Degradation Kinetics and Disinfection By-Product Formation of Iopromide during UV/Chlorination and UV/Persulfate Oxidation

: As the detection of micropollutants in various water resources is commonly reported, developing an efﬁcient technology to remove them to maintain water safety has become a major focus in recent years. The degradation kinetics of iopromide, one of a group of iodinated X-ray contrast media (ICM), using advanced oxidation processes of ultraviolet/chlorination (UV/Cl 2 ) and UV/persulfate (UV/PS) oxidation were investigated in this research. The results show that iopromide degradation ﬁtted pseudo-ﬁrst-order kinetics, and the rate constants were calculated as 2.20 ( ± 0.01) × 10 − 1 min − 1 and 6.08 ( ± 0.10) × 10 − 2 min − 1 in UV/Cl 2 and UV/PS, respectively. In the two systems, the degradation rates were positively correlated with the initial concentrations of HOCl and PS, respectively. In the UV/Cl 2 system, the degradation rate of iopromide reached a maximum at pH 7, while in the UV/PS system, pH had only a slight effect on the degradation rate. Chloride in water had a negligible effect on iopromide degradation, whereas bromide inhibited iopromide degradation in the UV/Cl 2 system. The contributions of UV irradiation, • OH, and RCS to iopromide degradation during UV/Cl 2 treatment were calculated as 20.8%, 54.1%, and 25.1%, respectively. One carbonated and three nitrogenated disinfection by-products (C-DBP (chloroform) and N-DBPs (dichloroacetonitrile, trichloronitromethane, and trichloroacetone)) were detected at relatively high levels, along with three emerging iodinated DBPs (dichloroiodomethane, monochlorodiiodomethane, and triiodomethane). More C- and N-DBPs were generated in the UV/Cl 2 and UV/PS systems than in UV irradiation, while considerably higher I-DBPs were generated in UV irradiation than in the other two systems. Thus, it is essential to pay attention to DBP formation when UV/Cl 2 or UV/PS is used to treat iopromide in water. In order to better control the generation of carcinogenic and toxic I-DBPs, Cl 2 or PS combined with UV should be adopted for iopromide degradation, instead of UV alone, for providing safe drinking water to the public.


Introduction
Iodinated X-ray contrast media (ICM) such as iohexol, iopromide, iopamidol, iomeprol, and diatrizoate are widely used in radiological investigations for imaging body organs or blood vessels [1,2]. ICM cannot be effectively removed by conventional wastewater treatment processes and is discharged into the aqueous environment. Therefore, ICM have been detected in many rivers and streams (up to µg L −1 in many countries) [3,4]. Iopromide is one of the most commonly used X-ray contrast media, and it remains almost unchanged Water 2022, 14, x FOR PEER REVIEW wastewater treatment processes and is discharged into the aqueous envir fore, ICM have been detected in many rivers and streams (up to μg L −1 in m [3,4]. Iopromide is one of the most commonly used X-ray contrast media almost unchanged after entering the body [5]. Iopromide has been detecte bodies [6], in concentrations ranging from ng L −1 to μg L −1 in wastewater an [7]. It is reported that disinfectants can react with natural organic matter (N and iodide during drinking-water treatment to form various disinfecti (DBPs) [8]. Although ICM themselves have no adverse effects on living cre thought to produce iodine, which results in the formation of emerging iod DBPs), mainly including iodinated trihalomethanes (I-THMs) and iodo-ac I-DBPs are of higher cytotoxicity and genotoxicity than their chlorinated analogues, and therefore they have attracted widespread attention [10,1 DBPs in drinking water can cause taste and odor problems, especially C sensory threshold concentration of 1 μg L −1 [12]. A study indicated tha react with chlorine to form trichloroacetic acid (TCAA), chloroform (C chloroiodomethane (CHCl2I) [13]. Therefore, it is essential to remove IC water. The physical properties of iopromide are summarized in Table 1. Table 1. The physical properties of iopromide.

Compound Molecular Formula Chemical Structure Mole
Iopromide C18H24I3N3O8 According to the previous literature, the degradation efficiency of th idation process for ICM is not ideal [14][15][16]. With the development of adv processes (AOPs), UV/chlorine (UV/Cl2) has been successfully applied fo cropollutants in recent years [17]. Other AOPs including UV/H2O2 (UV/PS), and UV/peroxymonosulfate can also effectively degrade ICM d ation of •OH and •SO4 − possessing high redox potentials (1.8-2.7 and 2. tively) [16]. Wu et al. investigated diatrizoate degradation by UV/Cl2 and reactive chlorine species (RCS) are major contributors to diatrizoate degra pared with UV/Cl2, both iopamidol and diatrizoate could be degrade UV/H2O2 [18]. UV/PS has been proven to be more efficient than UV/H2O degradation, and •SO4 − is the dominant reactive species in UV/PS oxidati order rate constant of •SO4 − with diatrizoate was calculated as 1.90 × 10 However, to the best of our knowledge, there is currently no literature re radation kinetics of iopromide during UV/Cl2 and UV/PS AOPs.
Therefore, the purpose of this research was: (1) to investigate the degr of iopromide during chlorination, UV photolysis, UV/Cl2, and UV/persulf tigate the effects of different oxidant concentrations, pH values, and chlori concentrations on iopromide degradation, (3) to investigate the contribu radicals to the degradation of iopromide by UV/Cl2, and (4) to evaluate th DBPs formed in the sequential chlorine disinfection process.

791.11
According to the previous literature, the degradation efficiency of the traditional oxidation process for ICM is not ideal [14][15][16]. With the development of advanced oxidation processes (AOPs), UV/chlorine (UV/Cl 2 ) has been successfully applied for degrading micropollutants in recent years [17]. Other AOPs including UV/H 2 O 2 , UV/persulfate (UV/PS), and UV/peroxymonosulfate can also effectively degrade ICM due to the generation of •OH and •SO 4 − possessing high redox potentials (1.8-2.7 and 2.5-3.1 V, respectively) [16]. Wu et al. investigated diatrizoate degradation by UV/Cl 2 and found that the reactive chlorine species (RCS) are major contributors to diatrizoate degradation [1]. Compared with UV/Cl 2 , both iopamidol and diatrizoate could be degraded effectively by UV/H 2 O 2 [18]. UV/PS has been proven to be more efficient than UV/H 2 O 2 for diatrizoate degradation, and •SO 4 − is the dominant reactive species in UV/PS oxidation. The secondorder rate constant of •SO 4 − with diatrizoate was calculated as 1.90 × 10 9 M −1 s −1 [19,20]. However, to the best of our knowledge, there is currently no literature reporting the degradation kinetics of iopromide during UV/Cl 2 and UV/PS AOPs. Therefore, the purpose of this research was: (1) to investigate the degradation kinetics of iopromide during chlorination, UV photolysis, UV/Cl 2 , and UV/persulfate, (2) to investigate the effects of different oxidant concentrations, pH values, and chloride and bromide concentrations on iopromide degradation, (3) to investigate the contribution of various radicals to the degradation of iopromide by UV/Cl 2 , and (4) to evaluate the C-, N-, and I-DBPs formed in the sequential chlorine disinfection process.

Experimental Procedures
A bench-scale reactor equipped with a low-pressure Hg UV mercury lamp (11 W, 254 nm, 4P-SE, Philips, Shanghai, China) was used to conduct the experiments. The UV intensity was determined as 1.12 mW/cm 2 using a UV radiometer (UV-C luxometer, Photoelectric Instrument Factory of Beijing Normal University, Beijing, China). Before the beginning of each experiment, the UV lamp was warmed up for 30 min to obtain a stable UV emission. A total of 100 mL of iopromide solution (10 µM) was prepared for UV/Cl 2 and UV/PS oxidation experiments, with the pH adjusted to 7.0. Then, an appropriate amount of NaClO (100 mM) or PS (1 M) was added into the reactor to achieve the desired dosage (25-200 µM for NaClO and 1-5 mM for PS). During each experiment, 1 mL of sample was withdrawn at certain time intervals and transferred to a high-performance liquid chromatography vial containing 30 µL of Na 2 S 2 O 3 (0.1 M for UV/Cl 2 ) or excess methanol (for UV/PS) to terminate the reaction. All samples were analyzed using HPLC. Duplicate experiments were performed.
The contribution of radicals to the degradation of iopromide by UV/Cl 2 was examined by adding TBA or EtOH as a radical scavenger for the reaction, with an oxidant (100 µM)/iopromide (10 µM) molar ratio of 10 at pH 7.
The experiments for DBP formation were conducted in 45 mL screw-cap glass vials with PTFE Sep under a headspace-free condition in a dark environment at 25 ± 1 • C. After 3 d, the samples were quenched using Na 2 S 2 O 3 (for UV/Cl 2 ) or excess methanol (for UV/PS) and extracted using MtBE for DBP analysis using gas chromatography (GC).

Analytical Methods
The concentration of iopromide was analyzed using an HPLC system (Agilent 1200, Palo Alto, CA, USA) equipped with an XTerra MS column (5 µm, 250 mm × 4.6 mm, Waters, Milford, CT, USA) and a UV spectrophotometer at the 242 nm wavelength. A 10 µL sample was injected with 88% mobile phase and 12% acetic acid in ultra-pure water (pH 4) at a flow rate of 1.0 mL/min at 30 • C. The detection limit of iopromide was 1 µM.
The pH values of the solutions were detected by using a regularly calibrated pH meter (FE20 FiveEasy, Mettler Toledo, Switzerland). The concentration of chlorine was measured using the N, N-diethyl-p-phenylenediamine (DPD) colorimetric method [21]. The concentration of persulfate was determined with a spectrophotometer (UV-2800, Unico Instruments Co., Ltd., Shanghai, China) at a wavelength of 352 nm [22].
The formation of DBPs was analyzed according to the US EPA Method 551.1 [23]. Samples were quenched and extracted using MtBE for the analysis using a GC (GC-2010 Plus, Shimadzu, Japan) equipped with an Rtx-5 column (30 m × 0.25 mm internal diameter, 0.25 µm film thickness, J&W, Palo Alto, CA, USA).

Kinetics of Iopromide Degradation during UV Photolysis, UV/Cl 2 , and UV/PS Oxidation
In order to elucidate the correlation between iopromide degradation and time, different reaction systems including UV photolysis, chlorination, persulfate, UV/Cl 2 and UV/persulfate systems were investigated. As shown in Figure 1, UV/Cl 2 could degrade iopromide most effectively, followed by UV/PS and UV photolysis, but not chlorination and persulfate alone. The degradation fitted pseudo-first-order kinetics well, as expressed in Equation (1) Water 2022, 14, x FOR PEER REVIEW 4 of 11 The formation of DBPs was analyzed according to the US EPA Method 551.1 [23]. Samples were quenched and extracted using MtBE for the analysis using a GC (GC-2010 Plus, Shimadzu, Japan) equipped with an Rtx-5 column (30 m × 0.25 mm internal diameter, 0.25 μm film thickness, J&W, Palo Alto, CA, USA).

Kinetics of Iopromide Degradation during UV Photolysis, UV/Cl2, and UV/PS Oxidation
In order to elucidate the correlation between iopromide degradation and time, different reaction systems including UV photolysis, chlorination, persulfate, UV/Cl2 and UV/persulfate systems were investigated. As shown in Figure 1, UV/Cl2 could degrade iopromide most effectively, followed by UV/PS and UV photolysis, but not chlorination and persulfate alone. The degradation fitted pseudo-first-order kinetics well, as expressed in Equation (1) (1) By using the data in Figure 1, the kobs values of iopromide degradation during UV photolysis, UV/Cl2, and UV/PS can be calculated as 4.58 (± 0.02) × 10 −2 , 2.20 (± 0.01) × 10 −1 , and 6.08 (± 0.10) × 10 −2 min −1 , respectively. The kobs values for UV/Cl2 and UV/PS were greater than the value for UV irradiation alone due to the generation of •OH, •SO4 − , and RCS [24].

Effects of Different Initial HOCl and PS Concentrations on Iopromide Degradation by UV/Cl2 and UV/PS Oxidation
The effects of oxidant concentration on the degradation of iopromide in the UV/Cl2 and UV/PS processes were studied ( Figure S1). As displayed in Figure 2a, the degradation of iopromide with different initial chlorine concentrations during UV/Cl2 fitted a pseudofirst-order kinetics model well (R 2 > 0.99). The kobs of iopromide increased from 4.58 (± 0.20) × 10 −2 to 3.47 (± 0.11) × 10 −1 min −1 as the chlorine concentration increased from 0 (UV alone) to 200 μM. The lower-left insert in Figure 2a shows a linear relationship between kobs and the chlorine concentration, with R 2 = 0.973, which indicates that the rate of iopromide degradation during UV/Cl2 is first order with respect to chlorine concentration. By using the data in Figure 1, the k obs values of iopromide degradation during UV photolysis, UV/Cl 2 , and UV/PS can be calculated as 4.58 (± 0.02) × 10 −2 , 2.20 (± 0.01) × 10 −1 , and 6.08 (± 0.10) × 10 −2 min −1 , respectively. The k obs values for UV/Cl 2 and UV/PS were greater than the value for UV irradiation alone due to the generation of •OH, •SO 4 − , and RCS [24].

Effects of Different Initial HOCl and PS Concentrations on Iopromide Degradation by UV/Cl 2 and UV/PS Oxidation
The effects of oxidant concentration on the degradation of iopromide in the UV/Cl 2 and UV/PS processes were studied ( Figure S1). As displayed in Figure 2a, the degradation of iopromide with different initial chlorine concentrations during UV/Cl 2 fitted a pseudo-first-order kinetics model well (R 2 > 0.99). The k obs of iopromide increased from 4.58 (± 0.20) × 10 −2 to 3.47 (± 0.11) × 10 −1 min −1 as the chlorine concentration increased from 0 (UV alone) to 200 µM. The lower-left insert in Figure 2a shows a linear relationship between k obs and the chlorine concentration, with R 2 = 0.973, which indicates that the rate of iopromide degradation during UV/Cl 2 is first order with respect to chlorine concentration. As the chlorine concentration increases, free chlorine can produce a series of radicals including •OH and RCS under UV irradiation, enhancing the rate of iopromide degradation, as expressed in Equations (2) and (3): where k uv/chlorine and k uv represent the observed pseudo-first-order reaction rate constants of iopromide degradation during UV/Cl 2 and UV photolysis, respectively, and k •OH and k RCS represent the pseudo-first-order reaction rate constants of iopromide degradation by •OH and RCS, respectively.
Water 2022, 14, x FOR PEER REVIEW 5 of 11 As the chlorine concentration increases, free chlorine can produce a series of radicals including •OH and RCS under UV irradiation, enhancing the rate of iopromide degradation, as expressed in Equations (2) and (3): where kuv/chlorine and kuv represent the observed pseudo-first-order reaction rate constants of iopromide degradation during UV/Cl2 and UV photolysis, respectively, and k•OH and kRCS represent the pseudo-first-order reaction rate constants of iopromide degradation by •OH and RCS, respectively. On the other hand, iopromide degradation also fitted pseudo-first-order kinetics well at different PS concentrations, as displayed in Figure 2b. The kobs for iopromide degradation increased considerably from 6.14 (± 0.18) × 10 −2 to 1.09 (± 0.02) × 10 −1 min −1 as the PS dosage increased from 1 mM to 5 mM, by increasing the formation of oxidizing radicals (•OH and •SO4 − ), as shown in Equations (4) and (5) [25].

Contributions of Radicals to Iopromide Degradation by UV/Cl2
It has been reported that during UV/Cl2, •OH and •Cl were the dominant radicals for the degradation of benzoic acid, while the effects of other reactive species such as •Cl2 − and •O − were negligible [26]. To determine the contributions of radicals in iopromide degradation in the UV/Cl2 system, radical scavenging experiments were performed with the addition of excessive TBA and EtOH. TBA reacts with •Cl with a rate constant of 1.8 × 10 10 M −1 min −1 , whereas the rate constant for the reaction of TBA with •Cl2 − is 2.1 × 10 4 M −1 min −1 [27,28]. As shown in Figure 3, the kobs values of iopromide during UV/Cl2 decreased to On the other hand, iopromide degradation also fitted pseudo-first-order kinetics well at different PS concentrations, as displayed in Figure 2b. The k obs for iopromide degradation increased considerably from 6.14 (± 0.18) × 10 −2 to 1.09 (± 0.02) × 10 −1 min −1 as the PS dosage increased from 1 mM to 5 mM, by increasing the formation of oxidizing radicals (•OH and •SO 4 − ), as shown in Equations (4) and (5) [25].

Contributions of Radicals to Iopromide Degradation by UV/Cl 2
It has been reported that during UV/Cl 2 , •OH and •Cl were the dominant radicals for the degradation of benzoic acid, while the effects of other reactive species such as •Cl 2 − and •O − were negligible [26]. To determine the contributions of radicals in iopromide degradation in the UV/Cl 2 system, radical scavenging experiments were performed with the addition of excessive TBA and EtOH. TBA reacts with •Cl with a rate constant of 1.8 × 10 10 M −1 min −1 , whereas the rate constant for the reaction of TBA with •Cl 2 − is 2.1 × 10 4 M −1 min −1 [27,28]. As shown in Figure 3, the k obs values of iopromide during UV/Cl 2 decreased to 1.64 (± 0.06) × 10 −1 and 1.01 (± 0.02) × 10 −1 min −1 in the presence of 100 mM TBA and EtOH, respectively, while the degradation of iopromide dropped from 84.2% to 74.7% and 56.7%, respectively, after 8 min of reaction time. As displayed in Figure 4, the calculated contributions of UV irradiation, •OH, and RCS to the degradation of iopromide were 20.8%, 54.1%, and 25.1%, respectively (Equations (6)- (8)). The radicals formed in the UV/Cl 2 system enhanced the iopromide degradation. It has been reported that the rate constant of the reaction of •OH with iopromide was 3.34 ± (0.14) × 10 9 M −1 s −1 using γ-irradiation [29], while there has been little research on the rate constant of •Cl with iopromide. Therefore, the role of •Cl should not be neglected in the degradation of iopromide.

Effects of Solution pH on Iopromide Degradation by UV/Cl 2 and UV/PS
As illustrated in Figure 5a, at pH 7 the rate of iopromide degradation in UV/Cl 2 reached a maximum and then decreased as the pH increased further to pH 9. In the UV/Cl 2 system, chlorine is dissociated into OCl − , which can form •OH and •Cl under UV irradiation (Equations (9) and (10)) [26,30]. As the solution pH changed, chlorine speciations were affected [31]. The acid dissociation constant (pKa) of HOCl is 7.5, so the solution was dominated by HClO at pH ≤ 7.5. Compared to OCl − , HOCl is more effective for inactivating Escherichia coli and other pathogens in drinking water [32]. However, in the previous experiments, chlorination could not degrade iopromide effectively (Figure 1 at pH 7). Although HOCl has no degradation effect on iopromide, the difference in HOCl/OCl − distribution in water can affect the UV absorbance and quantum yield of the solution. [33]. Under 254 nm irradiation, the quantum yields of HOCl and OCl − photolysis at room temperature were determined to be 1.45 and 0.97, respectively [34].
reached a maximum and then decreased as the pH increased further to pH 9. In the UV/Cl2 system, chlorine is dissociated into OCl − , which can form •OH and •Cl under UV irradiation (Equations (9) and (10)) [26,30]. As the solution pH changed, chlorine speciations were affected [31]. The acid dissociation constant (pKa) of HOCl is 7.5, so the solution was dominated by HClO at pH ≤ 7.5. Compared to OCl − , HOCl is more effective for inactivating Escherichia coli and other pathogens in drinking water [32]. However, in the previous experiments, chlorination could not degrade iopromide effectively (Figure 1 at pH 7). Although HOCl has no degradation effect on iopromide, the difference in HOCl/OCl − distribution in water can affect the UV absorbance and quantum yield of the solution. [33]. Under 254 nm irradiation, the quantum yields of HOCl and OCl − photolysis at room temperature were determined to be 1.45 and 0.97, respectively [34]. In Figure 5b, pH had a minor effect on the degradation of iopromide via UV/PS oxidation, which can be explained as follows. (1) According to the literature, under acidic and neutral conditions •SO4 − is the predominant radical [35]. (2) At pH ≥ 7, the •SO4 − can be converted to •OH according to Equations (11) and (12) [36]. The redox potential of •SO4 − and •OH is 2.5-3.1 V and 1.8-2.7 V, respectively [37]. Although the difference from Figure 5b is not significant, the reaction rate constant under alkaline conditions is smaller.

Effects of Inorganic Ions on Iopromide Degradation by UV/Chlorination
Clexists widely in the water matrix, with concentrations ranging from 0 to 20 mM in surface water and wastewater effluent [38]. Thus, in this study, the influence of chloride on the degradation of iopromide was studied at concentrations of 0-20 mM. As seen in can be converted to •OH according to Equations (11) and (12) [36]. The redox potential of •SO 4 − and •OH is 2.5-3.1 V and 1.8-2.7 V, respectively [37]. Although the difference from Figure 5b is not significant, the reaction rate constant under alkaline conditions is smaller.

Effects of Inorganic Ions on Iopromide Degradation by UV/Chlorination
Clexists widely in the water matrix, with concentrations ranging from 0 to 20 mM in surface water and wastewater effluent [38]. Thus, in this study, the influence of chloride on the degradation of iopromide was studied at concentrations of 0-20 mM. As seen in Figure 6a, the presence of Cl − has a negligible effect on the degradation of iopromide. The explanation for this phenomenon involves the reactions between Cl − and •OH to form •ClOH − [39]. In addition, Cl − can react with •Cl to produce •Cl 2 − [40]. The produced •ClOH − and •Cl 2 − can be decomposed into •OH and •Cl in a reversible manner [41].
Water 2022, 14, x FOR PEER REVIEW 8 of 11 Figure 6a, the presence of Cl − has a negligible effect on the degradation of iopromide. The explanation for this phenomenon involves the reactions between Cl − and •OH to form •ClOH − [39]. In addition, Cl − can react with •Cl to produce •Cl2 − [40]. The produced •ClOH − and •Cl2 − can be decomposed into •OH and •Cl in a reversible manner [41]. Brcan also compete for radicals with the target compounds. The concentration range of Brin this study was 0-20 mM. As displayed in Figure 6b, the degradation of iopromide was inhibited due to the presence of Br − , because bromide can react rapidly with •OH to form •BrOH − and act like a radical scavenger. The rate constant of the reaction of Brwith •OH is 6.6 × 10 11 M −1 min −1 [42]. Accordingly, the kobs value decreased from 2.20 (± 0.01) × 10 −1 min −1 to 6.24 (± 0.24) × 10 −2 min −1 as the bromide concentration increased from 0 to 20 mM.

Formation of DBPs during Iopromide Degradation in Different Oxidation Systems
In order to evaluate the formation of DBPs after iopromide oxidation in the sequential chlorine disinfection process, which is the most commonly applied disinfection process, various C-, N-, and I-DBPs were analyzed. The results are displayed in Figure 7. In the three systems, one C-DBP (CHCl3) and three N-DBPs (dichloroacetonitrile (DCAN), trichloronitromethane (TCNM), and trichloroacetone (TCAN)) were detected at high concentrations. The concentrations of CHCl3 and TCNM after UV/Cl2 were greater than those for UV irradiation alone, while the concentrations of DCAN and TCAN were similar in the UV and UV/Cl2 systems. In the UV/Cl2 system, the amounts of TCM and TCNM were greater than for UV irradiation. A similar phenomenon was also observed in the research of Qin et al. for the treatment of ronidazole in a UV/Cl2 system [43]. The concentrations of C-and N-DBPs in the UV/PS system were lower than those for UV irradiation and the UV/Cl2 system. On the other hand, three I-DBPs were detected (dichloromonoiodomethane (CHCl2I), chlorodiiodomethane (CHClI2), and triiodomethane (CHI3)), and much higher amounts were formed in the system with UV alone compared to the UV/Cl2 and UV/PS oxidation systems. Because the benzene ring of iopromide contains iodine atoms, the organic iodine can be a precursor of I-DBPs in the drinking-water chlorine disinfection process [44]. The reason for the higher amounts of I-DBPs formed with UV alone than in UV/Cl2 and UV/PS can be explained by the conversion of I − into stable  3 IO in UV/Cl2 and UV/PS [45]. Accordingly, UV/Cl2 and UV/PS oxidation can reduce the generation of toxic Brcan also compete for radicals with the target compounds. The concentration range of Brin this study was 0-20 mM. As displayed in Figure 6b, the degradation of iopromide was inhibited due to the presence of Br − , because bromide can react rapidly with •OH to form •BrOH − and act like a radical scavenger. The rate constant of the reaction of Brwith •OH is 6.6 × 10 11 M −1 min −1 [42]. Accordingly, the k obs value decreased from 2.20 (± 0.01) × 10 −1 min −1 to 6.24 (± 0.24) × 10 −2 min −1 as the bromide concentration increased from 0 to 20 mM.

Formation of DBPs during Iopromide Degradation in Different Oxidation Systems
In order to evaluate the formation of DBPs after iopromide oxidation in the sequential chlorine disinfection process, which is the most commonly applied disinfection process, various C-, N-, and I-DBPs were analyzed. The results are displayed in Figure 7. In the three systems, one C-DBP (CHCl 3 ) and three N-DBPs (dichloroacetonitrile (DCAN), trichloronitromethane (TCNM), and trichloroacetone (TCAN)) were detected at high concentrations. The concentrations of CHCl 3 and TCNM after UV/Cl 2 were greater than those for UV irradiation alone, while the concentrations of DCAN and TCAN were similar in the UV and UV/Cl 2 systems. In the UV/Cl 2 system, the amounts of TCM and TCNM were greater than for UV irradiation. A similar phenomenon was also observed in the research of Qin et al. for the treatment of ronidazole in a UV/Cl 2 system [43]. The concentrations of C-and N-DBPs in the UV/PS system were lower than those for UV irradiation and the UV/Cl 2 system. On the other hand, three I-DBPs were detected (dichloromonoiodomethane (CHCl 2 I), chlorodiiodomethane (CHClI 2 ), and triiodomethane (CHI 3 )), and much higher amounts were formed in the system with UV alone compared to the UV/Cl 2 and UV/PS oxidation systems. Because the benzene ring of iopromide contains iodine atoms, the organic iodine can be a precursor of I-DBPs in the drinking-water chlorine disinfection process [44]. The reason for the higher amounts of I-DBPs formed with UV alone than in UV/Cl 2 and UV/PS can be explained by the conversion of I − into stable IO − 3 in UV/Cl 2 and UV/PS [45]. Accordingly, UV/Cl 2 and UV/PS oxidation can reduce the generation of toxic I-DBPs effectively in the sequential chlorine disinfection process after iopromide degradation compared to UV irradiation, which is beneficial for providing safe drinking water to the public. I-DBPs effectively in the sequential chlorine disinfection process after iopromide degradation compared to UV irradiation, which is beneficial for providing safe drinking water to the public.

Conclusions
Iopromide could not be degraded by Cl2 or PS oxidation alone, but could be degraded by UV irradiation, UV/Cl2, and UV/PS oxidation. The degradation of iopromide fitted pseudo-first-order kinetics with the rate constants calculated as 4.58 (± 0.02) × 10 −2 , 2.20 (± 0.01) × 10 −1 , and 6.08 (± 0.10) × 10 −2 min −1 , in UV, UV/Cl2, and UV/PS systems, respectively. In UV/Cl2, kobs was positively correlated with the initial concentration of HOCl. The presence of Cl − had only a slight effect on iopromide degradation, while Br − inhibited it. The contributions of UV irradiation, •OH, and RCS to the degradation of iopromide were estimated to be 20.8%, 45.9%, and 33.3%, respectively. The degradation rate of iopromide first increased and then decreased with increasing pH, with a maximum at pH 7. During UV/PS oxidation, the rate constant of iopromide degradation increased significantly with an increase in PS concentration, while the pH had a negligible effect. Compared with UV irradiation, more C-and N-DBPs were produced in the UV/Cl2 and UV/PS systems, while much higher amounts of I-DBPs were produced in UV irradiation than in the other two systems. In order to better control the generation of carcinogenic and toxic I-DBPs, Cl2 or PS in combination with UV should be adopted for iopromide degradation, instead of UV alone, for providing safe drinking water to the public.

Supplementary Materials:
The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1: UV visible spectrum of iopromide degradation during chlorine.

Conclusions
Iopromide could not be degraded by Cl 2 or PS oxidation alone, but could be degraded by UV irradiation, UV/Cl 2 , and UV/PS oxidation. The degradation of iopromide fitted pseudofirst-order kinetics with the rate constants calculated as 4.58 (± 0.02) × 10 −2 , 2.20 (± 0.01) × 10 −1 , and 6.08 (± 0.10) × 10 −2 min −1 , in UV, UV/Cl 2 , and UV/PS systems, respectively. In UV/Cl 2 , k obs was positively correlated with the initial concentration of HOCl. The presence of Cl − had only a slight effect on iopromide degradation, while Br − inhibited it. The contributions of UV irradiation, •OH, and RCS to the degradation of iopromide were estimated to be 20.8%, 45.9%, and 33.3%, respectively. The degradation rate of iopromide first increased and then decreased with increasing pH, with a maximum at pH 7. During UV/PS oxidation, the rate constant of iopromide degradation increased significantly with an increase in PS concentration, while the pH had a negligible effect. Compared with UV irradiation, more C-and N-DBPs were produced in the UV/Cl 2 and UV/PS systems, while much higher amounts of I-DBPs were produced in UV irradiation than in the other two systems. In order to better control the generation of carcinogenic and toxic I-DBPs, Cl 2 or PS in combination with UV should be adopted for iopromide degradation, instead of UV alone, for providing safe drinking water to the public.

Informed Consent Statement: Not applicable.
Data Availability Statement: All analyzed data in this study have been included in the manuscript.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.