Comparative Study of the Oxidative Degradation of Di ﬀ erent 4-Aminobenzene Sulfonamides in Aqueous Solution by Sulﬁte Activation in the Presence of Fe(0), Fe(II), Fe(III) or Fe(VI)

: This study is focused on advanced oxidation technologies (AOTs) using the combined e ﬀ ect of Fe(0–VI) / sulﬁte systems, that produce mainly SO 4 •− radicals, to remove di ﬀ erent 4-aminobenzene sulfonamides (SAs), namely sulfamethazine, sulfadiazine, sulfamethizole, from aqueous solutions. Results obtained showed that neither sulﬁte nor iron alone is able to degrade SAs; however, the combined e ﬀ ect depends on the oxidation state of iron species whose e ﬀ ectiveness to activate sulﬁte to promote the degradation of SAs increased following this order: Fe(III) < Fe(II) < Fe(0) < Fe(VI). Using Fe(VI) / sulﬁte, the complete removal of SAs was obtained in 5 min largely surpassing the e ﬀ ectiveness of the other three systems. The sulfonamides’ removal percentage was markedly inﬂuenced by sulﬁte concentration and dissolved oxygen, which improved the generation of oxidant radicals. Response surface methodology was applied, and a quadratic polynomial model was obtained, which allowed us to determine the percentage of SAs degradation as a function of both the iron species and sulﬁte concentrations. The study of the inﬂuence of the water matrix on these AOTs revealed an inhibition of SAs’ removal percentage when using ground water. This is probably due to the presence of di ﬀ erent anions, such as HCO 3 − , Cl − , and SO 42 − in relatively high concentrations. According to the byproducts identiﬁed, the proposed degradation pathways include hydroxylation, SO 2 extrusion, and di ﬀ erent bond-cleavage processes. Cytotoxicity of degradation byproducts, using MTS assay with HEK 293 and J774 cell lines for the ﬁrst time, did not show an inhibition in cell proliferation, sustaining the safety of the process.


Introduction
Several techniques for the treatment of water pollutants by organic compounds, such as biological treatment, adsorption, membrane filtration, traditional and advanced oxidation technologies (AOTs), have been developed in the last decade [1][2][3]. The AOTs are commonly characterized by the generation of reactive species, including hydroxyl, superoxide, and sulfate radicals, as well as singlet oxygen, and have proven to be highly efficient degrading organic pollutants [4][5][6]. demonstrated the efficiency of the Fe(VI)/sulfite system to degrade a pollutant mixture in 30 s at pH 9.0, whereas Fe(VI) alone only achieved less than 6% removal. Sulfamethoxazole (5 µM), included in this mixture, was degraded at 70% with [Fe(VI)] = 50 µM and [sulfite] = 250 µM at pH 9.0 in 30 min. Feng et al. [33] showed accelerated ferrate oxidation of trimethoprim (TMP), enhancing the oxidation to 100% with the addition of one-electron and two electron transfer reductants (SO 3 2− and S 2 O 3 2− ). According to the above references [22][23][24][25][26], the reactions involved in iron/sulfite complex systems can be resumed as follows (Reactions (1)- (28)): Fe 2+ + HSO − 3 ↔ FeHSO + 3 ; k 4 = 1 × 10 4 M −1 (4) 4FeHSO FeSO HSO − 5 + HSO − 3 → 2SO 2− 4 + 2H + ; k 26 = 2.6 × 10 6 M −1 s −1 (26) Fe 3+ + HSO − 3 → FeSO + 3 +H + ; k 28 = 2.82 × 10 2 s −1 (28)  removal of contaminants [28]. The effectiveness of a degradation process in water treatment also implies that the degraded byproducts do not harm the environment. In this context, most toxicological studies have used Vibrio fischeri bacteria tests [16,36], although the application of line cells is becoming more relevant as toxicity models in humans [37]. According to this background, AOTs were applied in the present study to degrade three different 4-aminobenzene sulfonamides (sulfamethazine, sulfadiazine, and sulfamethizole) by sulfite activation in the presence of Fe(0), Fe(II), Fe(III), or Fe(VI). As far as we know, this is the first time that these systems of water treatment are used to degrade the above sulfonamides. Thus, the main objectives were (i) to compare the treatment efficiencies of each system for sulfonamide removal; (ii) to investigate the effects of some operational conditions, such as sulfite concentration, iron concentrations, initial pH, and water matrix; (iii) to model the sulfonamide degradation increasing process optimization by response surface methodology (RSM); (iv) to identify the byproducts obtained and investigate the mechanism for sulfonamide degradation in these systems, and (v) to determine the cytotoxicity of the degradation products resulting from each system studied by application of line cells.

Materials
All chemicals used in this study, namely sulfamethazine (SMZ) (> 99%), sulfadiazine (SDZ) (> 99%), sulfamethizole (SML) (> 99%), Fe(0) (> 95%), ferrous sulfate (> 99%), ferric sulfate (> 95%), potassium ferrate (> 90%), sodium sulfite (> 95%), hydroxylamine hydrochloride (NH2OH·HCl) (99%), hydrochloric acid (pure grade), sodium hydroxide (pure grade), and tert-butyl alcohol (> 99%), were of analytical grade and supplied by Sigma-Aldrich, Merck KGaA, Darmstadt, Germany. All the solutions were prepared using ultrapure water obtained in an equipment Milli-Q ® (Milli-pore). Some selected chemical properties of the SAs are shown in Table 1. removal of contaminants [28]. The effectiveness of a degradation process in water treatment also implies that the degraded byproducts do not harm the environment. In this context, most toxicological studies have used Vibrio fischeri bacteria tests [16,36], although the application of line cells is becoming more relevant as toxicity models in humans [37]. According to this background, AOTs were applied in the present study to degrade three different 4-aminobenzene sulfonamides (sulfamethazine, sulfadiazine, and sulfamethizole) by sulfite activation in the presence of Fe(0), Fe(II), Fe(III), or Fe(VI). As far as we know, this is the first time that these systems of water treatment are used to degrade the above sulfonamides. Thus, the main objectives were (i) to compare the treatment efficiencies of each system for sulfonamide removal; (ii) to investigate the effects of some operational conditions, such as sulfite concentration, iron concentrations, initial pH, and water matrix; (iii) to model the sulfonamide degradation increasing process optimization by response surface methodology (RSM); (iv) to identify the byproducts obtained and investigate the mechanism for sulfonamide degradation in these systems, and (v) to determine the cytotoxicity of the degradation products resulting from each system studied by application of line cells.

Experimental Procedure
The SAs degradation experiments with sulfite and different iron species [Fe(0), Fe(II), Fe(III), and Fe(VI)] were conducted in a cylindrical glass reactor with 1 L capacity. Most of the experiments were performed at pH 3 to guarantee the maximum concentration of radicals in the medium and, in removal of contaminants [28]. The effectiveness of a degradation process in water treatment also implies that the degraded byproducts do not harm the environment. In this context, most toxicological studies have used Vibrio fischeri bacteria tests [16,36], although the application of line cells is becoming more relevant as toxicity models in humans [37]. According to this background, AOTs were applied in the present study to degrade three different 4-aminobenzene sulfonamides (sulfamethazine, sulfadiazine, and sulfamethizole) by sulfite activation in the presence of Fe(0), Fe(II), Fe(III), or Fe(VI). As far as we know, this is the first time that these systems of water treatment are used to degrade the above sulfonamides. Thus, the main objectives were (i) to compare the treatment efficiencies of each system for sulfonamide removal; (ii) to investigate the effects of some operational conditions, such as sulfite concentration, iron concentrations, initial pH, and water matrix; (iii) to model the sulfonamide degradation increasing process optimization by response surface methodology (RSM); (iv) to identify the byproducts obtained and investigate the mechanism for sulfonamide degradation in these systems, and (v) to determine the cytotoxicity of the degradation products resulting from each system studied by application of line cells.

Experimental Procedure
The SAs degradation experiments with sulfite and different iron species [Fe(0), Fe(II), Fe(III), and Fe(VI)] were conducted in a cylindrical glass reactor with 1 L capacity. Most of the experiments were performed at pH 3 to guarantee the maximum concentration of radicals in the medium and, in

Experimental Procedure
The SAs degradation experiments with sulfite and different iron species [Fe(0), Fe(II), Fe(III), and Fe(VI)] were conducted in a cylindrical glass reactor with 1 L capacity. Most of the experiments were performed at pH 3 to guarantee the maximum concentration of radicals in the medium and, in the case of Fe(III), to avoid its precipitation. pH media were adjusted by the addition of HCl and NaOH (2N) as required in each case. SAs were first mixed with sulfite in constant agitation at room temperature (298 K), followed by the addition of different iron species. An initial sample was extracted to verify the initial concentration, and consecutive samples were withdrawn along time. One milliliter of hydroxylamine hydrochloride (1000 mg L −1 ) was added to quench the reaction. The samples were analyzed after centrifuging at 12,000 rpm for 10 min (Eppendorf Centrifuge 5424) to avoid sampling of possible iron oxide or hydroxide precipitates. Reaction follow-up time was set at 60 min. All samples were preserved in cold storage for further analysis. All experiments were carried out in triplicate.

Iron-Based/Sulfite Systems
To study the influence of the experimental variables, experiments were conducted by changing the iron species (1.79 × 10 −2 M) from Fe(0) to Fe(VI), whereas the sulfite and SAs concentrations were fixed at 5 × 10 −3 and 5.38 × 10 −5 M, respectively, also considering either the presence or absence of dioxygen for the Fe(0)/sulfite system. The influence of iron species concentration (5.37 × 10 −4 M-1.79 × 10 −3 M) was studied with initial concentrations for SAs and sulfite of 5.38 × 10 −5 M and 6.25 × 10 −5 M, respectively, whereas the influence of sulfite concentration (3.12 × 10 −3 M-6.25 × 10 −5 M) was performed using 5.38 × 10 −5 M of SAs and 1.79 × 10 −3 M for the corresponding iron species. In addition, considering the different iron species dominant at pH 3, we determined the SAs reaction rate constants as a function of the reaction rates observed in each case, according to the following Equation (29): k obs = k[Fe(n)] being n = 0, II, III or VI (29) where k obs is the reaction rate observed and corresponds to the slope of the straight line obtained when Ln(C/C 0 ) (C and C 0 are the final and initial SAs concentrations, respectively) is plotted against the reaction time, k (min −1 ) is the reaction rate constant with the different iron species, and [Fe(n)] is the initial concentration of the iron species added in each case. Iron and sulfonamide concentrations were chosen accordingly to properly follow the experiments and also improve their reproducibility.

Fe (VI) Determination in Aqueous Solution
The Fe(VI) concentration was determined using a UV-1600PC Spectrophotometer to ensure the stability of the standard solution, as mentioned by other authors [38]. The calibration curve was obtained with different Fe(VI) concentrations (1-50 ppm) obtained from a potassium ferrate stock solution (500 ppm). The absorbance was measured at a maximum absorption wavelength of λ max = 510 nm.

Determination of Sulfonamides in Aqueous Solution
SAs concentration was determined by reverse-phase high-performance liquid chromatography (HPLC), using a liquid chromatographer (Thermo-Fisher, Pittsburgh, USA) equipped with UV-visible detector and automatic autosampler with capacity for 120 vials. A PHENOMENEX Kinetex C 18 column was used (4.6 × 150 mm, 2.6 µm particle size). The mobile phase was 70% formic acid/H 2 O solution (0.1%, v/v) and 30% acetonitrile in isocratic mode at a flow of 0.35 mL min −1 . The detector wavelength was set at 270 nm.

Determination of Byproducts Degradation
Degradation byproducts were identified by using an Acquity ultra-pressure liquid chromatographer (UPLC) (Waters) equipped with a CORTECS™ C 18 column (2.1 × 75 mm, 2.7 µm) (Waters). The mobile phase in gradient mode (Baseline: 0% B, T8: 95% B, T8.1: 0% B) was Channel A, water with 0.1% formic acid and Channel B, acetonitrile with 0.1% formic acid, at a flow of 0.4 mL min −1 . Injection volume was 10 µL and column temperature 40 • C. The UPLC system was coupled to an SYNAPT G2 HDMS Q-TOF high-resolution mass spectrometer (Waters) equipped with electrospray ionization (ESI) source. Analysis parameters were determined in positive ionization mode, acquiring spectra in a mass range (m/z) between 50 and 1200 amu.

Determination of Byproducts Cytotoxicity
Degradation byproduct cytotoxicity was evaluated by using an MTS assay to determine the viability percentage of human embryonic kidney cells (HEK-293) and J774.2 cell line from BALB/C monocyte-macrophage provided by the CIC cell bank of the University of Granada. A number of 10,000 HEK-293 or J774.2 cells were incubated for 24 h, subsequently changing the medium and adding the SAs 60 min degraded samples; after incubation for 24 h more, 20 µL of MTS tetrazolium dye was added, and the absorbance at 490 nm was measured after 2 h using an INFINITENANOQUA, reading a given sample 9 times. All experiments were conducted in duplicate.

Collection and Characterization of Natural Waters
Groundwater (GW) and surface water (SW) were collected from well water and a river in Granada (Spain). The samples were characterized following the procedure mentioned elsewhere [36], filtrated, and cold-stored until used. The more representative parameters, namely pH, total organic carbon (TOC), and the concentration of anions (Cl − , NO 3 − , SO 4 2− , and HCO 3 − ) were measured.

Sulfonamide Degradation by the Fe(0)/Sulfite System
Table S1 (Supplementary Material) shows the results obtained for the Fe(0)/sulfite system at different concentrations of Fe(0) and sulfite. The degradation was negligible for the three SAs used when Fe(0) or sulfite alone were employed for a reaction time of 60 min. However, in the presence of both species together, the degradation percentages for SMZ, SDZ, and SML were 56.3%, 59.1%, and 62.1%, respectively (Table S1, Exp. 10, 20, and 30). These experiments were conducted with the same concentration of Fe(0) and sulfite. Therefore, differences in degradation percentage obtained can be attributed to differences in the molecular structure of the different SAs used.
To verify the role of dissolved molecular oxygen in SAs degradation, we selected the system Fe(0)/sulfite. The SMZ degradation kinetics obtained the Fe(0)/sulfite couple was conducted using deoxygenated (nitrogen-purged) and naturally oxygenated (non-purged) water. Figure 1 depicts the results obtained at pH 3. In deoxygenated water, SMZ can be degraded by only 10%; this value is much lower than that obtained when the process was carried out in oxygenated water (56.3%). Consequently, the key role of dioxygen during the degradation process is then confirmed. Previous studies found that Fe(0) in the presence of dioxygen in acidic media can generate reactive species, such as HO • radicals, and ferrate through Reactions (30)-(33) [39]. At the same time, the sulfate radical can react with water to form HO • (Reaction (34)).

Sulfonamide Degradation by the Fe(II)/Sulfite System
To investigate the degradation of SAs in aqueous solution by the Fe(II)/sulfite system, several experiments were carried out at pH 3 varying the dose of Fe(II) and sulfite (Table S2). It was found that Fe(II)/sulfite system can achieve good oxidation of all SAs by means of FeSO3 + species (Reactions (5), (6), and (28)) ( Table S2, Exp. 39, 48, and 57), since Fe(II) could be easily oxidized to Fe(III) in the presence of O2 (Reaction (5) and (35)).
On the other hand, in the presence of an excess of sulfite, the rapid formation of Fe(III) is followed by a slower redox process, during which Fe(II) oxidation is independent of its concentration [40]. SO3 •− radicals, which could be produced by Fe(II) and Fe(III) reaction with sulfite (Reactions (4)-(28)), should then raise the sulfonamide degradation.

Sulfonamides Degradation by the Fe(III)/Sulfite System
Degradation of SAs by the Fe(III)/sulfite system is shown in Table S3. Comparing SDZ, SMZ, and SML degradations for different conditions, SDZ is the most recalcitrant and is only degraded 28.0% (Table S3, Exp. 75). Fe(III) first serves as a coordinating metal cation leading to the formation of the FeSO3 + complex with SO3 2− , and as a radical initiator for SO3 •− , SO5 •− , and SO4 •− accompanied by the generation of Fe(II) [29]. The continuous generation of SO4 •− and HO • is largely subject to the cyclic redox process of Fe(II)/Fe(III) (Reactions (4)-(23)) [34], and could not be studied separately. Wang et al. [41] indicated that the rapid conversion of Fe(II) to Fe(III) limits the ultimate oxidizing capability. Consequently, the Fe(III) concentration has a significant influence on the degradation of SAs (Table S3).

Sulfonamide Degradation by the Fe(II)/Sulfite System
To investigate the degradation of SAs in aqueous solution by the Fe(II)/sulfite system, several experiments were carried out at pH 3 varying the dose of Fe(II) and sulfite (Table S2). It was found that Fe(II)/sulfite system can achieve good oxidation of all SAs by means of FeSO 3 + species (Reactions (5), (6), and (28)) (  (5) and (35)).
On the other hand, in the presence of an excess of sulfite, the rapid formation of Fe(III) is followed by a slower redox process, during which Fe(II) oxidation is independent of its concentration [40]. SO 3 •− radicals, which could be produced by Fe(II) and Fe(III) reaction with sulfite (Reactions (4)-(28)), should then raise the sulfonamide degradation.

Sulfonamides Degradation by the Fe(III)/Sulfite System
Degradation of SAs by the Fe(III)/sulfite system is shown in  (23)) [34], and could not be studied separately. Wang et al. [41] indicated that the rapid conversion of Fe(II) to Fe(III) limits the ultimate oxidizing capability. Consequently, the Fe(III) concentration has a significant influence on the degradation of SAs (Table S3).

Sulfonamides Degradation by the Fe(VI)/Sulfite System
Results shown in Figure 2 pointed out that the degradation rate of Fe(VI)/sulfite system is not only the highest but also the more efficient compared with the other studied systems. Indeed, SAs can be degraded in the first 5 min. Owing to, in acidic media, the redox potential of Fe(VI) species (E • = +2.20 V) is the highest of the common oxidants used in water treatment [42][43][44][45]. To determine the efficiency of Fe(VI)/sulfite system on the SAs degradation, several experiments were carried out (

Sulfonamides Degradation by the Fe(VI)/Sulfite System
Results shown in Figure 2 pointed out that the degradation rate of Fe(VI)/sulfite system is not only the highest but also the more efficient compared with the other studied systems. Indeed, SAs can be degraded in the first 5 min. Owing to, in acidic media, the redox potential of Fe(VI) species (E° = +2.20 V) is the highest of the common oxidants used in water treatment [42][43][44][45]. To determine the efficiency of Fe(VI)/sulfite system on the SAs degradation, several experiments were carried out (Table S4), concluding that sulfite could be activated by Fe(VI) to completely degrade sulfonamides. For instance, for a fixed Fe(VI) concentration (1.79 × 10 −2 M), an increase in the sulfite concentration from 5 × 10 −3 to 1 × 10 −2 M (Table S4, Exp. 92-93), enhanced the degradation percentage from 83.1% to 100%.

Comparison of Different Iron-Based/Sulfite Systems on the SAs Removal
To verify the feasibility of Fe(0─VI)/sulfite systems as degradation agents for sulfonamides, sulfamethazine (SMZ) was used as a model compound. The oxidation of SMZ by sulfite in the presence of iron along its different oxidation states is shown in Figure 2. In the sole presence of sulfite, the removal of SMZ observed is negligible. In most systems, the maximum degradation is almost achieved within 10 min, and a very rapid degradation occurred in the first 5 min of reaction. After 60 min of reaction, the degradation percentages were 56.3%, 43.7%, 35.7%, and 86.9% for Fe(0)/sulfite, Fe(II)/sulfite, Fe(III)/sulfite, and Fe(VI)/sulfite, respectively. Consequently, the Fe(VI)/sulfite one presents the greatest removal and, moreover, it is considered as a green oxidant for water and wastewater treatment [42,43]. The redox potential of Fe(VI) is significantly higher (+2.20 V) in an acid medium than in an alkaline medium (+0.72V) [44]; thus, under the studied pH (pH 3), SO4 •− can be produced through a process similar to Reactions (36)-(38) [45]. Thus, SO3 •− generated combines with oxygen to form SO5 •− (Reactions (8)- (9)) that enables the formation of SO4 •− . The formed SO4 •− quickly reacts with water to produce hydroxyl radicals through Reaction (39).

Comparison of Different Iron-Based/Sulfite Systems on the SAs Removal
To verify the feasibility of Fe(0-VI)/sulfite systems as degradation agents for sulfonamides, sulfamethazine (SMZ) was used as a model compound. The oxidation of SMZ by sulfite in the presence of iron along its different oxidation states is shown in Figure 2. In the sole presence of sulfite, the removal of SMZ observed is negligible. In most systems, the maximum degradation is almost achieved within 10 min, and a very rapid degradation occurred in the first 5 min of reaction. After 60 min of reaction, the degradation percentages were 56.3%, 43.7%, 35.7%, and 86.9% for Fe(0)/sulfite, Fe(II)/sulfite, Fe(III)/sulfite, and Fe(VI)/sulfite, respectively. Consequently, the Fe(VI)/sulfite one presents the greatest removal and, moreover, it is considered as a green oxidant for water and wastewater treatment [42,43]. The redox potential of Fe(VI) is significantly higher (+2.20 V) in an acid medium than in an alkaline medium (+0.72V) [44]; thus, under the studied pH (pH 3), SO 4 •− can be produced through a process similar to Reactions (36)-(38) [45]. Thus, SO 3 •− generated combines with oxygen to form SO 5 •− (Reactions (8)- (9)) that enables the formation of SO 4 •− . The formed SO 4 •− quickly reacts with water to produce hydroxyl radicals through Reaction (39).
Water 2019, 11, 2332 9 of 18 Among the reasons mentioned above (Section 2.2), we have also selected pH 3 in all experiments because, from Fe(0) at acidic conditions in dioxygen atmosphere, Fe(II) is generated, which is quite stable under these conditions [46,47]. Therefore, the stability of the species Fe(II)/Fe(III), which usually appears cyclically during these reactions, is maintained. Thus, the removal of SMZ could be attributed to the different radical species formed including SO 4 [48,49]. These results confirmed that sulfite could be activated by iron-based species to accelerate the degradation of recalcitrant contaminants that show a sluggish reactivity with sulfite alone.

Optimization of SAs Degradation Process Conditions
Response surface methodology (RSM) is an effective method to optimize process conditions, and it can determine the influence of various factors and their interactions on the indexes under investigation (response value) during technological operations. It can be used to fit a complete quadratic polynomial model through a central composite experiment, and it can lead to a more adequate experiment design and result expression [50].
The effect of both sulfite (x) and iron-base (y) concentrations on percentage removal of SAs (f(x,y)) was assessed by using a three-level full factorial design with one center point. A total of nine experiments were generated, taking the percentage removal of SAs as response and the concentrations of sulfite and iron-base as independent variables. The experimental responses were fitted to the equations given in Table S5.
Results obtained for our systems showed a similar surface response for the three SAs (SMZ, SDZ, and SML). In Figure 3, as an example, only the Fe(0)/sulfite system is depicted. As it can be seen, the maximum SAs degradation is obtained for concentrations of Fe(0) and sulfite of about 10 −2 and 5 × 10 −3 M, respectively, except for SMZ, for which the slope rises with the sulfite concentration without presenting a maximum in the range of the examined concentrations. The response surfaces for the rest of the systems, namely Fe(II), Fe(III), and Fe(VI), are depicted in Figures S1-S3 in the Supplementary Material.
From the data, the dependence of SAs degradation percentage with sulfite and iron-based concentration (f(x,y)) can be calculated using the equations showed in Table S5, which allow us to improve the design and reaction conditions for any system, for example, determining a desired degradation percentage if necessary. Table 2 shows the chemical composition of the water matrix used in this study, and Figure 4 depicts the SAs degradation by using the different iron-based/sulfite systems and types of water. According to Figure 4, the degradation percentage diminishes for all the iron-based systems following the order UW > SW >> GW. However, whereas there is a slight decrease in the degradation of SAs for the Fe(0)-and Fe(VI)-sulfite systems, this variation is more acute for the Fe(III)-sulfite system. This is due, in part, to the partial precipitation of Fe(III) species at the high pH value observed in the GW matrix (Table 2). presenting a maximum in the range of the examined concentrations. The response surfaces for the rest of the systems, namely Fe(II), Fe(III), and Fe(VI), are depicted in Figures S1-S3 in the Supplementary Material.

Influence of the Water Matrix on Iron-Based/Sulfite Systems
From the data, the dependence of SAs degradation percentage with sulfite and iron-based concentration (f(x,y)) can be calculated using the equations showed in Table S5, which allow us to improve the design and reaction conditions for any system, for example, determining a desired degradation percentage if necessary.  Table 2 shows the chemical composition of the water matrix used in this study, and Figure 4 depicts the SAs degradation by using the different iron-based/sulfite systems and types of water. According to Figure 4, the degradation percentage diminishes for all the iron-based systems following the order UW > SW >> GW. However, whereas there is a slight decrease in the degradation of SAs for the Fe(0)-and Fe(VI)-sulfite systems, this variation is more acute for the Fe(III)-sulfite  (40)-(53)) [38,51], diminishing the degradation percentage. However, this diminution is less intense than expected, probably due to new radicals appearing in the scavenging reactions (Reactions (40)-(53)). These secondary radicals, despite being less oxidant, have a longer half-life, which increases their possibility of reaction with pollutant molecules.

Degradation Byproducts and Pathways
Six major products were found in HPLC/MS analysis after treatment with the iron-based/sulfite systems (Table S6). Owing to the fact that all the systems generate some kind of byproduct, it is more adequate to pay attention to their retention times. Therefore, regarding the retention time (rt) values and correlated masses found, it could be found that lighter degradation compounds were found for SDZ, and the heavier (unique product detected also accompanied by a higher rt value) was found for SML. In principle, this could well correlate with a higher degradation for SDZ, leading to smaller molecules. Casually, among these sulfonamides, SDZ has a simpler structure, which is likely related to its lower resistance against radical attack, given the structural similarities between all of them. Furthermore, the aromatic thiadiazole moiety present in SML is relatively stable in acid media (not in basic solutions) in part due to its π-deficient character that enhances the occurrence of nucleophilic aromatic substitution processes. Thus, detected byproducts ordered by rt were: (i) From SDZ (molecular mass: 250.28), at 1.14 min, a peak with m/z value of 209.0780 u in full scan positive-mode ESI (M + H + ) was detected, corresponding with a mass of 208.2355 u, meaning a loss of 42 u. Basically, this mass could be assigned to the loss of an oxygen atom of the sulfamide group to give a sulfoxide and the breaking of the pyrimidine moiety to loss two carbon atoms (C 2 H 2 ) followed by a ring-closure to form a diazete moiety ( Figure 5).

Degradation Byproducts and Pathways
Six major products were found in HPLC/MS analysis after treatment with the iron-based/sulfite systems (Table S6). Owing to the fact that all the systems generate some kind of byproduct, it is more adequate to pay attention to their retention times. Therefore, regarding the retention time (rt) values and correlated masses found, it could be found that lighter degradation compounds were found for SDZ, and the heavier (unique product detected also accompanied by a higher rt value) was found for SML. In principle, this could well correlate with a higher degradation for SDZ, leading to smaller molecules. Casually, among these sulfonamides, SDZ has a simpler structure, which is likely related to its lower resistance against radical attack, given the structural similarities between all of them. Furthermore, the aromatic thiadiazole moiety present in SML is relatively stable in acid media (not in basic solutions) in part due to its π-deficient character that enhances the occurrence of nucleophilic aromatic substitution processes. Thus, detected byproducts ordered by rt were: (i) From SDZ (molecular mass: 250.28), at 1.14 min, a peak with m/z value of 209.0780 u in full scan positive-mode ESI (M + H + ) was detected, corresponding with a mass of 208.2355 u, meaning a loss of 42 u. Basically, this mass could be assigned to the loss of an oxygen atom of the sulfamide group to give a sulfoxide and the breaking of the pyrimidine moiety to loss two carbon atoms (C2H2) followed by a ring-closure to form a diazete moiety ( Figure 5). (ii) From SDZ, at 1.90 min, a peak with m/z value of 187.0974 u was detected, corresponding to a mass of 186.2132 u, meaning a loss of 64 u. This mass could be assigned to the loss of SO2, which could be considered as an extrusion-like process, although another likely process consists of the breaking of the pyrimidine moiety to give an N-methyl sulfonamide ( Figure 5). be considered as an extrusion-like process, although another likely process consists of the breaking of the pyrimidine moiety to give an N-methyl sulfonamide ( Figure 5).
(iii) From SDZ, at 3.48 min, a peak with m/z value of 189.1124 u was detected, corresponding to a mass of 188.2291 u, meaning a loss of 62 u. This mass could be assigned to the loss of the pyrimidine moiety and the hydroxylation of the remaining compound to give a hydroxylamine derivative. This nicely explains the high change in the rt between this byproduct and the previous case, despite the similar mass between them.
(iv) From SMZ (molecular mass: 278.33), at 3.53 min, a peak with m/z value of 215.1283 u was detected, corresponding to a mass of 214.2664 u, meaning a loss of 64 u. Again, a loss mass that could be assigned to the loss of SO 2 , or, more likely, to follow a similar mechanism to the first case involving the breaking of the pyrimidine ring to give a quite stable guanidine moiety ( Figure 5).
(v) From SMZ, at 3.94 min, a peak with m/z value of 359.0470 u was detected, corresponding to a mass of 358.3934 u, meaning a gain of 80 u, which fairly corresponds to the incorporation of five hydroxyl groups leading to the corresponding perhydroxylated SMZ derivative ( Figure 5). This could give us a good indication about the ultimate mechanism operating on these derivatives under experimental conditions, since radical sulfate initiates the radical formation on the aromatic rings of sulfamides as π electrons are energetically more accessible, and, finally, hydroxyl radicals trap these organic radicals leading to the corresponding hydroxylated derivatives.
(vi) From SML (molecular mass: 270.33), at 5.63 min, a peak with m/z value of 301.0067 u was detected, corresponding to a mass of 300.3142 u, meaning a gain of 30 u. As in the previous case, this could be assigned to the incorporation of two oxygen atoms under their carbonyl form, which could be ascribed to the phenyl moiety (in this case the thiadiazole moiety is more stable) in opposite positions to facilitate the electronic delocalization trough the corresponding quinonic-like form ( Figure 5).
Regarding the different iron-based systems, in the Fe(0), Fe(II), and Fe(III)/sulfite systems, a peak with m/z of 214 was detected at retention time (rt) of 3.5 min. These byproducts were commonly founded in degradation systems of SAs as Fenton, UV/persulfate, or gamma irradiation-Fe(II) [52][53][54]. Figure 6 shows the percentage of cellular viability of HEK293 and J774 cells as a function of SAs type in different systems. Neither SAs byproducts nor the SAs themselves exhibited inhibition of the growth of human embryonic cells (HEK293) compared to controls (100% of viability). A previous study [55,56] showed that a drug mix, including sulfamethoxazole as SAs, inhibited 30% of HEK293 cell proliferation compared to the control sample. According to the percentage of cellular viability, the byproducts obtained in this study do not present cytotoxicity to cell HEK293. To establish the relative toxicity of SAs compounds, mouse macrophages cells (J774) were also used as a model cell type. The J774 macrophage cell line was chosen due to proven applicability in particulate cytotoxicity and intracellular drug effectiveness studies [57]. For this cell line the percentage cellular viability was less than 75% for the SMZ:Fe(III)/sulfite systems with 74.92% (if the cell viability is less than 75%, the substance is considered to exert a toxic effect on the cells and, if it is greater than 75%, the substance does not exert a toxic effect on the cells). However, most of the byproducts were not cytotoxic to this type of cell, and, therefore, the systems used in this work are viable to use. Water 2019, 11, x FOR PEER REVIEW 14 of 18

Conclusions
Based on this study, we can conclude that the effectiveness of the iron species to activate sulfite anion to promote the degradation of SAs increased following this order: Fe(III) < Fe(II) < Fe(0) < Fe(VI). In fact, the Fe(VI)/sulfite system showed the highest effectiveness being able to completely degrade SAs in a very short period of time (5 min) at acidic pHs (pH 3).
Iron species with sulfite could subsequently form SO3 •− which can be converted to SO4 •− in an oxygen atmosphere. The ability to degrade SAs in deoxygenated water diminishes drastically, confirming oxygen's key role to generate reactive species.
Applying the response surface methodology, we have obtained quadratics equations for the systems studied in this research. This mathematical modeling allows calculating the degradation percentage based on a bidimensional grid containing the doses of iron and sulfite added to the system.
A remarkable influence of the water matrix was observed in this study. Consequently, bicarbonate, halide ions, or sulfate present in wastewater and groundwater were able to inhibit the SAs degradation.
Six major products were identified through HPLC/MS analysis for SMZ, SDZ, and SML. The general oxidation pathways proposed are the elimination of SO2, C-N bond cleavage on the aromatic ring, different atomic reorganization processes, and hydroxylation reactions promoted by HO• radicals. None of these byproducts inhibited the growth of cell lines HEK293 and J774. Consequently,

Conclusions
Based on this study, we can conclude that the effectiveness of the iron species to activate sulfite anion to promote the degradation of SAs increased following this order: Fe(III) < Fe(II) < Fe(0) < Fe(VI). In fact, the Fe(VI)/sulfite system showed the highest effectiveness being able to completely degrade SAs in a very short period of time (5 min) at acidic pHs (pH 3).
Iron species with sulfite could subsequently form SO 3 •− which can be converted to SO 4 •− in an oxygen atmosphere. The ability to degrade SAs in deoxygenated water diminishes drastically, confirming oxygen's key role to generate reactive species. Applying the response surface methodology, we have obtained quadratics equations for the systems studied in this research. This mathematical modeling allows calculating the degradation percentage based on a bidimensional grid containing the doses of iron and sulfite added to the system.
A remarkable influence of the water matrix was observed in this study. Consequently, bicarbonate, halide ions, or sulfate present in wastewater and groundwater were able to inhibit the SAs degradation.
Six major products were identified through HPLC/MS analysis for SMZ, SDZ, and SML. The general oxidation pathways proposed are the elimination of SO 2 , C-N bond cleavage on the aromatic ring, different atomic reorganization processes, and hydroxylation reactions promoted by HO• radicals. None of these byproducts inhibited the growth of cell lines HEK293 and J774.
Consequently, the proposed system Fe(VI)/sulfite is an economic, readily synthesized, and easily handled material with good properties for utilization in SAs degradation in contaminated water.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4441/11/11/2332/s1, Figure S1. Response surface obtained for the degradation of SAs as a function of Fe(II) and sulfite concentrations.