Efficient Degradation of Iopromide by Using Sulfite Activated with Mackinawite

Iopromide (IOP), an iodinated X-ray contrast medium (ICM), is identified as a precursor to iodide disinfection byproducts that have high genotoxicity and cytotoxicity to mammals. ICM remains persistent through typical wastewater treatment processes and even through some hydroxyl radical-based advanced oxidation processes. The development of new technologies to remove ICMs is needed. In this work, mackinawite (FeS)-activated sulfite autoxidation was employed for the degradation of IOP-containing water. The experiment was performed in a 500 mL self-made temperature-controlled reactor with online monitoring pH and dissolved oxygen in the laboratory. The effects of various parameters, such as initial pH values, sulfite dosages, FeS dosages, dissolved oxygen, and inorganic anions on the performance of the treatment process have been investigated. Eighty percent of IOP could be degraded in 15 min with 1 g L−1 FeS, 400 μmol L−1 sulfite at pH 8, and high efficiency on the removal of total organic carbon (TOC) was achieved, which is 71.8% via a reaction for 1 h. The generated hydroxyl and oxysulfur radicals, which contributed to the oxidation process, were identified through radical quenching experiments. The dissolved oxygen was essential for the degradation of IOP. The presence of Cl− could facilitate IOP degradation, while NO3− and CO32− could inhibit the degradation process. The reaction pathway involving H-abstraction and oxidative decarboxylation was proposed, based on product identification. The current system shows good applicability for the degradation of IOP and may help in developing a new approach for the treatment of ICM-containing water.


Introduction
Iopromide(N,N -bis(2,3-dihydroxypropyl)-2,4,6-triiodo-5-(2-methoxyacetamido)-Nmethylisophthalamide, IOP), an iodinated X-ray contrast medium (ICM), has been widely used for enabling medical imaging of blood vessels and soft tissues through X-ray examinations [1]. ICMs are metabolically stable in the human body and more than 90% is excreted in the first 24 h through urine or excrement [2]. In previous investigations, ICMs have been detected frequently in the effluent and surface water of sewage treatment plants [3], and even in tap water [4]. In recent years, ICMs have been identified as a precursor to iodide disinfection byproducts (I-DBPs) and have received increasing attention [5]. Studies have shown that ICMs are a major source of iodine during the formation of iodo-trihalomethanes (I-THM) and iodic acid disinfection intermediates (iodo-acids) [6]. These two classes of substances have high genotoxicity and cytotoxicity to mammals [7]. The metabolites or transformation products of ICM in the environment are likely to be detrimental to the environment and human health. The mineralization of ICMs, especially in water treatment systems, is important to avoid any potential harm caused by pernicious byproducts. However, ICMs are recalcitrant to traditional wastewater treatment processes and even in common drinking-water treatment plants [8]. The development of new degradation strategies to remove ICMs is needed.
Advanced oxidation processes (AOPs) have been widely tested for the degradation of ICMs. A variety of systems, including O 3 /H 2 O 2 , UV/TiO 2 , and UV/chloramines systems, have been applied for the degradation of ICMs [9]. However, not all of these can achieve effective degradation. The rate of degradation of IOP (k = 0.18 min −1 ) was observed to be lower than that of other pharmaceutical products in the presence of 0.5 g L −1 or 1 g L −1 TiO 2 with a xenon lamp or UV-A light irradiation [10], and the mineralization rate was also modest (<15%) in a laboratory with simulated wastewater. Effective removal was also not achieved via ozonation or O 3 /H 2 O 2 oxidation [11]. AOPs based on sulfate radicals (SR-AOPs) have shown good efficiency in removing various pollutants [12] and have therefore been used for the oxidation of ICMs, such as iohexol [13] and IOP [14]. It has been discovered that SO 4 •− has a higher redox potential (2.5-3.1 V, normal hydrogen electrode) than HO• (1.9-2.7 V, normal hydrogen electrode) [15]. SO 4 •− exhibits higher selectivity for oxidation than HO• [16]. SO 4 •− are usually generated by activated persulfate (PS) or peroxymonosulfate (PMS) via transition metals [17], heat [18], ultraviolet irradiation [19], alkali [20], and ultrasonication [21]. However, PS and PMS have several drawbacks, such as high cost, inherent toxicity, and residue problems [22]. Recently, the use of activated sulfite to generate SO 4 •− has attracted considerable attention. Sulfite is considered to be a promising alternative to PS or PMS because of its low toxicity and competitive price, where the price of sulfite is about 50% lower than PS and almost 10 times lower than PMS. In general, sulfite could be activated by various transitional metals, including Fe(II/III/VI), Mn(II/III/VII), Co(II), Cu(II), and Cr(III/VI), which involves one-electron transfer from S(IV) to metal ions (M n+ ), thereby generating SO 3 •− [23]. Then, SO 3 •− quickly reacts with oxygen to give SO 5 •− , and SO 4 •− could be formed through reduction of SO 5 •− with bisulfite, as shown in reaction (1)(2)(3)(4)(5)(6)(7)(8) [24,25].
Among the transition metals used for sulfite catalysis, the use of iron is most widely reported [26] due to its abundance in the environment, high catalysis efficiency, low toxicity, and low cost [27]. Fe(III) and SO 4 2− formed in the reaction are electron acceptors that are also beneficial for the biodegradation of organic pollutants [28]. However, ferrous ion has some drawbacks as catalysts, including its acidic or near neutral circumstances, slow circulation between Fe(II) and Fe(III), and accumulation of iron sludge [29]. The heterogeneous catalysts for the activation of sulfite have attracted intense interest for water treatment applications. Recently, mackinawite (FeS), a tetragonal ferrous sulfide crystal generated through the dissimilatory bacterial reduction of sulfate [30], has attracted increasing attention for use in water treatment. It has been tested as a catalyst for activating PS to efficiently remove 2,4-dichlorophenoxyacetic acid, p-chloroaniline [31], trichloroethene, Orange-G, and tetracycline [32], while its applications in activating sulfite are much less. Chen et al. reported a high degradation efficiency (95%) on propranolol by FeS-activated sulfite systems at pH 6 but insufficient mineralization efficiency (less than 20%) [33]. Whether this treatment method can effectively remove IOP and achieve high TOC removal efficiency would be interesting to study. Moreover, the optimal treatment parameters are also very important for enhancing the mineralization, which still needs further studies.
Fe 2+ + HSO 5 − → SO 4 •− + Fe 3+ +OH − (k 7 = 10 4 -10 7 L mol −1 s −1 ) In this work, the potential of the FeS/sulfite system for the degradation of IOP, selected as a model pollutant, was investigated. The operational parameters, including initial pH values, sulfite dosage, and FeS dosage on the degradation process, were optimized. The reactive species that contributed to the degradation process were identified through a radical and oxygen quenching experiment to reveal the mechanism of IOP degradation. The degradation products were identified and the proposed primary oxidation pathway was estimated. Finally, the influences of inorganic anions (Cl − , CO 3 2− and NO 3 − ) that existed commonly in natural water were also investigated.

Characterization of FeS
The morphology of FeS was investigated by using the scanning electron microscope (SEM). As can be clearly seen in Figure 1a,b, the surface of FeS has aggregated micrometerscale FeS particles. The stacking of these smaller particles causes the FeS surface to exhibit a thin plate-like structure with rugged surfaces and pleated edges [34]. This special sheet-like structure provides a large number of sites for the attachment and reaction of pollutants and free radicals. The phase structure of FeS was studied by X-ray diffraction (XRD) analysis. The recorded XRD pattern of the FeS particles is shown in Figure S1a. FeS was dovetailed with tetragonal structure planes with space group P4/nmm (129), corresponding to the standard card of FeS (JCPDS No. 14-0117) [35] assigned to the (101), (110), (111), (200), and (201) planes.
Results from the energy dispersive spectrometer (EDS) revealed that the content of S in FeS particles is 34.52%, as shown in Figure S1b. This finding suggests that the S content in the ore primarily corresponds to FeS formation. Therefore, the nominal purity of raw FeS is 50%, and its iron content is 62.44%. Excessive Fe (12.44%) and O (1.39%) contents may have originated from amorphous iron oxide formation [28]. The content of C (4.22%) in the raw FeS is small. The oxidation or degradation reaction was found to usually start from the outside of the FeS particles and proceed inward [36].
Fourier transform infrared spectroscopy (FTIR) was employed to investigate the functional groups, which is shown in Figure S1c ranging from 400 to 4000 cm −1 . At 412-425 cm −1 , there are very obscure bands that are assigned to Fe(II)-S stretching vibration [37]. The characteristic peaks at around 600 cm −1 are related to the stretching vibrations of S-S and the Fe-S bond [38]. A significant peak at about 1137 cm −1 is attributed to the vibration of Fe=S [37]. This peak is diminished observably after the degradation process which reveals that the reaction between FeS and sulfite may destroy the Fe=S bond on the surface of FeS. The peak at 1615 cm −1 indicates H 2 O bending vibrations [39], while some reports consider this peak is also probably assigned to C=NH [40]. The peak at 3420 cm −1 represents the stretching mode of the surface-bonded H 2 O molecules [39], which indicated there still exists water in the sample even after being vacuum dried.

Control Experiments
The degradation of IOP via activation of sulfite by FeS is shown in Figure 2a. The results show that in the presence of 400 µmol L −1 of sulfite alone, no degradation of IOP occurred in 15 min, while a small amount of about 10% IOP disappeared in the presence of only 1 g L −1 of FeS. This slight disappearance may be caused by the reducibility of FeS to reduce various halogenated organic compounds such as trichloroethylene, tetrachloroethylene [41], and p-chloroaniline via surface deprotonation or electron transfer, especially in alkaline pH [42]. Eighty percent of IOP could be degraded in 15 min with 1 g L −1 of FeS and 400 µmol L −1 of sulfite. This degradation was due to the intense oxidizability of SO 4 •− and •OH generated from the autoxidation of sulfite catalyzed by FeS. From the results of EDS, there are small fractions of oxygen that may be from the oxidized surface FeS. In order to exclude the catalysis effect of Fe 2 O 3 , 1 g L −1 of Fe 2 O 3 was employed as the catalyst and the results show that only 14% of IOP was degraded, which can reveal the low effect of Fe 2 O 3 on the catalysis of sulfite. The reaction of Fe(II)-sulfite in alkaline pH is different from that in acidic or near-neutral pH, which involves the coordination of the Fe(II)-OH complex and SO 3 2− to form the S(IV)-Fe(II)-OH complex [43]. The formed S(IV)-Fe(II)-OH complex could be oxidized into a S(IV)-Fe(III)-OH complex by the dissolved oxygen (DO) in an aqueous solution. SO 3 •− could be released into the aqueous solution with one-electron transfer from Fe(III) to S(IV) of the S(IV)-Fe(III)-OH complex, resulting in the formation of a S(IV)-Fe(II)-OH complex again. SO 3 •− is oxidized into SO 5 •− as the initiation of the oxysulfur species evolves. SO 4 •− and SO 4 2− are generated through the disproportional oxidation of sulfite by SO 5 •− [43]. Results in Figure 2a also show a two-stage kinetic curve, including a fast stage in which about 75% of IOP could be degraded in the first 2 min and a slow one thereafter. This curve is a typical degradation kinetics curve that exists for degradation via autoxidation of sulfite, and it was also observed in our previous studies [44]. This curve is considered as being caused by the fast consumption of sulfite and DO [45]. In the fast stage (first 2 min), the concentration of sulfite and DO is relatively high, which leads to the production of a higher amount of the oxysulfur species. Then, DO and sulfite concentrations drop sharply, and the concentrations of the reactive radical species decrease, which retard the degradation of IOP. The initial IOP degradation rate increased with the initial IOP concentration, which followed the pseudo-first-order kinetics shown in Figure 2b,c.
Fe ions could be released from the surface of FeS through reaction (9) [46], which may also form activated sulfite. To confirm this, the concentrations of released Fe ions in the present system were measured using the 1,10-phenanthroline method [47] and the limitation of detection and quantification were determined as 0.022 mg L −1 and 0.074 mg L −1 . The total dissolved Fe ion concentration in 15 min was less than 0.4 ± 0.05 mg L −1 (Fe 2+ concentration was less than 0.36 ± 0.04 mg L −1 ) as shown in Figure 2d. In order to further examine the role of Fe 2+ in the activation of sulfite, the degradation of IOP by the dissolved Fe 2+ /sulfite system was performed. Results in Figure 2e showed no degradation was found within 15 min. Therefore, FeS/sulfite system is considered to prove heterogeneous activation processes.

Effects of Initial pH
pH is a significant parameter in degradation using iron as catalysts. In order to study the effects of the initial pH values, experiments were carried out at five pH values of 5, 7, 8, 9, and 11. The results in Figure 3 indicate that at pH 8, the best performance of degradation of 80% IOP is realized, while there are no significant differences in the degradation efficiency at the other pH values of 5, 7, and 9 (60.4%, 61.4%, and 57.2%, respectively). No degradation could be found at pH 11. The overall effects of pH on the iron-catalyzed autoxidation process of sulfite are rather intricate, where pH can influence the speciation of S(IV) and Fe(II)/Fe(III) and the standard reduction potential of the oxidants. We can explicate the differences in degradation efficiency on the basis of the distribution of S(IV) and Fe(II) at different pH values. The fractions of species and equilibriums shown in Figures S2 and  S3 present the distribution of Fe(II) and S(IV) species are both pH-dependent. In alkaline circumstances (pH > 7), the Fe(II) and S(IV) species exist as the ferrous hydroxide complex and SO 3 2− ions, respectively. Then, the S(IV)-Fe(II)-OH complex can be formed, initiating the chain reaction of oxysulfur species in the alkaline solution as mentioned previously. When the pH values increase to 11, iron exists as Fe(OH) 2 (cr), which cannot form a complex with S(IV), resulting in no IOP degradation. On the other hand, in acidic conditions, the bisulfite ion is the predominant species and can form a complex with ferrous ion on the surface of FeS to form FeHSO 3 + , and this initiates the auto-oxidation of oxysulfur species that can contribute to the degradation of IOP, as shown in reactions (1)(2)(3)(4)(5)(6)(7)(8). However, the excess hydrogen ions can reduce the concentration of sulfite as the precursor of oxysulfur radicals shown in reaction (10), and this can result in a lower IOP-degradation efficiency. Moreover, the pH changes during the degradation process are monitored as shown in Figure S4. Results show, at initial pH values 5 and 7, the pH value declined apace and eventually stay stable at about 3.5, which leads to approximate degradation efficiencies of IOP under weak acid to the neutral condition. While the initial pH is 8, the final pH is 4.4 after the reaction. The pH could affect the conversion of SO 4 •− to HO• through reactions (14) and (15) and result in the different distributions of SO 4 •− and HO•, which could lead to a different degradation efficiency of IOP.

Effects of FeS Dosage
The effects of the FeS dosage on IOP degradation were investigated next by adding 0.4, 1.0, 1.6, 2.0, and 2.4 g L −1 of FeS to 400 µmol L −1 sulfite. The results, shown in Figure 4b, indicate that the initial reaction rate in the first 2 min increased with the Fe dose, which confirmed the catalytical effects of Fe(II) on S(IV). However, the total degradation efficiencies of IOP in 15 min were not fitted with that tendency of in first 2 min. As shown in Figure 4a, the efficiency of IOP degradation increased as the FeS dose was raised from 0.4 to 1.0 g L −1 , and decreased about 10% with the increase in the FeS dose from 1.0 g L −1 to 2.0 g L −1 . Further, at a FeS dose of 2.4 g L −1 , only 63% of IOP was degraded. This indicates that FeS plays a significant catalytic role, producing SO 4 •− , while excess FeS can scavenge SO 4 •− , as shown in reaction (11) [48]. This finding is consistent with the reported strong interaction between SO 4 •− and Fe 2+ ions where excessive Fe(II) acts as a sulfate-free radical scavenger in solution [49].
Fe 2+ + SO 4 •− → Fe 3+ + SO 4 2− k = 4.6 × 10 9 M −1 s −1 (11)  Figure S5 presents the effects of sulfite concentrations on the degradation of IOP. The results show that the IOP degradation efficiency raised from 61.5% to 80% as the initial sulfite concentration was increased from 200 µmol L −1 to 400 µmol L −1 , which could be attributed to increased sulfite content as the precursor of the oxysulfur radicals. However, when the initial concentration of sulfite was changed from 400 µmol L −1 to 700 µmol L −1 , the IOP degradation rate decreased from 73.7% to 54.3%. This is because excessive sulfite will reduce the concentrations of SO 4 •− through reaction (12) [50] and (13) [51], resulting in a low IOP degradation efficiency.

Role of Radicals and Dissolved Oxygen
Generally, SO 4 •− , SO 5 •− , and SO 3 •− are used as the main reactive oxysulfur radicals in the sulfite activation process. HO• is also considered as a reactive radical that contributes to the degradation of IOP, which could be generated by the reactions between SO 4 •− /H 2 O or SO 4 •− /HO − as shown in reactions (14) and (15) [52]. In order to confirm the contributions of different active species, EtOH and TBA are employed as the scavenger.
EtOH is usually used as an effective scavenger for both SO 4 •− ((1.6-7.7) × 10 7 M −1 s −1 ) and HO• (1.9 × 10 9 M −1 s −1 ) [53]. The rate constant for the reaction between TBA and HO• is (3.8-7.6) × 10 8 M −1 s −1 , which is 1000 times higher than that for the reaction between TBA and SO 4 •− ((4-9.1) × 10 5 M −1 s −1 ) [16]. A moderate amount of TBA could inhibit HO• alone rather than SO 4 •− . As shown in Figure 5a, in the presence of 1 mmol L −1 of EtOH, a degradation of only 17.2% IOP occurs, but with no scavenger, the degradation rate is 80%. This means this 17.2% degradation was contributed to the SO 3 •− /SO 5 •− since the rate constant between SO 3 •− /SO 5 •− and ethanol are very low (k ≤ 10 3 M −1 s −1 ) [54] and 67.6% of IOP were degraded with 1 mmol L −1 of TBA indicating that about 50.4% of the degradation was attributed to SO 4 •− and that HO• was responsible for about 12.4% of the degradation. As shown in reaction (5), the process of generating SO 5 •− by oxidation of SO 3 •− is an important step in the production of SO 4 •− using the present system. DO should be one of the significant factors influencing the degradation of IOP. In order to exclude DO from the present system, the reaction solution was purged with nitrogen to eliminate DO. Figure 5b shows that in the case of nitrogen purging, there was no significant change in the IOP concentration. For 15 min of reaction time, the IOP degradation rate was only 11%, which indicates that the effects from SO 3 •− are weak since SO 5 •− cannot be generated from SO 3 •− without DO. However, when the present system was purged with air, approximately 83% of the IOP was degraded in 15 min. This small enhancement can be attributed to the increased DO concentration leading to the formation of a larger amount of SO 5 •− , and subsequently, of a larger amount of SO 4 •− .

Effects of Inorganic Anions
The effects of common inorganic anions that exist in natural water including Cl − , NO 3 − , and CO 3 2− on the degradation of IOP were investigated. As shown in Figure S6a, the IOP degradation efficiency was slightly enhanced with the addition of Cl − (10 and 100 µmol L −1 ). This might be because Cl • could be generated via Cl − competing for the reactive oxidizing species, as shown in reactions (16) and (17). Cl • is selective and can thus degrade some organic pollutants with high rate constants of reaction [53]. However, with raising concentrations of Cl − , an increasing amount of SO 4 •− would be consumed. Cl 2

•−
and ClHO − with less reactivity might be produced through reactions (18)(19)(20), so if the Cl − levels are too high, this might lead to some inhibition of organic-pollutant degradation [31]. This explains why the IOP degradation rate is greater when 10 µmol L −1 of chloride ion is added to the reaction system than when 100 µmol L −1 is added.
Although NO 3 − indicates an inhibiting influence on the degradation of IOP, the effect was very slight, as shown in Figure S6b. This might be because of the adsorption of NO 3 − on the surface of FeS and its reduction [55], thus competing with the active sites for sulfite activation. In Figure S6c, the IOP degradation efficiency decreased with increasing concentration of CO 3 2− from 10 to 100 µmol L −1 in the FeS/sulfite system. For instance, the IOP degradation efficiency was only 56.9% with 100 µmol L −1 CO 3 2− , but it can achieve approximately 80% in the absence of CO 3 2 . This result can be reasonably explained by the fact that CO 3 •− , which is less reactive than SO 4 •− , can be generated through a competitive reaction with SO 4 •− and CO 3 2- [56].

Removal of TOC
According to the above discussion, the efficiency of mineralization is an important factor for the removal of IOP. In the current system, the removal of TOC was evaluated, which is shown in Figure 6. The TOC reduction achieved 51.4% after 15 min; it then reached 71.8% after 1 h and tended to remain stable thereafter. Chan et al. reported nearcomplete mineralization of IOP using the UV/peroxydisulfate system after approximately 70-80 min [14], which proves the applicability of SR-AOPs for the mineralization of IOP.

Determination of Degradation Products
Three primary degradation products (DPs) formed by IOP degradation were identified by LC-MS. All DPs and their fragments, determined from mass spectra, are presented in Table S1 and Figure S7, and these are denoted as DP789 (empirical formula C 18  based on their own molecular weight. The fragments of DPs are consistent with those found from photoinduced transformation [57] or electrochemical treatment [5] of IOP. There are two similar glycol structures at the end parts of IOP, namely part A and part B as shown in Figure 7, which are most likely to be oxidized, initiating the degradation of IOP. Further, DP789 is formed by the loss of two hydrogens from IOP, which could be formed by the oxidation of the terminal hydroxyl group to the aldehyde or the oxidation of the central hydroxyl group of the molecular chain to the keto moieties at the molecular chain containing the hydroxyl group at both ends [57]. Regardless of which parts are the first to be oxidized, the products, including DP760 and DP728, indicate that both of the glycol structures are finally oxidized to the aldehyde group and form DP728. The above-mentioned oxidation process could be formed by either HO• or SO 4 •− . HO• is preferred to process abstraction of H atoms from the α-H and hydroxyl group, as IOP continually forms aldehyde derivatives or keto moieties derivatives, while SO 4 •− is usually preferred to process electron transformation. However, some organic compounds with less electron-rich groups can also be decomposed by SO 4 •− through the process of Habstraction [58]. The oxidation process of IOP is very likely initiated through H-abstraction by SO 4 •− since α-carbon is vulnerable to the attack by SO 4 •− . The carbon-centered radicals could be formed during the H-abstraction process and could be then rapidly converted to peroxide radicals through oxygen addition. Subsequently, hydroperoxide radicals (HO 2 •− ) are eliminated from the formed peroxide radicals [59] to give aldehyde derivatives or keto moieties, and these are subsequently oxidized into aldehyde derivatives such as DP728.

Experiment
All batch experiments were carried out in a 550 mL beaker and in an open atmospheric condition. A 500 mL solution with IOP and sulfite at the desired concentration was added into the beaker and constantly stirred at a speed of 400 revolutions per minute with a polytetrafluoroethylene-coated electric stirrer. A predetermined amount of 0.5 g FeS was dosed into the IOP-sulfite solution. The pH value of the solution was adjusted using 0.25 M sodium hydroxide or sulfuric acid as quickly as possible. Each batch of solutions was maintained at a controlled temperature of 25 ± 2 • C during the entire experiment. At specific time intervals, a 2 mL sample was withdrawn from the vessel and then filtered using a polyethersulfone filter (0.45 µm), and then 1.125 mL of the filtrate was placed in a 2 mL vial with 0.375 mL of methanol as the terminating agent before the quantification. The error bars in each figure represent the standard deviation for at least thrice repeated experiments.

Characterization
The surface morphologies and chemical compositions of FeS particles were performed using an ultra-high resolution scanning electron microscope (SEM, SU810, Hitachi, Tokyo, Japan) and an energy-dispersive spectrometer (EDS, Bruker xflash 6l60). X-ray diffraction analysis (XRD, D-8, Bruker-axs, Langewiesen, Germany) was performed for 2θ from 20 • to 80 • . The FTIR analysis was carried out with Thermo Scientific Nicolet iS20 ranging from 400 to 4000 cm −1 .

High-Performance Liquid Chromatography (HPLC)
In this experiment, the IOP concentration was measured via high-performance liquid chromatography (Agilent 1260 Infinity, Agilent chromatography, Perris, CA, USA). The chromatographic separation was performed using a reversed-phase C18 column (250 mm × 4.6 mm, AkzoNobel Kromasil, Bohus, Sweden). The solution of the mobile phase was a mixture of acetonitrile and water at a ratio of 8:92 (v/v) flowing at 1 mL min −1 . The wavelength of the UV detector was 238 nm, and the column temperature was maintained at 25 • C.

Liquid Chromatography-MS
The samples used to identify the byproducts of IOP degradation require pretreatment before the quantified analysis. The samples with byproducts were concentrated via solid-phase extraction (SPE) method using an HLB solid-phase extraction column (200 mg sorbent, 6 mL, Jiangsu Green Union Scientific Instrument Co., LTD., Jiangsu, China). A solution containing 6 mL of CH 3 OH with 0.25% (v/v) formic acid, followed by 5 mL of ultrapure water was used for preliminary cleaning of the cartridge. Samples of 500 mL were percolated at a flow rate of 10 mL min −1 . A total of 6 mL of 0.25% (v/v) formic acid in CH 3 OH was added for elution. The eluates were concentrated to 1 mL with a gentle nitrogen flux. The final concentrated samples were analyzed by LC-MS. The chromatographic separations were carried out using an Agilent ZORBAX SB-C18 (150 × 2.1 mm × 3.5 µm particle size) column at a flow rate of 300 µL min −1 and monitored using an MS analyzer (6530 Q-TOF, Agilent Technologies, Santa Clara, CA, USA). The column temperature was maintained at 30 • C. The solution of the mobile phase was acetonitrile (with 0.05% formic acid, v/v) when run on ESI positive mode.

The Other Analytical Methods
The total organic carbon (TOC) was measured on an Analytik Jena N/C 300 (Analytik Jena AG, Langewiesen, Germany) TOC analyzer. The pH values were measured via an FE28 pH meter (Mettler Toledo, Switzerland) equipped with a LE438 probe. The concentrations of dissolved oxygen were determined using a JPB-607A dissolved-oxygen analyzer (China).

Conclusions
The FeS/sulfite system exhibits excellent efficiency in the oxidation of IOP, and can achieve over 70% mineralization in 1 h. The optimal dosage of FeS and sulfite are 1 g L −1 and 400 µmol L −1 , at pH = 8. By performing radical quenching experiments, hydroxyl and oxysulfur radicals were identified as the main contributors to the oxidation of IOP. Three primary products were determined, and the possible pathway for IOP degradation, involving H-abstraction and oxidative decarboxylation, was estimated. The results of this study indicate the system's potential for the degradation and mineralization of IOP. This work is helpful for developing a new approach for the removal of ICM-containing water and extending the applications of SR-AOPs on the water treatment field.
Supplementary Materials: The following are available online, Figure S1: (a) The XRD pattern of raw FeS; (b) EDS images and chemical compositions on the surface of raw FeS; (c) FTIR spectra of raw and residual FeS; Figure S2: Species distribution of 0.01 mM Fe(II) in aqueous solutions at pH in the range of 3-12, Figure S3: Species distribution of 0.4 mM S(IV) in aqueous solutions at pH in the range of 1-12, Figure S4: Changes of pH at different intial pH values, Figure S5: Effects of initial sulfite concentrations on the degradation of IOP by the FeS/sulfite system on IOP degradation, Figure S6 Table S1. Mass spectra results for and its degradation products.