Activation of Bisulfite with Pyrophosphate-Complexed Mn(III) for Fast Oxidation of Organic Pollutants

Aqueous complexes of Mn(III) ion with ligands exist in various aquatic systems and many stages of water treatment works, while HSO3− is a common reductant in water treatment. This study discloses that their encounter results in a process that oxidizes organic contaminants rapidly. Pyrophosphate (PP, a nonredox active ligand) was used to prepare the Mn(III) solution. An approximate 71% removal of carbamazepine (CBZ) was achieved by the Mn(III)/HSO3− process at pH 7.0 within 20 s, while negligible CBZ was degraded by Mn(III) or HSO3− alone. The reactive species responsible for pollutant abatement in the Mn(III)/HSO3− process were SO4•− and HO•. The treatment efficiency of the Mn(III)/HSO3− process is highly related to the dosage of HSO3− because HSO3− acted as both the radical scavenger and precursor. The reaction of Mn(III) with HSO3− follows second-order reaction kinetics and the second-order rate constants ranged from 7.5 × 103 to 17 M−1 s−1 under the reaction conditions of this study, suggesting that the Mn(III)/HSO3− process is an effective process for producing SO4•−. The pH and PP:Mn(III) ratio affect the reactivity of Mn(III) towards HSO3−. The water background constituents, such as Cl− and dissolved organic matter, induce considerable loss of the treatment efficiency in different ways.


Introduction
Under the stress of population expansion and increasing urbanization, water scarcity strongly increases [1,2]. The situation is becoming worse due to the water quality deterioration. To address this issue, various strategies were proposed, such as recycling of waste to reduce the pollution of environment [3][4][5]. However, many micropollutants are not disposed reasonably and are inevitably discharged into water. These micropollutants, including pharmaceuticals and personal care products (PPCPs), pesticides, biocides, and many others, lead to human health concern [6,7]. Degrading/removing micropollutants has become one of the most important tasks in supplying safe drinking water. However, some of them resist conventional treatment processes, including coagulation, sedimentation, filtration, chlorination, and biological treatment [8]. For these recalcitrant micropollutants, their oxidative degradation is often achieved by the advanced oxidation processes, which produce highly reactive species such as HO • and SO 4 •− through activating the radical precursor with different strategies [9][10][11][12].
Over the past decades, SO 4 •− -based AOPs have drawn significant attention as a viable alternative to traditional HO • -based AOPs in water and wastewater treatment [10] due to their various virtues, including (i) the higher redox potential (E 0 SO •− 4 /SO 2− 4 = +2.60-+ 3.10 V NHE > E 0 HO • /HO − = +1.90-+ 2.70 V NHE [13], (ii) lower costs of storage and transportation of persulfate, the common SO 4 •− precursor, than H 2 O 2 , the common HO • precursor, (iii) the Among the proposed transition metal or metal oxides, Mn(IV) is the thermodynamically favored oxidation state in surface waters and is nearly ubiquitous, and thus gains considerable attention [35]. Mn(IV) was commonly considered to exist in the form of MnO 2 . However, it was found that treatment efficiency of the MnO 2 /HSO 3 − process is highly dependent on the morphology of the employed MnO 2 . Sun et al. [29] reported that the colloidal MnO 2 activated HSO 3 − process led to the removal of pollutants at the timescale of tens of seconds. Differently, Wang et al. [31] found that amorphous MnO 2 activated HSO 3 − process resulted in the half-time of pollutant degradation at several hours. It was demonstrated that the reactivity of MnO 2 towards contaminants is highly related to the content of Mn(III), and Mn(III) was considered to possess higher redox activity than MnO 2 [36][37][38]. However, the influence of Mn(III) on activating HSO 3 − is unknown. In addition to being absorbed by MnO 2 particles, Mn(III) stabilized by ligands (free Mn(III) ion is prone to disproportionation) is widespread in natural waters and water treatment processes. Trouwborst et al. [39] found that soluble Mn(III) could be stabilized by natural ligands and its concentration was as high as 5 µM, which constituted up to 100% of the total dissolved Mn pool in the Black Sea. Madison et al. [40] disclosed that soluble Mn(III) is primarily produced via oxidation of Mn(II) diffusing upwards from anoxic sediments in the Laurentian Trough (Quebec, Canada). Mn(III) was also reported to be produced through the interaction of influent Mn oxide solids with natural organic matter (NOM) in the clarifier sludge of a water treatment plant in England [41]. Thus, the added HSO 3 − seems very likely to encounter with Mn(III), and single electron abstraction transfer from HSO 3 − to Mn(III) was expected, leading to the following formation of SO 4 •− (Equations (1)- (3)). However, the performance of the Mn(III)/HSO 3 − process for pollutant abatement was unclear, impeding not only the understanding of the already-constructed MnO 2 /HSO 3 − process but also the design and development of a new Mn(III)/HSO 3 − advanced oxidation process. To address this, efforts were needed to evaluate the potential of the Mn(III)/HSO 3 − process in pollutant abatement, which is the aim of this study. Firstly, the efficiency of pollutant removal by this process as the function of reaction conditions (e.g., pH, dosage of Mn(III), HSO 3 − , dissolved oxygen, and ligand:Mn(III) ratio) was investigated. Then, the kinetics of Mn(III) reacting with HSO 3 − were analyzed. Further, the reactive species responsible for pollutant degradation in the Mn(III)/HSO 3 − process were differentiated. Finally, the influence of common background water constituents (e.g., Cl − and NOM) on pollutant degradation by the Mn(III)/HSO 3 − process was elaborated. Carbamazepine (CBZ) is one of the pharmaceuticals most frequently detected in the aqueous environment and was selected as the probe contaminant in this study [42].

Chemicals and Materials
Nitrobenzene (NB), pyrophosphate (PP), manganese sulfate (MnSO 4 ), potassium permanganate (KMnO 4 ), methanol (MeOH), and tert-butanol (TBA) were purchased from Sinopharm Chemical Reagent Co. Ltd., (Beijing, China). Sodium thiosulfate pentahydrate (Na 2 S 2 O 3 •5H 2 O), sodium bisulfate (NaHSO 3 , S(IV)), and sodium chloride (NaCl) were received from Macklin Biochemical Co., Ltd., Shanghai, China. Bisphenol A (BPA), CBZ, and 5,5-dithiobis(2-nitrobenzoic acid) (DTNB) were obtained from Aladdin Industrial Co., Ltd., Shanghai, China. All the above chemicals are of analytical grade and can be used directly without further purification. Humic acid (industry pure) was obtained from Sigma-Aldrich. The concentration of humic acid stock solution was determined with total carbon analyzer and used to simulate dissolved organic matter (DOM) in water. All of the experimental solutions in this study were prepared by dissolving chemicals in ultrapure water (18.2 MΩ-cm) produced using a ULUPURE water purification system, and stored at 4 • C in the dark. The stock solutions of NaHSO 3 (250 mM) were freshly prepared every day to avoid oxidation by oxygen. The species of S(IV) depend on pH (pK a,HSO -3 = 7.2), and HSO 3 − was used to represent S(IV) in this study. Free Mn(III) ion is prone to disproportionation to yield MnO 2 and Mn(II) under environmentally relevant conditions. Most of the past studies about Mn(III) were conducted with insoluble Mn(III)-rich Mn oxides or in the presence of excess ligands. In this study, PP, a nonredox active ligand, was employed to stabilize Mn(III), and the stability constants were reported to be at least 10 6 . The stable Mn(III) stock solution was prepared with the following procedure: a mixture of 10 mL of KMnO 4 (50 mM), 1.25 mL of Na 2 S 2 O 3 (0.2 M), and 100 mL of PP (0.25 M), diluted to 1 L with ultrapure water with rapid mixing. Then, a pale red solution was obtained and remained stable for several days. The reaction that took place under these conditions can be described by Equation (4):

Experimental Procedures
Batch experiments: The experiments on pollutant oxidation by the Mn(III)/HSO 3 − process were conducted in 0.2 L glass bottles at 20 ± 1.0 • C. A piece of glass was used as the cover to prohibit the volatilization of organics. Reactions were initiated by quickly spiking Mn(III) into solutions containing target compounds, HSO 3 − , and the constituents(s) of interest. After reaction of 20 s, 1 mL of sample was rapidly transferred into a 2 mL sample vial containing 10 µL of Na 2 S 2 O 3 stock solution. For the experiments where the effect of oxygen was to be examined, the reaction solution was sparged with N 2 for at least 30 min before adding Mn(III) stock solution. All experiments were run in duplicates or triplicate, and the data were averaged.
Stopped-flow kinetic experiments: The reduction kinetics of Mn(III) by HSO 3 − under different conditions were conducted with a stopped-flow spectrophotometer (SFS, Model SX 20, Applied Photophysics Ltd., Leatherhead, UK). Two working solutions were prepared before the experiments. One working solution contained 100 µM Mn(III) and 5 mM PP, and the other contained 500 µM HSO 3 − and substrates of interest. Reactions were initiated by simultaneously injecting an equal volume of two working solutions. 258 nm was used to detect the decay of Mn(III) (ε Mn(III)−PP at 258 nm = 6750 M −1 s −1 ) [28].

Analytical Methods
The concentrations of organic pollutants were analyzed by a high-performance liquid chromatograph (Thermo Scientific U3000 series) equipped with a UV detector and a C18 column (4.6 × 250 mm, 5 µm particle size, Thermo Scientific, Waltham, MA, USA) at 35 • C. The injection volume was 10 µL and the flow rate was 1.0 mL/min with a mobile phase of acetonitrile-formic acid aqueous solution at pH 3.5. The concentration of HSO 3 − was monitored using the modified 5.5 -dithiobis(2-nitrobenzoic acid) colorimetric method [43]. The concentration of dissolved oxygen (DO) was measured online with a JPB-607A portable DO meter (Leici, Shanghai, China). Chlorine and chlorate were detected by ion chromatograph (DIONEX AQUION RFIC). The concentrations of organic pollutants were analyzed by a high-performance liquid chromatograph (Thermo Scientific U3000 series) equipped with a UV detector and a C18 column (4.6 × 250 mm, 5 µm particle size, Thermo Scientific, Waltham, MA, USA) at 35 °C. The injection volume was 10 µL and the flow rate was 1.0 mL/min with a mobile phase of acetonitrile-formic acid aqueous solution at pH 3.5. The concentration of HSO3 − was monitored using the modified 5.5′-dithiobis(2-nitrobenzoic acid) colorimetric method [43]. The concentration of dissolved oxygen (DO) was measured online with a JPB-607A portable DO meter (Leici, Shanghai, China). Chlorine and chlorate were detected by ion chromatograph (DIONEX AQUION RFIC). Figure 1 shows the influence of HSO3 − on CBZ removal by Mn(III) at pH 7.0. Negligible CBZ was degraded by Mn(III) alone ( Figure 1) and HSO3 − alone (not shown). Once HSO3 − coexisted with Mn(III), CBZ was degraded rapidly, suggesting the generation of reactive species in the Mn(III)/HSO3 − process. The removal of CBZ by the Mn(III)/HSO3 − process is highly related to the dosage of HSO3 − . The removal of CBZ increased followed by decrease with increasing HSO3 − dosage from 20 to 2000 µM, and the maximal removal of 71% was obtained at HSO3 − dosage of 250 µM. According to the valence states of manganese, the only product of Mn(III) reduction by HSO3 − was Mn(II), which is inert to CBZ. The reactive species in the Mn(III)/HSO3 − process are possibly derived from HSO3 − evolution. Based on the reported mechanisms of HSO3 − oxidation by dissolved oxygen in the presence of transition metals ( Figure S1), various radicals (e.g., SO3 •− , SO4 •− , SO5 •− , HO • ) are expected to be involved in the Mn(III)/HSO3 − process. SO3 •− and SO5 •− are very weak, with redox potential of 0.63 and 1.1 V vs. NHE [44], and CBZ is recalcitrant to these two radicals [45]. The remaining candidates of reactive species that accounted for the CBZ oxidation by the Mn(III)/HSO3 − process are SO4 •− and HO • . Quenching experiments were conducted to verify the presence of these two radicals. MeOH and TBA were employed as the scavengers of radicals because TBA is only reactive to HO • (k HO • -TBA = 3.8 -7.6 × 10 8 M −1 s −1 ) [14], while MeOH is an effective quenching agent for both HO • and SO4 •− (k HO • -MeOH ) = 8.0 − 10 × 10 8

Pollutant Removal by the Mn(III)-Activated HSO3 − Process
. As shown in Figure S2, the removal of CBZ was significantly depressed by MeOH and TBA, indicating that HO • was the main radical responsible for CBZ degradation. This is contradictory to the prevailing viewpoint that SO4 •− serves as the main oxidant in the sulfite-based AOPs under acidic condition [46]. Figure S1 shows that the oxidation of HSO3 − in the presence of dissolved oxygen is a chain reaction. Wang et al. [31] found that alcohols could react with the involved radicals and interrupt the chain reaction, resulting in the decreased formation of HO • and SO4 •− . Thus, the quenching experiments might fail to achieve radical differentiation. Relative rate method was applied to analyze the role of HO • and SO4 •− in pollutant abatement. In this  [44], and CBZ is recalcitrant to these two radicals [45]. The remaining candidates of reactive species that accounted for the CBZ oxidation by the Mn(III)/HSO 3 − process are SO 4 •− and HO • . Quenching experiments were conducted to verify the presence of these two radicals. MeOH and TBA were employed as the scavengers of radicals because TBA is only reactive to HO • (k HO • −TBA = 3.8 -7.6 × 10 8 M −1 s −1 ) [14], while MeOH is an effective quenching agent for both HO • and SO 4 . As shown in Figure S2, the removal of CBZ was significantly depressed by MeOH and TBA, indicating that HO • was the main radical responsible for CBZ degradation. This is contradictory to the prevailing viewpoint that SO 4 •− serves as the main oxidant in the sulfite-based AOPs under acidic condition [46]. Figure S1 shows that the oxidation of HSO 3 − in the presence of dissolved oxygen is a chain reaction. Wang et al. [31] found that alcohols could react with the involved radicals and interrupt the chain reaction, resulting in the decreased formation of HO • and SO 4 •− . Thus, the quenching experiments might fail to achieve radical differentiation. Relative rate method was applied to analyze the role of HO • and SO 4 •− in pollutant abatement. In this method, a low concentration of probe compounds (CBZ, NB) was added, which was thought not to interrupt the chain reaction of HSO 3 − oxidation. The degradation kinetics of these two compounds can be described with Equations (5) and (6).
dt are defined as the time-integrated concentration of HO • and SO 4 •− , respectively. k HO • −NB , k HO • −CBZ , and k SO •− 4 −CBZ are the second-order rate constants when the probe compound is reacting with the corresponding reactive species. With the experimentally determined removal of probe compounds (Figures 1 and S3), along with literature-reported second-order rate constants of the radicals towards CBZ and NB (Table S1), the contribution of HO • and SO 4 •− to the degradation of CBZ was calculated. It should be noted that negligible NB was removed by Mn(III) alone ( Figure S3), indicating that NB was recalcitrant to Mn(III). As shown in Figure 1, both HO • and SO 4 •− contributed to CBZ degradation, and SO 4 •− played the major role [46]. Figure S4 shows that relative contribution of HO • and SO 4 •− to CBZ degradation are stable and independent of the dosage of HSO 3 − . The increase of CBZ removal with increasing HSO 3 − dosage from 0 to 250 µM should be attributed to the increased formation rate of radicals. However, besides serving as the radical precursor, HSO 3 − also plays the role of radical scavenger. As shown in Figure S1, both HO • and SO 4 •− can be reduced by HSO 3 − with the second-order rate constants of 2.7 × 10 9 and 3.1 × 10 8 M −1 s −1 , respectively. Consequently, the removal of CBZ decreased with further increase of HSO 3 − dosage from 250 to 2000 µM. Figure S5 shows that CBZ removal slightly increased with elevating Mn(III) dosage from 25 to 100 µM, which could be ascribed to the increased chain initiation reaction rate (i.e., the reaction of Mn(III) + HSO 3 − → Mn(II) + SO 3 •− ). Figure 2A shows a typical time course of Mn(III) reduction by HSO 3 − at pH 7.0, where the initial concentrations of Mn(III) and HSO 3 − were 50 and 2000 µM, respectively. The concentration of HSO 3 − was assumed to be constant within 6 s under the reaction condition of this study. The loss of Mn(III) followed the pseudo-first-order kinetics, suggesting that the reaction is first-order with respect to Mn(III). For the constant initial concentration of Mn(III) (50 µM), the pseudofirst-order rate constant (k obs , s −1 ) varied linearly with HSO 3 − concentration ( Figure 2B), demonstrating a first-order dependence on HSO 3 − concentration. Measured k obs values for the experiments with different dosages of HSO 3 − are shown in Figure S6. The reaction kinetics of HSO 3 − with the PP-complexed Mn(III) can be described as where k is the apparent second-order rate constant, which is determined to be 295 M −1 s −1 at pH 7.0. The fast reaction of Mn(III) with HSO 3 − resulted in the considerable generation of radicals over a short period of time, accounting for the rapid degradation of CBZ under the reaction condition of Figure 1. Meanwhile, HSO 3 − was rapidly exhausted in the presence of excessive dissolved oxygen. Though the sampling time was set to 20 s, the reaction was expected to have ended earlier.
To diminish the scavenging effect of HSO 3 − and extend its lifetime, a multiple-dosing mode of HSO 3 − was adopted. As shown in Figure 3, the degradation efficiency of CBZ achieved 81% when 125 µM of HSO 3 − was added at 20 and 40 s, respectively,~10% higher than that with a single dosing of 250 µM of HSO 3 − . To diminish the scavenging effect of HSO3 − and extend its lifetime, a multiple-dosing mode of HSO3 − was adopted. As shown in Figure 3, the degradation efficiency of CBZ achieved 81% when 125 µM of HSO3 − was added at 20 and 40 s, respectively, ~10% higher than that with a single dosing of 250 µM of HSO3 − .

Influence of Pyrophosphate:Mn(III) Ratio of on Pollutant Abatement by Mn(III)/HSO3 − Process
Klewicki and Morgan [47] found that PP lends a kinetic stabilization to Mn(III), which is the function of PP:Mn(III) ratios. Jiang et al. [48] disclosed that addition of PP into the MnO4 − solution enhanced the oxidation of BPA, which was attributed to the contribution of PP-stabilized Mn(III) formed in situ upon MnO4 − reduction. However, the enhancement decreased with increasing the molar ratio of Mn(III):PP from 1:10 to 1:50 at pH 6.0, indicating the decreased reactivity of Mn(III)-PP with elevating PP concentration. The influence of the concentration of PP on CBZ removal by the Mn(III)/HSO3 − process was investigated and the results are shown in Figure 4A. Though increase of PP concentration benefits the stability of Mn(III), it negatively influences CBZ removal. Figure 4B shows the reduction kinetics of PP-stabilized Mn(III) by HSO3 − as the function of PP concentration. Assuming that the concentration of HSO3 − remained constant during the reaction, the second-order rate constants of Mn(III) reacting with HSO3 − in the presence of 0.2, 0.5, 1.0, and 2.5 mM of PP were calculated to be 3335, 1490, 745, and 243 M −1 s −1 , respectively, confirming the negative effect of PP on Mn(III) reactivity. Thus, a lower dosage of PP benefits the reactivity of Mn(III) and is thus in favor of CBZ degradation. Accordingly, the role of Mn(III) in natural water should be related to the species and concentrations of ligands. It should be noted that too-low concentrations of PP might lead to the disproportionation of Mn(III) spontaneously in a short time. Thus, 2.5 mM of PP was applied to prepare the stock solution of Mn(III) and conduct the following experiments.  To diminish the scavenging effect of HSO3 − and extend its lifetime, a multiple-dosing mode of HSO3 − was adopted. As shown in Figure 3, the degradation efficiency of CBZ achieved 81% when 125 µM of HSO3 − was added at 20 and 40 s, respectively, ~10% higher than that with a single dosing of 250 µM of HSO3 − .

Influence of Pyrophosphate:Mn(III) Ratio of on Pollutant Abatement by Mn(III)/HSO3 − Process
Klewicki and Morgan [47] found that PP lends a kinetic stabilization to Mn(III), which is the function of PP:Mn(III) ratios. Jiang et al. [48] disclosed that addition of PP into the MnO4 − solution enhanced the oxidation of BPA, which was attributed to the contribution of PP-stabilized Mn(III) formed in situ upon MnO4 − reduction. However, the enhancement decreased with increasing the molar ratio of Mn(III):PP from 1:10 to 1:50 at pH 6.0, indicating the decreased reactivity of Mn(III)-PP with elevating PP concentration. The influence of the concentration of PP on CBZ removal by the Mn(III)/HSO3 − process was investigated and the results are shown in Figure 4A. Though increase of PP concentration benefits the stability of Mn(III), it negatively influences CBZ removal. Figure 4B shows the reduction kinetics of PP-stabilized Mn(III) by HSO3 − as the function of PP concentration. Assuming that the concentration of HSO3 − remained constant during the reaction, the second-order rate constants of Mn(III) reacting with HSO3 − in the presence of 0.2, 0.5, 1.0, and 2.5 mM of PP were calculated to be 3335, 1490, 745, and 243 M −1 s −1 , respectively, confirming the negative effect of PP on Mn(III) reactivity. Thus, a lower dosage of PP benefits the reactivity of Mn(III) and is thus in favor of CBZ degradation. Accordingly, the role of Mn(III) in natural water should be related to the species and concentrations of ligands. It should be noted that too-low concentrations of PP might lead to the disproportionation of Mn(III) spontaneously in a short time. Thus, 2.5 mM of PP was applied to prepare the stock solution of Mn(III) and conduct the following experiments. Klewicki and Morgan [47] found that PP lends a kinetic stabilization to Mn(III), which is the function of PP:Mn(III) ratios. Jiang et al. [48] disclosed that addition of PP into the MnO 4 − solution enhanced the oxidation of BPA, which was attributed to the contribution of PP-stabilized Mn(III) formed in situ upon MnO 4 − reduction. However, the enhancement decreased with increasing the molar ratio of Mn(III):PP from 1:10 to 1:50 at pH 6.0, indicating the decreased reactivity of Mn(III)-PP with elevating PP concentration. The influence of the concentration of PP on CBZ removal by the Mn(III)/HSO 3 − process was investigated and the results are shown in Figure 4A. Though increase of PP concentration benefits the stability of Mn(III), it negatively influences CBZ removal. Figure Figure S1 shows that the combination of oxygen with SO3 •− led to the formation of SO5 •− which is a critical step for the following generation of HO • and SO4 •− . Hence, DO is indispensable for pollutant abatement in the Mn(III)/HSO3 − process. As shown in Figure  5  The influence of pH on the degradation of CBZ in the Mn(III)/HSO3 − process was investigated, and the results are shown in Figure 6. Over 95% of CBZ was removed at pH 5.0, while the removal of ciprofloxacin dropped progressively from 81% to 52% as pH increased from 6.0 to 9.0. The reduction kinetics of Mn(III) by HSO3 − under the pH range of 5.0-9.0 were displayed in Figure S7. The second-order rate constants of Mn(III) reacting with HSO3 − at pH 5.0-9.0 were calculated and are listed in Table S2. The reaction rate of Mn(III) with HSO3 − is sensitive to pH and decreased monotonously with increasing pH. The pKa of HSO3 − is 7.2, and the species of HSO3 − shifted to SO3 2− with increasing pH from 5.0 to 9.0. The electron density on the SO3 2− is much higher than that on the HSO3 − and the oxidation of SO3 2− is expected to be easier than that of HSO3 − . Thus, the declined reaction rate of Mn(III) with HSO3 − with increasing pH should be mainly ascribed to the decreased   Figure S1 shows that the combination of oxygen with SO 3 •− led to the formation of SO 5 •− which is a critical step for the following generation of HO • and SO 4 •− . Hence, DO is indispensable for pollutant abatement in the Mn(III)/HSO 3 − process. As shown in Figure 5, decrease of DO concentration from 8.0 to 0.9 mg/L depressed the removal efficiency of CBZ by 42%. According to the stoichiometric ratio of 1:1 in the reaction O 2 with SO 3

Influence of Dissolved Oxygen and pH on Pollutant Degradation by Mn(III)/HSO 3 − Process
•− ,~28 µM of SO 5 •− was expected to be formed in the presence of 0.9 mg/L of DO, much lower than that in the presence of 8.0 mg/L of DO. Though the transformation efficiency from SO 5   Figure S1 shows that the combination of oxygen with SO3 •− led to the formation of SO5 •− which is a critical step for the following generation of HO • and SO4 •− . Hence, DO is indispensable for pollutant abatement in the Mn(III)/HSO3 − process. As shown in Figure  5  The influence of pH on the degradation of CBZ in the Mn(III)/HSO3 − process was investigated, and the results are shown in Figure 6. Over 95% of CBZ was removed at pH 5.0, while the removal of ciprofloxacin dropped progressively from 81% to 52% as pH increased from 6.0 to 9.0. The reduction kinetics of Mn(III) by HSO3 − under the pH range of 5.0-9.0 were displayed in Figure S7. The second-order rate constants of Mn(III) reacting with HSO3 − at pH 5.0-9.0 were calculated and are listed in Table S2. The reaction rate of Mn(III) with HSO3 − is sensitive to pH and decreased monotonously with increasing pH. The pKa of HSO3 − is 7.2, and the species of HSO3 − shifted to SO3 2− with increasing pH from 5.0 to 9.0. The electron density on the SO3 2− is much higher than that on the HSO3 − and the oxidation of SO3 2− is expected to be easier than that of HSO3 − . Thus, the declined reaction rate of Mn(III) with HSO3 − with increasing pH should be mainly ascribed to the decreased The influence of pH on the degradation of CBZ in the Mn(III)/HSO 3 − process was investigated, and the results are shown in Figure 6. Over 95% of CBZ was removed at pH 5.0, while the removal of ciprofloxacin dropped progressively from 81% to 52% as pH increased from 6.0 to 9.0. The reduction kinetics of Mn(III) by HSO 3 − under the pH range of 5.0-9.0 were displayed in Figure S7. The second-order rate constants of Mn(III) reacting with HSO 3 − at pH 5.0-9.0 were calculated and are listed in Table S2. The reaction rate of Mn(III) with HSO 3 − is sensitive to pH and decreased monotonously with increasing pH. The pK a of HSO 3 − is 7.2, and the species of HSO 3 − shifted to SO 3 2− with increasing pH from 5.0 to 9.0. The electron density on the SO 3 2− is much higher than that on the HSO 3 − and the oxidation of SO 3 2− is expected to be easier than that of HSO 3 − . Thus, the declined reaction rate of Mn(III) with HSO 3 − with increasing pH should be mainly ascribed to the decreased reactivity of Mn(III), which is consistent with previous studies on the relationship of Mn(III) reactivity with pH [28,49,50]. Consequently, the degradation efficiency of CBZ is prominent at low pH due to the higher formation rate of radicals. reactivity of Mn(III), which is consistent with previous studies on the relationship of Mn(III) reactivity with pH [28,49,50]. Consequently, the degradation efficiency of CBZ is prominent at low pH due to the higher formation rate of radicals. Table S3 summarizes the treatment efficiency of CBZ by different HSO3 − -based AOPs. Compared to other reported HSO3 − -based AOPs (Table S3), the Mn(III)/HSO3 − process is more superior in pollutant removal.

Influence of Chloride and DOM on Pollutant Degradation by Mn(III)/HSO3 −
Previous studies have demonstrated that Cl − , one of the most common background water constituents, shifts the distribution of SO4 •− -based AOPs due to the high reactivity of Cl − towards SO4 •− [51,52]. Consequently, the treatment efficiency of SO4 •− -based AOPs decreased with increasing the dosage of Cl − . Figure 7 shows the removal of CBZ by the Mn(III)/HSO3 − process in the presence of different concentrations of Cl − . Consistent with previous studies [51][52][53], addition of 1 and 10 mM of Cl − retarded the degradation of CBZ, and the negative effect was more obvious with higher dosage of Cl − . Table S4 summarizes the principal reactions in the Cl − /SO4 •− system along with their rate constants obtained from the literature. The conversion of radicals by Cl − complicates this process and results in the generation of multiple secondary radicals (e.g., HO • , Cl • , Cl2 •− , and ClO • ). It should be noted that ClO • was derived from the radicals (e.g., HO • , Cl • , and Cl2 •− ) with HOCl. The accumulation of HOCl was found to be negligible, which might be ascribed to the rapid consumption of HOCl by the added HSO3 − ( Figure S8). Thus, ClO • was neglected in this study. In addition, due to the fast consumption of SO4 •− by Cl − , the concentration of SO4 •− decreased significantly in the presence of Cl − at the level of mM and its contribution to pollutant abatement was commonly ignored [51]. Therefore, HO • , Cl • , and Cl2 •− were considered to be the major reactive species responsible for pollutant degradation.

Influence of Chloride and DOM on Pollutant Degradation by Mn(III)/HSO 3 −
Previous studies have demonstrated that Cl − , one of the most common background water constituents, shifts the distribution of SO 4 •− -based AOPs due to the high reactivity of Cl − towards SO 4 •− [51,52]. Consequently, the treatment efficiency of SO 4 •− -based AOPs decreased with increasing the dosage of Cl − . Figure 7 shows the removal of CBZ by the Mn(III)/HSO 3 − process in the presence of different concentrations of Cl − . Consistent with previous studies [51][52][53], addition of 1 and 10 mM of Cl − retarded the degradation of CBZ, and the negative effect was more obvious with higher dosage of Cl − . Table S4 summarizes the principal reactions in the Cl − /SO 4 •− system along with their rate constants obtained from the literature. The conversion of radicals by Cl − complicates this process and results in the generation of multiple secondary radicals (e.g., HO • , Cl • , Cl 2 •− , and ClO • ). It should be noted that ClO • was derived from the radicals (e.g., HO • , Cl • , and Cl 2 •− ) with HOCl. The accumulation of HOCl was found to be negligible, which might be ascribed to the rapid consumption of HOCl by the added HSO 3 − ( Figure S8). Thus, ClO • was neglected in this study. In addition, due to the fast consumption of SO 4 •− by Cl − , the concentration of SO 4 •− decreased significantly in the presence of Cl − at the level of mM and its contribution to pollutant abatement was commonly ignored [51]. Therefore, HO • , Cl • , and Cl 2 •− were considered to be the major reactive species responsible for pollutant degradation. reactivity of Mn(III), which is consistent with previous studies on the relationship of Mn(III) reactivity with pH [28,49,50]. Consequently, the degradation efficiency of CBZ is prominent at low pH due to the higher formation rate of radicals. Table S3 summarizes the treatment efficiency of CBZ by different HSO3 − -based AOPs. Compared to other reported HSO3 − -based AOPs (Table S3), the Mn(III)/HSO3 − process is more superior in pollutant removal.

Influence of Chloride and DOM on Pollutant Degradation by Mn(III)/HSO3 −
Previous studies have demonstrated that Cl − , one of the most common background water constituents, shifts the distribution of SO4 •− -based AOPs due to the high reactivity of Cl − towards SO4 •− [51,52]. Consequently, the treatment efficiency of SO4 •− -based AOPs decreased with increasing the dosage of Cl − . Figure 7 shows the removal of CBZ by the Mn(III)/HSO3 − process in the presence of different concentrations of Cl − . Consistent with previous studies [51][52][53], addition of 1 and 10 mM of Cl − retarded the degradation of CBZ, and the negative effect was more obvious with higher dosage of Cl − . Table S4 summarizes the principal reactions in the Cl − /SO4 •− system along with their rate constants obtained from the literature. The conversion of radicals by Cl − complicates this process and results in the generation of multiple secondary radicals (e.g., HO • , Cl • , Cl2 •− , and ClO • ). It should be noted that ClO • was derived from the radicals (e.g., HO • , Cl • , and Cl2 •− ) with HOCl. The accumulation of HOCl was found to be negligible, which might be ascribed to the rapid consumption of HOCl by the added HSO3 − ( Figure S8). Thus, ClO • was neglected in this study. In addition, due to the fast consumption of SO4 •− by Cl − , the concentration of SO4 •− decreased significantly in the presence of Cl − at the level of mM and its contribution to pollutant abatement was commonly ignored [51]. Therefore, HO • , Cl • , and Cl2 •− were considered to be the major reactive species responsible for pollutant degradation. CBZ, NB, and BPA were employed to analyze the distribution of these radicals. BPA was recalcitrant to Mn(III) ( Figure S9). The degradation kinetics of the probe compound by the Mn(III)/HSO 3 − process in the presence of Cl − can be descried by Equations (5), (8) and (9).
where k Cl • −CBZ and k Cl •− 2 −CBZ are the second-order rate constants of CBZ oxidation by Cl •− , respectively. Based on the removal of these probe compounds (Figures 7 and S10) and the listed second-order rate constants in Table S1, the time-integrated concentrations of HO • , Cl • , and Cl 2 •− were calculated and are shown in Figure S11.  Figure  S11). This could be ascribed to the high reactivity of Cl • towards Cl − (eq s2 in Table S4), leading to the conversion of Cl • to Cl 2 •− in the presence of high concentration of Cl − . The increased concentration of Cl 2 •− with increasing dosage of Cl − further confirmed this speculation. The time-integrated concentration of Cl 2 •− is much higher than that of other radicals, resulting in the predominant contribution of Cl 2 •− in CBZ abatement (Figures 7 and S11). Besides oxidizing CBZ, Cl 2 •− is also expected to transform HSO 3 − to SO 3 •− , resulting in the enhanced consumption rate of HSO 3 − . Consequently, the removal efficiency of CBZ decreased.
Chlorate is a typical byproduct of the persulfate-based AOPs in the presence of chloride [54]. However, negligible chlorate was detected in the Mn(III)/HSO 3 − process in the presence of 1 and 10 mM of chloride ( Figure S8), suggesting that chlorate formation is not a concern in the Mn(III)/HSO 3 − process. DOM is one of the most common background water constituents in natural water and often shows negative effects on pollutant abatement by AOPs [55]. The influence of DOM on CBZ removal by the Mn(III)/HSO 3 − process was shown in Figure 8. A total of 1.0 mg/L of DOM, the representative concentration in natural water, decreased CBZ removal by 13%. DOM could serve as the radical scavenger (Table S5), and resulted in the decrease of CBZ removal. In addition, the conversion of radicals by DOM might interrupt the chain propagation of HSO 3 − oxidation, and thus decrease the consumption of HSO 3 − . As shown in Figure S12, DOM retarded the oxidation of HSO 3 − . As mentioned above, that HSO 3 − also plays the role of radical scavenger, the retarded consumption of HSO 3 − in the presence of DOM might take some responsibility in decreasing CBZ removal by the Mn(III)/HSO 3 − process under the reaction condition of this study.

Conclusions
PP-complexed Mn(III) was proven to be effective in activating HSO3 − in this study. Due to the high reactivity of Mn(III) towards HSO3 − , considerable concentrations of SO4 •− and HO • were generated rapidly, resulting in the effective removal of pollutants. The dosage of HSO3 − was critical for pollutant abatement due to the scavenging effect of HSO3 − . Increasing pH and PP:Mn(III) ratio depressed the reactivity of Mn(III) towards HSO3 − and thus inhibited pollutant removal. Similar to other SO4 •− -based advanced oxidation processes, a significant loss of the treatment efficiency induced by Cl − was observed. In addition, Cl − complicated the chemistry of the Mn(III)/HSO3 − system and introduced reactive chlorine species. Another common water constituent, DOM, also showed negative effect on the treatment efficiency of the Mn(III)/HSO3 − process. This study broadened the HSO3 − -based advanced oxidation processes. However, water parameters and constituents need careful consideration in the application of the Mn(III)/HSO3 − process for pollutant abatement in real water. Different ligands might exist in water and influence the reactivity of Mn(III) towards HSO3 − , which also needs attention in conducting the Mn(III)/HSO3 − process.

Conclusions
PP-complexed Mn(III) was proven to be effective in activating HSO 3 − in this study. Due to the high reactivity of Mn(III) towards HSO 3 − , considerable concentrations of SO 4 •− and HO • were generated rapidly, resulting in the effective removal of pollutants. The dosage of HSO 3 − was critical for pollutant abatement due to the scavenging effect of HSO 3 − . Increasing pH and PP:Mn(III) ratio depressed the reactivity of Mn(III) towards HSO 3 − and thus inhibited pollutant removal. Similar to other SO 4 •− -based advanced oxidation processes, a significant loss of the treatment efficiency induced by Cl − was observed. In addition, Cl − complicated the chemistry of the Mn(III)/HSO 3 − system and introduced reactive chlorine species. Another common water constituent, DOM, also showed negative effect on the treatment efficiency of the Mn(III)/HSO 3 − process. This study broadened the HSO 3 − -based advanced oxidation processes. However, water parameters and constituents need careful consideration in the application of the Mn(III)/HSO 3 − process for pollutant abatement in real water. Different ligands might exist in water and influence the reactivity of Mn(III) towards HSO 3 − , which also needs attention in conducting the Mn(III)/HSO 3 − process.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ijerph19159437/s1. Figure S1: The reported mechanism for the evolution of radicals in the transition metal/HSO 3 process; Figure S2: Influence of radical scavengers on CBZ removal; Figure S3: Influence of HSO 3 dosage on NB degradation; Figure S4 Figure S8: The spectra of ion chromatography of reaction solution; Figure S9: Degradation of BPA by Mn(III) alone; Figure S10: Influence of Clon the degradation of BPA and NB by Mn(III)/HSO 3 process; Figure S11: Influence of Clon the distribution of radicals; Figure S12: Influence of DOM on the consumption of HSO 3 in the presence of Mn(III); Table S1: Second-order rate constants of pollutant oxidation by radicals;  References [29,45,46,51,[55][56][57][58][59][60][61][62] are cited in the supplementary materials.  Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.