoOOH@MXene composites. Furthermore, the crystallite size of 1.3CoOOH@MXene was calculated according to the Debye-Scherrer equation (Ghanbari et al. 2021a): D = Kλ/βcosθ, where D is the crystallite size in the direction perpendicular to the lattice planes, K is a numerical factor frequently referred to as the crystallite-shape factor, λ is the wavelength of the X-rays, β is the width (full-width at half-maximum) of the X-ray diffraction peak in radians, and θ is the Bragg angle (Hargreaves 2016). The calculated average crystalline size of 1.3CoOOH@MXene was 19.3 nm, indicating the formation of nano-sized crystals.
FTIR spectra was then performed to investigate the difference of functional groups on CoOOH, MXene, and 1.3CoOOH@MXene. From Fig. 2f, for the CoOOH, three distinguishable peaks at 581.5, 1839, and 3430 cm −1 are associated with the Co-O double bond, Co-OH group, and H-bonded hydroxyl group (-OH), respectively (Li et al. 2015a). As for the MXene, the peak at 1059 cm −1 can be assigned to the C-F bond, and the peak at 1172 cm −1 represents the C-O group. The peaks at 1631 and 3446 cm −1 belong to the bending vibration of -OH and the asymmetric stretching of -OH, respectively (Xiu et al. 2018). For the 1.3CoOOH@MXene nanocomposite, the primary characteristic peaks of the original CoOOH and MXene are all observed, and no additional functional groups are detected. This further reveals the successful construction of CoOOH@MXene.
The BET surface area of CoOOH, MXene, and various CoOOH@MXene composites was determined by measuring the N 2 adsorption-desorption isotherms, and the results are illustrated in Fig. 2g and Table 1. All samples possess a typical type-IV isotherm with a hysteresis loop, evidencing the presence of mesopores. Specifically, MXene shows the BET surface area of 6.89 m 2 /g, with a pore volume of 0.012 cm 3 /g. The incorporation of CoOOH greatly increased the BET surface area, 160.99, 183.32, and 143.13 m 2 /g for 0.5CoOOH@MXene, 1.3CoOOH@MXene, and 2.1CoOOH@MXene, respectively, with corresponding pore volumes of 0.11, 0.24, and 0.23 cm 3 /g, respectively. The higher BET surface area of the composite was probably due to the good dispersibility of CoOOH particles on the MXene substrate. A higher BET surface area can supply sufficient active sites, thereby achieving higher catalytic activity . This is consistent with the results of degradation experiments in the following part.

Performance of CoOOH@MXene for SMX degradation
The catalytic performance of CoOOH@MXene was studied by the effect of catalyst activation of PMS to remove SMX. Figure 3a shows the removal capacity of SMX in different reaction systems. PMS was ineffective for SMX degradation because of its low tendency toward producing enough active radicals. The MXene also could not activate PMS effectively. And the 1.3CoOOH@MXene could hardly remove SMX either due to the terrible adsorption capacity. By contrast, with the occurrence of PMS and CoOOH@MXene composites, an abrupt drop of SMX concentration occurred, demonstrating that the CoOOH@MXene was efficient activator for PMS, that benefiting the SMX degradation. From the figure, the removal rate reached 99% in the 1.3CoOOH@MXene activated PMS system within 10 min. In addition, the remaining PMS concentration during the reaction progress was recorded in Fig. S1a. Specific oxidant efficiency (SOE), the molar ratio of removed SMX per consumed PMS, was also calculated by Eq. (1) and applied to compare the performance of different reaction systems (Ghanbari et al. 2021b). The higher SOE represents higher efficiency of PMS consumption. In Fig. S1b, the SOE values were 0.064, 0.063, 0.062, and 0.057 for 0.5CoOOH@ MXene/PMS, 1.3CoOOH@MXene/PMS, 2.1CoOOH@ MXene/PMS, and CoOOH/PMS, respectively. The highest SOE value was obtained in 0.5CoOOH@MXene/PMS system rather than in 1.3CoOOH@MXene/PMS system, probably because SMX no longer significantly degraded despite PMS continuing to be consumed in the final stage of the reaction ).
The degradation process is matched well with the pseudofirst-order kinetics, which can be expressed through the following correlation equation: − ln(C/C 0 ) = k obs t, where C 0 represents the initial SMX concentration, C represents the SMX concentration at a certain moment during the reaction, and k obs represents the apparent kinetic rate constant of SMX degradation (min −1 ). As shown in Fig. 3b, CoOOH has the lowest k obs value (0.054 min −1 ), while k obs achieved by 0.5CoOOH@MXene, 1.3CoOOH@MXene, and 2.1CoOOH@MXene are 0.175, 0.334, and 0.211 min −1 , respectively. The 1.3CoOOH@MXene/PMS displayed the highest reaction rate constant (0.334 min −1 ) for SMX degradation, which was much higher than the reported catalysts (Table S1).
The excellent catalytic activity of the composites was attributed to that MXene inhibited the agglomeration of small size of CoOOH. It was noted that 1.3CoOOH@MXene possessed the highest BET surface area, which contributed

Fig. 1 The synthesis route of CoOOH@MXene
Environmental Science and Pollution Research (2022) 29:52995-53008 52998 to PMS activation for SMX degradation as well. Yet, excessive CoOOH loading resulted in an observable declined degradation efficiency, which was probably caused by the overlapping of active sites (Shi et al. 2014). Because of the highest catalytic activity, 1.3CoOOH@MXene was utilized for the follow-up experiments. Additionally, the TOC measurements were carried out to evaluate the mineralization degree of SMX through the oxidizing process. From Fig. S2, the TOC removal efficiency achieved by the 1.3CoOOH@ MXene/PMS system was almost 30.9% in 10-min reaction time, which was mainly due to the conversion of SMX into intermediate products rather than fully oxidization into CO 2 and H 2 O. Increasing the reaction time could improve the mineralization degree of the targeted pollutant.

Effects of operating parameters on catalytic activity
So as to further examine the catalytic activity of 1.3CoOOH@MXene, the effects of initial pH, catalyst dosage, PMS concentration, SMX concentration, and coexisting inorganic anions on the SMX degradation by the catalytic system were studied. At first, experiments on the SMX degradation were performed at pH = 3.0-9.0 as shown in Fig. 3c. With the solution pH increased from 3.0 to 9.0, the removal efficiency gradually enhanced. Compared with the acidic condition, the alkaline condition seemed to be more conducive to the SMX removal. At pH of 9.0, SMX could be absolutely eliminated in less than 5 min, while the degradation was only 71.4% when the initial pH was set at 3.0. The obtained k obs values for pH of 3.0, 5.0, 7.0, and 9.0 were found to be 0.28, 0.35, 0.43, and 0.71 min −1 , respectively (the inset in Fig. 3c).
In general, the SMX removal at different pH was influenced by the pK a of PMS (9.4), the point of zero charge (pH PZC ) of 1.3CoOOH@MXene, and the pK a of SMX (pK a1 = 1.6, pK a2 = 5.7) (Zhang et al. 2020c). Firstly, with increasing solution pH to 9.0, the amount of SO 5 2− in the system would sharply increase. SO 5 2− is easier to be activated to produce SO 4 •− than HSO 5 − (Hong et al. 2020). On the other hand, as displayed in Fig. S3 the pH PZC of 1.3CoOOH@MXene was 4.7. That is to say, at pH below 4.7, the surface charge of the catalyst was positive, while at pH value higher than 4.7, the catalyst surface would be negatively charged. Under acidic pH, plenty of hydrogen bonds were produced between H + and O-O groups of HSO 5 − , preventing the contact between PMS and the positively charged surface of 1.3CoOOH@MXene. This resulted in a lower removal of SMX (Yan et al. 2019). Besides, the excess H + in the solution would consume the generated free radicals as well. With regard to alkaline conditions, OH − participated in the activation process of PMS, thereby accelerating the formation of active species ). Finally, the solution pH also has a considerable influence on the existing species of SMX. At pH below 5.7, SMX exists primarily in non-protonated form. Once the pH surpasses 5.7, SMX exists in a deprotonated form. As reported, the reactivity of the deprotonated SMX was higher than the non-protonated SMX , and this was helpful for the degradation. Fig. S4a exhibits the SMX degradation at various catalyst dosages (0.1, 0.2, 0.3, and 0.4 g/L). At a fixed PMS concentration (0.5 mM), in 5 min, the degradation rate increased from 59.8 to 99.4% as the catalyst dose improved from 0.1 to 0.4 g/L. A higher catalyst dose could definitely supply more active centers for PMS activation, thereby enhancing the PMS utilization rate. The influence of PMS concentration on the SMX removal is depicted in Fig. S4b. The degradation enhanced gradually with the improvement of PMS concentration at a fixed catalyst dosage. At PMS concentration of 0.3 mM, the degradation rate reached 95.4% in 10 min; when the PMS concentration was 0.9 mM, it only took less than 5 min to achieve a nearly complete elimination of SMX. Higher PMS concentration corresponds to a faster degradation owing to more number of active species produced by PMS. In view of the economic aspect, 0.2 g/L of 1.3CoOOH@ MXene and 0.5 mM of PMS were chosen in the following tests. Fig. S4c represents initial SMX concentration (10, 20, 30, and 40 μM) on the performance of 1.3CoOOH@ MXene in the catalytic system. As the pollutant concentration increased from 10 to 40 μM, the degradation rate was restrained step by step. This may be attributed to the insufficient amount of oxidizing species .
Similarly, as the concentration of F − improved from 0 to 30 mM, the degradation rate in 1 min enhanced significantly from 32.1 to 97.4%. Nevertheless, the TBA quenching tests in the following paragraphs showed that the system lacked • OH radicals. Hence, • OH may be produced Environmental Science and Pollution Research (2022) 29:52995-53008 53000 on the surface of the catalyst. F − could release free radicals on the catalyst surface into the solution, thereby promoting the degradation of SMX ). Fig. S5c shows that the existence of HCO 3 − decreased the SMX degradation. Bicarbonate can react with • OH and SO 4 •− to bicarbonate radicals (CO 3 •− ) (Eqs. (4), (5)), which has low reactivity with SMX (Sharma et al. 2015).
As for H 2 PO 4 − , it showed a imperceptible influence on the SMX degradation in the 1.3CoOOH@MXene activated PMS system (Fig. S5d).

Stability and reusability test
Stability and reusability of catalysts are essential for practical application in view of the economic aspect.

3
In order to study the structure stability of 1.3CoOOH@ MXene, Co ion dissolution concentration under different pH values was assessed. From Fig. S6a, the Co ion dissolution concentration at pH 9.0 was 14.8 μg/L, and it increased gradually with the decrease of pH. Yet, the leaching Co ion concentration was much lower than the legal limit of 1.0 mg/L (GB 8978-2002) stipulated by the China Environmental Protection Agency. This indicated the impressive stability of the as-prepared 1.3CoOOH@MXene.
Repeating tests were carried out to assess the reusability of 1.3CoOOH@MXene as well. At the end of each cycle, the catalyst was collected by centrifugation, washed with large amount of ethanol and ultrapure water, and then dried it at 60 °C overnight for the next round of cycling experiments. As illustrated in Fig. 3d, there was no significant degradation loss during the consecutive runs. In the fourth run, the removal rate was even higher than 90%, demonstrating the superior recyclability of 1.3CoOOH@MXene. In addition, the recovered catalyst were also characterized by XRD patterns (Fig. S6b). No significant difference was detected between the fresh and used catalyst, proving its excellent physicochemical stability.

Radical identification
To recognize reactive oxygen species (ROS) involved in the system of 1.3CoOOH@MXene/PMS, a series of quenching experiments were performed. In general, methanol (MeOH) can capture both • OH and SO 4 •− , whereas tert-butyl alcohol (TBA) is used as a quencher of • OH . p-benzoquinone (BQ) is applied to scavenge O 2 •− (Zhang et al. 2020b), and furfuryl alcohol (FFA) has a high reaction rate constant for 1 O 2 . As shown in Fig. 4a, almost no inhibition role of TBA in the SMX removal was observed even if its concentration was 500 times as that of PMS. On the contrast, the addition of 250 mM MeOH greatly declined the degradation rate to 40%, indicating that SO 4 •− was mainly involved in SMX oxidation reaction. However, excess MeOH (250 mM) could not completely inhibit the degradation, implying the involvement of some other ROS. Both BQ and FFA showed severe suppression on the oxidation process, especially when the FFA concentration was 10 mM the degradation was significantly inhibited to 10.7%. The above phenomenon suggested that 1 O 2 played the vital role for SMX degradation, while O 2 •− and SO 4 •− took the second place, while • OH just provided limited activity. To valid the involved ROS intuitively, the electron paramagnetic resonance (EPR) spectroscopy was carried out with spin-trapping agents of 5-dimethyl-pyrroline-N-oxide (DMPO) or 2,2,6,6-tetramethyl-4-piperidine (TEMP). From Fig. 4b, no conspicuous signals were found in the presence of PMS alone. Whereas strong representative 1:2:1:2:1:2:1 heptet signals were observed, which was assigned to the combined signals of DMPO-OH and DMPO-SO 4 adducts. This further evidenced the existence of • OH and SO 4 •− . Simultaneously, the EPR spectra corroborated the inference that O 2 •− was produced on the basis of the characteristic peaks in Fig. 4c. Additionally, a 1:1:1 triplet signal related to TEMP-1 O 2 could also be detected (Fig. 4d) (Wang et al. 2016). According to the above analyses,

Activation mechanism analysis
To better understand the catalytic mechanism of the as-synthesized CoOOH@MXene composites in PMS activation, the valance states and species transformation of Co and O before and after the oxidation reaction were further analyzed through X-ray photoelectron spectroscopy (XPS). The fullscale XPS spectra of 1.3CoOOH@MXene shown in Fig. 5a illustrates the main element of Co, O, Ti, and C. Figure 5b depicts the high-resolution XPS spectrum of Co 2p. Two peaks at 781.5 and 796.5 eV corresponds to Co 2p 3/2 and Co 2p 1/2 , respectively (Gong et al. 2017). In the deconvoluted spectrum, the peak of Co 2p 3/2 at 781.6 eV and Co 2p 1/2 at 797.3 eV corresponded to Co(III), and the peak of Co 2p 3/2 at 783.5 eV and Co 2p 1/2 at 798.5 eV signified Co(II) . Meanwhile, the peaks spotted at 787.5 and 802.5 eV were attributed to satellite peaks of Co 2p 3/2 and Co 2p 1/2 spin-orbits . The overall Co intensity of 62% and 38% were assigned to Co(III) and Co(II) on the fresh 1.3CoOOH@MXene, while the relative proportion of Co(III) declined to 57% and Co(II) increased to 43% on the used one. The increased proportion of Co(II) was attributed to the reduction of Co(III) to Co(II), which demonstrated the participation of the redox pairs of Co(III)-Co(II)-Co(III) in the SMX degradation in the catalytic system. The XPS spectra of O 1 s (Fig. 5c) could be resolved into three individual peaks, 532.9, 531.3, and 529.9 eV representing the molecular water adsorbed on the surface (O ads ), metal hydroxide (Co-OH), and metal oxide (Co-O), respectively (Zhu et al. 2018). After the catalytic reaction, the O ads amount dropped from 43 to 31%, indicating the participation of H 2 O. Surprisingly, the relative proportion of Co-OH sharply grew from 28 to 42% because of the formation of CoOH + , which was the most effective oxygen species to activate PMS and played a crucial role in the activation process (Song et al. 2018).
On the basis of the above discussion, a tentative catalytic mechanism for PMS activation by 1.3CoOOH@MXene was unveiled (Fig. 6). At first, PMS attached on the catalyst surface reacted with Co(III) to produce SO 5 •− and Co(II) via Eq. (6) (Ghanbari and Moradi 2017). In the meantime, the reaction between Co(II) and HSO 5 − resulted in the formation of • OH and SO 4 •− (Eqs. (7), (8))

Possible degradation pathways and toxicity evaluation of intermediates
To clarify the degradation process of SMX in the 1.3CoOOH@MXene activated PMS system, LC-MS was  Fig. 7 Possible degradation pathways of SMX in the 1.3CoOOH@MXene/PMS system Environmental Science and Pollution Research (2022) 29:52995-53008 53005 employed to recognize degradation intermediates. The MS spectra are shown in Fig. S7, and ten possible intermediates obtained are listed in Table S2. According to the identified intermediates and previous reports, four feasible degradation pathways were put forward in Fig. 7. In route 1, the amine group on the benzene ring in SMX was oxidized to form nitroso-SMX (P3) (Du et al. 2018). Then, P3 was then oxidized to generate P7 and P10 via deep hydroxylation and cleavage of unsaturated bonds, followed by the production of P8 and P9 (Liu et al. 2017a). In route 2, the S-N bond was attacked by reactive oxygen species to yield P5 and P10 (Lai et al. 2018). In route 3, the opening of isoxazole ring occurred, leading to the generation of P4. Then, the C-N bond of P4 was cleaved by an electron-transfer reaction to produce P5 and P6 (Bao et al. 2019). In route 4, the hydroxylation of heterocyclic ring occurred to induce P2. The methyl group on heterocycle of isoxazole was further oxidized to P1 due to the attack of SO 4 •− and • OH (Mohatt et al. 2011). Furthermore, the ecological structure activity relationship (ECOSAR) program was applied to predict the biotoxicity of SMX and its oxidation intermediates. From Fig. S8 and Table S3, the LC 50 value of SMX toward fish and daphnia is calculated as 410.762 and 1.872 mg/L, and the EC 50 value of SMX toward green algae is 6.615 mg/L. The corresponding ChVs for fish, daphnia, and green algae are 2.337, 0.086, and 10.402 mg/L, respectively. Thus, SMX can be categorized as an extremely toxic organic pollutant . After the oxidation process, the acute and chronic toxicities of all intermediates to three typical aquatic organisms are less than that of SMX (except the LC50 values of P3, P5, and P10). This reveals that activating PMS by CoOOH@ MXene to degrade SMX is of great significance to the safety of water environment.

Conclusions
Novel CoOOH@MXene with different mass ratio of CoOOH were successfully prepared and applied to activate PMS for the SMX degradation. The CoOOH@MXene displayed impressive catalytic activity. The catalytic performance followed the order: 1.3CoOOH@MXene (0.33 min −1 ) > 2.1CoOOH@ M X e n e ( 0 . 2 1 m i n − 1 ) > 0 . 5 C o O O H @ M X e n e (0.18 min −1 ) > CoOOH (0.054 min −1 ), which was dependent on their specific surface area. Increasing initial solution pH promoted the SMX degradation, with corresponding rate constants of 0.28, 0.35, 0.43, and 0.71 min −1 for pH of 3, 5, 7, and 9, respectively. Increasing PMS concentration and 1.3CoOOH@MXene dose improved the degradation as well. The addition of Cl − and F − greatly accelerated the degradation rate of SMX, while the presence of HCO 3 − lowered the degradation. Moreover, the as-synthesized showed excellent recyclability and stability in the consecutive runs with low cobalt ion leaching. Four degradation pathways were proposed according to the recognized degradation intermediates. According to the quenching tests and EPR analysis, • OH, SO 4 •− , O 2 •− , and 1 O 2 radicals all participated in the SMX degradation, and 1 O 2 played a primary role. Overall, 1.3CoOOH@MXene could be a promising candidate for activating PMS for the SMX degradation.