Identification of intermediates and transformation pathways derived from photocatalytic degradation of five antibiotics on ZnIn2S4

Identification of intermediates and transformation pathways derived from photocatalytic degradation of five antibiotics on ZnIn2S4

Chemical Engineering Journal 304 (2016) 826–840 Contents lists available at ScienceDirect Chemical Engineering Journal journal homepage: www.elsevie...

4MB Sizes 0 Downloads 3 Views

Chemical Engineering Journal 304 (2016) 826–840

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Identification of intermediates and transformation pathways derived from photocatalytic degradation of five antibiotics on ZnIn2S4 Bo Gao a,⇑, Shaonan Dong a, Jiadong Liu a, Lifen Liu b, Qiqi Feng a, Na Tan a, Tingting Liu a, Longli Bo a, Lei Wang a a b

School of Environmental and Municipal Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, PR China Key Laboratory of Industrial Ecology and Environmental Engineering, MOE, School of Environ Sci Technol, Dalian University of Technology, Dalian 116024, PR China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Photocatalytic degradation of five

Tetracycline hydrochloride Chloramphenicol Erythromycin Rifampicin Lincomycin hydrochloride

12

Concentration (mg/l)

antibiotics by ZnIn2S4 under visible light.  Identification of intermediates using LCMS-IT-TOF.  Numerous byproducts generated during photocatalytic degradation of antibiotics.  Degradation pathways for five antibiotics were proposed.

8 adsorption

4

0 -30

a r t i c l e

i n f o

Article history: Received 29 April 2016 Received in revised form 6 July 2016 Accepted 7 July 2016 Available online 7 July 2016 Keywords: Visible light LCMS-IT-TOF Reaction kinetics Degradation pathway

0

30

60 90 120 Reaction time (min)

A comprehensive study of degradation efficiency and transformation pathways derived from photocatalytic degradation of five antibiotics including tetracycline hydrochloride, chloramphenicol, rifampicin, lincomycin hydrochloride and erythromycin on ZnIn2S4 under visible light irradiation was investigated. All the five antibiotics cannot be detected after 90 min photocatalytic degradation by ZnIn2S4. The calculated pseudo-first-order constants (kr) were in the order of tetracycline hydrochloride (0.0858 min1) > erythromycin (0.0846 min1) > lincomycin hydrochloride (0.0285 min1). However, degradation of rifampicin and chloramphenicol cannot be described by pseudo-first-order reaction kinetics. The electron paramagnetic resonance (EPR) results indicated that the main reactive oxygen species in this study was  superoxide radical (O 2 ) and minor active species was hydroxyl radical ( OH). Thirty-four, thirty, twenty, sixteen and eighteen kinds of intermediate species were identified by LCMS-IT-TOF during the 180 min photocatalytic degradation of rifampicin, erythromycin, chloramphenicol, lincomycin hydrochloride and tetracycline hydrochloride, respectively. The proposed photocatalytic degradation pathways provided detailed evolution process of five antibiotics, which would be meaningful for the researches on transformation behavior of antibiotics. Ó 2016 Elsevier B.V. All rights reserved.

The frequently detection of various antibiotics in groundwater, drinking water, surface water, sediment and agricultural land [1]

E-mail address: [email protected] (B. Gao). http://dx.doi.org/10.1016/j.cej.2016.07.029 1385-8947/Ó 2016 Elsevier B.V. All rights reserved.

180

a b s t r a c t

1. Introduction

⇑ Corresponding author.

150

had received increasing concern on antibiotics pollution. Recently, a survey research [2] had indicated that the total emission for 36 antibiotics was estimated to be 53800 tons in China. Tetracyclines, macrolides, chloramphenicols and lincomycin are the main antibiotic categories detected in environment media [2]. Rifampicin is on the World Health Organization’s list of essential medicines, the most important medications needed in a basic health system [3].

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

Erythromycin (ERY) as a macrolide antibiotic is useful for the treatment of a number of bacterial infections and also is an alternative antibiotic for patients that are allergic to penicillin [4]. In addition, it is used for animals to promote the growth of food-producing animals and prevent bacterial infections [4]. The erythromycin and its derivatives belong to the third most frequently used antibiotics in human and animal therapy, as a result, these compounds can be detected frequently in municipal and industrial wastewater [4]. Chloramphenicol (CHL) as an effective broad spectrum antibiotic with excellent antibacterial property had been widely used in veterinary clinics since the 1950s [5]. The CHL had been detected as emerging pollutant in natural waters because that its excellent antibacterial property had prevented the removal efficiencies of conventional treatment technologies such as activated sludge [5]. What’s more, a concentration of up to 28 ng l1 of CHL had been detected in urban water supplies of Shanghai, China [6]. The presence of CHL may potentially threaten ecological system as well as human health [7]. Lincomycin hydrochloride (LCM) is an antibiotic active against the grampositive bacteria, which was widely used for veterinary purposes [8]. Tetracycline hydrochloride (TE) was usually present in aquatic environment due to the great amount of usage, especially the overuse and misuse in developing countries [9], which would result in potential environmental problems. Therefore, tetracycline hydrochloride (TE), erythromycin (ERY), chloramphenicol (CHL), lincomycin hydrochloride (LCM) and rifampicin (RIF) were selected as the target antibiotics in this study. The residual antibiotics and their metabolites in environment could cause chronic toxicity and endocrine disruption to aquatic life and human beings, and increased bacterial resistance [10]. The conventional biological treatment used in municipal or industrial wastewater treatment plants was not effective to eliminate these compounds completely [11]. The increased public concern over these antibiotics had prompted the need to develop effective advanced treatment processes to completely remove antibiotics from secondary clarifier effluents before their final discharge. In addition, it is necessary to develop effective pretreatment process to degrade antibiotics to biodegradable products before entering the municipal wastewater treatment plants. The advanced oxidation processes (AOPs) such as photocatalysis [12–14], ozonation [4,15], photoelectro-Fenton [16,17], activated persulfate oxidation [5] and electrocatalytic destruction [18,19], photocatalysis integrated with adsorption [20], and Fenton integrated with flocculation [21] were often adopted to decompose these antibiotics. The photocatalytic technology has been accorded great significance to address diverse environmental issues [22–24]. There was research [11] indicated that the application of photocatalytic oxidation could result in decreased toxicity of antibiotics. Moreover, photocatalytic oxidation was used to degrade trace organics such as antibiotics, drugs, analgesics, surfactants or herbicides to harmless substrate [25]. Photocatalysis is a rising technology with vast potential in pretreatment of high concentration antibiotics and advanced treatment of trace antibiotics. It is well-known that the shorter the wavelength, the higher the energy, so that most semiconductor photocatalysts can be stimulated by ultraviolet light. However, the ultraviolet light accounts for less than 5% of solar energy [26], it is very important to develop highly active photocatalysts with visible light, as the visible light (390 nm–780 nm) occupies 45% of the solar spectrum [26]. Ternary sulfide ZnIn2S4 with unique layered structure has suitable band gap corresponding to visible-light absorption, which is a good candidate as an eco-friendly visible-light-driven photocatalyst [27] and has been widely investigated for the photocatalytic degradation of refractory organics [28]. Our previous work [29] had found that layered microsphere ZnIn2S4 revealed high adsorption capacity, efficient photocatalytic activity and high stability for photocat-

827

alytic degradation of refractory halogenated compounds. Based on the above considerations, in this paper, self-prepared ZnIn2S4 was chosen as photocatalyst for degradation of antibiotics. The photocatalytic degradation intermediate products of antibiotics are very complicated due to the large variety in environment and intricate molecular structure of antibiotics. The aim of this study was to identify the degradation products resulting from photocatalytic degradation of the selected five antibiotics on ZnIn2S4 under visible light and thus to propose the photocatalytic degradation pathway. Intermediate compounds formed during photocatalytic degradation were monitored through sensitive technique high performance liquid chromatography-Mass spectrometry-Ion trap-Time of flight (LCMS-IT-TOF). The significant advantage of LCMS-IT-TOF used in this experiment is the perfect combination of ion trap and time of flight mass spectrometry, which can achieve high-resolution MSn detection and instantaneous changeover of negative and positive ion mode. The degradation pathways of the selected antibiotics during photocatalytic degradation process were proposed based on the formation of new intermediates and disappearance of some intermediates at 20 min, 90 min, 120 min and 180 min. 2. Experimental 2.1. Materials The selected five antibiotics including tetracycline hydrochloride, chloramphenicol, rifampicin, lincomycin hydrochloride and erythromycin were analytical purity and all purchased from AladdinÒ. The initial concentration of the above mentioned five antibiotics was 10 mg l1. The reagents such as Zn(NO3)26H2O, CH3CSNH2 (TAA) were of analytical grade and supplied by Tianjin Secco Romeo Chemical Reagent Co., Ltd. In(NO3)35H2O and acetic acid were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. The reagents for chromatographic analysis such as methanol, acetonitrile were chromatographically pure and purchased from Fisher Chemical Reagent Co., Ltd. The water used in this experiment was ultrapure water with the specific resistance of 18.2 MX cm and with total organic carbon less than 1 lg l1. All the materials were used as received without further purification. The preparation of ZnIn2S4 was according to our previous research [29], which 0.5 mmol In(NO3)35H2O, 0.25 mmol Zn (NO3)26H2O and an excessive CH3CSNH2 (TAA, 4 mmol) were dissolved in 50 ml deionized water and stirred for complete dissolution. Then the mixture was transferred to a 70 ml Teflon-lined autoclave and maintained at 80 °C for 6 h. After cooling to ambient temperature naturally, yellow product was obtained. The yellow product was washed twice with deionized water by suction filtration and finally dried at 60 °C for 10 h. After grinding and sieving through 320 mesh sieve, the obtained photocatalyst was used for photocatalytic experiment. 2.2. Photocatalytic degradation of antibiotics The photocatalytic degradation of antibiotics was carried out in the three layers of cold trap type photochemical reactor which purchased from Shanghai Binlon Instrument Co., Ltd. The light source was positioned in the innermost layer of the reactor, which was encircled with circulating cooling water in the interlayer of the reactor in order to control reaction temperature. Outer layer of the reactor was filled with reaction solution and the valid volume was 500 ml. The schematic diagram of photocatalytic experiment set-up was shown in Fig. S1. A 100 W iodine-gallium lamp (PHILIPS) with light intensity of 50 mW cm2 was used as the visible

828

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

light source in this experiment. A 25 mg ZnIn2S4 photocatalyst was dispersed in a 500 ml aqueous solution containing 10 mg l1 targeted antibiotics (tetracycline hydrochloride, chloramphenicol, rifampicin, lincomycin hydrochloride or erythromycin). Aeration (0.1 L min1) was adopted at the bottom of the reactor for thorough mixing. Prior to irradiation, the suspension was maintained in the dark for 30 min to ensure the establishment of the adsorption/desorption equilibrium. After adsorption experiment, the suspension was continuously irradiated for 180 min. Samples were withdrawn from the photoreactor at specific time intervals and then separated from photocatalyst by filtration through 0.22 lm millipore (acetate cellulose) membrane for further analysis. In order to verify photodegradation stability of ZnIn2S4, chloramphenicol was selected to carry out the cycle experiments. 10 mg ZnIn2S4 photocatalyst was mixed into 200 ml of chloramphenicol solution and then started the photocatalytic reaction. After 1 h reaction, the 1st run was finished and the suspension was separated by filtration through 0.22 lm millipore membrane to recover the suspended catalysts. The supernatant was used to analyze the concentration of chloramphenicol. The collected used photocatalysts were dispersed in 200 ml fresh chloramphenicol solution to carry out the second run. In this way, the experiments were repeated for 5 times. The production of free reactive radicals was identified using an electron paramagnetic resonance (EPR) spectrometer (Bruker, EMM-6/1/P/L, EMM2098). A 12 mmol l1 BMPO was added in the solution containing ZnIn2S4 photocatalyst. After the mixture was illuminated under a 100 W iodine-gallium lamp for 1 min, the free radicals present in solutions were analyzed immediately by EPR. The microwave frequency and power was set at 9.85 GHz and 6.32 mW, respectively. 2.3. The analysis of antibiotics The quantitative determination of antibiotics during the photocatalytic degradation was determined by high performance liquid chromatography (HPLC, Jasco LC-2000) equipped with UV detector and C18 column of Agilent 5 TC-C18(2) 250  4.6 mm. Unless otherwise specified, the parameters of liquid chromatography for all the samples were set as follows, the sample injection volume was 20 ll, the flow rate of mobile phase was 0.5 ml min1 and chromatographic column temperature was 40 °C. The liquid chromatographic analysis of tetracycline hydrochloride was carried out with a mixture of 25% ultrapure water supplemented with 1% glacial acetic acid and 75% methanol and with a UV detector wavelength of 280 nm. A mobile phase of methanol and ultrapure water (60:40, v/v) and UV detector wavelength of 278 nm were employed for chloramphenicol analysis. The mobile phase for rifampicin analysis was acetonitrile and ultrapure water (60:40, v/v) and the detection wavelength was set at 254 nm. For the analysis of lincomycin hydrochloride, the mobile phase was acetonitrile and ultrapure water (60:40, v/v) and the absorbance detection wavelength was 214 nm. A mobile phase of 0.5 mmol/l H2SO4 solution and acetonitrile (20:80, v/v) and UV detector wavelength of 200 nm were employed for erythromycin analysis. The detailed chromatographic analysis condition of HPLC was shown in Table S1. The data of antibiotic concentration in the photocatalytic degradation experiments were average of three tests and the error bars in the figures represented the standard deviation. The qualitative detection of degradation intermediates was performed by LCMS-IT-TOF (Shimadzu) with an electrospray ionization (ESI) source. The HPLC packed column used in this LCMS-ITTOF was Shim-pack VP-ODS 250 L  2.0 and P/N 228-34937-95 (Shimadzu). The parameters of liquid chromatography for all the samples to be tested were as follows, the flow rate was 0.3 ml min1, sample injection volume was 5 ll and chromato-

graphic column temperature was 40 °C. The mobile phase and detector wavelength were the same as the above-mentioned chromatographic condition of HPLC (Jasco LC-2000). The detailed chromatographic analysis condition of LCMS-IT-TOF was shown in Table S2. Unless otherwise specified, the mass spectrometer was set at the second-stage (MS2), and the ion accumulation time was 10 ms and 60 ms for the first-stage and the second-stage, respectively. The scanning acquisition mass-to-charge ratio (m/z) range was set at 80–450, 80–350, 80–750, 80–850 and 80–420 for tetracycline hydrochloride, chloramphenicol, erythromycin, rifampicin and lincomycin hydrochloride, respectively.

3. Results and discussion 3.1. Photocatalytic degradation of antibiotics Prior to irradiation, dark adsorption was carried out to evaluate the adsorption amount of five antibiotics on ZnIn2S4. After 30 min adsorption, the adsorption amount on ZnIn2S4 showed difference among the five antibiotics (Fig. S2). The adsorption removal of rifampicin, erythromycin and tetracycline hydrochloride was 13%, 24% and 22%, respectively. Specially, ZnIn2S4 showed excellent adsorption performance for lincomycin hydrochloride and 42% removal rate was obtained after 30 min adsorption. No obvious removal of chloramphenicol (0.5%) was observed after 30 min adsorption by ZnIn2S4. Though the highest adsorption removal rate (42%) was obtained on ZnIn2S4, lincomycin hydrochloride displayed the slowest photocatalytic degradation rate. There was no adsorption of chloramphenicol on ZnIn2S4, but the following photocatalytic degradation was improved. This indicated that the antibiotics adsorbed on the surface of photocatalyst might exert complicated effect on the subsequent photocatalytic reaction. The reasons for the difference in adsorption might be related to chemical structure of antibiotics and interactions between ZnIn2S4 and antibiotics. The work of D. D. Dionysiou [30] had found that electrostatic interactions between substrate and the catalyst would influence the adsorption and photocatalytic efficiency of the system. All the five antibiotics cannot be detected after 90 min photocatalytic reaction, but the photocatalytic degradation rate showed diversity (Fig. S2). The photocatalytic degradation of chloramphenicol and rifampicin followed the similar best degradation tendency. The degradation curve of erythromycin was similar to the degradation curve of tetracycline hydrochloride. The photocatalytic degradation of lincomycin hydrochloride on ZnIn2S4 was inferior to the other four antibiotics. In order to quantitatively evaluate the difference in degradation efficiency, pseudo-first-order rate equation was adopted to describe the photocatalytic degradation of five antibiotics.

dC ¼ kr C dt ‘‘kr” is the pseudo-first-order rate constant (min1) which can be determined by plotting ln(C0/Ct) versus reaction time t (min). At the start of photocatalytic reaction, the initial concentration was marked as C0, and the concentration at reaction time t was marked as Ct. As shown in Fig. S3, photocatalytic degradation of three antibiotics including erythromycin, tetracycline hydrochloride and lincomycin hydrochloride can be described by pseudofirst-order rate equation, and the rate constants (kr) of which were 0.0846 min1, 0.0858 min1 and 0.0285 min1, respectively, which were consistent with the degradation curve in Fig. S2. The linear dependence of photocatalytic degradation of lincomycin hydrochloride fitting by first-order reaction kinetics was not good enough (R = 0.9796), which might be attributable to the excellent

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

adsorption of lincomycin hydrochloride on ZnIn2S4. Moreover, the photocatalytic degradation of chloramphenicol and rifampicin cannot be described by pseudo-first-order reaction kinetics. According to the structure of chloramphenicol, there are three CAN bonds in the structure. Compared with the CAC bond, the CAN bond was more likely to be broken during photocatalytic degradation based on the data of bond energy in Table S3. In addition, the CAOH structure in the branch chain of chloramphenicol could be easily oxidized into [email protected] structure under photocatalytic

829

degradation. Therefore, chloramphenicol was degraded promptly in the initial 10 min. There are three CAN bonds and one NAN bond in the structure of rifampicin. It is relatively easy to break the NAN bond in rifampicin structure to generate some simpler intermediates because of the relatively weak bond energy of NAN bond, which can explain the rapidity of its degradation. Although the structure of lincomycin hydrochloride contains relatively lower bond energy of CAS bond, the photocatalytic degradation rate of lincomycin hydrochloride was inferior to

Fig. 1. Single mass spectra of rifampicin and 34 intermediates produced during 180 min photocatalytic degradation on ZnIn2S4.

830

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

HO

O

HO

HO

HO

OH

O

OH

O

HO

OH

OH

OH

OH

OH

HO

OH

O

NH

NH

NH

NH

O

O

O O

O

OH

O

652

O

O

HO

HO

O

626 O

HO

HO

O

O

O

O

OH OH

OH

O

OH OH

OH

OH

OH

N

OH

N

O

N

O

N OH

750

NH

N

O

N O

O

OH

N

OH

792

O

NH

N

O

OH O

OH

O

NH

N

O

O

OH

NH

NH

665

HO

O

O

O

O

OH

OH O

822

O

O

706

HO O

HO

HO

HO O

HO

O OH

OH

OH

OH

OH

OH

OH

OH

O

O

O

NH

O

N

O

O

O

NH

N

NH O

NH O

OH

OH

OH

599

O

O

719

O

O

N

OH

605

O

634

O

O HO

O

HO

HO

O HO

OH

OH

OH

OH

OH

O

HO

O

NH

OH

O

HO

O OH

NH

O

O

O

O

OH

O

O

OH

591

O

O

NH

NH O

OH

557

O

O

OH O

585

OH

OH

OH

HO

OH

OH

HO

O

OH

581

HO

OH

OH OH

O

NH

OH

NH

O

NH

O O

O

O O

OH

O

OH O

532

OH

457

O

417

O

OH

544

O

O

HO

HO

OH OH

OH

OH

OH

OH

OH

OH

OH

NH NH O O

O

O

O O

O

410

O

486

O

514

NH

NH O

OH

OH

OH O

O O

OH O

OH

OH

OH

OH

491

O

489

OH

OH

OH

HO OH

HO

OH

OH

OH

OH

NH O

O

O O

O

O

OH

OH

OH

O

344

O

OH

O

317

OH

428

O

OH O

O

O

O NH

259

OH

466 NH O

OH

HO

HO

OH O

190

HO

OH

OH

OH

270

NH O

NH

O

O O

OH O

OH

394

NH

O

HO

O

254 O

HO

217

451

Fig. 2. Proposed photocatalytic degradation pathways of rifampicin on ZnIn2S4 under visible light (expressed by molecular weight (MW)).

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

831

Fig. 3. Single mass spectra of erythromycin and 30 intermediates produced during 180 min photocatalytic degradation on ZnIn2S4.

chloramphenicol and rifampicin. On the one hand, there was previous research indicated that lincomycin was difficult to be decomposed even by strong oxidants such as NaIO4 and H2O2 [8]. On the other hand, there was 42% lincomycin hydrochloride adsorbed on the surface of ZnIn2S4 at the initial 30 min adsorption phase. The adsorbed lincomycin hydrochloride on ZnIn2S4 would be preferentially degraded at the beginning of the photocatalytic degradation. And the generated intermediates might be adsorbed on ZnIn2S4 and some adsorption sites on ZnIn2S4 surface might be occupied, so that the adsorption of lincomycin hydrochloride was decreased. This further resulted in the concentration of free lincomycin hydrochloride in bulk solution was reduced slowly. The calculation of photocatalytic degradation rate was based on the residual substrate concentration in bulk solution. Therefore, lincomycin hydrochloride showed a little slower photocatalytic degradation rate. In order to verify photodegradation stability of ZnIn2S4, chloramphenicol was selected to carry out the 5 cycles experiment. The chromatogram of photocatalytic degradation after 1, 2, 3, 4 and 5 cycle were shown in Fig. S4. The results showed that chloramphenicol was completely removed even after 5 cycles, which

indicated that ZnIn2S4 was stable in the advanced oxidation process.

3.2. The mechanism and transformation pathways for antibiotics decomposition by ZnIn2S4 In order to determine the main active species responsible for photocatalytic degradation of antibiotics, the electron paramagnetic resonance (EPR with BMPO) technique was employed to probe the reactive oxygen species. From EPR spectrum in Fig. S5, the main peaks engendered in ZnIn2S4 system under visible light  irradiation was the superposition of O 2 and OH spectra and the  was stronger than that of OH. Through the fitting intensity of O 2 analysis, the main reactive oxygen species in this study was superoxide radical (O 2 ) and minor active species was hydroxyl radical (OH). Considering the detected radicals, there were three main degradation mechanism might be involved in the photocatalytic degradation of antibiotics by ZnIn2S4. The first was the bond cleav age of the NAN, CAN, CAOH and CAC via attack of O 2 and OH radicals. The second one was the substitution reaction by hydroxyl

832

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840 O

OH

O

O

O O

OH OH

OH

HO

O

OH

OH

HO

O

HO

O

O

N

OH

HO

176

OH

OH

O

O O

OH

O

N

OH

OH

OH O

O

O

O

O

O

O

HO

O

N

O

O

O

O

733

691

588 OH

O

O

O

O

382 175

OH

O

OH

OH

O

O

O

OH

HO

OH

HO

OH OH

OH

O

O

O O

HO

N

O

O

O

O

H

OH O

OH

O

O

O

O

O

O

OH

O

OH

O

OH

O

O O

530 O

O

719

O

O

319

O

O

O

636

490

O

O OH OH

OH

HO

HO

O

OH

O

OH

OH

HO

O

OH

O

O

OH

OH

OH O

O

O

O

OH

O

516

O

HO

OH

O

O

O O

O

O

256

O O

O

O

O

O O

O

472

O

OH

HO OH

O

O

643

590

O

HO

OH

O

O

OH

O

O

O

OH O

O

502

OH

O O

O

OH

O

O

O O

O

222

O

HO

O

O

O

HO O

O

O

O O

440

O

O

O

O O

607

O

578

O

OH

O

466

O

OH

O

O

OH

O

O

OH

O

OH O

HO

OH

O

O

O

O

O

O

O

O

HO

O

OH

O

O

419

O O

O

294 O

425

O

O

481

368

O

OH

O

O

OH

O

O

OH O

O

O O

OH

HO

O

326

O

OH

232 O

O

O

O

360 414

HO

HO

OH OH

Fig. 4. Proposed photocatalytic degradation pathways of erythromycin on ZnIn2S4 under visible light (expressed by molecular weight (MW)).

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

radical (OH). The third one was the oxidation of CAOH to [email protected] and  the oxidative process through O 2 and OH radicals. The bond energy of main chemical bond appeared in the structure of antibiotics was shown in Table S3. The chemical bond with lower bond energy was broken in the first place under photocatalytic degradation. The intermediate products resulting from photocatalytic degradation of antibiotics by ZnIn2S4 under visible light irradiation at 20 min, 90 min, 120 min and 180 min were analyzed using LCMS-IT-TOF. The HPLC spectra of photocatalytic degradation of rifampicin, erythromycin, chloramphenicol, lincomycin and tetracycline at different reaction time were provided in the Support information in Figs. S6, S8, S10, S12 and S14, respectively, the total ion chromatogram (TIC) spectra of which were also pro-

833

vided in the Support information in Figs. S7, S9, S11, S13 and S15. The single mass spectra of detected intermediates were shown in Figs. 1, 3, 5, 7 and 9. The calculation for identification of generated intermediates can be performed with the information about the mass-to-charge ratio (m/z) of the generated fragment ion, taking into account that the generated ions were generated by loss or gain of a proton. The formation and disappearance of each intermediate species at different reaction time were listed in Tables S4–S8. Based on the abovementioned information, the tentative transformation pathways derived from photocatalytic degradation of antibiotics were proposed in Figs. 2, 4, 6, 8 and 10 and detailed discussions of the five antibiotics were respectively expounded in the following sections.

Fig. 5. Single mass spectra of chloramphenicol and 20 intermediates produced during 180 min photocatalytic degradation on ZnIn2S4.

834

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840 OH

OH

OH OH OH

OH O

O

HN

O

O

O

HN N

270

O

254

HN

OH

O

O

N

N O

OH

N

O

323

OH

195

O

Cl

OH

Cl

OH

OH OH O

OH

HN

O

O

N

N

O

O

N

OH

183

Cl

O

Cl

NH2

293

OH

HN

128

HO

O

278

Cl

O

OH

HO

Cl

OH

244 HO

Cl

Cl

HO

184

O

Cl

O

OH

179

O HN

Cl

O

Cl

OH

129

O

HN

O O

HO

OH O

223

Cl

O

OH

C

C

O

O OH

N O

H2C

HO

167 H2 C

HO

C C

115

138

HO

OH

OH

111 O

CH

CH

122

OH

90

O

CH2 O

O

134

106

OH Fig. 6. Proposed photocatalytic degradation pathways of chloramphenicol on ZnIn2S4 under visible light (expressed by molecular weight (MW)).

3.2.1. Photodegradation pathway analysis of rifampicin In this study, the photocatalytic degradation pathway of rifampicin was proposed in Fig. 2. According to the single mass spectra (MS) in Fig. 1 and additional profiles of formation/disappearance of each intermediate in Table S4, there were four protonated intermediates corresponding to mass spectrum at m/z 793, 586, 487, 429 and five deprotonated intermediates corresponding m/z of 633, 598, 580, 556, 456 were detected after 20 min photocatalytic degradation on ZnIn2S4. The characteristic peak of protonated rifampicin at m/z 823 after 20 min reaction was detected. As shown in Fig. 2, the product ion with m/z 793 came from rifampicin degradation by elimination of methanol, which was further fragmentation produced the compound with m/z 633 via breakage of NAN bond (822 ? 792 ? 634 expressed by relative molecular mass). On the other hand, the rifampicin was partially decomposed to the detected intermediate with m/z 598 by detachment of nitrogenous ring via cleavage of NAN bond and cleavage of nitrogenbearing heterocyclic starting at CAN bond. Subsequent demethylation, dehydration, detachment of acetyl and chain scission reaction occurred, which further produced simpler compounds with m/z 586, 580, 556, 487, 456 and 429. And the degradation pathways expressed by molecular weight (MW) were 822 ? 599; 599 ? 585, 581, 557; 557 ? 457, 486; 486 ? 428. After 90 min reaction, the characteristic peak of RIF was not detected, which was in conformity with its conversion efficiency in Fig. 1 that 100% RIF was transformed within 70 min. From Table S4, three protonated products with m/z 793, 586, 487 and four deprotonated intermediates with m/z 633, 598, 580, 556 detected at 20 min were disappeared and completely transformed to the newly detected products at 90 min. Three protonated inter-

mediates with m/z 719, 591, 545 and six deprotonated ones with m/z 750, 705, 604, 531, 490, 465 were formed at 90 min. The photocatalytic degradation mechanism involved in this stage mainly included cleavage of nitrogen-bearing heterocyclic starting at CAN bond, detachment of acetoxyl group, methoxyl, detachment of nitrogenous ring via cleavage of NAN bond. The possible degradation pathways at this stage expressed by molecular weight (MW) were as follows: 822 ? 719 ? 605 ? 491; 822 ? 706; 792 ? 750; 634 ? 591 ? 532 ? 466; 585 ? 544; 581 ? 491. Partial products detected at 90 min with m/z 719, 591, 545 (protonated) and 750, 705, 604, 531, 490 (deprotonated) were completely decomposed as photocatalytic reaction progress. Some simpler compounds with m/z 665, 515 (protonated) and 651, 488, 409 (deprotonated) appeared at 120 min. The major transformation process for further degradation of intermediates was the detachment of nitrogenous ring via cleavage of NAN bond, cleavage of nitrogen-bearing heterocyclic starting at CAN bond, further dehydration and chain scission. The photocatalytic degradation pathways expressed by molecular weight (MW) at this stage were 750 ? 652; 706 ? 665; 605 ? 489; 532 ? 514; 491, 544 ? 410. The photocatalytic reaction was finished after 180 min reaction under visible light irradiation in this study. Eleven new endproducts with m/z 418, 318, 255, 260, 218 (protonated) and 625, 450, 393, 343, 269, 189 (deprotonated) were detected. The degradation mechanism involved in this stage included the cleavage of nitrogen-bearing heterocyclic ring starting at CAN bond, chain scission via substitution reaction by hydroxide radical, continuous dehydration and cleavage of oxygen heterocyclic ring via hydroxyl radical attack. In Brief, the photocatalytic degradation pathways

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

835

Fig. 7. Single mass spectra of lincomycin and 16 intermediates produced during 180 min photocatalytic degradation on ZnIn2S4.

expressed by molecular weight (MW) at this stage were 514, 466 ? 451; 652 ? 626; 457 ? 417; 428 ? 394; 410 ? 344 ? 317 ? 259; 410 ? 270 ? 254 ? 217 ? 190. There were 34 kinds of intermediate detected during 180 min photocatalytic degradation of RIF. Thereinto, 21 kinds of intermediate were completely decomposed and subsequently degraded to simpler compounds, and other 13 kinds of product were left over in the reaction mixture as end-products after 180 min photocatalytic reaction. 3.2.2. Photodegradation pathway analysis of erythromycin The chemical structure of ERY is complex, which contains a 14membered lactone ring, cladinose and desosamine deoxy sugars [4], yielding the molecular sum formula of C37H67O13N and molecular weight of 733 gmol1. The characteristic peak of ERY was detected in positive mode with m/z 734. Photocatalytic degradation of ERY under visible light irradiation resulted in 31 degradation products recognizable by LCMS-IT-TOF (Fig. 3) in the whole 180 min reaction. By contrast, six degradation products were generated by ozone treatment of ERY in previous study [4]. As shown in Fig. 4, there were broadly three pathways for ERY degradation in this paper. The first one was to generate the products with molecular weight of 175, 176 and 382 arising from elimination of desosamine sugar (175) and cladinose sugar (176) and leaving

behind lactone ring (382) [4]. This indicated that the extra oxygen in the ERY was bonded to the desosamine sugar ring and cladinose sugar ring, respectively. From the mass-to-charge ratio in Table S5, we can see that the desosamine and cladinose sugar were detected even after 180 min photocatalytic reaction, which indicated that these two intermediates didn’t continue to degrade. At the same time, the lactone ring was continued to degrade by deethylation and dehydration to generate the deprotonated intermediate with m/z 318 which was further degraded to deprotonated products with m/z 255 and 221 through ring-opening reaction arising from hydroxyl radical attack. According to the degradation by ozone in previous study [4], the second pathway was primary degradation by de-methylation of the tertiary amine to generate the secondary amine with molecular weight of 719 after 20 min photocatalytic reaction. This primary degradation product was further decomposed to simpler intermediates detected in the following reaction. The degradation mechanism such as the elimination of secondary amine, demethylation, deethylation and dehydration, substitution reaction by hydroxyl radical and ring-opening reaction were involved in this process. The photocatalytic degradation pathways involved in the further degradation (shown in Fig. 4) expressed by molecular weight were as follows: (1) 719 ? 530 ? 516 ? 502 ? 466 ? 368 ? 326; (2) 719 ? 643 ? 607 (607 ? 578 ? 419 ? 232) ? 481 ? 414 (425) ? 360; (3) 719 ? 636 ? 590. The third

836

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840 HO

HO

HO O

O

OH

N

O

SH

HO

OH

316

OH

HO O

O

OH

N

N

O

N H

HO

OH

O

OH

O

N H O

OH

406

N

268

OH N H

HO

OH

374

O N

S

N H

N H

N H

288

O

N

N

N

HO

O

O

294

N H

N H HO

OH

256

OH

HO

346

OH OH

O

O

N

N

O

O

HO

S

N

O

N H

N H

N

N H

OH HO

OH

390

222

O

N H

238

OH

OH

260 O N O

O N H

N

O N

NH 2

HO

344

OH

N H

OH

170

O

184 O N

N

O N H

328

HO

OH

127 Fig. 8. Proposed photocatalytic degradation pathways of lincomycin on ZnIn2S4 under visible light (expressed by molecular weight (MW)).

pathway was ring-opening reaction of lactone ring via radical ion attack and resulted in the generation of a deprotonated intermediate with m/z 690 (733 ? 691), which was further degraded to the detected deprotonated intermediates with m/z of 587, 471, 439, 293 and protonated product with m/z of 491. The photocatalytic degradation pathway was 691 ? 588 ? 490 ? 472 ? 440 ? 294. 3.2.3. Analysis of photodegradation products and pathways of chloramphenicol The chemical structure of CHL includes nitrobenzene ring, propylene glycol and dichloro-acetamide, which has a molecular sum formula of C11H12Cl2N2O5 and molecular weight of 323 g mol1 [18]. As summarized in Table S6, the mass-to-charge ratio of 321 (deprotonated) detected at 0 and 20 min was identified as the characteristic peak of CHL and vanished after 90 min, which was in accordance with its conversion efficiency in Fig. S2 that 100% CHL was transformed within 70 min. According to the structure of intermediates derived from mass spectra peak (Fig. 7), many hydroxylated intermediates were generated after photocatalytic degradation because of the attack of hydroxyl radical. By referring to previous study [18], five pathways for photocatalytic degradation of CHL on ZnIn2S4 within 180 min were proposed in Fig. 6. The degradation was initiated by cleavage of CAN bond yielding protonated product with m/z 195 by losing of dichloroacetamide, which was further dehydrated to form protonated end-product with m/z 180 after 180 min reaction. On the one hand,

another protonated intermediate with m/z 294 was formed by deprivation of methanol from propylene glycol branch chain of CHL, which was further hydroxylated with denitration to produce protonated intermediate with m/z 245 detected at 120 min. Subsequent oxidation of the lateral groups of which via hydroxyl radical attack resulted in the formation of 4-hydroxybenzoic acid (m/z 138 (deprotonated)). On the other hand, dichloro-acetamide (m/z 128 (protonated)) was formed via further degradation of intermediate with m/z 294 as the reaction time prolonged to 180 min, which was further transformed to protonated products with m/z 130, 112 and deprotonated product with m/z of 89. In addition, CHL was also decomposed via dechloridation under hydroxyl radical attack to yield products with m/z 271 (protonated) and 253 (deprotonated). Further hydroxylation of the intermediate (m/z 253) led to formation of a deprotonated product with m/z 182 and further denitration via hydroxyl radical attack resulted in the formation of protonated product with m/z 185. The 4-nitrobenzoic acid (m/z 166 (deprotonated)) was generated by elimination of methanol via hydroxyl radical attack. It was noticed that 4-nitrobenzoic acid and the protonated intermediate with m/z 185 were also changed into 4-hydroxybenzoic acid (m/z 138 (deprotonated)). Further dehydroxylation and cleavage of benzene ring of 4hydroxybenzoic acid led to formation of P-hydroxybenzaldehyde (m/z 122 (deprotonated)) and ring-opening protonated product with m/z 116, respectively. Moreover, CHL can also be degraded through denitration yielding protonated intermediate with m/z

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

837

Fig. 9. Single mass spectra of tetracycline and 18 intermediates produced during 180 min photocatalytic degradation on ZnIn2S4.

279. Subsequent dechloridation via hydroxyl radical attack generated a protonated product with m/z 223, which transferred to protonated intermediate with m/z 134 due to dehydration and cleavage of C-N bond. The intermediates with m/z 122 and 134 were both transformed to benzaldehyde (m/z 105 (deprotonated)).

3.2.4. Phototransformation pathways of lincomycin hydrochloride Lincomycin hydrochloride (LCM) is an antibiotic classified as a constituent of the lincosamide group, which consists of a pyranose ring, an amide moiety and a pyrrolidine ring [31]. The protonated m/z of 407 was the characteristic peak of lincomycin originated

838

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840 H3C

CH3

H3 C N

HO

HO

CH3 N

CH3

HO

CH3

OH

OH

HO

NH2

NH2

NH2

NH2 O

HO

OH

OH

OH O

O

OH

O

OH

246

OH

O

O

OH

444

O

O

O

HO

O

181

383

CH3

CH3

H3 C

O

CH3

N

HO

HO

OH HO NH2

OH

HO

OH O

O

OH

OH

O

306

O

O

OH

O

340

230

O

O

154

CH3

H3 C

HO

CH3

CH3

N OH

HO

OH

NH2

OH

O

OH

OH

158

292 O

O

O

O

122

O

190

O

214 OH HO

OH

OH

OH

O

OH

94

O

240

128

CH2 CH3

CH2

74

OH

O

176 CH2

CH2

CH2

OH

CH3

60

OH

Fig. 10. Proposed photocatalytic degradation pathways of tetracycline on ZnIn2S4 under visible light (expressed by molecular weight (MW)).

from the hydrolysis of lincomycin hydrochloride in aqueous solution. Based on the information of detected intermediates (Fig. 7) and additional profiles of formation/disappearance of each intermediate species over reaction time (Table S7), the transformation pathways followed by photocatalytic degradation of LCM was proposed in Fig. 8. There were four degradation pathways and sixteen main species involved in the 180 min photodegradation of LCM on ZnIn2S4 under visible light. The first pathway was initiated by demethylation from methylthio-moiety and dehydration from the pyranose ring, which resulted in the formation of deprotonated product with m/z 373. By referring to previous research [31], the formation of a protonated species with m/z 289 involved in the cleavage of the pyranose ring and substitution of sulfhydrylgroup with a hydroxyl group. The deprotonated intermediate (268) with m/z 268 derived from the loss of a molecule of water from amide moiety and oxidation of hydroxyl group to aldehyde. Subsequently, the deprotonated product with m/z 221 was generated through detachment of formaldehyde and molecule of water, which was further decomposed to a protonated product with m/z 171 by detachment of pyrrolylene. The pyrrolidine ring with protonated m/z 128 as end-product came from m/z 171 by elimination of amide group. In addition, a durable deprotonated intermediate with m/z 293 was generated from deprotonated product with m/z 373 by detachment of sulfhydryl-group on pyranose ring, loss of a molecule of water from amide moiety and ethyl from pyrrolidine ring. The second pathway started with detachment of

hydroxyl group from amide moiety and produced a deprotonated species with m/z 389, which subsequently underwent detachment of methylthio-group with formation of deprotonated m/z 343. Further detachment of hydroxyl group from pyranose ring resulted in the formation of a deprotonated end-product with m/z 327 that remained in solution after 180 min reaction. There were two deprotonated intermediates with m/z 345, 259 and one protonated product with m/z 184 produced through the third pathway, which involved detachment of methylthio-group and demethylation, the cleavage of pyranose ring via hydrolytic attack, detachment of ethanol-based group and dehydration, respectively. The fourth transformation pathway suggested from previous research [31] was initiated through pyranose ring opening with detachment of C3H6OS and formation of protonated product with m/z 317. Subsequent substitution of ethylene glycol by hydroxyl group along with dehydration led to the production of protonated m/z 257, which further dehydration resulted in the formation of protonated endproduct with m/z 239. 3.2.5. Analysis of photodegradation products and pathways of tetracycline hydrochloride Based on the LC–MS information of byproducts (Fig. 9) and by referring to previous proposed pathway [21], the main degradation process of TE was proposed in Fig. 10. The deprotonated m/z of 443 was the characteristic peak of tetracycline originated from the hydrolysis of tetracycline hydrochloride in aqueous solution. The

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

deprotonated product with m/z of 382 was generated via loss of Ndimethyl group due to the relative low bond energy of CAN and loss of hydroxyl group. On the other hand, the deprotonated product with m/z 245 came from photocatalytic degradation of TE by cleavage of the third ring under attack of hydroxyl radical, whose further fragmentation produced the protonated ions with m/z 230 and 215 via consecutive loss of hydroxyl groups. Subsequent loss of N-dimethyl group and elimination of amide group resulted in the formation of protonated product with m/z 129, which further decomposed to protonated end-products with m/z 75 and 61 via ring opening reaction. The cleavage of the third ring of deprotonated intermediate with m/z 382 via hydroxyl radical attack along with oxidation and detachment of primary amine led to the formation of protonated products with m/z 182 and 154, respectively. Further loss of hydroxyl group and water molecule produced the deprotonated end-product with m/z 121. In addition, the protonated product with m/z 340 was generated through detachment of amide group from intermediate of m/z 382. The detachment of water molecule and hydroxyl group involved in the formation of protonated products with m/z 306 and 191. The demethylation resulted in yielding deprotonated product with m/z 291 and protonated product with m/z 177. The ring opening reaction and the cleavage of ring along with hydroxylation involved in the formation of protonated product with m/z 241 and deprotonated products with m/z 157, 93. 4. Conclusions To sum up, there were up to dozens of intermediate products identified during the photocatalytic degradation of each antibiotic. The major transformation process for rifampicin was the detachment of nitrogenous ring via cleavage of N-N bond, cleavage of nitrogen-bearing heterocyclic starting at C-N bond, demethylation, detachment of acetoxyl group, acetyl and methoxyl. The photocatalytic degradation mechanism suggested for erythromycin mainly included the elimination of desosamine sugar and cladinose sugar and leaving behind lactone ring, demethylation, deethylation and ring-opening reaction. The transformation pathways involved in the degradation of chloramphenicol included detachment of dichloro-acetamide, hydroxylation with denitration, dechloridation, decarbonylation, deprivation of methanol and cleavage of benzene ring. Degradation of lincomycin hydrochloride contained detachment of methylthio-group, pyrrolylene and amide group, demethylation, hydroxylation and cleavage of pyranose ring. Moreover, the degradation of tetracycline hydrochloride depended on detachment of N-dimethyl group, primary amine group and amide group and ring-opening reaction. Acknowledgments This study was supported by the science and technology foundation for talents and young researcher from Xi’an University of Architecture and Technology (Project No. RC1441, RC1440, QN1515 and QN1516). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2016.07.029. References [1] L. Zhou, G. Ying, S. Liu, J. Zhao, B. Yang, Z. Chen, H. Lai, Occurrence and fate of eleven classes of antibiotics in two typical wastewater treatment plants in South China, Sci. Total Environ. 452–453 (2013) 365–376.

839

[2] Q. Zhang, G. Ying, C. Pan, Y. Liu, J. Zhao, Comprehensive evaluation of antibiotics emission and fate in the river basins of China: source analysis, multimedia modeling, and linkage to bacterial resistance, Environ. Sci. Technol. 49 (2015) 6772–6782. [3] 19th WHO Model of Essential Medicines, Retrieved May 10, 2015. [4] D.B. Luiz, A.K. Genena, E. Virmond, H.J. José, R.F.P.M. Moreira, W. Gebhardt, H. Fr. Schröder, Identification of degradation products of erythromycin A arising from ozone and advanced oxidation process treatment, Water Environ. Res. 82 (2010) 797–805. [5] M. Nie, Y. Yang, Z. Zhang, C. Yan, X. Wang, H. Li, W. Dong, Degradation of chloramphenicol by thermally activated persulfate in aqueous solution, Chem. Eng. J. 246 (2014) 373–382. [6] J. Lei, X.L. Hu, D.Q. Yin, H.C. Zhang, Z.Y. Yu, Occurrence, distribution and seasonal variation of antibiotics in the Huangpu River, Shanghai, China, Chemosphere 82 (2011) 822–828. [7] C. Yan, Y. Yang, J. Zhou, M. Liu, M. Nie, H. Shi, L. Gu, Antibiotics in the surface water of the Yangtze Estuary: occurrence, distribution and risk assessment, Environ. Pollut. 175 (2013) 22–29. [8] A.D. Paola, M. Addamo, V. Augugliaro, E. García-López, V. Loddo, G. Marcì, L. Palmisano, Photodegradation of lincomycin in aqueous solution, Int. J. Photoenergy (2006) 1–6. [9] L. Yue, S. Wang, G. Shan, W. Wu, L. Qiang, L. Zhu, Novel MWNTs-Bi2WO6 composites with enhanced simulated solarphotoactivity toward adsorbed and free tetracycline in waterLongfei, Appl. Catal. B 176 (2015) 11–19. [10] A.C. Johnson, V. Keller, E. Dumont, J.P. Sumpter, Assessing the concentrations and risks of toxicity from the antibiotics ciprofloxacin, sulfamethoxazole, trimethoprim and erythromycin in European rivers, Sci. Technol. 511 (2015) 747–755. [11] B. Czech, I. Jos´ko, P. Oleszczuk, Ecotoxicological evaluation of selected pharmaceuticals to Vibrio fischeri and Daphnia magna before and after photooxidation process, Ecotoxicol. Environ. Saf. 104 (2014) 247– 253. [12] J. Guo, S. Ouyang, P. Li, Y. Zhang, T. Kako, J. Ye, A new heterojunction Ag3PO4/ Cr-SrTiO3 photocatalyst towards efficient elimination of gaseous organic pollutants under visible light irradiation, Appl. Catal. B 134–135 (2013) 286– 292. [13] J. Ren, S. Ouyang, H. Chen, N. Umezawa, D. Lu, D. Wang, H. Xu, J. Ye, Effective mineralization of organic dye under visible-light irradiation over electronicstructure-modulated Sn(Nb1xTax)2O6 solid solutions, Appl. Catal. B 168–169 (2015) 243–249. [14] M. Zhou, D. Han, X. Liu, C. Ma, H. Wang, Y. Tang, P. Huo, W. Shi, Y. Yan, J. Yang, Enhanced visible light photocatalytic activity of alkaline earth metal ionsdoped CdSe/rGO photocatalysts synthesized by hydrothermal method, Appl. Catal. B 172–173 (2015) 174–184. [15] A.G. Gonçalves, J.J.M. Órfão, M.F.R. Pereira, Ozonation of erythromycin over carbon materials and ceria dispersed on carbon materials, Chem. Eng. J. 250 (2014) 366–376. [16] S. Garcia-Segura, E.B. Cavalcanti, E. Brillas, Mineralization of the antibiotic chloramphenicol by solarphotoelectro-Fenton. From stirred tank reactor to solar pre-pilot plant Sergi, Appl. Catal. B 144 (2014) 588–598. [17] I.R. Bautitz, R.F.P. Nogueira, Photodegradation of lincomycin and diazepam in sewage treatment plant effluent by photo-Fenton process, Catal. Today 151 (2010) 94–99. [18] J. Chen, Y. Xia, Q. Dai, Electrochemical degradation of chloramphenicol with a novel Al doped PbO2 electrode: performance, kinetics and degradation mechanism, Electrochim. Acta 165 (2015) 277–287. [19] N. Oturan, J. Wu, H. Zhang, V.K. Sharma, M.A. Oturan, Electrocatalytic destruction of the antibiotic tetracycline in aqueous medium by electrochemical advanced oxidation processes: effect of electrode materials, Appl. Catal. B 140–141 (2013) 92–97. [20] H. Wang, X. Yuan, Y. Wu, G. Zeng, H. Dong, X. Chen, L. Leng, Z. Wu, L. Peng, In situ synthesis of [email protected](Ti) core-shell microparticle for the removal of tetracycline from wastewater by integrated adsorption and visible-lightdriven photocatalysis, Appl. Catal., B, Accepted Manuscript. [21] Y. Fu, L. Peng, Q. Zeng, Y. Yang, H. Song, J. Shao, S. Liu, J. Gu, High efficient removal of tetracycline from solution by degradation and flocculation with nanoscale zerovalent iron, Chem. Eng. J. 270 (2015) 631–640. [22] N. Zhang, M.Q. Yang, S. Liu, Y. Sun, Y.J. Xu, Waltzing with the versatile platform of graphene to synthesize composite photocatalysts, Chem. Rev. 115 (2015) 10307–10377. [23] S. Liu, Z.R. Tang, Y. Sun, J.C. Colmenares, Y.J. Xu, One-dimension-based spatially ordered architectures for solar energy conversion, Chem. Soc. Rev. 44 (2015) 5053–5075. [24] C. Han, M.Q. Yang, B. Weng, Y.J. Xu, Improving the photocatalytic activity and anti-photocorrosion of semiconductor ZnO by coupling with versatile carbon, Phys. Chem. Chem. Phys. 16 (2014) 16891–16903. [25] R.L. Fernández, J.A. McDonald, S.J. Khan, P. Le-Clech, Removal of pharmaceuticals and endocrine disrupting chemicals by a submerged membrane photocatalysis reactor (MPR), Sep. Purif. Technol. 127 (2014) 131–139. [26] H. Yu, X. Quan, Y. Zhang, N. Ma, S. Chen, H. Zhao, Electrochemically assisted photocatalytic inactivation of Escherichia coli under visible light using a ZnIn2S4 film electrode, Langmuir 24 (2008) 7599–7604. [27] L. Yuan, M.Q. Yang, Y.J. Xu, A low-temperature and one-step method for fabricating ZnIn2S4-GR nanocomposites with enhanced visible light photoactivity, J. Mater. Chem. A 2 (2014) 14401–14412.

840

B. Gao et al. / Chemical Engineering Journal 304 (2016) 826–840

[28] W.Y. Lim, M. Hong, G.W. Ho, In situ photo-assisted deposition and photocatalysis of ZnIn2S4/transition metal chalcogenides for enhanced degradation and hydrogen evolution under visible light, Dalton Trans. 45 (2016) 552–560. [29] B. Gao, L. Liu, J. Liu, F. Yang, Photocatalytic degradation of 2,4,6tribromophenol over Fe-doped ZnIn2S4: stable activity and enhanced debromination, Appl. Catal. B 129 (2013) 89–97.

[30] C. Zhao, M. Pelaez, X. Duan, H. Deng, K. O’Shea, D. Fatta-Kassinos, D.D. Dionysiou, Role of pH on photolytic and photocatalytic degradation of antibiotic oxytetracycline in aqueous solution under visible/solar light: Kinetics and mechanism studies, Appl. Catal. B 134–135 (2013) 83–92. [31] P. Calza, C. Medana, E. Padovano, F. Dal Bello, C. Baiocchi, Identification of the unknown transformation products derived from lincomycin using LC-HRMS technique, J. Mass Spectrom. 47 (2012) 751–759.