Abstract

Benzotriazoles (BTs) attract increasing concerns because of abundant presence in environmental water bodies. In this study, degradation of 1H-benzotriazole (1H-BT) was performed by a customized vacuum ultraviolet (VUV) device emitting 185 + 254 nm (VUV/UV-C) irradiation. Degradation of 1H-BT presented an apparent rate constant reached 8.17 × 10−4 s−1. Degradation mechanisms included 185 + 254 nm photodegradation and radical reaction. The later one may be the predominant one, which presented a k·OH-1H-BT at (7.3 ± 0.8) × 109 M−1 s−1. Effects of anions revealed that VUV interception and radical trapping were the dominant restraining factors. Degradation of 1H-BT can be attributed to VUV induced radical-based oxidation. Radical-induced addition, substitution and fracture generated abundant hydroxylated and open-loop products during 10–45 min. Identification using reactive oxygen species and apoptosis in Escherichia coli was conducted. Variations of these two indicators revealed that the incomplete degradation products presented higher toxicities than 1H-BT, and a further mineralization reduced their toxicities. In the pure water solution with little impurities, VUV can induce efficient degradation of 1H-BT, suggesting its potential for eliminating and detoxifying MPs.

INTRODUCTION

Benzotriazoles (BTs) have been used as corrosion inhibitors, drug precursors and antifreezes in various industrial products. Studies have documented a variety of toxic in vitro/in vivo effects of BTs in organisms (Gatidou et al. 2019). The annual output of BTs reached 9,000 tons in the USA (Hart et al. 2004). A proportion of BTs ends up in sewer systems after application. Conventional biological treatment processes cannot completely eliminate BTs due to their low biodegradability. For example, removal efficiencies of different BTs in some wastewater treatment plants (WWTPs) applied conventional biological treatment processes varying from 10% to 80% (Hollender et al. 2009; Reemtsma et al. 2010). Residual BTs and their degradation products are released into environment water bodies, which are or may be the sources for drinking water. Municipal WWTPs are therefore considered as a source of BTs to environmental waters (Molins-Delgado et al. 2017), and it is desirable to develop novel treatment technologies for efficient eliminating BTs.

A series of advanced oxidation processes (AOPs), such as photocatalysis, ozonation, ultrasound oxidation and Fenton oxidation, were tested for eliminating micro-pollutants (MPs) in water (Wang et al. 2019; Zhang et al. 2019). UV-based AOPs (UV-AOPs), including UV/H2O2, UV/S2O82− and UV/TiO2, have been proved to degrade BTs efficiently (Borowska et al. 2016). But these UV-AOPs require the addition of chemical reagents and complicated operation. In light of that, additive-free vacuum ultraviolet (VUV) may be a novel potential UV-AOP. Emission of VUV falls within the range of 100–200 nm, which induces homolytic dissociation of H2O molecules to form OH (Zoschke et al. 2014; Kozmér et al. 2016). The reactions of MPs in VUV systems therefore involve a combination of direct photolysis and indirect oxidation. Thus, VUV can degrade various MPs efficiently, such as phenols (Oppenlander & Gliese 2000), pesticides (Duca et al. 2017), etc. Although the low penetration of VUV emission (Weeks et al. 1963) limits its application in full-scale water treatments so far, VUV is still a potential technology for MPs removal in preparation of ultrapure water or in decentralized water treatment systems with thin layer water flow.

Determination of photochemical kinetics parameters will provide useful information for improving the efficiency of VUV. Up to now, the kinetics parameters of several targeted MPs during VUV treatment have been examined. Xie et al. (2018) created kinetic models to predict the organic matter degradation kinetics in VUV devices. Li et al. (2016) applied a VUV/UV reaction system to determine the key kinetics parameters for organic matter degradation. Their model took into account the propagation of VUV irradiation and oxidation of OH. Although important information about the degradation kinetics has been reported by these models using simple organic matter, further research is still needed for more comprehensive investigation in regarding to the degradation kinetics and mechanism of environmental MPs. Furthermore, only a few researches selected actual contaminants as the targets, as most of these literature aimed at degradation of dyes, which was light absorption and could be transformed easily. Researches about the degradation products and pathways of targeted contaminants during VUV treatment were rarely reported.

To date, the reaction kinetics and mechanisms of BTs in the VUV/UV-C system are still unclear, not to mention the degradation products determination and toxicological characteristics. In this study, a customized VUV/UV-C device was developed to explore degradation kinetics and mechanisms of 1H-benzotriazole (1H-BT). Irradiation screening and radicals scavenging based on the VUV feature were performed. Transformation products were measured by high resolution mass spectrometry (HRMS). In addition, the toxicity variation was elucidated by toxicological analysis based on Escherichia coli.

MATERIALS AND METHODS

Chemical reagents

1H-BT (99%), ethyl alcohol (EtOH), tert-butyl alcohol (TBA) and ascorbic acid (99%) were applied (Sigma-Aldrich). AR NaNO3, NaCl, Na2CO3, Na2SO4, KH2PO4 and NaHCO3 were all 98% (Sinopharm). 18.2 MΩ ultrapure water was applied. A standard strain Escherichia coli ATCC11303 was obtained from microbial culture collection center in Guangdong Province, China.

VUV/UV-C irradiation module and degradation experiments

A VUV/UV-C collimated light irradiation module, which contained a VUV/UV-C light source and a framework, was assembled (Fig. S1). This module was gas-tight with a continuous N2 protection during reaction. A low-pressure argon-mercury lamp (GPH 212T5VH/4, 90 cm length) emitting peak values at 185 nm and 254 nm which was purchased from Heraeus (Germany). Customized circular quartz vessels similar to Petri dishes with two volumes at 20 mL (diameter at 6 cm) and 80 mL (diameter at 10 cm) were applied. Intensity of 254 nm was measured by a 254 nm photometer. By measuring the formation of formaldehyde during VUV treatment of methanol in excess, intensity of 185 nm irradiation was determined (Xie et al. 2018). Furthermore, some modification factors were also considered based on the method described by Bolton & Linden (2003). The reflection factor (0.975), water factor and Petri factor (0.9) were included, and the typical UV fluences at 254 and 185 nm on the solution surface were finally calculated to be ∼6.98 × 10−4μeinstein cm−2 s−1 (∼0.329 mW cm−2) and ∼2.63 × 10−5μeinstein cm−2 s−1 (∼0.017 mW cm−2), respectively. Thus, the actual intensity percentage of 185 nm was 5%.

The phosphate buffer, which contains KH2PO4, K2HPO4 or even NaOH, may have absorption of VUV irradiation, adding this buffer solution could have negative effect on the degradation reaction using VUV treatment. Thus, for most experiments in the current study, no adjustment of pH was conducted. The prepared 1H-BT ultrapure water solution ([1H-BT]0 = 8.39 μM) had a pH value at 6.8–7.5. If necessary, adjustment of pH was performed using buffered solutions with NaOH, KH2PO4 and H3PO4. 20 mL of 1H-BT solution ([1H-BT]0 = 8.39 μM) was added into the quartz vessels. The temperature was maintained at 25 ± 2 °C. The reaction vessel was shaken orbitally (60 rpm). Reaction started by turning on the VUV/UV-C light source. At the pre-set times, 0.5 mL solution was obtained, and then transferred into amber tubes (4 °C). For quenching experiments, radical scavenger, such as ascorbic acid, EtOH or TBA was used (100 mM). In wavelength screening experiment, coverage of JGS-2 quartz beyond the quartz vessel was used. JGS-2 quartz can screen light irradiation <200 nm, thus, only 254 nm irradiation can be utilized after its coverage. For toxicological experiments, 10 groups of VUV/UV-C treated solution (each 80 mL) were prepared. At one pre-set time, 20 mL solution was obtained from one treated group, and then this solution was discarded. Similar sampling procedure was performed at the next pre-set time with another group solution.

Kinetics and molar absorption coefficient calculation

Second-order rate constant (k·OH-CIP) between 1H-BT and ·OH was determined based on our previous study (Chen et al. 2019). This calculation does not consider the direct photolysis. Thus, there would be an overestimation of k·OH-CIP based on the calculation of this equation. Molar absorption (extinction) coefficients (ɛ) were also obtained followed the method in our previous study (Chen et al. 2019). Apparent quantum yields of 1H-BT photolysis by 254 nm irradiation was calculated based on a reported procedure (Xie et al. 2018).

Determination of 1H-BT and degradation products

Detailed analysis procedures were similar to our previous study (Ye et al. 2018).

Microbial culture and pollutant exposure

E. coli ATCC11303 was cultured with 100 mL lysogeny broth medium and were orbitally shaken (150 r min−1, 37 °C, 12 h). A centrifugation at 6,000 g (10 min) was conducted to separate the cultured cells, and then they were washed by sterile phosphate buffer. 20 mL M9 medium was used to prepare a bacterial suspension containing 1.0 g L−1 separated cells. Solutions with culture medium (the control group), 8.39 μM 1H-BT (reaction time at 0 min) and the mixture of degradation products after VUV/UV-C reactions (reaction times at 5, 10, 20, 30, 45, 60 min, the treated group, reaction conditions were the same as earlier) were mixed with the bacterial suspension (volume ratio at 1:1). These mixed suspensions were maintained in the dark (37 °C, 160 r min−1). These two sample groups were collected for reactive oxygen species (ROS), membrane potential (MBP) and apoptosis measurements after 6-h exposure (Li et al. 2018). More detailed toxicology analysis procedure is described in Text S1 (available with the online version of this paper). Pollutant exposure experiments were repeated in three independent biological replications.

RESULTS AND DISCUSSION

Basic degradation efficiency and mechanism

UV-C (254 nm) and VUV/UV-C (185 + 254 nm) reactions for 1H-BT degradation were performed with an identical UV photon fluence at 0.35 mW cm−2 (254 nm). Irradiation of UV-C (200–280 nm) can induce direct photolysis of 1H-BT (Borowska et al. 2016), which was also observed in the current sole UV-C treatment (Figure 1). In UV-C treatment (0.35 mW cm−2), the photolysis of 1H-BT presented first-order-reaction kinetics. A rate constant (k254-d) at 1.17 × 10−4 s−1 was observed. VUV/UV-C irradiation resulted in a faster degradation. ∼99% 1H-BT was degraded in 60-min VUV/UV-C reaction. A pseudo-first-order reaction kinetics was observed with a kapp-vuv at 8.17 × 10−4 s−1 (Figure 1). The improvement after using VUV/UV-C may be attributed some other degradation processes.

Figure 1

Degradation efficiency of 1H-BT. Experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 uM, total UV irradiation intensity = 0.329 mW cm−2, 185 VUV irradiation intensity = 0.017 mW cm−2. Sole UV indicates 254 nm, UV + VUV indicates 185 + 254 nm, JGS-2 indicates UV + VUV with JGS-2 quartz cover, EtOH indicates experiment with ethanol, TBA indicates experiment with tertiary butanol. All the experiments were carried out in triplicate with error bars representing the standard error of the mean.

Figure 1

Degradation efficiency of 1H-BT. Experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 uM, total UV irradiation intensity = 0.329 mW cm−2, 185 VUV irradiation intensity = 0.017 mW cm−2. Sole UV indicates 254 nm, UV + VUV indicates 185 + 254 nm, JGS-2 indicates UV + VUV with JGS-2 quartz cover, EtOH indicates experiment with ethanol, TBA indicates experiment with tertiary butanol. All the experiments were carried out in triplicate with error bars representing the standard error of the mean.

To reveal the mechanisms within this promotion, optical screening with JGS-2 quartz (cut-off wavelength <200 nm) and additions of typical radical scavengers were performed. Using JGS-2 quartz, the transformation efficiency of 1H-BT obviously reduced (kapp-vuv at 1.00 × 10−4 s−1, Figure 1). This indicated that the improving efficiency after using VUV/UV-C may be due to 185 nm VUV. For radicals scavenging, degradation of 1H-BT was inhibited by 10 mM ascorbic acid, and kapp-vuv decreased to 1.50 × 10−4 s−1. As ascorbic acid is reductive and can react with many oxidants, this inhibition implied that oxidation was the dominant contributor of 1H-BT degradation. Whereas ascorbic acid has a high ɛ (∼3,130 M−1 cm−1) at 185 nm, implying that it can also screen VUV irradiation. To further identify the reactive mechanism, EtOH and TBA, with high reaction rate constants to ·OH (k·OH) (Buxton 1988), were applied. Furthermore, ɛ of EtOH and TBA at 185 nm are ∼31.31 M−1 cm−1 and ∼14.72 M−1 cm−1, indicating their weak UV screening effects. In the presences of these two scavengers, the kapp-vuv declined to 1.50 × 10−4 s−1 and 3.83 × 10−4 s−1, respectively (Figure 1). Therefore, ·OH oxidation generated from splitting of H2O was the predominant mechanism for VUV processes. Based on Figure 1, EtOH and TBA at 100 mM could not completely inhibit the indirect photolysis of 1H-BT, which suggested that some other reactive species, such as ·O and O2·− (Kozmér et al. 2016), also existed in the VUV system.

Degradation kinetics

According to the section ‘Basic degradation efficiency and mechanism’, 1H-BT degradation procedure included the direct photolysis (185 and/or 254 nm) and oxidation by ·OH in the VUV system. In the kinetics experiments, all the results were based on one concentration of 1H-BT ([1H-BT]0 = 8.39 μM) and one temperature (25 ± 2 °C). Thus, reaction rate of 1H-BT under VUV/UV-C treatment can be deduced by:  
formula
(1)
where k185-d and k254-d are the 185 nm and 254 nm photolysis rate constants, respectively. k·OH-1H-BT is the second order reaction rate between ·OH and 1H-BT. In addition, the degradation kinetics in the sole UV-C treatment with similar UV photon fluence (254 nm) can be calculated by:  
formula
(2)

Determining these reaction rate constants will provide basic kinetics information, as well as further verify the reaction mechanisms.

Even pure water has significant absorption of 185 nm irradiation (1.8 cm−1; (Weeks et al. 1963)). Approximate 90% 185 nm ultraviolet is absorbed by 5 mm thickness of ultrapure water (Han et al. 2004). The depth of reaction solution in the current study reached 15 mm, and the intensity at 185 nm only constituted 6% for the total intensity. Thus, the 185 nm direct photolysis could be the minor contributor of 1H-BT removal. Thus, k185-d can be neglected, and Equation (1) can be reduced to:  
formula
(3)

Kinetics about the 254 nm irradiation in VUV/UV-C was obtained based on the calculation in sole UV-C system. Under 0.35 mW cm−2 UV-C irradiation, degradation of 1H-BT presented a photolysis with a k254-d at 1.17 × 10−4 s−1 (Table 1). Based on this k254-d and the calculation procedure reported by Xie et al. (2018), the apparent quantum yields of 1H-BT photolysis by 254 nm irradiation was calculated to be about 2.13 × 10−3 mol einstein−1.

Table 1

Kinetic parameters of sole UV and VUV systems

Sole UV (0.329 mW cm−2)
VUV (0.017 mW cm−2)
k254 (s−1)Ф254-d (mol einstein−1)kapp (s−1)k·OH-1H-BTA (M−1 s−1)
(1.17± 0.3) × 10−4 2.13 × 10−3 (8.17 ± 0.51) × 10−4 (7.3 ± 0.8) × 109 
Sole UV (0.329 mW cm−2)
VUV (0.017 mW cm−2)
k254 (s−1)Ф254-d (mol einstein−1)kapp (s−1)k·OH-1H-BTA (M−1 s−1)
(1.17± 0.3) × 10−4 2.13 × 10−3 (8.17 ± 0.51) × 10−4 (7.3 ± 0.8) × 109 

Experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 uM, total UV irradiation intensity = 0.329 mW cm−2, 185 VUV irradiation intensity = 0.017 mW cm−2.

Based on this k254-d, 254 nm induced photolysis of 1H-BT under the similar intensity in VUV/UV-C system followed an identical rate with a k254-d at 1.17 × 10−4 s−1. The total apparent reaction rate constant for VUV (kapp-VUV) was 8.17 × 10−4 s−1, and it was obviously higher than k254-d, suggesting that ·OH reaction was the dominant degradation process in VUV/UV-C system.

To obtain the rate constant (k·OH-1H-BT), a completive reaction with para-chlorobenzoic acid was performed. Following a similar calculation process reported by Huber et al. (2003), k·OH-1H-BT was (7.3 ± 0.8) × 109 M−1 s−1, higher than that between para-chlorobenzoic acid and ·OH (∼5.0 × 109 M−1 s−1). Of note, under 0.35 mW cm−2 sole 254 nm irradiation, the kobs was 1.17 × 10−4 s−1, implying that photolysis contributed only a small part for 1H-BT degradation during VUV/UV-C treatment. Thus, the photolysis (185 + 254 nm) can be ignored during completive reaction calculation, and the calculated k·OH-1H-BT may be overestimated, and it was only an approximate value. Leitner & Roshani (2010) used ibuprofen as a targeted contaminant and found that k·OH-1H-BT was (1.07 ± 0.45) × 1010 M−1 s−1 in UV/H2O2 system under pH = 6.25. Borowska et al. (2016) calculated the kinetics parameters in a UV/H2O2 system, and k·OH-1H-BT was reported to be 2.70 × 109 M−1 s−1 under pH = 7.0. These different k·OH-1H-BT may be due to the different experiment circumstances and targeted compounds. Based on our obtained reaction constant and Equation (3), [·OH] was ∼9.59 × 10−13 M in the current reaction.

Influencing factors

UV irradiation intensity is a basic factor for VUV/UV-C system. As the 185 nm intensity increased from 0.017 mW cm−2 to 0.096 mW cm−2, kapp raised from 8.2 × 10−4 s−1 to 5.0 × 10−3 s−1. A proximate linear correlation was observed between the VUV intensity and kapp (Figure 2(a)). Raising VUV intensity increased the generation of ·OH, inducing a faster reaction rate. To improve the experimental efficiency, reaction with 185 nm intensity at 0.048 mW cm−2 (kapp = 2.4 × 10−3 s−1, t0.5 = 288 s) was used in the subsequent reactions.

Figure 2

Degradation of 1H-BT under different influence factors. (a) Irradiation intensity, (b) pH, (c) Cl, (d) NO3, (e) SO42−, (f) H2PO3, (g) CO32−, (h) HCO3. Experimental conditions (if not specify): solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 μM, total UV irradiation intensity = 0.940 mW cm−2, 185 VUV irradiation intensity = 0.048 mW cm−2. All the experiments were carried out in triplicate with error bars representing the standard error of the mean.

Figure 2

Degradation of 1H-BT under different influence factors. (a) Irradiation intensity, (b) pH, (c) Cl, (d) NO3, (e) SO42−, (f) H2PO3, (g) CO32−, (h) HCO3. Experimental conditions (if not specify): solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 μM, total UV irradiation intensity = 0.940 mW cm−2, 185 VUV irradiation intensity = 0.048 mW cm−2. All the experiments were carried out in triplicate with error bars representing the standard error of the mean.

pH value is a crucial factor for radical-based treatment (Figure 2(b)). When pH value changed from 5.0 to 9.0, the kapp maintained at high values in the range of 2.3–2.5 × 10−3 s−1. pH value is usually sustained in neutral in various water treatments. Thus, the pH optimum herein suggested that VUV would be advantageous for 1H-BT removal in DWTPs or WWTPs. But the kapp decreased to 0.4 × 10−3 s−1 at pH = 11.0. Since pH 11 is close to the pKa of H2O2 (Crittenden et al. 1999), the increase of pH value induced more deprotonation of H2O2 to HO2 (Equation (4)) (Boczkaj & Fernandes 2017). The equilibrium shifting to the left side of Equation (4) would reduce the stationary concentration of ·OH. Furthermore, HO2 has higher absorption coefficient than H2O2, increasing the inner filter effect, resulting in a low oxidation reaction rate. Furthermore, this may be due to the ionization of 1H-BT under alkaline condition, and it presents a different reactivity both for photolysis and for ·OH oxidation. Similar inhibition was also observed in the degradation of tetracycline with VUV treatment (Borowska et al. 2016; Yao et al. 2017). Under an acidic condition (pH = 3.0), the kapp decreased to 1.9 × 10−3 s−1. The reason for this phenomenon is not yet clear (Buxton 1988).  
formula
(4)

Anions are also common constituents in natural waters. The effects of Cl, NO3, SO42−, H2PO4, CO32− and HCO3 were evaluated. Four anions, Cl, NO3, CO32− and HCO3, had inhibition on the 1H-BT degradation, while SO42− and H2PO4 had negligible effects (Figure 2). Generally, these different performances can be attributed to their distinct UV/VUV screening capacities and ·OH scavenging capacities, thus, the ɛ of these anions at 254 nm and 185 nm (Table S1), as well as the k·OH-anions (Table S2), should be integrated to reveal the intrinsic mechanisms.

Cl had a significant effect on the reaction (Figure 2(c)). When Cl concentration increased from 10 to 500 mg L−1, the kapp declined from 1.8 × 10−3 s−1 to 0.2 × 10−3 s−1. Table S2 shows the k·OH-anions of anions, and Cl has a k·OH-anions = 4.3 × 109 M−1 s−1 at pH = 2 (Buxton 1988). However, the k·OH-anions of Cl at pH = 7 is still unknown. Cl has a ɛ at 3,307 ± 164 M−1 cm−1, indicating that it has a significant absorption against 185 nm irradiation. Thus, the inhibition mechanism of Cl may include a dominant screening effect and somewhat ·OH quenching. NO3 can absorb 185 nm VUV dramatically (ɛ at 4,736 ± 216 M−1 cm−1). Furthermore, the k·OH-anions of NO3 was reported to be 1.0 × 105 M−1 s−1. The effect of NO3 was enhanced when its concentration increased (10–500 mg L−1, Figure 2(d)), suggesting that its effect can also be attributed to the significant absorption of 185 nm VUV (Duca et al. 2017).

CO32− and HCO3 are strong scavengers for ·OH. High reaction rates have been confirmed between ·OH and CO32−, as well as HCO3 (Table S2). However, their ɛ values are relatively low (∼1,148 M−1 cm−1 for CO32− and ∼464 M−1 cm−1 for HCO3). Based on Figure 2(g), even low concentration CO32− had a severe inhibition. For example, after adding 10 mg L−1 CO32−, kapp decreased from 2.4 × 10−3 s−1 to 0.7 × 10−3 s−1. High concentrations of CO32− and HCO3 (100–500 mg L−1) both contributed to critical decline of reaction efficiency (Figure 2(h)). Similar inhibition was observed in a study aimed at 2,4-dichlorophenoxyacetic acid degradation using VUV (Imoberdorf & Mohseni 2011b).

Compared to the above anions, H2PO4 and SO42− had slight effects on the reaction. H2PO4 has a low reaction rate with ·OH and a low absorption of VUV, as well as SO42−. Therefore, presences of H2PO4 and SO42− had slight effects (Figure 2(e) and 2(f)). Consequently, absorption of 185 nm and ·OH quenching both contributed to the inhibition of VUV reaction.

Mineralization, degradation intermediates and pathways

Ultrapure water solution containing 8.39 μM 1H-BT presented 604 μg L−1 total organic carbon (TOC). As shown in Figure 3(c), ∼84% TOC was removed in 60 min VUV/UV-C treatment, indicating a continuous mineralization of 1H-BT. High mineralization (>80%) was reported in several VUV degradation studies aimed at NOM removal (Imoberdorf & Mohseni 2011a, 2014). During VUV/UV-C treatment, 1H-BT can be transformed into various products. After the product identification procedure (Text S2, available online), seven stable products were identified. Extracted ion chromatograms (EICs), evolution tendencies and generation pathways are presented (Table 2, Figures 3 and 4).

Table 2

BTA organic products in VUV treatments

NameProduct AProduct BProduct CProduct D
Proposed structure     
Molecular formula C6H5N3C6H5N3O2 C6H5N3O3 C5H5N3O4 
[M + H] + : theoretical m/z 136.0505 152.0454 168.0404 172.0353 
[M + H] + : observed m/z 136.0500 152.0454 168.0402 172.0349 
Retention time (min) 2.61 1.62 1.31 0.88 
EIC     
NameProduct EProduct FProduct G
Proposed structure     
Molecular formula C4H3N3O4 C4H3N3O2 C4H3N3O3  
[M + H] + : theoretical m/z 158.0196 126.0298 142.0247  
[M + H] + : observed m/z 158.0190 126.0290 142.0243  
Retention time (min) 1.07 0.70 1.19  
EIC     
NameProduct AProduct BProduct CProduct D
Proposed structure     
Molecular formula C6H5N3C6H5N3O2 C6H5N3O3 C5H5N3O4 
[M + H] + : theoretical m/z 136.0505 152.0454 168.0404 172.0353 
[M + H] + : observed m/z 136.0500 152.0454 168.0402 172.0349 
Retention time (min) 2.61 1.62 1.31 0.88 
EIC     
NameProduct EProduct FProduct G
Proposed structure     
Molecular formula C4H3N3O4 C4H3N3O2 C4H3N3O3  
[M + H] + : theoretical m/z 158.0196 126.0298 142.0247  
[M + H] + : observed m/z 158.0190 126.0290 142.0243  
Retention time (min) 1.07 0.70 1.19  
EIC     

Experimental conditions: solution temperature 25 ± 2 °C, pH 6.5–7.2, [1H-BT]0 = 8.39 uM, 254 UV irradiation intensity = 0.35 mW cm−2, 185 VUV irradiation intensity = 0.022 mW cm−2.

Figure 3

Evolution curves of degradation products and total organic carbon. (a),(b) Relative intensity variations of products, (c) variation of total organic carbon. Experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 μM, total UV irradiation intensity = 0.940 mW cm−2, 185 VUV irradiation intensity = 0.048 mW cm−2. For total organic matter analysis, experiment was carried out in triplicate with error bars representing the standard error of the mean.

Figure 3

Evolution curves of degradation products and total organic carbon. (a),(b) Relative intensity variations of products, (c) variation of total organic carbon. Experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 μM, total UV irradiation intensity = 0.940 mW cm−2, 185 VUV irradiation intensity = 0.048 mW cm−2. For total organic matter analysis, experiment was carried out in triplicate with error bars representing the standard error of the mean.

Figure 4

Proposed generative pathways of degradation products. Experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 μM, total UV irradiation intensity = 0.329 mW cm−2, 185 VUV irradiation intensity = 0.017 mW cm−2.

Figure 4

Proposed generative pathways of degradation products. Experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 μM, total UV irradiation intensity = 0.329 mW cm−2, 185 VUV irradiation intensity = 0.017 mW cm−2.

The dominant degradation pathways may include ·OH-based addition, cleavage, hydroxylation, carbonylation and carboxylation (Figure 3). Molecular structure and bond energy level of 1H-BT molecule may affect the reaction pathways. The electrophilic benzene ring was susceptible to •OH attack, which resulted in the electrophilic substitution to generate product A. A further substitution by another hydroxyl led to the generation of product B (Figure 4). The hydroxylated benzene ring was active, which would be susceptible to further •OH attack. A series of substitution, cleavage, hydroxylation and carboxylation of products A and B resulted in transformations to products C, D and G. Another pathway may follow the cleavage and carbonylation of carboxylated benzene ring, resulting in the generation of products F and E.

Relative intensity evolution curves of degradation products further supported the proposed degradation pathways of these two pathways (peak area; Figure 3). For example, products A and B were two simple hydroxylated products with maximal peak intensities at 10 min, implying their abundant generation in the early stage. At 20 min, product F with two aldehyde groups was dominant. Products G and E had high intensities at the later stage (30–45 min). These two products were accumulated during reaction, which may be due to their saturated oxidized structure. These abundant products reconfirmed an incomplete mineralization of 1H-BT during 10–45 min VUV/UV-C treatment. As the irradiation dosage increased (60 min reaction or more), mineralization ratio raised.

Toxicology analysis

E. coli is one species of common bacteria in environment. Thus, it was used as the representative model bacteria to reveal the effects of 1H-BT and its degradation products on the bacteria physiology. Three toxicity endpoints, including MBP, ROS and apoptosis, were evaluated. For bacteria with cell wall, e.g. E. coli, MBP is the difference in electric potential between the interior and the periplasm of a cell. MBP plays an important role in ATP generation and substance transport. Ion transporters embedded in a membrane transport ions across the membrane, resulting in the establishment of concentration gradients across the membrane (Yang et al. 2017). A slight variation of MBP in E. coli exposed to degraded 1H-BT was observed (Figure 5(a)), suggesting that the transfer of H+ and other ions across the membrane via the cellular electron transport chain was stable. The degradation of 1H-BT into hydroxylated products may have little effect on the ions transport functions of E. coli.

Figure 5

Toxicology analysis. (a)–(c) Variations of membrane potential, reactive oxygen species and apoptosis of Escherichia coli after exposure to culture medium (control), 8.39 μM of 1H-BT (reaction time at 0 min) and the degradation products mixture (reaction time at 5, 10, 20, 30, 45, 60 min); typical original graphs of (d) membrane potential, (e) reactive oxygen species and (f) apoptosis after the exposure to 1H-BT. VUV experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 μM, total UV irradiation intensity = 0.329 mW cm−2, 185 VUV irradiation intensity = 0.017 mW cm−2.

Figure 5

Toxicology analysis. (a)–(c) Variations of membrane potential, reactive oxygen species and apoptosis of Escherichia coli after exposure to culture medium (control), 8.39 μM of 1H-BT (reaction time at 0 min) and the degradation products mixture (reaction time at 5, 10, 20, 30, 45, 60 min); typical original graphs of (d) membrane potential, (e) reactive oxygen species and (f) apoptosis after the exposure to 1H-BT. VUV experimental conditions: solution temperature 25 ± 2 °C, pH 6.8–7.5, [1H-BT]0 = 8.39 μM, total UV irradiation intensity = 0.329 mW cm−2, 185 VUV irradiation intensity = 0.017 mW cm−2.

For ROS and apoptosis, similar tendencies were observed (Figures 5 and S2). The unit of Y-axis is A.U., which is the intensity of fluorescence from exposed cells. The highest level (∼12.5 A.U.) was observed in samples after 20 min treatment, which was nearly twice as much of the lowest (∼6 A.U.) in control group with culture medium. After further degradation (60 min), this value decreased to ∼6 A.U. (Figure 5(b)). Of note, promotion of electron transport chains on membrane can also produce superoxide and result in enhanced oxidative stress. However, considering that no evident MBP variation was induced, the increasing ROS here was primarily generated inside cells, but not from the electron transport chain. The exposure of cells to ROS damages cellular components, including DNA, membrane lipids and proteins (Ezraty et al. 2017). The increasing ROS proved the adaptation of E. coli to stress driven by incomplete degradation products. Thus, it was supposed that products A, B, F and G with high abundance during 10–45 min had higher toxicity to E. coli than 1H-BT. The low ROS of 60 min sample indicated that the toxicity of early products was declined.

Increasing apoptosis during this time range (Figure 5(c)) was another direct evidence. The apoptosis is precisely regulated in response to various stimuli, including ROS, endoplasmic reticulum stress, lipopolysaccharides, and influx of calcium ions (Pleinis et al. 2017). The increasing toxicity of incomplete degradation products, which induced more generation of ROS, was responsible for the aggravation of apoptosis. These data indicated that the incomplete degradation of 1H-BT generated high toxic products, which had adverse effects for E. coli. To the contrary, the ROS of further mineralization 1H-BT samples (60 min) were reduced to a similar low level like that of control samples, while similar apoptosis result was also observed. These data suggested that the further mineralization reduced the overall toxicity of early degradation products. Furthermore, some technologies can be integrated to form a hybrid treatment method, such as cyclodextrin cavity encapsulation (Zhou et al. 2015, 2017), etc., which may have the potential to further remove the toxic products after VUV treatment.

CONCLUSION

Degradation mechanisms of 1H-BT in VUV/UV-C system involved 185 + 254 nm photolysis and ·OH reaction. Radical-induced addition, substitution and fracture of 1H-BT cyclic structure were predominant and produced various compounds with high abundances during 10–45 min. Toxicological analysis suggested that these incomplete degradation products had higher toxicities than 1H-BT, and the further degradation reduced the overall toxicity. Variable pH values, NOM and anions all had effects on VUV system, and irradiation screening and radicals scavenging were two dominant inhibition factors. In the pure water solution with little impurities, VUV can induce efficient degradation of 1H-BT, suggesting its potential for eliminating and detoxifying MPs. Considering its poor penetration and susceptibility to anions, the application of VUV still required further improvements.

ACKNOWLEDGEMENTS

This project was supported by the National Natural Science Foundation of China (Grant No. 51778270).

REFERENCES

REFERENCES
Bolton
J. R.
,
Linden
K. G.
2003
Standardization of methods for fluence (UV dose) determination in bench-scale UV experiments
.
Journal of Environmental Engineering-Asce
129
(
3
),
209
215
.
Chen
Y.
,
Ye
J.
,
Chen
Y.
,
Hu
H.
,
Zhang
H.
,
Ou
H.
2019
Degradation kinetics, mechanism and toxicology of tris(2-chloroethyl) phosphate with 185 nm vacuum ultraviolet
.
Chemical Engineering Journal
356
,
98
106
.
Crittenden
J. C.
,
Hu
S.
,
Hand
D. W.
,
Green
S. A.
1999
A kinetic model for H2O2/UV process in a completely mixed batch reactor
.
Water Research
33
(
10
),
2315
2328
.
Duca
C.
,
Imoberdorf
G.
,
Mohseni
M.
2017
Effects of inorganics on the degradation of micropollutants with vacuum UV (VUV) advanced oxidation
.
Journal of Environmental Science and Health Part A – Toxic/Hazardous Substances & Environmental Engineering
52
(
6
),
524
532
.
Ezraty
B.
,
Gennaris
A.
,
Barras
F.
,
Collet
J. F.
2017
Oxidative stress, protein damage and repair in bacteria
.
Nature Reviews Microbiology
15
(
7
),
385
396
.
Gatidou
G.
,
Anastopoulou
P.
,
Aloupi
M.
,
Stasinakis
A. S.
2019
Growth inhibition and fate of benzotriazoles in Chlorella sorokiniana cultures
.
Science of the Total Environment
663
,
580
586
.
Han
W. Y.
,
Zhang
P. Y.
,
Zhu
W. P.
,
Yin
J. J.
,
Li
L. S.
2004
Photocatalysis of p-chlorobenzoic acid in aqueous solution under irradiation of 254 nm and 185 nm UV light
.
Water Research
38
(
19
),
4197
4203
.
Hart
D. S.
,
Davis
L. C.
,
Erickson
L. E.
,
Callender
T. M.
2004
Sorption and partitioning parameters of benzotriazole compounds
.
Microchemical Journal
77
(
1
),
9
17
.
Hollender
J.
,
Zimmermann
S. G.
,
Koepke
S.
,
Krauss
M.
,
McArdell
C. S.
,
Ort
C.
,
Singer
H.
,
von Gunten
U.
,
Siegrist
H.
2009
Elimination of organic micropollutants in a municipal wastewater treatment plant upgraded with a full-scale post-ozonation followed by sand filtration
.
Environmental Science & Technology
43
(
20
),
7862
7869
.
Huber
M. M.
,
Canonica
S.
,
Park
G. Y.
,
Von Gunten
U.
2003
Oxidation of pharmaceuticals during ozonation and advanced oxidation processes
.
Environmental Science & Technology
37
(
5
),
1016
1024
.
Imoberdorf
G.
,
Mohseni
M.
2011a
Degradation of natural organic matter in surface water using vacuum-UV irradiation
.
Journal of Hazardous Materials
186
(
1
),
240
246
.
Imoberdorf
G. E.
,
Mohseni
M.
2011b
Experimental study of the degradation of 2,4-D induced by vacuum-UV radiation
.
Water Science and Technology
63
(
7
),
1427
1433
.
Imoberdorf
G.
,
Mohseni
M.
2014
Comparative study of the effect of vacuum-ultraviolet irradiation on natural organic matter of different sources
.
Journal of Environmental Engineering
140
(
3
),
04013016
.
Kozmér
Z.
,
Arany
E.
,
Alapi
T.
,
Rózsa
G.
,
Hernádi
K.
,
Dombi
A.
2016
New insights regarding the impact of radical transfer and scavenger materials on the OH-initiated phototransformation of phenol
.
Journal of Photochemistry & Photobiology A Chemistry
314
,
125
132
.
Leitner
N. K. V.
,
Roshani
B.
2010
Kinetic of benzotriazole oxidation by ozone and hydroxyl radical
.
Water Research
44
(
6
),
2058
2066
.
Li
Y.
,
Li
C. S.
,
Qin
H. M.
,
Yang
M.
,
Ye
J. S.
,
Long
Y.
,
Ou
H. S.
2018
Proteome and phospholipid alteration reveal metabolic network of Bacillus thuringiensis under triclosan stress
.
Science of the Total Environment
615
,
508
516
.
Molins-Delgado
D.
,
Tavora
J.
,
Diaz-Cruz
M. S.
,
Barcelo
D.
2017
UV filters and benzotriazoles in urban aquatic ecosystems: the footprint of daily use products
.
Science of the Total Environment
601
,
975
986
.
Pleinis
J. M.
,
Davis
C. W.
,
Cantrell
C. B.
,
Qiu
D. Y.
,
Zhan
X. Z.
2017
Purification, auto-activation and kinetic characterization of apoptosis signal-regulating kinase I
.
Protein Expression and Purification
132
,
34
43
.
Wang
T.
,
Zhou
Y.
,
Cao
S.
,
Lu
J.
,
Zhou
Y.
2019
Degradation of sulfanilamide by Fenton-like reaction and optimization using response surface methodology
.
Ecotoxicology and Environmental Safety
172
,
334
340
.
Weeks
J. L.
,
Meaburn
G. M. A. C.
,
Gordon
S.
1963
Absorption coefficients of liquid water and aqueous solutions in the far ultraviolet
.
Radiation Research
19
(
3
),
559
567
.
Xie
P. C.
,
Yue
S. Y.
,
Ding
J. Q.
,
Wan
Y.
,
Li
X. C.
,
Ma
J.
,
Wang
Z. P.
2018
Degradation of organic pollutants by vacuum-ultraviolet (VUV): kinetic model and efficiency
.
Water Research
133
,
69
78
.
Yao
H.
,
Pei
J.
,
Wang
H.
,
Fu
J.
2017
Effect of Fe(II/III) on tetracycline degradation under UV/VUV irradiation
.
Chemical Engineering Journal
308
,
193
201
.
Ye
J. S.
,
Zhou
P. L.
,
Chen
Y.
,
Ou
H. S.
,
Liu
J.
,
Li
C. S.
,
Li
Q. S.
2018
Degradation of 1H-benzotriazole using ultraviolet activating persulfate: mechanisms, products and toxicological analysis
.
Chemical Engineering Journal
334
,
1493
1501
.
Zhou
Y.
,
Gu
X.
,
Zhang
R.
,
Lu
J.
2015
Influences of various cyclodextrins on the photodegradation of phenol and bisphenol a under UV light
.
Industrial & Engineering Chemistry Research
54
(
1
),
426
433
.
Zhou
Y.
,
Zhang
R.
,
Chen
K.
,
Zhao
X.
,
Gu
X.
,
Lu
J.
2017
Enhanced adsorption and photo-degradation of bisphenol A by β-cyclodextrin modified pine sawdust in an aquatic environment
.
Journal of the Taiwan Institute of Chemical Engineers
78
,
510
516
.
Zoschke
K.
,
Bornick
H.
,
Worch
E.
2014
Vacuum-UV radiation at 185 nm in water treatment – a review
.
Water Research
52
,
131
145
.

Supplementary data