Insight into catalytic activation of bisulfite for lomefloxacin degradation by simple composite of calcinated red mud

Antibiotic was detected in many environments, and it had posed a serious threat to human health. The advanced oxidation process has been considered an effective way to treat antibiotics. In this work, using industrial waste red mud (RM) as raw material, a series of modified RM (MRM-T; T donates the calcination temperature) was obtained via a facile calcination method and applied to activate sodium bisulfite (NaHSO3) for the lomefloxacin (LOM) degradation. Among all MRM-T, MRM-700 exhibited superior catalytic activity, and approximately 89% of LOM (10 mg/L) was degraded at 30 min through the activation of NaHSO3 ([NaHSO3] = 0.5 g/L) by MRM-700 ([MRM-700] = 0.9 g/L). Moreover, the kinetic constant of LOM removal in the MRM-700/NaHSO3 system (0.082 min−1) was 16.4 times higher than that of the RM-raw/NaHSO3 system (0.005 min−1). The as-synthesized product of MRM-700 was characterized by N2 adsorption–desorption isotherms, X-ray diffraction (XRD), scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), and Raman spectra. The result indicated that the catalyst possessed excellent pore structure, high specific area, and abundant Fe3+ sites, and the lattice of Fe2O3 was doped after calcination, both of which were favorable for the activation of NaHSO3. The quenching experiment proved that •SO4− and •OH− active species were produced in MRM-700/NaHSO3 system, and •SO4− played a dominant role in LOM removal. In addition, the potential LOM degradation pathway was analyzed via UPLC-MS technology and density functional theory (DFT) calculation, and the toxicity of the treated LOM solution was tested by the culture of mung bean sprouts. This study not only provided a feasible strategy for the valuable use of RM to activate NaHSO3 but also offered a cost-effective catalyst for the efficient removal of pollutants in wastewater.


Introduction
Antibiotics were developed and utilized so that humans and animals could be protected from infectious diseases. In contemporary society, the common antibiotics include fluoroquinolones, sulfonamides and tetracyclines, macrolide, and β-lactam You et al. 2021). Among these antibiotics, fluoroquinolones (FQs) have been frequently used in humans and animals in recent years due to their broad antibacterial spectrum, high bioavailability, and long half-life (Zhang et al. 2019a). Lomefloxacin (LOM), a representative of fluoroquinolone, is an effective drug for treating human bacterial infections, such as treatment and prevention of underlying ocular infections (Guo et al. 2020a). However, overuse of antibiotics cannot be fully metabolized by humans and animals, leading to the release of antibiotics into the environment. Thus, many studies have shown that fluoroquinolones were detected in municipal wastewater treatment plants (Voigt et al. 2020;Guo et al. 2020b;Mackul'ak et al. 2015). According to the reference, LOM was detected in the Yellow River, Haihe River, and Liaohe River at a concentration of 299 ng/g, which posed a potential threat to the ecosystem and human health (Zhou et al. 2011). The problem of overuse of LOM has been recognized and gained more and more attention. So, it is important to find a way to deal with the LOM remaining in water bodies.
Currently, many methods are used to treat antibiotics in wastewater, such as adsorption (Tian et al. 2022), biological treatment, membrane processes , persulfate oxidation (Zhang et al. 2019a, b), and photocatalytic degradation (Ebrahimi & Akhavan 2022;Yousefi et al. 2021;Ghiyasiyan-Arani et al. 2017). Because of a promising, efficient, and eco-friendly technology (Ganiyu et al. 2022), sulfite-based advanced oxidation processes (S-AOPs) are valued and applied in wastewater treatment . Sulfate radicals (•SO 4 − ) based heterogeneous S-AOPs show strong oxidation performance (Zhen et al. 2021) and have been successfully applied to the treatment of pollutants in water, such as antibiotics , pesticides , and persistent organic pollutants . In a word, S-AOPs have the following advantages: (i) Compared with hydroxyl radicals (•OH − ) (E 0 = 2.7 V), sulfate radicals (•SO 4 − ) have a higher oxidation-reduction potential (E 0 = 2.5-3.1 V vs. NHE) (Hu et al. 2021); (ii) sulfate radicals have a wider operative pH range ); (iii) sulfate radicals have a longer half-time (Li et al. 2016a). Commonly, persulfate (PS) or peroxymonosulfate (PMS) can be activated to generate sulfate radicals ) through heating (Balcioglu et al. 2016), light radiation (Khanet al. 2017), transition metal (Qi et al. 2021), alkali activation (Cassidy et al. 2015), and electrochemistry activation (Li et al. 2016b). However, the widespread availability of PS and/or PMS in S-AOPs is relatively limited due to the relatively high cost and persistent toxicity of PS and PMS (Crincoli & Huling 2021). Different from persulfate and peroxymonosulfate, sulfite, having the advantages of low cost and nontoxicity, shows a good future in treating wastewater. Sodium bisulfite (NaHSO 3 ) of sulfite is an industrial by-product of the chemical industry (Qi et al. 2021). Besides its common advantages of sulfite of low cost and nontoxicity, NaHSO 3 was abundant and easily obtained for use (Wu et al. 2021a, b). But the utilization of NaHSO 3 in S-AOPs is limited due to the low activation efficiency. Thus, it is very important for updating the technology for activating sulfite . Mei et al. (2020) synthesized Fe 2 O 3 as the catalyst, which can be used to activate NaHSO 3 for the degradation of orange II. Yang et al. (2020) obtained a Fe@Fe 2 O 3 core-shell nanomaterial, and more than 90% of orange II was degraded by the Fe@Fe 2 O 3 -NaHSO 3 system at 30 s. Based on the reports, it was safely concluded that Fe 2 O 3 is an ideal material to activate NaHSO 3 .
Red mud (RM), an industrial solid waste, is generated in the Bayer process of alumina production (Jahromi & Agblevor 2018). RM has the characteristics of high alkalinity and heavy metal ions, both of which are potentially harmful to ecological and personal health . It has been reported that the growth rate of RM is maintained at 120 million tons per year (Archambo et al. 2021). However, RM also as a resource is now widely used in many industries, including the most significant consumption being as an additive in cement and construction (Archambo & Kawatra 2021;Samal 2021). Furthermore, as a mixture of various minerals, red mud, containing many elements such as Fe, Al, and Ti (Guru et al. 2018), is often used as a raw material for the preparation of efficient catalysts (Das, 2021;Russkikh et al. 2020). Ba et al. (2022) prepared an RM-based Fenton-like catalyst (ACRM bp ), which was obtained by calcination of the acidified RM together with bagasse. The ACRM bp showed excellent activity in the Fenton-like reaction. Using RM and coconut shells as the sources of iron and carbon, respectively, Guo et al. (2021c) synthesized biochar-supported red mud catalyst (RM-BC (HP) ) through the method of hydrothermal synthesis under acidic conditions followed high-temperature calcination, and the synthesized RM-BC (HP) was an efficient PS activator for the removal of acid orange. Shi et al. (2020) prepared a catalyst from RM under different calcination temperatures, and the prepared catalyst showed excellent performance for the degradation of TC with its electron transfer capability without adding any other agents. In a word, RM containing a variety of metal elements can be used to prepare the catalyst for special functions via calcination without adding other metal oxides.
In this study, using RM as raw material, the catalyst of modified RM was successfully obtained by a simple calcination method without the addition of other chemicals. The as-prepared catalyst of modified RM was designed as MRM-T, where T represents the calcination temperature. The characterization of MRM-T was studied through XRD, SEM, BET, XPS, and Raman spectra. Furthermore, the obtained catalyst was applied to activate sodium bisulfite for the degradation of lomefloxacin. The main purposes of this paper were listed as follows: (1) to investigate the reasons for the increased activity of MRM-T after calcination; (2) to explore the mechanism of NaHSO 3 activation by the catalyst and determine the main free radicals in the reaction system; (3) to propose the possible degradation pathways of LOM via LC-MS analysis and DFT calculation; (4) to present a systematic investigation about the operating parameters on the degradation of lomefloxacin.

Preparation and characterization of catalyst
The main chemical composition of RM and the mass fractions are shown in Table 1 (Li et al. 2019). Before use, the RM was dried at 100 °C. Then, the dried RM was screened by a 60 mesh screen and recorded as RM-raw. The calcination method for obtaining the MRM-T sample was based on the previous studies of our group (Li et al. 2019). The sample of RM-raw was firstly calcined in a muffle furnace at different temperatures (300 °C, 500 °C, 700 °C, and 900 °C) for 2 h. When the process of calcination finished, the as-prepared catalyst of MRM-T was obtained. According to the calcination temperatures of 300 °C, 500 °C, 700 °C, and 900 °C, the obtained catalysts were expressed as MRM-300, MRM-500, MRM-700, and MRM-900, respectively. Moreover, to compare the catalytic ability of MRM-T, the reference catalyst of pure Fe 2 O 3 was prepared through the calcination of commercial Fe 2 O 3 at 700 °C for 2 h, and the as-prepared reference catalyst was designed as MFe 2 O 3 -700.
X-ray different (XRD) patterns of the sample were obtained with a SmartLab 3KW diffractometer (Rigaku, Japan). BET was determined using a Micromeritics ASAP 2460 Version 3.01. Scanning electron microscopy (SEM) micrographs of the sample were recorded with a ZEISS Sigma 300 scanning electron microscope. X-ray photoelectron spectra (XPS) were performed on Thermo Scientific K-Alpha. Raman spectra were recorded using a WITec Alpha 300R HR Raman spectrometer with 532 nm as the excitation source.

Catalytic activity measure
The activity of the catalyst was evaluated through LOM degradation in the Fenton-like system and the reaction facility described in our previous work . Batch experiments were carried out in a 500-mL beaker located in the mechanical shaker. In each experiment, 90 mg of catalyst powder was dispersed into 100 mL of 10 mg/L LOM solution (initial pH = 5). The reaction was initiated by adding 50 mg of NaHSO 3 to the mixture, and the mixed suspension was shaken for 30 min at 160 rpm by a constant temperature shaker. After the reaction, the reacted suspension was centrifuged to separate the catalyst. The remaining LOM was determined using a UV-vis spectrophotometer (VIS-722, SHYK, China) at the maximum wavelength of 281 nm. The degradation ratio of LOM in the reaction solution was calculated by the following formula Eq. (1); the first-order model (Langmuir-Hinshelwood model, Eq. (2)) was used to fit the experimental data of LOM degradation to gain the reaction kinetics.
where C 0 and C t represent the initial concentration of LOM and the concentration of LOM after the reaction (mg/L), respectively, and k represents the pseudo-first-order rate constant (min −1 ).
For pH effect experiments, the value of the initial pH of reaction solution was adjusted by NaOH and HCl (0.1 mM) to 3, 5, 7, and 9, respectively. All the experiments were performed in duplicate to ensure the veracity of the results. In addition, the degradation intermediates of LOM were determined by using an electrospray ionization mass spectrometry (ESI-MS, THERMO FISHER, USA).

Toxicity test
Using MRM-700 as catalyst, the raw LOM solution (10 mg/L) was firstly treated by the activation of NaHSO 3 under the following conditions: reaction time = 30 min, initial pH of 5 (unadjusted), [NaHSO 3 ] = 0.5 g/L, and [catalyst] = 0.9 g/L. After treatment, the treated LOM solution was used to carry out the toxicity testing. The toxicity of the experimental solution was examined based on the length of the mung beans (2) −ln C t ∕C 0 = kt  . Firstly, the mung bean seeds were washed with deionized water. Then, the mung beans were placed glass petri dish. Following, a layer of paper towels was covered with mung beans, and the paper towels were kept damp until the mung beans sprouted. Next, the normally growing mung bean sprouts were selected for further incubation in dark for 4 days in three different Petri dishes by using the treated LOM solution, the initial LOM solution, and deionized water, respectively. After the incubation of 4 days, the seedlings were sampled from each petri dish, and the radicle length was measured.

Reusability and stability of MRM-700
The reusability of MRM-700 was tested by a cycling experiment with five repetitions. A total of 90 mg of catalyst and 50 mg of NaHSO 3 were dispersed in 100 mL of 10 mg/L LOM solution. Then, the reaction was carried out for 30 min. For the reacted suspension, after standing for about 20 min, the supernatant was removed by a glass pipette equipped with a needle. And then, using the deposited MRM-700 as the catalyst, the new degradation reaction for the next test cycle was started after 100 mL of LOM solution and 50 mg of NaHSO 3 were added. Furthermore, during each test cycle, the total Fe ions (C tol , mg/L) and the ferrous ions (C Fe 2+ , mg/L) leached into the suspension were measured using the method that has been presented in our previous report ).

Theoretical calculations
Based on the density functional theory (DFT), the LOM molecule was geometrically optimized by Gaussian 09 program, and the Fukui index was performed by Multiwfn 3.6.
The detailed methods were shown in text S2 in Supporting Information.

Characterization
To investigate the effect of calcination on the pore structure, the N 2 adsorption-desorption isotherms and the corresponding pore size distribution curves for RM-raw, MRM-700, and MRM-900 were recorded, respectively. As shown in Fig. 1(a), both the isotherms of RM-raw and MRM-700 were typical IV adsorption isotherms with a well-defined H3-type hysteresis Heydariyan et al. 2022), which was the characteristic of mesoporous materials with some parallel plate slits (Busto et al. 2016;Li et al. 2020). As shown in Fig. 1(b) and Table 2, MRM-700 had higher mesopore content than RM-raw. According to Table 2, compared with the RM-raw, MRM-700 has a larger surface area (10.89 m 2 g −1 ). The BJH pore size distribution showed that MRM-700 had an average diameter of 17.15 nm and a narrow pore volume of 0.021 cm 3 g −1 . Thus, compared  with RM-raw, MRM-700 exhibited a better pore structure, which could expose more active sites on the catalyst surface and provide enough space for the reaction between contaminants and free radicals. The XRD patterns of RM-raw, MRM-300, MRM-500, MRM-700, and MRM-900 are presented in Fig. 2. As shown in Fig. 2(a), the diffraction peaks were observed at 2θ = 24.2°, 33.2°, 35.7°, 40.9°, 49.5°, 54.1°, 62.6°, and 64.1°, corresponding to (0 1 2), (1 0 4), (1 1 0), (1 1 3), (0 2 4), (1 1 6), (2 1 4), and (3 0 0) planes of α-Fe 2 O 3 (hematite: JCPDS No. 33-0664), respectively (Zhang et al. 2019b;Wu et al. 2021a, b). The intensity of the diffraction peaks of α-Fe 2 O 3 in MRM-T was significantly stronger than that of RM-raw (Panahi-Kalamuei et al. 2015;Davar et al. 2010), due to the transformation of FeOOH (goethite: JCPDS No. 29-0713) to α-Fe 2 O 3 (Li et al. 2019). The ferric oxide contained in the prepared composite, as a transition metal oxide, plays an activating effect on HSO 3 − , thus the formation of more α-Fe 2 O 3 in the MRM-T could improve the catalytic activity of the catalyst. And the α-Fe 2 O 3 was the main substance for activating NaHSO 3 due to the fact that the other characteristic peaks were not observed in the XRD pattern of MRM-700 (Fig. 2a). Moreover, as shown in Fig. 2(b), the peaks of MRM-700 shifted to a higher angle, indicating that the lattice of α-Fe 2 O 3 was doped by other atoms in the RM (Guo et al. 2020a Fig. 2(a), the three compositions of cancrinite, calcite, and gehlenite found in RM-raw have been changed in the MRM-T sample due to calcination. As shown in Fig. 2(a), cancrinite and calcite were not observed in MRM-700 because both of them were decomposed under the calcination temperature of 700 °C (Fernandez & Carlos 2016). As a result, MRM-700 had a larger surface area than RM-raw, which was consistent with the result of BET testing. Moreover, some analyses of the XRD patterns of MRM-900 are presented in Text S1 in Supporting Information.
The SEM images of RM-raw and MRM-700 are displayed in Fig. 3. As depicted from SEM images, the RM-raw sample displayed irregular shapes and loose arrangement structures. Compared with RM-raw (Fig. 3a), MRM-700 ( Fig. 3b) showed some agglomeration (Shi et al. 2020), but the particles of MRM-700 were very fine and could increase the overall surface area of the catalyst (Hajjaji et al. 2016). On the other hand, during the process of calcination, gas generated from the thermal decompositions of calcite and other unstable organics, all of which were the constituents of RM-raw, was conducive to the formation of loose structure of the MRM-700 catalyst (Fernandez & Carlos 2016). Furthermore, as displayed in Fig. 3(c), the corresponding elemental mapping of the MRM-700 composite illustrated that the Fe and O elements were distributed in the composite, confirming the distribution of Fe 2 O 3 in the MRM-700 composite. Based on the above reasons, for the obtained catalyst of MRM-700, with its high surface area and loose structure, more active sites could be exposed on the catalyst surface, which would provide a platform for surface reaction on the solid catalyst Amiri et al. 2017). And, the information obtained from the SEM result was consistent with the BET testing, indicating that MRM-700 truly held a loose structure and higher surface area than RM-raw.
The purity and type of different phases could be identified by the characteristic phonon modes in the Raman spectra (Fouad et al. 2019). As shown in Fig. 4(a), the Raman a b Fig. 2 a XRD picture of RM-raw, MRM-300, MRM-500, MRM-700, and MRM-900; b enlarged view of plane spectrum was recorded in the range of 100-700 cm −1 to further confirm the structures of MRM-700 and RM-raw, and the Raman spectrum of Fe 2 O 3 had six different characteristic phonon modes that were categorized as two A 1g and four E g modes. The two modes of A 1g were located at 221 and 487 cm −1 , and the five E g modes were located at 289 cm −1 , 405 cm −1 , 605 cm −1 , and 651 cm −1 , respectively . And, these modes were consistent with Fe 2 O 3 Raman active modes, indicating that the hematite was present both in the MRM-700 and RM-raw. Moreover, compared with the RM-raw, the peaks of MRM-700 showed a slight shift toward low energy in Fig. 4a, signifying the other atoms in the RM-raw doped into Fe 2 O 3 (Idrees et al. 2022), which was consistent with the result of XRD testing. In addition, the peak of MRM-700 at 651 cm −1 was not observed due to the higher crystallinity compared to that of RM-raw (Fouad et al. 2019).
The element compositions of RM-raw and MRM-700 were studied via X-ray photoelectron spectroscopy (XPS). As shown in Fig. 4(b), Fe 2p, O 1s, Al 2p, Si 2p, and Ti 2p were observed in RM-raw and MRM-700, respectively. In addition, it could be observed that the intensity of the O peak was enhanced in MRM-700, compared with that in RM-raw. This was due to the fact that the metal oxides have been produced in MRM-700, which was consistent with the result of XRD.
The O 1s XPS spectra of RM-raw and MRM-700 are shown in Fig. 4(c). For RM-raw, the three peaks located at 533.16 eV, 531.36 eV, and 529.44 eV belonged to O-H, Fe-O, and lattice oxygen, respectively Guo et al. 2020b). However, different from the O 1s XPS spectrum of RM-raw, the peak of O-H disappeared in the highresolution spectrum of O 1s for MRM-700, indicating the O-H disappearance in MRM-700. The phenomenon mentioned above was also consistent with the modified method of calcination.
High-resolution XPS spectra of Fe 2p were conducted to investigate the surface valence states of Fe for RM-raw and MRM-700, respectively. As shown in Fig. 4(d), for RM-raw and MRM-700, the Fe 2p XPS spectrum could be deconvoluted into eight peaks, as reported by Ioannidi et al. (2020). The peak positions obtained are listed in Table S2 in Supporting Information. For the RM-raw (Wang et al. 2016;Xu et al. 2016;Moreira et al. 2017;Li et al. 2022), the Fe 2p 3/2 peaks of Fe 3+ appeared at 709.5 and 710.7 eV, and the Fe 2p 1/2 peaks of Fe 3+ appeared at 723.2 and 725.8 eV. In addition, the Fe 2p 1/2 peak of Fe 2+ appeared at 715.7 eV. Moreover, the satellite peaks appeared at 718.6 and 730.5 eV for Fe 3+ and 712.9 eV for Fe 2+ . It was confirmed that the Fe 2+ and Fe 3+ existed in the RM-raw. Furthermore, for RMraw and MRM-700 catalysts, the ratios of Fe 2+ /(Fe 2+ + Fe 3+ ) and Fe 3+ /(Fe 2+ + Fe 3+ ) could be determined by calculating the area ratios of Fe 2+ and Fe 3+ to the total area of Fe 2+ and Fe 3+ , respectively. According to Fig. 4(d), the proportion of Fe 3+ and Fe 2+ to the total Fe (i.e., Fe 2+ + Fe 3+ ) in the RMraw were about 85.5 and 14.5%, respectively. On the one hand, this result could be consistent with the result of XRD that Fe 3+ existed in the RM-raw. On the other hand, compared with the XRD, the low content of Fe 2+ can be detected in the XPS due to the high sensitivity of XPS. After calcination, as shown in Fig. 4(d), for MRM-700, the Fe 2p 3/2 peaks of Fe 3+ appeared at 708.9 and 710.1 eV, and the Fe 2p 1/2 peaks of Fe 3+ appeared at 722.6 and 724.5 eV. In addition, the Fe 2p 1/2 peak of Fe 2+ appeared at 715.6 eV. Moreover, the satellite peaks that appeared at 718.0 and 731.5 eV for Fe 3+ and 712.2 eV for Fe 2+ were also found in MRM-700. However, compared with the Fe 2p 3/2 peak of Fe 3+ of RMraw appeared at 709.5 and 710.7 eV and the Fe 2p 1/2 peak of Fe 3+ of RM-raw appeared at 723.2 and 725.8 eV, the Fe 2p 3/2 and Fe 2p 1/2 peaks of Fe 3+ of MRM-700 decreased to 708.9 eV, 710.1 eV, 722.6 eV, and 724.5 eV, respectively, which were attributed to the formation of the doping lattice of Fe 2 O 3 . Moreover, as shown in Fig. 4(d), the proportions of Fe 3+ and Fe 2+ to the total Fe in the MRM-700 were about 89.4 and 10.6%, respectively. Obviously, the proportion of Fe 3+ to the total Fe of MRM-700 (89.4%) has increased from the initial value of 85.5% of RM-raw, indicating that the content of Fe 3+ was increased after calcination and calcination could expose more Fe 3+ sites on the surface of the prepared catalyst. In summary, the amount of Fe 3+ sites was increased and the lattice of Fe 2 O 3 was doped after calcination, both of which would facilitate the reaction of activation NaHSO 3 (Shao et al. 2021;Yousefi et al. 2019).

Influence of reaction parameters
The Fig. 5(a) shows the effect of initial solution pH on the LOM degradation under the reaction conditions as follows: [NaHSO 3 ] = 0.5 g/L and [MRM-700] = 0.9 g/L. When the pH value was 5 or 3, more than 85% of LOM was removed only at about 30 min. The degradation reactions    5) and (6) (Dou et al. 2020;Mei et al. 2020). According to the reference (Dou et al. 2020), •SO 4 − lasted longer than •OH − in the reaction system, so more •SO 4 − could contact with more LOM in a given time period. For the above reason, •SO 4 − showed a higher degradation rate during the degrading of LOM than •OH − . Different from the acidic medium, when the pH of the reaction system increased to 7 and 9, the degradation ratio of LOM decreased (Fig. 5(a)). The decrease in degradation ratio could be ascribed to the fact that part of •SO 4 − could easily react with OH − in alkaline and even neutral conditions, as shown in Eq. (7). Though •OH − generated with the equalmolar loss of •SO 4 − via Eq. (7), the lower standard redox potential of •OH − (2.8 V) than that of •SO 4 − (3.0 V) might be the major cause of the decrease in degradation ratio at pH 7 and 9 (Sun et al. 2019). Moreover, •SO 4 − could react with the molecule of water via Eq. (8), in which both •OH − and H + are generated with the quenching of •SO 4 − . It should be noted that the generated H + from Eq. (8) could be neutralized by the OH − of the alkaline solution. In other words, the equation of Eq. (8) would be intensified in an alkaline medium than in an acidic and alkaline medium. For the reasons mentioned above, when the pH value was 9, the degradation ratio at the reaction time of 30 min was reduced to ca. 70%. However, at the reaction time of 30 min, compared with that at pH 3 or pH 5 in acidic media, the decrease of the obtained degradation ratio at pH 9 was a relatively low drop. Thus, the MRM-700/NaHSO 3 system had an excellent performance of the application in the wide pH range of 3 to 9.
(3) The effect of MRM-700 dosages on the degradation of LOM is shown in Fig. 5(b). As shown in Fig. 5(b), when the dosage of MRM-700 was 0.3 g/L, only 55% of LOM was degraded at 30 min. As the amount of catalyst dosage increased to 0.6 and 0.9 g/L, greater degradation efficiencies have been gained in the MRM-700/NaHSO 3 system. For example, at the catalyst dosages of 0.6 and 0.9 g/L, the degradation ratios of LOM at 30 min reached 67 and 89%, respectively. The catalyst of MRM-700 was the source of ≡Fe 3+ in the reaction system, and the ≡Fe 3+ was the main species that catalyzed NaHSO 3 to generate •SO 4 − or •OH − (Eqs. (3)- (7)). With the increase of catalyst dosages, enough ≡Fe 3+ species could be provided by the added MRM-700 catalyst, which would react with NaHSO 3 to generate a sufficient amount of •SO 4 − and •OH − via Eqs. (3)-(6) for the degradation of LOM (Wei et al. 2015). However, when the MRM-700 dosage was increased to 1.2 g/L, the degradation efficiency of the reaction system catalyzed by the added MRM-700 catalyst significantly declined, which was compared to the obtained degradation efficiency of the reaction system at the MRM-700 dosage of 0.9 g/L. For example, at the catalyst dosages of 1.2 g/L, the degradation ratio of LOM at 30 min declined to 73% from the value of 89% at the catalyst dosages of 0.9 g/L. With the excess catalyst, severe aggregation of MRM-700 composition in the reaction system could form, leading to the decrease of active sites of MRM-700 catalyst .
As the main source of •SO 4 − in the MRM-700/NaHSO 3 system, NaHSO 3 plays an important role in the degradation of LOM. It could be inferred that more •SO 4 − would be generated with a higher concentration of NaHSO 3 . In this work, the effect of NaHSO 3 on the degradation of LOM has been investigated. As shown in Fig. 5(c), when the NaHSO 3 concentration increased from 0.3 to 0.5 g/L, the degradation efficiency of the reaction system with the added NaHSO 3 significantly increased. At the reaction time of 30 min, the degradation ratios obtained under the conditions of [NaHSO 3 ] = 0.3 g/L and [NaHSO 3 ] = 0.5 g/L were about 16 and 87%, respectively. However, when the NaHSO 3 concentration further increased to 0.7 g/L, the degradation efficiency of the MRM-700/NaHSO 3 system showed a slight decrease. For example, at the reaction time of 30 min, the degradation ratio obtained under the conditions of [NaHSO 3 ] = 0.7 g/L was about 56%. With the NaHSO 3 concentration of 0.7 g/L, the decline of degradation efficiency of the MRM-700/NaHSO 3 system might be due to the fact that the •SO 4 − produced by excess NaHSO 3 was probably scavenged by the unfavorable consumption, as shown in Eqs. (9) and (10) (Ji et al. 2021;Das & Mohanty 2021).
According to the previous report (You et al. 2021;Guo et al. 2020b), the ions of NO 3 − , SO 4 2− , HCO 3 − , and Cl − are commonly presented in wastewater and even in natural water. As the common co-existing ions in water, NO 3 − , SO 4 2− , HCO 3 − , and Cl − can influence the degradation of LOM. Thus, it was necessary to examine the effect of co-existing ions on the degradation of LOM. The degradation efficiency of the MRM-700/NaHSO 3 system in the presence of different ions has been investigated. As shown in Fig. 5(d), compared with the degradation ratio of 90% obtained in the blank system without any added ions, the co-existing ions of NO 3 − , SO 4 2− , HCO 3 − , and Cl − in the reaction system, all of which was about 1 mmol/L, decreased the degradation ratio of LOM, and the degradation ratios obtained at 30 min for the ions of NO 3 − , SO 4 2− , HCO 3 − , and Cl − were about 76%, 77%, 4.5%, and 70%, respectively. It was found that the presence of NO 3 − , SO 4 2− , and Cl − exhibited a relatively weak negative effect on the degradation of LOM. The coexisting ions in the reaction system could react with •SO 4 − , thereby inhibiting the decomposition of LOM by the radical of •SO 4 − . However, as shown in Fig. 5(d), HCO 3 − played a dramatic negative effect on the degradation of LOM in the MRM-700/NaHSO 3 system. The intensive inhibitory effect from HCO 3 − was attributed to the fact that HCO 3 − as a radical scavenger had a strong ability to quench the radicals in advanced oxidation processes (Hajjaji et al. 2016).
To explore the importance of the synergistic reaction between NaHSO 3 and the modified red mud, the activation of bisulfite by RM-raw, MRM-700, and MFe 2 O 3 -700 has been carried out, respectively. Furthermore, to objectively evaluate the effect of adsorption by solid catalyst on the removal of LOM, both RM-raw and MRM-700 were used for the comparison experiments without adding NaHSO 3 . As shown in Fig. 6(c), the adsorption ratios of LOM at 30 min by RM-raw and MRM-700 were only 10 and 11%, respectively, indicating that the adsorption effect from the solid catalyst of modified red mud was not significant. Moreover, as shown in Fig. 6(c), without any solid catalyst, namely only in the presence of NaHSO 3 alone in the degradation system, the degradation ratio of LOM was almost zero. Thus, the refractory organic LOM was hardly degraded without the catalyst. However, when MRM-700 and NaHSO 3 existed at the same time, namely constituting the sulfite-based advanced oxidation process of the MRM-700/NaHSO 3 , the degradation of LOM could be easily realized and the degradation ratio quickly increased up to 89% at the reaction time of 30 min. Thus, MRM-700/NaHSO 3 system serving as a Fenton-like system had significant catalytic activity for the degradation of LOM (Mei et al. 2020). Additionally, it was important to study the role of Fe 2 O 3 in the reaction system. In our work, MFe 2 O 3 -700 as the comparative catalyst was selected to activate NaHSO 3 to remove LOM, which was the sulfite-based advanced oxidation process of MFe 2 O 3 -700/NaHSO 3 system. In the constituted MFe 2 O 3 -700/NaHSO 3 system, the dosage of MFe 2 O 3 -700 was 0.34 g/L, which was approximately equal to the iron content of RM in terms of Fe 2 O 3 used in the RM-raw/NaHSO 3 system. As displayed in Fig. 6(c), the degradation ratio of LOM at 30 min was about 0%, 27%, and 89% for MFe 2 O 3 -700/NaHSO 3 system, RM-raw/ NaHSO 3 system and MRM-700/NaHSO 3 system, respectively. The results showed that MFe 2 O 3 -700 cannot effectively activate NaHSO 3 to degrade LOM. Furthermore, RM-raw/NaHSO 3 system showed a higher degradation ratio of LOM (ca. 27%) than the MFe 2 O 3 -700/NaHSO 3 system (ca. 0%), indicating that some properties originating from RM contributed to the catalytic activity of the catalyst, such as the adsorption property and morphological structure. Moreover, compared with that obtained in the RM-raw/NaHSO 3 system (27%), the degradation ratio of LOM obtained in the MRM-700/NaHSO 3 system significantly increased to 89% at the time of 30 min. The results indicated that the increased catalytic capacity of MRM-700 did not depend simply on the content of Fe 2 O 3 in the prepared catalyst, and the reasons would be discussed in detail in the "Possible reaction mechanism" section.  Figure 7 provides a photo of mung bean sprouts incubated in the toxicity test. It could be seen that the average lengths of the mung bean sprouts incubated by deionized water, the initial LOM solution, and the treated LOM solution were around 21.1 cm, 13.2 cm, and 19.4 cm, respectively. The results indicated that after the treatment by the MRM-700/NaHSO 3 system, the treated LOM solution showed a decreased biological toxicity compared with the initial LOM solution. Moreover, compared with the mung bean sprouts incubated with deionized water, the mung bean sprouts incubated by the treated LOM solution only showed a slight reduction in the average length, which might be attributed to the residual toxicity in the treated solution. In conclusion, the toxicity of the LOM solution could be effectively reduced by the system of MRM-700/NaHSO 3 .

Stability and reusability of MRM-700
Reusability and stability are important factors to consider as part of catalyst performance. To evaluate the stability and reusability of MRM-700, MRM-700 was repeated three times. The result is shown in Fig. 8. The MRM-700 catalyst showed a high activity in the degradation of LOM, and even in the third cycle, the relatively high degradation ratio of about 70% was achieved easily at the reaction time of 30 min. It should be noted that the slight decrease in the degradation efficiency during the cycling test might be mainly due to two factors: (1) the degradation intermediates accumulated on the surface of the catalyst, thus blocking the active site of the catalyst; (2) the loss of catalyst, which happened with the removal of the supernatant. Moreover, the testing reaction took place under weakly acidic conditions, resulting in the partial leaching of iron ions. The iron concentration in the reaction solution was also measured after each reaction cycle to evaluate the stability of the MRM-700. The leaching amount of iron from the MRM-700 during the cycling experiment is shown in Table 3. The iron leaching concentrations were all lower than 2 mg/L of the European Union discharge directive . Additionally, the mass percentage of Fe 2 O 3 in the RM was    Table 1). And based on the cycling experiment, the mean value of the leaching amount of total Fe ions was about 0.334 mg/L), which was calculated based on the result of Table 3. For each cycle in use, the mass percentage of Fe leached from MRM-700 was found to be ~ 0.14 wt% of the total Fe in the catalyst. Moreover, with such a little leaching amount of Fe ions, the influence of the leaching iron ions on the LOM degradation could be negligible (Xu et al. 2016). In other words, the degradation of LOM in the MRM-700/NaHSO 3 system mainly came from the heterogeneous catalysis under the action of the MRM-700 catalyst. The result indicated that MRM-700 was the safe and stable catalyst material when it was used in the S-AOP system for wastewater treatment.

Possible reaction mechanism
To investigate the reactive radical species in the S-AOP system, radical quenching experiments were carried out by adding two different scavengers into the degradation system. In the work, the scavengers of isopropanol (IPA) and methanol (MeOH) were applied to quench •OH − , and •SO 4 − , respectively. As presented in Fig. 9, with the addition of IPA and the addition of MeOH, both of which were 10 mmol/L in the reaction system, the degradation ratio of LOM decreased to 84 and 18%, respectively. As can be seen that both •OH − and •SO 4 − were involved in the MRM-700/NaHSO 3 system. The degradation of LOM was significantly inhibited by the addition of MeOH, suggesting that •SO 4 − was the main reactive species involved in the reaction of the MRM-700/NaHSO 3 system. The intermediates of LOM were investigated by UPLC-MS testing to analyze the degradation pathway of LOM. The possible structure and mass-to-charge ratio (m/z) of transformation products during the reaction process are listed in Table S3. The degradation intermediates of LOM of UPLC-MS chromatograms are shown in Fig. S2 in Supporting Information. Moreover, to further clarify the degradation pathway of LOM, DFT calculation was used to explain the location of reactive sites. According to the references Zhang et al. 2020a, b;Fu et al. 2021), the optimized LOM geometry (Fig. 10(a)), the highest occupied molecular orbital (HOMO), and lowest unoccupied molecular orbital (LUMO) (Fig. 10(b)) and Fukui index (Fig. 10(c)) were obtained and analyzed. HOMO and LUMO were often used to describe the ability of pollutants to be oxidized and reduced, respectively. And, Fukui indexes of f − , f + , and f 0 have described the site of the electrophilic attack, nucleophilic attack, and radical attack on the LOM molecule, respectively. Manifestly, the atom 18 N exhibited the highest f − , and f 0 was also relatively high, indicating that 18 N was more susceptible to be attacking. Furthermore, the 15O (f + = 0.05045, f 0 = 0.040085) and 11O (f − = 0.08043, f + = 0.08004, f 0 = 0.080235) with a high value of Fukui index were also easily attacked by •SO 4 − and •OH − . Even though the Fukui index values of 1C (f + = 0.05143, f 0 = 0.03812), 6C (f − = 0.054, f 0 = 0.03064), and 8C (f + = 0.10114, Fig. 11 Possible degradation pathway of LOM in MRM-700/ NaHSO 3 system f 0 = 0.059935) were high, it was difficult to be attacking for these atoms due to the saturated state and steric hindrance [11] ). In addition, the atoms (20C, 21C, 25C, and 9C) with a small Fukui index might be attacked by free radicals.
According to DFT calculation, LC-MS results, and relevant studies (You et al. 2021;Guo et al. 2020b;Ma et al. 2020;Zhang et al. 2020a, b;Li et al. 2012), the degradation pathway of LOM in MRM-700/NaHSO 3 system is proposed and showed in Fig. 11. In the system of MRM-700/NaHSO 3 , the molecules of LOM were attacked by •SO 4 − and •OH − , then the intermediate P1 was formed due to the reaction of defluorination of LOM. And for the molecules of LOM, after the ringopening, defluorination, and hydroxyl substitution, the intermediate P2 was formed. According to the Fukui index, the N18 and N22 of molecules of LOM were easy to be attacked by free radicals. Hence, the molecules of LOM also could be transformed into the intermediate P3 and the intermediate P6 after the cleavage of the C-N bond. Furthermore, the intermediate P4 could be generated from the molecule of LOM through the defluorination and hydroxy substitution of LOM. In addition, after the attack by free radicals on the C8, C9, N18, and N22, the molecules of LOM could be converted into the intermediate P5 due to the cleavages of the carbon-carbon bond and the C-N bond. In addition, both OH − and SO 4 − could react with -NH 2 groups to generate the N-centered radicals. Thus, the molecule of LOM could be converted into the intermediate P7 by the coupling of N-center free radicals. As listed in Part A of Fig. 11, the intermediates mentioned above with the large molecular structure have been confirmed via the UPLC-MS (Fig. S2). The above-mentioned intermediates could be further degraded to the intermediate products with a simple molecular structure, including the intermediate P8, intermediate P9, and intermediate P10 (as shown in Part B), all of which were also confirmed by the UPLC-MS (Fig. S2). Finally, with the continued activation of NaHSO 3 by MRM-700, the above intermediates with simple molecular structures could be eventually mineralized into CO 2 , H 2 O, F + , etc.
To understand the mechanism of LOM degradation by MRM-700, the XPS spectra of Fe in fresh and used MRM-700 were compared. According to the result of XPS in Sect. 3.1.4, it could be seen that Fe 3+ and Fe 2+ were presented in fresh MRM-700, and the proportions of Fe 3+ and Fe 2+ to the total Fe in the fresh MRM-700 were 89.4 and 10.6%, respectively. For the used MRM-700, the Fe 2p 3/2 peaks of Fe 3+ appeared at 709.4 and 710.7 eV, and the Fe 2p 1/2 peaks of Fe 3+ appeared at 723.0 and 724.6 eV, as shown in Fig. 12. In addition, the Fe 2p 1/2 peak of Fe 2+ of the used MRM-700 appeared at 714.0 eV. Furthermore, the satellite peaks appeared at 718.0 and 732.0 eV for Fe 3+ and 712.0 eV for Fe 2+ in the XPS spectra of the used MRM-700. The aforementioned results implied that both Fe 3+ and Fe 2+ existed in the used MRM-700 (You et al. 2021). It should be noted that for the used MRM-700, the proportions of Fe 3+ and Fe 2+ to the total Fe were 82.7 and 17.3%, respectively. Compared with the fresh MRM-700, the change of the proportions of Fe 3+ and Fe 2+ to the total Fe after using indicated that the Fe 3+ /Fe 2+ redox couple was involved in the NaHSO 3 activation that was catalyzed by MRM-700 (Wei et al. 2015). Moreover, according to the results of the "Catalytic degradation of LOM by MRM-T" section, the MRM-700 catalyst held a significant increase in the catalytic ability over the RM-raw. There were two reasons for the improvement of catalytic capacity. On the one hand, the method of calcination could provide a looser structure, more surface area, and an active site (Fe 2 O 3 ) for MRM-700 than RM-raw, as the characterized results in the "Characterization" Possible mechanism of degradation LOM in MRM-700/ NaHSO 3 system section. The improvement in the physical and chemical properties of the prepared catalyst by calcination was conducive to the increase of catalytic ability. On the other hand, RM, as industrial solid waste, contained a variety of metal and non-metal ions except for Fe 2 O 3 . According to the results of the XRD and XPS, the heteroatom doping was shown after calcination. The heteroatom doping improved the electronic structure of Fe 2 O 3 and boosted the Fe 2+ /Fe 3+ cycle (Eqs. ( 3) and (6)). Thus, the activity of MRM-700 was increased . Finally, the proposed degradation mechanism of LOM by MRM-700/ NaHSO 3 system is shown in Fig. 13. HSO 3 − was accumulated on the surface of MRM-700, and then the reactions of Eqs.

Conclusion
In this study, the MRM-700 was successfully obtained by the calcination method and then employed as a catalyst to activate NaHSO 3 for degradation LOM. Approximately 89% of LOM (10 mg/L) was easily degraded at 30 min under the following reaction conditions: initial pH of 5 (unadjusted), [NaHSO 3 ] = 0.5 g/L, and [MRM-700] = 0.9 g/L. The MRM-700/NaHSO 3 system had an excellent performance of the application in the wide pH range of 3 to 9, and NO 3 − , SO 4 2− , and Cl − in the MRM-700/NaHSO 3 system did not weaken the catalytic activity of MRM-700. In addition, the underlying mechanism for the activating of NaHSO 3 by MRM-700 was studied through the quenching experiment. It was found that •SO 4 − and •OH − were the main reactive radicals for the degradation of LOM in the reaction system. And, based on the results of UPLC-MS and DFT calculation, the possible LOM degradation pathway was proposed. Then, the MRM-700 exhibited excellent catalytic activity for NaHSO 3 activation, which could be attributed to excellent pore structure, high specific area, abundant Fe 3+ sites, and the doping lattice of Fe 2 O 3 . Especially, doping improved the electronic structure of Fe 2 O 3 and boosted the Fe 2+ /Fe 3+ cycle. This study not only provided a route toward the reuse of RM by a simple calcination method without the addition of other chemicals but also developed a cost-effective, stable, and efficient catalyst for Fenton-like catalytic applications.
Funding Financial supports from the Guangxi Science Foundation Funded Project (2021GXNSFAA075006), the Open Project of Guangxi Key Laboratory of Bio-refinery (GXKLB20-01), and the Innovation Project of Guangxi Graduate Education (202110593194) are gratefully acknowledged.
G u a n g x i S c i e n c e F o u n d a t i o n F u n d e d Project,2021GXNSFAA075006,Linye Zhang,Innovation Project of Guangxi Graduate Education,YCSW2022051,Linye Zhang,Open Project of Guangxi Key Laboratory of Bio-refinery,GXKLB20-01,Linye Zhang,Undergraduates' Innovation and Entrepreneurship Program in Guangxi, 202110593134, Linye Zhang Data availability All data generated or analyzed during this study are included in this published article.