Introduction

With the rapid development of modern industry, more and more organic micro-pollutants have been frequently detected in aquatic environment (Shao et al. 2009; Tobergte and Curtis 2013). Most micro-pollutants have strong water solubility, long half-life, and bioaccumulation (Richardson and Ternes 2005; Huerta-Fontela et al. 2011). These organic micro-pollutants are of great concern because of their potential damage to water environment and human health (Richardson and Ternes 2005). It was pointed out that prolonged exposure to a common pesticide produced behavioral and neuropathological features of Parkinson’s disease, and musty odor compounds in drinking water greatly affected the esthetic quality of drinking water and consumer acceptability (Betarbet et al. 2000; Sung et al. 2005; Peter and Von Gunten 2007). Furthermore, some endocrine disruptors have been found in human breast milk, which has adverse effects on human reproduction, immunity, nerve development, and function (Main et al. 2005). Thus, the effective degradation of organic micro-pollutants in water treatment processes has become an important issue.

Conventional water and wastewater treatment processes (e.g., coagulation, filtration, and biological treatment) were shown to be unable to remove these micro-pollutants effectively (Huerta-Fontela et al. 2011). Fortunately, advanced oxidation processes (AOPs) have proven effective in water treatment for the removal of micro-pollutants (Haag and Yao 1992; Ahmed and Chiron 2014; Luo et al. 2015). The conventional AOPs rely on in situ generation of highly reactive and non-selective radicals (i.e., HO•) for the rapid oxidation of a wide range of organic contaminants. The AOPs include processes such as UV/H2O2, Fe2+/H2O2, and ozone. (Can and Çakır 2010; Yang et al. 2010; He et al. 2012; Rodríguez-Chueca et al. 2016). Many pharmaceuticals react with HO• at a secondary reaction rate of 108 to 1010 M−1 s−1, and the main mechanism is hydrogen abstraction, electrophilic addition, and electron transfer reactions (Haag and Yao 1992; Huber et al. 2003; Deng et al. 2013).

Recently, sulfate radical (SO4•−)–based AOPs have attracted significant attention for the degradation of various organic pollutants (Lutze et al. 2015). SO4•− is usually generated from activation of persulfate by heat, ferrous iron catalysis, and UV light (Yang et al. 2010; Bennedsen et al. 2012; Kattel et al. 2017). Compared with the non-selective HO• (1.8–2.8 V vs.NHE), SO4•− has a higher oxidizing power (2.5–3.1 V vs. NHE) and longer half-life (Luo et al. 2015). SO4•− degrade target pollutants mainly through electron transfer reactions, hydrogen abstraction, and addition mechanisms. However, the selectivity of SO4•− through electron transfer is stronger than HO• (Xu and Li 2010; Liu et al. 2012). According to some recent reports, SO4•− react with certain organic compounds at a higher rate than HO• (Xie et al. 2015; Liu et al. 2018). Thus, reactivity and energy efficiency of SO4•− -based oxidation processes may also be different from HO•-based oxidation processes.

Furthermore, various background components in water, such as CO32−/HCO3−, Cl−, and natural organic matter (NOM), would scavenge HO• and/or SO4•− with different rates, which may lead to various effects on the degradation of target organic compounds in HO•-based and SO4•−-based oxidation processes. Considering different substances have different molecular structures, although the degradation of micro-pollutants in AOPs has already been investigated, it is still necessary to compare the efficiency of HO• and SO4•−-based oxidation for specific pollutants under various conditions.

The purpose of this study was to systematically compare HO•-based and SO4•−-based AOPs for the mineralization of the emerging contaminants in different water matrices. UV/peroxydisulfate (UV/PDS) and UV/H2O2 systems were selected because they were among the most common SO4•−- and HO•-based AOPs, respectively (Shah et al. 2013; He et al. 2014; Xiao et al. 2016). The pesticide atrazine (ATZ), the antibiotic triclosan (TCS), and the odorant 2,4,6-trichloroanisole (TCA) were used as model pollutants. They all have long half-life and are most widely detected in the drinking water environment (Arnold et al. 1995; Chen et al. 2011; de la Casa-Resino et al. 2012; Vlachos et al. 2008; Vestner et al. 2010). Seriously, they pose a long-term threat to ecosystems and human health (Chalew and Halden 2009; Dann and Hontela 2011; Prat et al. 2011). The effect of key factors such as various oxidant concentrations, micro-pollutant initial concentration, different pH conditions, and the most common anions (i.e., Cl− and CO32−/HCO3−) on the mineralization efficiency was evaluated. The several reaction rate constants of those three pollutants towards SO4•− and HO• were also determined in this study. These results will be helpful to better understand the degradation of micro-pollutants in UV/PDS and UV/H2O2 processes, as well as the contribution of secondary radicals in the degradation processes.

Materials and methods

Chemicals and reagents

Chemical standard atrazine (ATZ, purity > 99%), 2,4,6-trichloroanisole (TCA, purity > 98%), triclosan (TCS, purity > 99%), H2O2 solution (35%, v/v), and benzoic acid were purchased from Sigma-Aldrich Chemical Co. Ltd. (USA). Peroxydisulfate (PDS), sodium hydroxide, perchloric acid, sodium thiosulfate, sodium carbonate, and sodium chloride were of ACS reagent grade and obtained from Sinopharm Chemical Reagent Co. Ltd. China. Suwannee River natural organic matter (NOM) was obtained from the International Humic Substances Society. Acetonitrile (ACN) and acetic acid of HPLC grade were purchased from Fisher Scientific for ATZ, TCA, and TCS determination. All solutions were prepared with Milli-Q water from a Millipore Water Purification System.

Analytical methods

Concentrations of ATZ, TCA, and TCS were analyzed using a high-performance liquid chromatography (HPLC, Waters 2695) system equipped with a symmetry C18 column (4.6 × 150 mm, 5 μm particle size, Waters) and a photodiode array detector (PDA, Waters 2998). The mobile phase consisted of water (1‰ acetic acid) and acetonitrile (40:60 for ATZ, TCA, and TCS) at a flow rate of 1.0 mL min−1 and the sample injection volume was 100 μL. The UV wavelengths for the measurement of ATZ, TCA, and TCS were set at 221 nm, 290 nm, and 230 nm, respectively. Solution pH was measured by a pH meter (PH3-3C, Shanghai precision Instrument CO. Ltd.). H2O2 and PDS concentrations were measured by the iodometric method (American Public Health Association et al. 1915). Multi 3100 N/C TOC analyzers (Analytic jena, German) were used to determine the concentration of natural organic matter (mgC·L−1) according to ISO-8245.

Experimental procedures

The photochemical process was conducted using a bench-scale UV apparatus, consisting of four low-pressure Hg UV lamps (254 nm, GPH212t5l/4, 10 W, Heraeus) housed in a shuttered box, with a vertical tube extending from the bottom. The light travels through the collimated tube down into a 100 mL sample, which is magnetically stirred in a cylindrical glass dish (6 cm diameter × 6 cm deep) 30 cm from the tube. The photon flux (I0, 253.7 nm) from the ultraviolet light source into the solution was determined by iodide-iodate chemical photometric method to be 1.291 × 10−7 Einstein·L−1 s−1 and UV irradiance used in this study was 0.16 mW cm−2 after calibration. All experiments were carried out in deionized water at a controlled temperature (20 ± 1 °C) and certain time (10 min). Solution pH was buffered at pH 5.0 to 9.0 using 10 mM phosphate and adjusted with NaOH or HClO4. Samples (1 mL) were withdrawn at certain time intervals and quenched by adding 20 μL of sodium thiosulfate. All experiments were conducted in duplicated with the relative standard deviation of less than 5%.

Determination of the second-order rate constants of SO4−• and HO• with ATZ, TCA, and TCS

In order to determine the second-order rate constants for the reaction between ATZ, TCA, and TCS with SO4•− and HO•, benzoic acid (BA) was used as a reference compound according to the competition kinetics, the concentration of the target pollutant (i.e., ATZ, TCS, and TCA) and the competitive reagent (i.e., BA) were in the same order of magnitude (i.e., 5 μM). The constant reaction rates between SO4•− and HO• with BA were 1.2 × 109 M−1 s−1 and 5.9 × 109 M−1 s−1 (Buxton et al. 1988; Guan et al. 2011), respectively. The reaction rate kinetic constants of ATZ, TCA, and TCS with SO4•− and HO• were determined during the UV/PDS and UV/H2O2 processes, respectively. 5 mM t-butanol (TBA) was added to the reaction to scavenge HO• produced in UV/PDS process, because the reaction rate constant for TBA and HO• (kHO • , TBA=6.0 × 108 M−1 s−1) (Buxton et al. 1988) is much higher than that with SO4•− (\( {k}_{{\mathrm{SO}}_4^{\bullet -},\mathrm{TBA}} \)=4.0 × 105 M−1 s−1) (Neta and Huie 1988). The target substances degradation kinetics can be calculated by the following Eq. (1) (Li et al. 2018; Lu et al. 2018):

$$ \ln \frac{{\left[ com\right]}_0}{{\left[ com\right]}_t}-{k}_{directi, uv}t=\frac{k_{SO_4^{\bullet -}/\mathrm{HO}\bullet, \kern0.5em \mathrm{com}}}{k_{SO_4^{\bullet -}/\mathrm{HO}\bullet, \kern0.5em \mathrm{BA}}}\mathit{\ln}\frac{{\left[ BA\right]}_0}{{\left[ BA\right]}_t} $$
(1)

where com is the ATZ, TCA, and TCS, respectively, kdirect,uv is the observed rate constants for those three compounds degradation by UV irradiation, \( {k}_{SO_4^{\bullet -}/\mathrm{HO}\bullet, \kern0.5em \mathrm{com}} \) is the second-order rate constants of SO4•− and HO• with ATZ, TCA, and TCS, and \( {k}_{SO_4^{\bullet -}/\mathrm{HO}\bullet, \kern0.5em \mathrm{BA}} \) is the second-order rate constants of SO4•− and HO• with BA.

Results and discussion

Reaction rate constants of SO4−• and HO• with ATZ, TCA, and TCS

The active substances in UV/PDS and UV/H2O2 processes including SO4−• and HO• play a major role in the degradation of target substances. The plot of \( \ln \frac{{\left[ com\right]}_0}{{\left[ com\right]}_t}-{k}_{directi, uv}t \)and \( \mathit{\ln}\frac{{\left[ BA\right]}_0}{{\left[ BA\right]}_t} \) yielded a straight line with a slope of \( \frac{k_{SO_4^{\bullet -}/\mathrm{HO}\bullet, \kern0.5em \mathrm{com}}}{k_{SO_4^{\bullet -}/\mathrm{HO}\bullet, \kern0.5em \mathrm{BA}}} \) in Fig. 1. The determined second-order rate constants of these three compounds with SO4•− and HO• (i.e., \( {k}_{SO_4^{\bullet -}} \) and kHO•) are shown in Table 1.

Fig. 1
figure 1

Determination of the reaction rate constants of SO4•− (a) and HO• (b) reacting with ATZ, TCA, and TCS in different solution pH value (pH = 5, pH = 7, pH = 9). Conditions: [ATZ] = [TCS] = [BA] = [TCA] = 5 μM, [PDS] = [H2O2] = 500 μM, 10 mM phosphate buffer, I0 = 1.291 × 10−7Einstein L−1 s−1, reaction time = 10 min

Full size image
Table 1 The second-order rate constants (M−1 s−1) of ATZ, TCA, and TCS with SO4•− and HO•
Full size table

According to the above results, the second-order rate constants of ATZ and TCA with SO4•− and HO• were not affected by pH in the range of pH 5 to 9, this mainly being because ATZ and TCA could not get ionized when the pH was in the range of 5 to 9 (the pKa of ATZ and the Kow of TCA is 1.68 and 4.11, respectively). However, the pKa of TCS is 7.9 (Karnjanapiboonwong et al. 2010), TCS mainly exists in the form of deprotonation when the pH is 9, and the second-order rate constants of SO4•− and HO• with the deprotonation form of TCS were obtained much higher than that of the molecular form of TCS.

Effect of oxidant dosage

Figure 2 shows the effects of oxidant dosage (100–500 mM) on the ATZ, TCA, and TCS degradation by the UV/H2O2 and UV/PDS processes, respectively. For UV/H2O2, when the H2O2 dosage increased from 100 to 500 μM, after 10 min of reaction, the ATZ, TCA, and TCS removal rates were increased from 40.2 to 63.2%, 65.7 to 77.8%, and 39.9 to 47.5%, respectively. For UV/PDS systems, the ATZ, TCA, and TCS removal rates were increased by 33.8%, 18.8%, and 21.3%, respectively. These results indicated that the ATZ, TCA, and TCS degradation rates increased with increasing oxidant dosage. Similar observations were also reported in previous studies where an increase in PDS and H2O2 concentration levels could correspondingly promote the degradation rate of carbamazepine in water (Deng et al. 2013). Notably, the removal rates of ATZ, TCA, and TCS by UV/PDS was about 20~30% higher than that by UV/H2O2 under the same oxidant dosage. According to the results above, the reaction rates of HO• with these three targets are the same order of magnitude as SO4•− (Table 1), but the molar adsorption coefficient and quantum yields of PDS (i.e., Φ = 0.7 mol·Einstein−1,ε = 21.1 M−1 cm−1) is higher than H2O2 (i.e., Φ = 0.5 mol·Einstein−1, ε = 18 M−1 cm−1) (Yang et al. 2010; Zhang et al. 2015). Therefore, the catalytic efficiency of PDS by UV irradiation was higher than that of H2O2 in this study, leading to the higher steady-state concentrations of SO4•− in UV/PDS than HO• in UV/H2O2 under the same conditions.

Fig. 2
figure 2

Effect of oxidants dosage for ATZ, TCA, TCS degradation by UV/PDS and UV/H2O2. Conditions: [ATZ] = [TCS] = 2 μM, [TCA] = 200 nM, 10 mM phosphate buffer at pH 7, I0 = 1.291 × 10−7Einstein·L−1 s−1, reaction time = 10 min

Full size image

Besides, some previous studies have found that the efficiency of UV/H2O2 was higher than that of UV/PDS. Tan et al. found that the degradation efficiency of acetaminophen in UV/H2O2 was higher than that in UV/PDS (Tan et al. 2014). It was mainly because the second-order rate constant of the reaction between acetaminophen with SO4•− was lower than that of the reaction between acetaminophen with HO•.

Effect of the initial ATZ, TCA, and TCS concentration

ATZ, TCA, and TCS degradation were evaluated at different initial concentrations in the UV/H2O2 and UV/PDS systems, respectively. As it could be seen in Fig. 3, the degradation rates of ATZ, TCA, and TCS in both systems decreased with the increase of the initial concentration. For example, when the ATZ concentration increased from 0.5 to 5 μM in UV/H2O2, the degradation of ATZ decreased from 67.1 to 35.8% after 10 min, while it decreased from 89.9 to 42.8% in the UV/PDS system. At the initial concentration of each target, the UV/PDS degradation efficiency was higher than that of UV/H2O2.

Fig. 3
figure 3

Effect of initial concentrations of ATZ, TCA, TCS degradation by UV/PDS and UV/H2O2. Conditions: [PDS] = [H2O2] = 200 μM, 10 mM phosphate buffer at pH 7, I0 = 1.291 × 10−7 Einstein·L−1 s−1, reaction time = 10 min

Full size image

These results were consistent with some previous reports, mainly because (i) the increase in micro-pollutant concentration reduced the UV transmission efficiency of the solutions, resulted in the decrease of HO• and SO4•− steady-state concentration; and (ii) the increasing intermediate oxidation products improved the scavenge effect on radicals. For instance, recent research indicated that the second-order rate constants of parent ATZ and its primary oxidation products (e.g., Atra-imine, DEA, DIA, terbuthylazine, and propazine) with HO• and SO4•− were comparable (i.e., 109 M−1 s−1) (Lutze et al. 2014). The intermediate oxidation products of TCA and TCS (e.g., chlorobenzoquinone, chlorophenol) have also been proved that they have strong ability to scavenge HO• and SO4•−.

Effect of NOM

In source water, the concentration of NOM is usually at the level of mgC·L−1. Figure 4 shows the effects of NOM on ATZ, TCA, and TCS degradation. As it could be seen, when the dosage of NOM increased from 0.2 to 5.0 mgC·L−1, the degradation of ATZ, TCA, and TCS was significantly inhibited by NOM in both UV/H2O2 and UV/PDS processes.

Fig. 4
figure 4

Effect of NOM for ATZ, TCA, TCS degradation by UV/PDS and UV/H2O2. Conditions: [PDS] = [H2O2] = 200 μM, [ATZ] = [TCS] = 2 μM, [TCA] = 200 nM, 10 mM phosphate buffer at pH 7, I0 = 1.291 × 10−7 Einstein·L−1 s−1, reaction time = 10 min

Full size image

The inhibitory effect of NOM on ATZ, TCA, and TCS degradation could be explained as the following: (i) NOM would exert an inner filter effect for the photolysis of H2O2 and PDS, the molar extinction coefficient of NOM was detected to be 0.11 L·mgC−1·cm−1 (Luo et al. 2016), (ii) NOM reacted with SO4•− and HO• as radical scavenger. In addition, previous studies reported that SO4•− or HO• reacted with certain organic compounds (e.g., aromatics, ketones, esters, and aliphatics) by electron transfer to generate organic free radicals. However, these organic free radicals have been shown to be much less reactive than SO4•− or HO•. Besides, Fig. 4 indicates that the removal rates of ATZ, TCA, and TCS in UV/PDS were higher than that in UV/H2O2 under the same NOM concentrations. One of the reasons was that the second-order rate constants of NOM reacting with HO• is one order of magnitude higher than that of SO4•−, as shown in Eqs. (2–3) (Lee et al. 2007; Lutze et al. 2015).

$$ {{\mathrm{SO}}_4}^{-}\bullet +\mathrm{NOM}\to \mathrm{products}\to k=6.8\times {10}^3\ \left(\mathrm{L}\ {\mathrm{mgC}}^{-1}\ {\mathrm{s}}^{-1}\right) $$
(2)
$$ \mathrm{HO}\bullet +\mathrm{NOM}\to \mathrm{products}\kern0.5em \mathrm{k}=2.5\times {10}^4\ \left(\mathrm{L}\ \mathrm{mg}{\mathrm{C}}^{-1}{\mathrm{s}}^{-1}\right) $$
(3)

Effect of pH

The degradation of ATZ, TCA, and TCS by UV/H2O2 and UV/PDS at various pH values were carried out, respectively. As shown in Fig. 5, when the solution pH increased from 5 to 9, a significant role in lowering ATZ and TCA removal was exhibited in UV/PDS (i.e., 86.1% vs 46.2% for ATZ, 99.9% vs 67.7% for TCA). However, the increasing pH had a slight inhibitory effect on the degradation of ATZ and TCA in UV/H2O2. Different from this, TCS degradation was inhibited both in these two processes when pH ranged from 5 to 8, but it was greatly promoted at pH 9. For example, the removal rates of TCS were 43% and 63.7% at pH 8 in UV/H2O2 and UV/PDS, respectively, while which were increased to 59.5% and 86.8% at pH 9.

Fig. 5
figure 5

Effect of pH for ATZ, TCA, TCS degradation by UV/PDS and UV/H2O2. Conditions: [PDS] = [H2O2] = 200 μM, [ATZ] = [TCS] = 2 μM, [TCA] = 200 nM, 10 mM phosphate buffer at pH 7, I0 = 1.291 × 10−7 Einstein·L−1 s−1, reaction time = 10 min

Full size image

Generally, the quantum efficiency of photodissociation of PDS and H2O2 was identical when the pH was range from 5 to 9, and the formation of SO4•− and HO• would not be affected by solution pH. As shown in Fig. 1, the reaction rate constants between radicals (i.e., SO4•− and HO•) with ATZ and TCA were stable at pH 5.0, 7.0, and 9.0. Actually, there were 10 mM phosphate buffer in these two processes, whose concentration was over at least 50 times more than the concentration of micro-pollutants. The reaction rate constant of SO4•− with HPO42− is 1.2 × 106 M−1 s−1, which is two orders of magnitude higher than that of SO4•− with its protonated form H2PO4− (7.2 × 104 M−1 s−1) (Luo et al. 2016). Therefore, the percent of HPO42− increased dramatically when pH increased from 5 to 9, leading to increasing the consumption of SO4•− in UV/PDS process. The reaction rate constant between HO• and HPO42− (1.5 × 105 M−1 s−1) is slightly higher than that between HO• and H2PO4− (2.0 × 104 M−1 s−1) (Luo et al. 2016). The scavenging capacity of phosphate buffer is less affected by solution pH.

Unlike ATZ and TCA, the pKa of TCS is 7.9 (Karnjanapiboonwong et al. 2010), the percentage of TCS anionic state in aqueous solution increased suddenly with increasing pH from 8.0 to 9.0. The anionic state of TCS has a higher molar absorption coefficient; the rate constants of anionic state of TCS with HO• and SO4•− (9.97 × 109 M−1 s−1 and 4.32 × 109 M−1 s−1, respectively) are higher than those of molecular state of TCS (5.3 × 109 M−1 s−1 and 0.96 × 109 M−1 s−1, respectively.).

Effect of CO32−/HCO3−

Figure 6 shows the removal rates of ATZ, TCA, and TCS under various CO32−/HCO3− concentrations in UV/H2O2 and UV/PDS, respectively. As it could be seen, the removal rates of ATZ and TCA decreased with increasing CO32−/HCO3− in both processes. For example, ATZ decreased from 44.8% and 67.6% to 25% and 35% in UV/H2O2 and UV/PDS processes, respectively, when CO32−/HCO3− concentrations increased from 0.5 to 10 mM. In particular, when CO32−/HCO3− concentrations was less than 2 mM, the removal rates of TCS dropped by 10% and 6% in UV/H2O2 and UV/PDS processes, respectively. As the CO32−/HCO3− dosage increased gradually, the degradation rate of TCS had a slight promotion effect both in the UV/H2O2 and UV/PDS processes.

Fig. 6
figure 6

Effect of CO32−/HCO3− for ATZ, TCA, TCS degradation by UV/PDS and UV/H2O2. Conditions: [PDS] = [H2O2] = 200 μM, [ATZ] = [TCS] = 2 μM, [TCA] = 200 nM, 10 mM phosphate buffer at pH 7, I0 = 1.291 × 10−7 Einstein·L−1 s−1, reaction time = 10 min

Full size image

Usually, the presence of CO32−/HCO3− could reduce the degradation efficiency of micro-pollutants (e.g., ATZ, TCA), which was because CO32−/HCO3− could scavenge both HO• and SO4•− at relative high reaction, and the inhibition effect on HO• was much stronger than that on SO4•− (e.g., the second-order rate of bicarbonate with HO• is 8.6 × 106 M−1 s−1 and that with SO4•− is 1.6 × 106 M−1 s−1) (Buxton et al. 1988; Zuo et al. 1999; Lai et al. 2017). Therefore, CO32−/HCO3− showed a stronger inhibitory effect on the degradation of targets in the UV/H2O2 process than UV/PDS process.

On the other hand, significant amounts of HO• and SO4•− react with CO32−/HCO3− to form the milder oxidant (CO3•−, 1.57 ± 0.03 V vs.NHE) (as shown in Eqs. 4–7) (Buxton al. 1988; Zuo et al. 1999). In comparison with HO• and SO4•−, CO3•−, as a secondary radical, has a stronger selectivity for target oxidation. The rate constants for the reactions involving CO3•− are in the range of 102~109 M−1 s−1 (Yan et al. 2019). Previous studies reported that the CO3•− was found to primarily react with some electron-rich compounds (e.g., phenols, amines, and sulfur compounds). These results in this study suggested that CO3•− was less reactive with ATZ and TCA, but the reactivity of CO3•− with TCS was relativity higher.

$$ \mathrm{HO}\bullet +{\mathrm{H}\mathrm{CO}}_3^{-}\to {\mathrm{H}}_2\mathrm{O}+{\mathrm{CO}}_3{\bullet}^{-}\kern0.5em \mathrm{k}=8.6\times {10}^6{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(4)
$$ \mathrm{HO}\bullet +{\mathrm{CO}}_3^{2-}\to {\mathrm{OH}}^{-}+{\mathrm{CO}}_3{\bullet}^{-}\kern0.5em \mathrm{k}=3.9\times {10}^8{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(5)
$$ {\mathrm{SO}}_4{\bullet}^{-}+{\mathrm{HCO}}_3^{-}\to {\mathrm{SO}}_4^{2-}+\mathrm{H}{\mathrm{CO}}_3{\bullet}^{-}\kern0.5em \mathrm{k}=1.6\times {10}^6{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(6)
$$ {\mathrm{SO}}_4{\bullet}^{-}+{\mathrm{CO}}_3^{2-}\to {\mathrm{SO}}_4^{2-}+{\mathrm{CO}}_3{\bullet}^{-}\kern0.5em \mathrm{k}=6.1\times {10}^6{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(7)

Effect of Cl−

The influence of Cl− on ATZ, TCA, and TCS degradation by UV/H2O2 and UV/PDS was investigated at Cl− concentrations ranging from 0.5 to 10 mM (Fig. 7). For UV/PDS process, Cl− had a significant inhibitory effect on the degradation of ATZ and TCS. When Cl− concentrations increased from 0 to 10 mM, after 10 min of reaction, ATZ and TCS degradation decreased by 24.0% and 33.1%, respectively. While, the inhibitory effect of Cl− on the degradation of TCA was relatively small. Cl− had little effect on the degradation of ATZ, TCA, and TCS by UV/H2O2 process. When Cl− concentrations increased from 0 to 10 mM, after 10 min of reaction, the removal rates of ATZ, TCA, and TCS had only dropped 4.9%, 4.3%, and 4.6%, respectively.

Fig. 7
figure 7

Effect of Cl− for ATZ, TCA, TCS degradation by UV/PDS and UV/H2O2. Conditions: [PDS] = [H2O2] = 200 μM, [ATZ] = [TCS] = 2 μM, [TCA] = 200 nM, 10 mM phosphate buffer at pH 7, I0 = 1.291 × 10−7 Einstein·L−1 s−1, reaction time = 10 min

Full size image

For UV/PDS process, the second-order rate constants of Cl− with SO4•− is 3.1 × 108 M−1 s−1 at pH = 7 (Neta and Huie 1988). Usually, the fast reaction of SO4•− with Cl− yields Cl• and Cl2•− as shown in Eqs. 8 and 13 (Fang et al. 2014). Similar to CO3•−, these second radical species have strong selectivity for the oxidation of organic matter (Kwon et al. 2018). For example, the secondary rate constant of Cl2•− with 2-propanol is only 1.5 × 105 M−1 s−1, while the secondary rate constant with hydroquinone is as high as 1.5 × 109 M−1 s−1 (Neta and Huie 1988; Fang et al. 2014).

On the other hand, the presence of Cl− in the UV/PDS process could promote the conversion of SO4−• to HO•. Wu et al. used Fe(III)-ethylenediamine-N,N ′-disuccinic acid to catalyze PDS to produce SO4•− to degrade 4-bromophenol (Wu et al. 2015). They found that when 10 mM Cl− was present, the steady-state concentration of SO4•− in the system was only 8.0 × 10−14 M, while the steady-state concentration of HO• could reach 3.7 × 10−12 M (Fang et al. 2014). In this study, the reaction rates of SO4•− and HO• with ATZ, TCA, and TCS were of the same order of magnitude at pH 7, so SO4•− to HO• conversion had little effect on target degradation.

In this study, the inhibitory effect of Cl− on ATZ and TCS degradation could be explained as the low reactivity with these two compounds of those secondary reactive chlorine radical species than that of SO4•− in UV/PDS process. Nevertheless, those secondary reactive chlorine radical species might have similar reactivity with TCA and SO4•−. For UV/H2O2 process, the presence of Cl− scavenging for HO• was apparently weak, which was ascribed to the fast-backward reaction process in the equilibrium reaction of Cl− with HO• to form ClOH•− (i.e., Eqs. 9) (Neta and Huie 1988; Fang et al. 2014).

$$ {\mathrm{SO}}_4^{-}\bullet +{\mathrm{CI}}^{-}\to \mathrm{Cl}\bullet +{\mathrm{SO}}_4^{2-}\kern0.5em \mathrm{k}=3.0\times {10}^8{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(8)
$$ \mathrm{HO}\bullet +{\mathrm{CI}}^{-}\to \mathrm{CIOH}{\bullet}^{-}\kern0.5em {\mathrm{k}}_{+}=4.3\times {10}^9\kern0.5em \mathrm{k}=6.0\times {10}^9{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(9)
$$ \mathrm{Cl}\bullet +{\mathrm{HO}}^{-}\to \mathrm{CIOH}{\bullet}^{-}\kern0.5em k=1.8\times {10}^{10}{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(10)
$$ \mathrm{CIOH}{\bullet}^{-}\kern0.5em \to \mathrm{HO}\bullet +{\mathrm{Cl}}^{-}\kern0.5em \mathrm{k}=6.0\times {10}^9{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(11)
$$ \mathrm{CIOH}{\bullet}^{-}\kern0.5em +{\mathrm{H}}^{+}\to \mathrm{Cl}\bullet +{\mathrm{H}}_2\mathrm{O}\kern0.75em {\mathrm{k}}_{+}=2.1\times {10}^{10}\kern0.5em \mathrm{k}=2.5\times {10}^5{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(12)
$$ \mathrm{Cl}\bullet +{\mathrm{Cl}}^{-}\to {\mathrm{Cl}}_2{\bullet}^{-}\kern0.5em \mathrm{k}=8.5\times {10}^9{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(13)
$$ {\mathrm{Cl}}_2{\bullet}^{-}+{\mathrm{OH}}^{-}\to {\mathrm{Cl}}^{-}+\mathrm{ClOH}{\bullet}^{-}\kern0.5em \mathrm{k}=4.5\times {10}^7{\mathrm{M}}^{-1}{\mathrm{s}}^{-1} $$
(14)

Conclusion

In this work, ATZ, TCA, and TCS removal efficiencies by UV/PDS and UV/H2O2 were evaluated under different experimental conditions. Increasing oxidant dosage and decreasing micro-pollutant initial concentration were observed to enhance the degradation of ATZ, TCA, and TCS. The presence of NOM and Cl− had inhibitory effects on ATZ, TCA, and TCS in both processes. pH affected both the contribution of SO4•− and HO• and dissociation species in solutions. UV/PDS degrading ATZ and TCA was affected by solution pH and was significant reduced in alkaline solutions, but the change was not obvious in UV/H2O2. For TCS, the degradation rate was found to be the lowest at pH 8 and increased dramatically at pH 9 in UV/PDS and UV/H2O2. The presence of CO32−/HCO3− had a small inhibitory effect on ATZ and TCA degradation but promoted the degradation of TCS significantly (> 2 mM) in UV/H2O2 and UV/PDS processes. Results of this study also demonstrate that UV/PDS effectively eliminates ATZ, TCA, and TCS with much higher efficacy than UV/H2O2.