Mineralization of sulfonamides from wastewater using ozone-based systems

Pharmaceutical compounds, and antibiotics in particular, are contaminants of serious concern in wastewater. They can be detrimental to human health when they reach agricultural soils or water sources owing to the formation of antibiotic-resistant bacteria and genes. Most antibiotics come from the pharmaceutical industry, hospitals, the aquaculture industry and the livestock industry. The continuous discharge of antibiotics even at low concentrations poses a potential threat to the environment and they have been detected in aquatic environments such as lakes, rivers, groundwater, and wastewater (Ngan et al. 2018). Li et al. (2013) detected 22 antibiotics in a municipal wastewater reclamation plant in Beijing, China. The three kinds of antibiotic that were present in greatest quantities were quinolones, sulfonamides, and macrolides with concentrations of 4,916, 2,911, and 365 ng/L, respectively, in the influent of the plant. The influent concentrations of sulfadiazine (SDZ), sulfamerazine (SMR) and sulfamethazine (SMZ) in a wastewater treatment plant (WTP) in Canada were found to be 42, 9.3 and 45 ng/L (Guerra et al. 2014), respectively; and those of SMZ and SMR in a WTP in Korea were 164 and 48 ng/L, respectively, while the corresponding effluent concentrations were 144 and 33 ng/L (Behera et al. 2011; Sim et al. 2013). Unfortunately, the efficiency of treatment of antibiotics in the WTP was limited by their antibacterial property, resulting in their direct discharge into the environment (Villar-Navarro et al. 2018; Osinska et al. 2020). Antibiotics in the environment may lead to the generation of an antibiotic-resistance gene, which can spread by horizontal gene transfer in microorganisms. Antibiotic-resistant bacteria or multi-antibiotic-resistant bacteria are thus generated (Pehrsson et al. 2016), posing a significant threat to human health and the ecosystem (Pruden et al. 2006; Hvistendahl 2012). Hence, SDZ, SMR and SMZ were used as the target compounds in this study.

Sulfonamides are extremely difficult to eliminate by coagulation, sedimentation, or filtration (Westerhoff et al. 2005). Accordingly, developing economical and efficient methods for degrading sulfonamides in wastewater has attracted considerable attention. Effective advanced treatment processes are needed to prevent the entry of sulfonamides into the natural environment. Advanced oxidation processes (AOPs) can generate highly reactive radicals, such as hydroxyl (HO•), sulfate (⁠⁠) and superoxide (⁠⁠) radicals, which can degrade sulfonamides and other refractory organic pollutants into biodegradable products with low toxicity (Velo-Gala et al. 2017; Sharma & Feng 2019). Klauson et al. (2019) used heterogeneous photocatalysis, ozonation, and both homogeneous and heterogeneous Fenton-like treatment to degrade sulfamethizole and found that their ozone-based processes had the greatest sulfamethizole decomposition efficiency. The effect of ozone can be improved by the presence of ultraviolent (UV) radiation (Silva et al. 2019) or catalysts, which may be heterogeneous and/or homogeneous (Kasprzyk-Hordern et al. 2003). These external approaches can also promote the decomposition of molecular ozone to HO• or other reactive oxidative radicals. Persulfate (PS) is low cost, is easy to use and exhibits oxidant stability in the ambient environment and it can be activated to form ⁠, which is strongly oxidizing. Therefore, PS was used here as the oxidant that is added to ozone-based systems to promote their oxidation efficiency.

Batista et al. (2014) used different processes (UV photolysis, UV/H₂O₂ and photo-Fenton) to degrade SDZ, SMR, and SMZ. Yang et al. (2018) found that the incomplete mineralization of sulfonamides increases their acute toxicity. Therefore, the mineralization efficiency of sulfonamides is very important. The mineralization of SDZ, SMR and SMZ by ozone-based systems has not yet been thoroughly investigated. This study concerns the mineralization of SDZ, SMR, and SMZ from aqueous solution using ozone, UV/ozone, ozone/PS and UV/ozone/PS systems. The objectives of this study were (i) to compare the mineralization efficiencies of SDZ, SMR and SMZ in various ozone-based systems; (ii) to evaluate the influences of PS addition on oxidation efficiency, and (iii) to determine the effects of pH and salinity on SMZ mineralization in UV/ozone and UV/ozone/PS systems.

The sulfonamides SDZ, SMR and SMZ were all purchased from Alfa Aesar. Table 1 presents their physiochemical properties. These three sulfonamides are of the same pharmacophore group. Sodium persulfate (Na₂S₂O₈) and phosphoric acid (H₃PO₄) were purchased from Sigma-Aldrich. Salinity was provided by NaCl, which was obtained from Taiyen. The pH of the solution was controlled by adding HNO₃ and NaOH using an automatic titrator, and both of these substances were purchased from Merck. All reagents were of analytical grade and used without further purification. Deionized water (D.I. water) was used throughout this study.

Ozone was produced using a corona discharge ozone generator (250 W, Ozone Solutions TG-20) with oxygen as the feed gas. Before the experiment, 1,500 mL of D.I. water was added to the reactor and then ozonized for 30 min, by which time a constant ozone concentration in aqueous solution was reached. Then, 500 mL sulfonamide solution was added to the reactor. Ozone gas was continuously fed into the reactor at a constant flow rate of 3.353 L/min at 298 K. The residual ozone concentration in the solution was measured by the indigo colorimetric method (Method 4500) (APHA 1992). Since the most abundant dissolved ions in seawater are sodium and chloride, NaCl was added to the solution to enable a photocatalytic study at the same salinity level (3.5%) as in seawater to elucidate the effect of salinity on SMZ mineralization. The initial concentrations of sulfonamide, PS, and NaCl were 20 mg/L, 5 mM, and 36.27 g/L, respectively. Mineralization experiments were performed in a 3 L hollow cylindrical glass reactor. An 8 W UV lamp (254 nm, 1.12 W/m², Philips) was placed inside a quartz tube as the light source. All systems were stirred continuously at 300 rpm and the temperature was maintained at 298 K. Aliquots with volume of 20 mL were withdrawn from the photoreactor at pre-specified intervals. The oxidant Na₂S₂O₈ and the acidifier H₃PO₄ were placed in an O.I. 1010 total organic carbon (TOC) analyzer. The decrease in TOC therein yielded the degree of mineralization of the sulfonamides. All experiments were conducted in duplicate, and mean values were reported.

Figure 1 presents the mineralization of sulfonamides in various ozone-based systems. After 120 min of reaction, the mineralization percentages of SDZ, SMR and SMZ in the ozone system (Figure 1(a)) were 52, 38 and 25%, respectively; those in the UV/ozone system (Figure 1(b)) were 79, 77 and 74%, respectively; those in the ozone/PS system (Figure 1(c)) were 57, 50 and 31%, respectively; and those in the UV/ozone/PS system (Figure 1(d)) were 97, 97 and 98%, respectively. The mineralization of these sulfonamides approximately followed pseudo-first-order kinetics, expressed as ln(C_t/C₀) = −kt, where t is reaction time, k is the pseudo-first-order rate constant, and C₀ and C_t are the concentrations of TOC at times t = 0 and t = t, respectively. Table 2 presents the pseudo-first-order mineralization rate constants, and their correlation coefficients, of three sulfonamides in the ozone-based systems. For SDZ, the k values in ozone, UV/ozone, ozone/PS and UV/ozone/PS systems were 0.0144; 0.0195; 0.0156 and 0.0902 min⁻¹, respectively; for SMR they were 0.0101; 0.0130; 0.0129 and 0.0837 min⁻¹, respectively; and for SMZ, they were 0.0058; 0.0101; 0.0069 and 0.0802 min⁻¹, respectively. Various studies have also shown that the degradation rates of antibiotics fit pseudo-first-order kinetics (Adil et al. 2020; Zhang et al. 2020).

In the ozone/PS system, the reactions that are specified by Equations (1) and (6)–(12) proceeded simultaneously in aqueous solution. In the UV/ozone/PS system, the reactions of Equations (1)–(5) proceeded simultaneously in aqueous solution. Figure 2 plots the residual ozone concentration in the ozone-based systems during SMZ mineralization. After the first 30 min of aeration of D.I. water with ozone, the dissolved ozone concentrations at pH 5 and pH 7 were 5.7–6.8 and 4.6–5.4 mg/L, respectively. After SMZ was added and allowed to react for 120 min, the dissolved ozone concentrations in the ozone, UV/ozone, ozone/PS and UV/ozone/PS systems at pH 5 were 6.0, 0.7, 5.6 and 0.3 mg/L (Figure 2(a)), respectively; those at pH 7 were 4.6, 0.6, 4.9 and 0.1 mg/L, respectively (Figure 2(b)). Under UV irradiation, ozone decomposed to form HO•, so the residual ozone concentrations in the UV/ozone and UV/ozone/PS systems were low. Based on an analysis of the reaction equations, the major radicals that mineralized sulfonamides in UV/ozone/PS are suggested to have been and and those in the UV/ozone system are suggested to have been HO•. Combining PS with ozone yielded additional in the ozone/PS system. The mineralization rates of all tested sulfonamides followed the order UV/ozone/PS > UV/ozone > ozone/PS > ozone (Table 2).

The chemical structures of SDZ, SMR and SMZ can be distinguished by the function-groups that are bonded to a pharmacophore-group. The function-groups in SDZ, SMR and SMZ all contained a pyrimidine ring. Moreover, the pyrimidine rings in SMR and SMZ bonded to one and two methyl groups, respectively. Yang et al. (2018) stated that the substituted groups of sulfonamides significantly affected their degradation rate. The Kow value of an organic compound is an important parameter that describes its hydrophilicity. The Kow values here followed the order SMZ > SMR > SDZ (Table 1). The degrees of hydrophilicity followed the order SDZ > SMR > SMZ, which is related to the bonding of methyl groups to the heterocyclic group. The heterocyclic group of SMZ contained more methyl substituents than those of SMR and SDZ so SMZ exhibited the greatest steric hindrance of radical addition and therefore the lowest degradation rate (Batista et al. 2014). The mineralization rates of sulfonamides followed the order SDZ > SMR > SMZ in ozone-based systems (Table 2), as also observed by Batista et al. (2014) (UV/H₂O₂ and photo-Fenton systems). Yin et al. (2018) used peroxymonosulfate to degrade sulfonamides and obtained degradation rates that followed the order SDZ > SMR > SMZ> sulfathiazole (STZ) > SMX. Zhang et al. (2016) used direct photolysis to degrade sulfonamides and found degradation rates that followed the order STZ > SMZ > SMR > SDZ. This study suggests that the photodegradation rate varies with the oxidation system and the sulfonamide species.

Solution pH of aqueous solution importantly affects the predominant radical species and the speciation of the sulfonamide. Figure 3 presents the mineralization of SMZ in various ozone-based systems at various pHs. After 120 min reaction, the SMZ mineralization percentages in ozone, UV/ozone, ozone/PS and UV/ozone/PS at pH 5 (Figure 3(a)) were 25%, 74, 31 and 98%, respectively; moreover, those at pH 7 (Figure 3(b)) were 56, 75, 60 and 98%, respectively. Ozone is a powerful oxidant for degrading refractory compounds and can function in direct pathway through molecular ozone or in an indirect pathway that involves the production of by ozone decomposition (Kasprzyk-Hordern et al. 2003). pH can significantly affect the pathway action of ozone because at a pH of less than 7, direct action dominates, while at a pH of greater than 7, is the predominant species.

Sulfonamides have two dissociation constants, one of which corresponds to protonation of the aniline N and the other of which is associated with the protonation of sulfonamide. Owing to its two pK_a-values, SDZ, SMR and SMZ had cationic, neutral or anionic forms (Table 1). However, under environmental conditions, only the latter two species are of interest. Deprotonation of the sulfonamide nitrogen (pH > pK_a2) reportedly facilitates its oxidation in ozonation due to the weakening of the electron-withdrawing effect of the sulfonamide S–N bond (Dodd et al. 2006). Protonated sulfonamides exhibit reduced reactivity toward electrophilic radicals, but deprotonated sulfonamides are easily oxidized by reactive radicals (Dodd et al. 2006). Furthermore, the fast degradation at basic pH may also be ascribed to the high formation efficiency (Kolthoff & Miller 1951). SMZ is predominantly protonated at its amine moiety at pH < 2.3 (pK_a1); it is neutral at pH 2.3–7.4 (pK_a2), and it is anionic at pH >7.4 owing to deprotonation of the –SO₂NH– group (Table 1). Protonated SMZ exhibits reduced reactivity with electrophilic radicals, but deprotonated SMZ is readily oxidized (Dodd et al. 2006). Hence, in all of the tested ozone-based systems, the SMZ mineralization rates followed the order pH 7 > pH 5. Adil et al. (2020) found that the mineralization efficiency of trimethoprim increased with the pH in ozone and ozone/PS systems; however, that of sulfamethoxazole decreased. The present research suggests that the effects of pH on the mineralization of sulfonamide by AOPs varied with the species of oxidant and sulfonamide.

Ozone, UV/ozone, ozone/PS and UV/ozone/PS systems were used to mineralize SDZ, SMR and SMZ. According to the reaction equations, the major radicals that mineralized the sulfonamides in UV/ozone/PS were and ⁠; those in UV/ozone system were HO•. Combining PS with ozone produced additional in the ozone/PS system. The sulfonamide mineralization rates followed the order UV/ozone/PS > UV/ozone > ozone/PS > ozone. Adding PS reduced the EEO value in ozone-based systems. The mineralization rates of sulfonamides followed the order SDZ > SMR > SMZ in ozone-based systems and pH 7 > pH 5. Chloride ions reduced the SMZ mineralization rates in UV/ozone and UV/ozone/PS because they scavenged radicals.

The authors would like to thank the Ministry of Science and Technology and National Kaohsiung University of Science and Technology, for financially supporting this research under Contract No. MOST 109-2221-E-992-050 and 109B01, respectively. Mr Ni of National Kaohsiung University of Science and Technology is appreciated for performing some of the experiments.

The authors have no conflicts of interest to declare that are relevant to the content of this article.

All relevant data are included in the paper or its Supplementary Information.