Heterogeneous activation of peroxymonosulfate by sillenite Bi25FeO40: Singlet oxygen generation and degradation for aquatic levofloxacin

Chemical Engineering Journal 343 (2018) 128–137

Contents lists available at ScienceDirect

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

Heterogeneous activation of peroxymonosulfate by sillenite Bi25FeO40: Singlet oxygen generation and degradation for aquatic levofloxacin

T

⁎

Yang Liua, Hongguang Guoa,b, , Yongli Zhanga, Weihong Tanga, Xin Chenga, Wei Lia a b

College of Architecture and Environment, Sichuan University, Chengdu 610065, China Department of Civil & Environmental Engineering, University of Washington, Box 352700, Seattle, WA 98195-2700, United States

H I G H L I G H T S

G RA P H I C A L AB S T R A C T

Sillenite Bi FeO catalyst was syn• thesized and characterized. of PMS using S-BFO • Activation showed a superior effect on degrada25

40

tion of LVF.

oxygen was identified as the • Singlet main reactive oxygen species using

•

EPR, HPLC–MS and DO monitoring experiments. Parallel generations of 1O2 including the PMS activation and the oxygen vacancy/O∗ production were proposed.

A R T I C L E I N F O

A B S T R A C T

Keywords: Sillenite bismuth ferrite Peroxymonosulfate Singlet oxygen Levofloxacin Mechanism

Sillenite bismuth ferrite (S-BFO) Bi25FeO40 was synthesized by a hydrothermal process and firstly adopted for the activation of peroxymonosulfate (PMS). Multiple characterization was conducted for the morphology and physicochemical features of S-BFO. Degradation of aquatic levofloxacin (LVF) was thoroughly evaluated by using a coupled process for the decontamination of the typical emerging organics. Some crucial parameters for the activation kinetics as well as economic recyclability were examined with the detailed mechanism proposed. For the first time, singlet oxygen (1O2) was identified as the main reactive oxygen species through radical scavenging experiments, electron paramagnetic resonance spectroscopy (EPR) and HPLC-MS determination. Based on the XPS and EPR results, a catalytic mechanism is proposed concerning the parallel generations of 1O2: (i) Bi5+ is replaced by Bi3+, resulting in the formation of an oxygen vacancy in the lattice and an active oxygen (O∗), which could produce 1O2 through the reactions with PMS; (ii) direct formation of 1O2 from radSO-5 generated by the interaction between Bi5+ and PMS. This study demonstrates a novel catalyst for heterogeneous activation of PMS via a non-radical mechanism, which could be alternatively adopted in the decontamination in surface/ground water.

⁎

Corresponding author at: College of Architecture and Environment, Sichuan University, Chengdu 610065, China. E-mail address: [email protected] (H. Guo).

https://doi.org/10.1016/j.cej.2018.02.125 Received 8 December 2017; Received in revised form 27 February 2018; Accepted 28 February 2018 1385-8947/ © 2018 Published by Elsevier B.V.

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

1. Introduction

activators have not been demonstrated via non-radical reactions concerning PMS. Fluoroquinolone antibiotics (FQs) is a kind of important and broadspectrum pharmaceuticals for human and veterinary purpose, with limited biodegradable ability [35]. Levofloxacin (LVF) is a typical fluoroquinolone used to treat severe bacteria and nosocomial infections, which has been recently encountered in the surface or ground water in many countries [36,37]. LVF is known to be resistant to conventional biological oxidation, and causing long-term concerns in the environment, since it has substantial activity against a broad array of gram-positive and gram-negative bacteria [38]. The continuous introduction of LVF into the environment can affect natural waters quality and potentially impact drinking water supplies, and induce proliferation of bacterial drug resistance in ecosystem system [39]. In this study, sillenite bismuth ferrite (S-BFO) was prepared by a hydrothermal synthesis technique, and adopted to activate PMS as a heterogeneous activator for the first time. The crystalline structure, morphology and textural property of the prepared S-BFO were thoroughly characterized by various characterizations. The catalytic performance of S-BFO was evaluated by activation of PMS for the removal of levofloxacin (LVF). In the heterogeneous activation, the critical impacting factors (including PMS concentration, catalyst dosage, pH and temperature) on the degradation of LVF were investigated. The interaction mechanism between bismuth ferrite and PMS was proposed by performing experiments using different radical scavengers and electron spin resonance (ESR) measurement. The results demonstrated new findings concerning the non-radical mechanism for PMS activation by S-BFO, and the great potential in the water decontamination for reluctant compounds.

BiFeO3 (2.0–2.7 eV) with a perovskite structure showed great potential on organic pollutant degradation and water splitting, because of its high chemical stability, ferroelectricity and ferromagnetic properties [1]. Differing from TiO2, the traditional catalyst in UV-activated photocatalysis, many studies have been reported on the visible light photodegradation performance of BiFeO3, including water decontamination for surface/ground water [2,3]. However, due to the synthesizing difficulty of single-phase perovskite-type bismuth ferrite, many impurity phases are derived in the preparation process [4]. Sillenite bismuth ferrite (Bi25FeO40, S-BFO) a typical byproduct in the synthesis of BiFeO3, shows great photocatalytic activity for the narrower energy gap (1.68 eV) and high superparamagnetic behavior, providing advantages for feasible separation and recovery [5]. Previous study has revealed that the formation of sillenite-type and perovskite-type bismuth ferrites mainly depend on the reaction-time [6]. During the past few years, the advanced oxidation processes (AOPs) based on sulfate radicals (%SO4−, 2.5–3.1 V) have been extensively utilized in decontaminations due to the high nonselective reactivity towards most organic pollutants [7]. Thermolysis [8], transition metals [9], ultrasound [10], bases [11], UV [12] and other oxidants (i.e. hydrogen peroxide, ozone) [13,14] are commonly used to active peroxymonosulfate (PMS) or persulfate (PS) to generate %SO4−. Due to its higher energy efficiency and being more economical, the transition metal activation technology has been reported in previous studies. Generally, the transition metal activation of PMS or PS can be achieved in homogeneous and heterogeneous systems. However, the heterogeneous catalysis system is advantageous over the homogeneous catalysis system due to its easily separation and reuse, requiring no secondary treatment and having a broader pH range [15]. Many studies have demonstrated superb degradation efficiency because of the variable chemical states and unoccupied orbitals of these metal-based catalysts and nonmetal catalysts for the activation of PMS by heterogeneous catalyst, such as Co- [16], Mn- [17], Fe- [18], Cu- [19] and Cbased [20]. Among these catalysts, ferrite and ferrite composites are relatively low cost, nontoxic, and have high potential to be engineered as effective PMS activator. Previous studies have demonstrated that Fe2O3 [21], Fe3O4 [22], FeS2 [23], MFe2O4 (M = Cu, Mn, Zn and Co) [24,25] and Bi2Fe4O9 [26] were frequently used for PMS activation. In addition, BiFeO3 and BiFeO3-based composites have been applied to catalytic activation of H2O2, PS or PMS due to the excellent stability and effective recoverability. For instance, Luo et al. reported about the using of BiFeO3 as a Fenton-like catalyst for degrading organic pollutants [27]. An et al. prepared BiFeO3 and graphene-BiFeO3 composite for photo-Fenton like degradation of tetrabromobisphenol A [28,29]. However, most of the incentives for the activation of PMS by ferrite or ferrite composites are extremely related with sulfate radicals, and the contribution of different types of active sites to the ferrite or ferrite catalyst is still not clear and conclusive. Moreover, to the best of our knowledge, there is no research reported on the interactions between PMS and S-BFO for the treatment of organic wastewater. Previous studies have demonstrated that various heterogeneous catalysts (such as CuO, carbons and nitrogen doped reduced graphene oxide) can effectively activate persulfate via a non-radical oxidation process for the removal of organics [30–32]. Recently, it was reported that the non-radical oxidation process could be observed on the activation of PMS via benzoquinone [33]. These studies revealed that other non-radical reactive oxygen species (ROS) do exist in the activation of PMS or PS, nevertheless, the crucial role still seems unclear in most of the studies. In our recent study, singlet oxygen (1O2), instead of %OH or %SO4−, was found as a vital species on the degradation of 2,4-dichlorophenol in the coupled carbon nanotube/persulfate system [34]. As a non-radical reactive oxygen species (ROS), singlet oxygen has been widely used for selective oxidation of organic substrates. Nevertheless, studies on the heterogeneous oxidation for Fe-based heterogeneous

2. Material and methods 2.1. Chemicals Peroxymonosulfate (2KHSO5.KHSO4.K2SO4, AR) and acetonitrile (HPLC grade) were acquired from Sigma-Aldrich (Shanghai, China). Levofloxacin (LVF, 98%), 2,2,6,6-Tetramethyl-4-piperidinol (TEMP, 98%) and 9,10-diphenylanthracene (DPA, 99%) were purchased from TCI Scientific Ltd. (Shanghai, China). 5,5-dimethyl-1-pyrrpline-N-oxide (DMPO) and furfuryl alcohol (FFA, 98%) were purchased from Aladdin Scientific Ltd. (Shanghai, China). Ferric (III) nitrate nonahydrate (Fe (NO3)3·9H2O), bismuth nitrate pentahydrate (Bi(NO3)3·5H2O), tert-butanol (TBA), ethanol, isopropanol (IPA) and other reagents of analytical grade quality were purchased from Kelong Chemical Reagent Co. Ltd. (Chengdu, China). All chemical reagents were used without further purification. 2.2. Synthesis of S-BFO Similar with perovskite-type bismuth ferrites, S-BFO could be synthesized via sol–gel [40], hydrothermal [41], and combustion [42]. In this work, S-BFO was synthesized by a hydrothermal synthesis technique at low temperature. In a typical synthesis procedure, 10 mmol Bi (NO3)3·5H2O and 10 mmol Fe(NO3)3·9H2O were dissolved in dilute 100 mL nitric acid solution (1 mol/L), and then, a 20 mL KOH (10.0 mol/L) solution was slowly added into the mixed solution under magnetical stirring. The final pH of mixture was about 13–14. Subsequently, the suspension was kept under stirring for 15 min, followed by centrifugation and repeated washing with deionized water more than 5 times until neutral pH. The obtained precipitant was dried in a vacuum drying oven at 80 °C for 12 h, and the pressure was remained at −0.09 ∼ −0.10 MPa. The product was then added into a 70.0 mL KOH (4.0 mol/L) solution, which was regarded as an alkali mineralizer, made with both ethanol and water (V:V = 4:3), and the mixture was vigorously stirred for 30 min to obtain a uniform suspension. The above mixture was transferred into a 100 mL stainless-steel autoclave 129

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

The XRD patterns of the catalyst are shown in Fig. 1. For both of fresh and used Bi25FeO40, diffraction peaks with 2θ at 24.6°, 27.7°, 30.2°, 32.8°, 35.2°, 37.5°, 39.7°, 41.8°, 43.7°, 45.5°, 52.4°, 53.8°, 55.7° and 61.7° were observed, which are in a good agreement with the standard data of sillenite Bi25FeO40 (S-BFO) (JCPDS card No. 46-0416). Previous study has confirmed that Bi25FeO40 crystallizes in cubic space group I23 (197) [41]. By using the MDI Jade 6.5, the lattice constant of

fresh and used Bi25FeO40 are calculated as: a = b = c = 10.1780 ± 0.0008 Å and a = b = c = 10.1794 ± 0.0011 Å, respectively, which are in accordance with the standard values (10.1812 Å). The result demonstrates that S-BFO has been successfully obtained under these conditions, and the fair stability of the obtained S-BFO is also demonstrated in Fig. 1 due to the almost unchanged crystal curve before and after the heterogeneous-catalytic reactions. However, in the XRD analysis, the intensity of crystallinity peaks slight decreased after degradation reaction attributed from the organics adsorbed onto the catalyst surface, which will be discussed in the later section. The microscopic morphology of S-BFO was investigated by SEM (Fig. S1). The high-magnification SEM image shows that the S-BFO is composed of large cuboid crystals with a relatively smooth surface, and the sizes of these polyhedron are in the range of 2–5 μm. A similar microscopic morphology of S-BFO was found in a previous study, which investigated the influence of additive agents for the synthesis of S-BFO with different morphologies [41]. In addition, S-BFO samples with other morphologies were also synthesized using hydrothermal process. For example, Zhang et al. reported that the Bi25FeO40 is composed of nanoparticle-assembled BFO tetrahedrons, with many particle-like building blocks [43]. Köferstein et al. prepared a tetrahedron-like Bi25FeO40 with a size of 30 μm, consisting of smaller particles between 0.2 and 2 μm [42]. It was reported that the gathered morphology was obtained due to the strong electrostatic attraction between the oxygen vacancy and O dangling bonds, and the S-BFO unit shows irregular growth, leading to the formation of many defects on the grain surface [41,44]. To identify characteristic functional groups associated with S-BFO, the FT-IR spectroscopy measurement was conducted. The FT-IR spectra of the S-BFO catalytic are shown in Fig. 2. Three strong and broad absorbance bands in the region 400–500 cm−1 and 500–700 cm−1 are shown for the S–BFO samples, which show the typical absorption bands for the Bi–O and Fe–O [45]. The broad band 455 cm−1 and 530 cm−1 are assigned to the mode of stretching vibrations of the Bi–O bending vibration, corresponding to the characteristic peaks of sillenite. Also, the absorbance bands at 577 cm−1 are from the stretching vibrations of Fe–O [46]. The peak at 1330 cm−1 was assigned to the stretching vibration of C–OH [2]. The bands located at 1500–1350 cm−1 indicate the existence of residual organic compounds (like ethanol) during the washing process. The broad band between 3600 and 3000 cm−1 is attributed to O–H stretching, and the peaks at 1634 cm−1 is assigned to H–O–H bending stretching of the adsorbed H2O [2]. Fig. 3 shows N2 adsorption/desorption isotherms and the associated pore size distribution of the S-BFO samples. It can be seen that the SBFO samples presented the type IV isotherm with a type H3 hysteresis

Fig. 1. X-ray diffraction patterns of fresh and used S-BFO.

Fig. 2. FT-IR spectra of S-BFO.

equipped with a Teflon lining, and heated for 20 h at 120 °C. The obtained red brown S-BFO powder was respectively washed with deionized water and ethanol several times, and then dried to constant weight at 60 °C for 24 h before use. 2.3. Characterization The experimental details of characterization of catalysts through XRD, SEM, BET, VSM, FT-IR and XPS were described in Supplementary Material Text S1. 2.4. Degradation of LVF Activation of PMS by the S-BFO catalyst was carried out for the degradation of LVF in water. A typical reaction was conducted in a 250mL glass beaker reactor with fixed amounts of LVF, PMS and catalyst with specified condition. At given time intervals, 2 mL samples were collected, and immediately quenched with FFA. After being centrifuged at 10,000 rpm for 5 min, the supernatant was analyzed by a HPLC system. A Waters e2695 high performance liquid chromatography (HPLC) system equipped with a 2489 UV-visible detector was used to monitor the concentration of LVF with Waters C18 reversed phase column (150 × 4.6 mm, 5 μm particle) at 295 nm. The mobile phase of acetonitrile and water (0.1% formic acid) (15/85) was maintained for 6.5 min at 0.5 mL/min. The column temperature was maintained at 40 °C. A PerkinElmer PinAAcle-900T atomic absorption spectrometer (AAS, USA), coupled with a heated graphite tube atomizer was used to determine the leached iron from the S-BFO catalyst. The PMS concentration was determined using the iodometric method [25]. Dissolved oxygen (DO) generation was measured by a portable dissolved oxygen meter (HACH, HQ30D). 3. Results and discussion 3.1. Characterization of S-BFO

130

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

Fig. 3. Nitrogen sorption isotherms and pore diameter distribution (inset) of S-BFO.

loop, indicating a mesoporous structure in accordance with the classification of IUPAC [47]. It was found that S-BFO presents a high surface area (SBET) of 13.26 m2/g, which is higher than previous reported (SBET = 9.73 m2/g) [48]. The comparable external surface area (Se) of 14.36 m2/g, indicates the lack of micropores in this sample. Lower than the perovskite structure BiFeO3, 0.0472 g/cm3 was obtained for the pore volume of S-BFO [49]. The porosity may be caused from the interspace between the cuboid crystals gathered. By linearly extrapolating the data at t = 0.266–0.739 nm, a straight line can be seen ascribed to the linear statistical thickness (shown in Fig. S2). Moreover, S-BFO displays a narrow hysteresis loop at a relative pressure P/P0 range of 0.90–0.99, and it has a rather narrow pore size distribution concentrated at around 3.0 nm. The magnetic property of S-BFO samples has been investigated by a vibrating sample magnetometer (VSM), the result is shown in Fig. S3. The saturation magnetization value was estimated to be 0.0905 emu/ mg for S-BFO sample. It demonstrates the facile separability of S-BFO which could be easily separated from the solution by an external magnetic field in 2 min, as shown in the lower right panel of Fig. S3 (inset). The elemental composites and chemical states of S-BFO were further elucidated with XPS. The full-scan XPS spectrum of S-BFO in Fig. 4a reveals the presence of only Bi, O and Fe, and a high-resolution XPS spectra was conducted to quantify and evaluate the contribution of each chemical species. As shown in Fig. 4b, the binding energy of Bi 4f (Bi-1), Bi 4f (Bi-2) are 158.6 eV and 163.9 eV, respectively, which are consistent with the literature values of Bi3+ and Bi5+ [42]. The previous studies have confirmed that the formula of S-BFO is Bi3 + 24(Bi5+Fe3+)O40, which suggest the Bi3+ in the tetrahedral position accompanied by oxygen vacancies (E = inert 6s2 electron pair), whereas Bi5+ and Fe3+ ions share the tetrahedral positions within the cage in the Bi25FeO40 [42,50]. The Fe2P spectrum is composed of a doublet structure due to multiple splitting, as shown in Fig. 4c. According to the curve fitting analysis of Fe2P3/2 and Fe2P1/2, the main peak at 710.8 eV (Fe-1, Fe2P3/2) combined with two satellite peaks at 718.3 eV (Fe-2, Fe2P3/2) and 724.1 eV (Fe-3, Fe2P1/2) corresponds to Fe3+ [51]. Furthermore, no peaks of Fe2+ were observed. Fig. 4. XPS spectra of S-BFO: (a) the survey spectrum; (b) high-resolution Bi4f; (c) highresolution Fe2p3/2 and Fe2p1/2.

3.2. Catalytic evaluation and stability of S-BFO The catalytic performance of S-BFO was considered by activation of various inorganic oxidants (PMS, PS, H2O2) for removal of LVF. The removal profiles of LVF under diverse catalyst reaction systems are shown in Fig. 5. Control experiments showed that the degradation of

LVF in single PMS was negligible. In addition, S-BFO causes little LVF adsorption with less than 3% adsorbed onto the surface. The simultaneous presence of S-BFO and various inorganic oxidants led to a 131

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

Levofloxacin is a zwitterion at physiological pH, possessing a carboxylic group with pKa = 5.5, a piperazinyl group with pKa = 8.0 and another proton accepting function with pKa = (6.8 ± 0.3) [55]. Under acidic conditions (pH < 5.5), LVF with positive charge becomes dominant (LVF+), and contribute the accelerative effect of LVF degradation due to the electrostatic attraction between LVF and S-BFO. The pHzpc of SBFO is 2.78 (data were not shown), which means that negative charge exists on the S-BFO surface in the pH range 3.0–11.0. In the neutral and weak-base condition (pH = 7.0 and 9.0), levofloxacin mainly exists in zwitterionic form, which is more readily subjected to ROS attack according to the Guo’ study [56]. Therefore, the removal rate increased with the ascending zwitterionic form. In the strong alkaline solution (pH = 11.0), LVF with negative charge (LVF-) became dominant and PMS also predominantly exists as an anionic species (pKa = 9.3), which could cause the drastic decline in LVF removal due to the higher electrostatic repulsion among S-BFO, LVF− and SO52− [57]. The results demonstrated that the advantage of S-BFO in degrading organic contaminant at pH 3.0–9.0. Fig. 6c illustrates the LVF degradation by S-BFO activated PMS at various catalyst dosages (ranging from 100 to 2000 mg/L). Additionally, the reaction solution pH change were shown in Fig. S4d. It noted that the solution pH was reduced to 3.4–4.2 when the catalyst dosage ranged from 100 to 2000 mg/L. A positive effect as clearly observed at the dosage ranging from 100 to 1000 mg/L due to the increasing interaction between S-BFO and PMS. While the removal rate was descended from 86.5% to 74.8% at the range of 1000 to 2000 mg/ L, which could attribute to the limited activated sites on the surface of the synthesized material and the amount of PMS. The effect of temperature on LVF degradation was also evaluated by varying the reaction temperatures, as exhibited in Fig. 6d. A large ascending trend in LVF removal could be observed when increase the temperature from 15 °C to 40 °C. Moreover, Fig. S4e shows the pH change over the course of the reactions. The stable solution pH at 3.4 indicates that the LVF degradation efficiency decreased was only accordance with the temperature variation. The reaction rate was evaluated by the pseudo-first order kinetics model:

Fig. 5. Comparison for the removal efficiency of LVF in various systems ([LVF] = 5 mg/L, [PMS/PS/H2O2] = 0.6750 mM, [catalyst] = 1000 mg/L, T = 25 °C, initial solution pH 5.8–6.3).

significant enhancement of LVF degradation as compared to the control tests. 86.5%, 56.7% and 18.3% of LVF was degraded within 60 min in the S-BFO/PMS, S-BFO/PS and S-BFO/H2O2 systems, respectively. The results indicate that the S-BFO is active for H2O2, PS and PMS activation. However, compared with S-BFO/H2O2 process, better removal of LVF is observed in S-BFO/PS and S-BFO/PMS, which may be caused by the following reasons: firstly, the reaction pH can crucially affect the interactions between catalyst and oxidants. As shown in Fig. S4a, the pH changing in different systems were monitored. For S-BFO/H2O2, the catalytic reaction occurred at weak alkalis and neutral conditions, which caused the inhibition for H2O2 decomposition [52]. Moreover, differing from the unselective performance of %OH in H2O2, the selective oxidation capability of ROS (i.e. %SO4−) for the target contaminants could also contribute the tendency of LVF degradation in SBFO/PS and S-BFO/PMS process. Of all the three oxidants, PMS showed the most intensive enhancement due to the stability and relatively higher oxidation potential. Previous studies have also demonstrated that PMS showed better performance on metallic activation compared to PS or H2O2 [53]. In order to elucidate the crucial effect of the PMS dosages on the degradation of LVF in the S-BFO/PMS process, the PMS concentration ranged 0.0675–1.3500 mM, as illustrated in Fig. 6a. As can be seen, the removal rate of LVF is 28.3%, 51.7%, 86.5% and 91.1% for concentrations of PMS at 0.0675, 0.1350, 0.3375, 0.6750 and 1.3500 mM, respectively, indicating that the higher PMS concentration could lead to enhancement of the decontamination. However, a limited increase could be observed with continuously increasing of the PMS dosage from 0.6750 to 1.3500 mM, because excessive PMS would inhibit the target compound degradation due to the interior reaction between HSO5− and ROS [54]. Thus, the optimum PMS concentration for the catalytic process can be explained by the competitive reactions which may adversely affect the generation of the reaction species. Otherwise, as shown in Fig. S4b, the reaction pH decreased with the PMS concentration increasing, due to the ionization of PMS molecules. Thus, the PMS concentration would play an important role of the reaction pH. The influence of pH on LVF removal was studied, with the results shown in Fig. 6b. Slight promotion could be observed with the solution pH increasing 3.0 from 9.0, despite of obvious inhibition at pH 11.0, suggesting that the pH dependence phenomenon was mainly caused by the catalytic reaction over the S-BFO surface. As shown in Fig. S4c, the solution pH in the reactions decreased to 2.9–3.4 in the initial pH range of 3.0–9.0. When the initial pH is 11.0, the solution pH has limited change through the whole reaction. The related results correlated well with the distribution of the deprotonated species of fluoroquinolones.

C −ln ⎛ t ⎞ = K obst C ⎝ 0⎠ ⎜

⎟

(1)

where the C0 is initial concentration of LVF, Ct is the concentration of LVF at time t, and Kobs is the observed reaction rate constant. In order to evaluate the activation energy of the S-BFO reaction, the relationship of Kobs and temperature was shown in the inset of Fig. 6d, which was fitted using the Arrhenius equation (Eq. (2)).

lnK obs = lnA−

Ea RT

(2)

where R is the universal gas constant (8.314 J/(mol K). A is a constant, Ea is the activation energy and T is the reaction temperature (K). Eventually, the Ea value for the S-BFO/PMS system was obtained as 30.49 kJ/mol on the basis of the Arrhenius equation. The Ea obtained in this work was slightly lower than the reported α-Mn2O3@α-MnO2/PMS system (32.1–68.8 kJ/mol) [58] or the CNT/PMS system (45.65 kJ/ mol) [59], and higher than the CuFe2O4/PMS/TBBPA (25.7 kJ mol−1) system [25]. The stability and reusability of S-BFO was evaluated in the degradation of LVF in consecutive runs. The recycling results are shown in Fig. S5. The S-BFO were separated from the suspension with a handheld permanent magnet and washed with ethanol and deionized five times sequentially. The removal of LVF reached 86.5%, 54.7%, 56.9%, 55.3% and 56.4% for the first, second, third, fourth and fifth recycle, respectively. The S-BFO synthesized in our study can maintain catalytic activity after the second recycling experiment. It demonstrates that the synthesized catalyst can be recycled and reused, but reused with the loss of some activity for the catalytic degradation of an organic compound with PMS. Fig. S6 shows the microscopic morphology of reused 132

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

Fig. 6. Effect of the related parameters on the degradation of LVF in the S-BFO/PMS system. (a) PMS concentration; (b) initial pH; (c) catalyst dosage; (d) temperature on LVF removal efficiency. ([LVF] = 5.0 mg/L, [PMS] = 0.6750 mM (for b, c, d), [catalyst] = 1000 mg/L (for a, b, d), T = 25 °C, pH = 5.8–6.3 (for a, c, d)).

3.3. Activation mechanism in PMS activation using S-BFO

S-BFO, which has no obvious change and the pH values still maintain at 3.3–3.4 after 60 min reaction on the catalyst reuses tests (Fig. S4f). The deactivation of S-BFO may be related to two reasons: i) the surface chemistry changing of the catalyst due to the residual organic intermediates and metallic transformation in the highly oxidative environment; ii) the inevitable loss of catalyst during the recycling and washing process. To assess the ability of S-BFO/PMS system to remove LVF under environmental relevant conditions, experiments were conducted with humic acid (HA) and river water (RW, Jiang’an river). As a representative of the natural organic matter (NOM), HA has a polymeric structure containing carboxyl and hydroxyl functional groups, and is reported to perform as a quencher for free radicals for its ubiquity in various waters [60]. Fig. S7 shows that the removal of LVF has gradually decreased from 86.5% to 61.1% with the increase of HA concentration from 0 to 10 mg/L. The inhibition of LVF removal efficiently was ascribed to the competition reaction with HA and LVF for ROS [61]. In addition, compared to ultrapure water (UPW), due to the quenching effect, the removal efficiency of LVF decreased to 51.3% in RW. Hence, the acceptable removal rate for LVF in HA solutions and RW, indicates an effective option for removing organic contaminants in the real experimental waters using S-BFO/PMS process.

3.3.1. Identification of the oxidizing species during PMS activation over SBFO It is well accepted that PMS can be activated to generate %SO4− or % OH, and to degrade organic compounds [7,15].To detect the major oxidants in the reactions, isopropanol (IPA) and tert-butanol (TBA) were selected as probe chemicals for the considerable reaction rates with free radicals. IPA is capable of quenching both sulfate radicals and hydroxyl radicals, whereas TBA is an effective scavenger for only %OH [62]. If the %SO4− or %OH is the primary oxidizing species, a significant inhibition on the degradation of LVF would be observed. As the results show in Fig. 7, the degradation only decreased by less than 10% at 60 min, which suggests that %SO4− and %OH do not play an important role in the reactions. Thus, other activate species in the system between S-BFO and PMS may exist, contributing to the degradation of LVF. Recent studies have demonstrated that singlet oxygen could be attributed as a crucial role in the reactions related to PS or PMS [34,63]. Therefore, in order to further confirm this, FFA and NaN3, were selected as probe chemicals for the considerable reaction rates with 1O2. As previous reported, NaN3 and FFA are effective scavenger for 1O2 (k1O2/ 9 −1 −1 s , k1O2/FFA = 1.2 × 108 M−1 s−1) [34]. As NaN3 = 1.0 × 10 M shown in Fig. 7, the degradation of LVF was well inhibited by the addition of NaN3 and FFA. When excess NaN3 and FFA were dosed, the degradation rate of LVF decreased from 86.5% to 27.2%, 41.6%, respectively. Furthermore, it is well-known that 1O2 is a selective oxidizing species, which can react with electron-rich compounds (eg., 133

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

were identified in the EPR spectrum at single DMPO, with or without PMS dosing, indicating that no spins had been trapped without DMPO. When DMPO was added in the S-BFO/PMS/LVF system, both DMPOOH adducts and DMPO-SO4 adducts were observed. At 5 min or later, % SO4− was detected along with %OH, suggesting that both radical species were formed. However, the intensities of the DMPO radical adducts signals indicated that the present concentration of %SO4− and %OH radicals in the reaction system might be lower compared to the UV/ PMS system [67]. As shown in Fig. 8b, differing from the DMPO addition system, a typical three-line EPR spectrum was observed once TEMP was dosed, demonstrating a direct evidence for 1O2 in the S-BFO/ PMS system. A relatively weak EPR spectrum was also identified in the single PMS system due to the generation of 1O2 from the self-decomposition of PMS [33,68] (Eq. (3)). These results indicated that it is 1O2 that constitutes the major active species in the S-BFO/PMS system. − 2− 2− 1 −1 S−1) HSO− 5 + SO5 → HSO4 + SO4 + O2 (k1 = 0.2 M

(3)

In addition, previous studies have demonstrated that 1O2 can react with a chemical probe 9, 10-diphenylanthracene (DPA) to generate the 9, 10-dibenzanthracene peroxides (DAPO2), showing an alternative identification method for 1O2 [69]. In this study, the formation of DAPO2 was monitored using the HPLC-MS technique (shown in Fig. S10). The chromatographic peak of DAPO2 was observed only in the SPBFP/PMS system (30 min), compared to the single PMS and S-BFO (0 min) systems. This result further confirms the existence of 1O2 on the degradation of LVF in the heterogeneous catalytic reactions between SBFO and PMS.

Fig. 7. Effects of scavengers on LVF degradation in the S-BFO/PMS system. ([LVF] = 5.0 mg/L, [PMS] = 0.6750 mM, [catalyst] = 1000 mg/L, T = 25 °C, pH = 5.8–6.3, [FFA] = [NaN3] = 10 mM, [TBA/IPA] = 200 mM).

phenols) but did not react with saturated alcohols (eg., TBA, IPA) [64]. Based on this, phenol or nitrobenzene (NB) was selected to verify the competitive effect, demonstrating the existence of 1O2, with the results shown in Fig. S8. The degradation rate obviously got slowed down by the addition of phenol or nitrobenzene (NB). The inhibition of phenol or NB confirms that 1O2 is likely produced in the heterogeneous catalytic reactions between S-BFO and PMS. In addition, the leached Fe was 0.00–0.01 mg/L at the reaction process as shown in Fig. S9. The results demonstrate that S-BFO has a potential applications for pollution degradation by activation of PMS.

3.3.3. Proposed mechanism of PMS activation by S-BFO To further elucidate the internal atmosphere variation, the dissolved oxygen (DO) change was monitored through the reaction. As shown in Fig. S11, it can be seen that the reaction solution dissolved oxygen (DO) increased in the catalyst reaction process. A previous study has demonstrated that singlet oxygen could decay quickly to triplet oxygen (3O2) and the increase of DO is indirect evidence for the 1O2 existence in the S-BFO/PMS coupled system [33]. Since 1O2 is attributed as the main active species in the S-BFO/PMS system, the oxygen could be the critical limitation factor. The generated 1 O2 may come from dissolved oxygen in solution or lattice oxygen in the S-BFO. To elucidate the oxygen, various gas environments (including O2, N2 and air bubbling) were adopted for comparison, and the results are illustrated in Fig. S12. Insignificant variation was observed for the LVF degradation in the different systems. Previous studies have

3.3.2. Chemical detection of active species In order to further investigate the ROS involved in the S-BFO/PMS system, EPR spectroscopy was conducted with DMPO and TEMP as the spin trap agents. DMPO is generally considered as a good probe for % SO4− and %OH in the published study [65], while TEMP can react with 1 O2 by spin trapping to yield paramagnetic 2,2,6,6-tetramethyl-4-piperidone-1-oxyl (TEMP-O) [34]. Previous studies have demonstrated that %SO4− and %OH can be discerned by determination of the signals of DMPO-OH adducts and DMPO-SO4 adducts, respectively [22,66]. As shown in Fig. 8a, no peaks

Fig. 8. EPR spectra of DMPO-OH and DMPO-SO4 adducts (a) and TEMP-1O2 adduct (b). ([LVF] = 5.0 mg/L, [PMS] = 0.6750 mM, [catalyst] = 1000 mg/L, T = 25 °C, pH = 5.8–6.3, [DMPO] = 80 mM, [TEMP] = 10 mM).

134

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

≡Fe3+ + HSO5− →%SO5− + ≡Fe2+ + H+ ≡Fe

2+

≡Fe

2+

≡Bi

5+

2 %SO5

+ HSO5 + HSO5 +

−

− −

→%SO4− →SO4

2HSO5−

+ H2 O

→2

−

+ ≡Fe

+ ≡Fe

%SO5−

→2HSO4−

3+

3+

+ 1.5

+ OH

(5)

−

+ %OH

+ ≡Bi 1

(4)

−

3+

+ 2H

O2

≡ Bi5 + + O2 − → ≡Bi3 + + Ovac + 0.5O2

(6) +

(7) (8) (9)

Ovac → O∗

(10)

O∗ + HSO5− → HSO−4 + 1O2

(11)

1O 2

(12)

→ 3O2

Bi25FeO40 + SO52 − → Bi25FeO39 + SO24− + 1O2

(13)

− 1 Bi25FeO40 + HSO− 5 → Bi25 FeO39 + HSO4 + O2

(14)

Firstly, PMS could be activated by Fe3+ on the surface of S-BFO, producing %SO4−, %SO−5 and %OH as the main species through the Eqs. (4)–(6) [53]. Bi5+ could similarly activate PMS directly to %SO−5, leading to the formation of 1O2 through Eq. (7), (8) [26,72]. Moreover, it was demonstrated that 1O2 could come from the Bi5+/Bi3+ redox couple in the crystal lattice. In the S-BFO structure, the [Fe3+O4] tetrahedral group substituted the [Bi3+O3] umbrella group with the transportation of Bi3+ to the oxygen hole, which increases the distance between Bi3+ and the other three oxygen atoms [73,74]. On the other hand, the solitary 6S2 electron pair on the electron orbit of Bi3+, also extends to the oxygen hole direction. Therefore, the fresh generated Bi3+ ions with large radius cause the lattice expansion, resulting in the generation of an oxygen vacancy. Hence, some lattice oxygen atoms could be released and then transformed into active oxygen (O∗). Once PMS was dosed into the solution, O∗ could also be converted to 1O2 (Eqs. (9)–(11)) [75]. The generation of 1O2, with higher energy than the ground state triplet oxygen, will decay rapidly to triple oxygen (3O2), as in Eq. (12). In short, the mechanism of catalytic reaction for PMS generating singlet oxygen can be attributed to Eqs. (13), (14). Based on the above discussion, a feasible mechanism for the degradation of LVF concerning the interactions between S-BFO and PMS is proposed in Fig. 10.

Fig. 9. XPS spectra of O1s envelope for the fresh and the used S-BFO.

found that the presence of DO showed the significant promotion for ROS generated [70]. Therefore, the generated 1O2 should come from lattice oxygen in the S-BFO. In the preparation of the S-BFO, parts of the Bi5+ ions were replaced by Bi3+ ions, resulting in formation of oxygen vacancies in the lattice. As shown in Fig. 9a, the binding energies of O1s in the XPS analysis before the reaction are 529.7 eV, 531.0 eV for O-1 and O-2 respectively. Comparing with this, after the catalyst reaction, the O1s envelope of the S-BFO obtained could be decontrolled into 3 peaks, with binding energies of 529.7 (O-1), 531.0 (O-2), and 532.3 eV (O-3), corresponding to lattice oxygen, chemisorption oxygen and physically adsorbed oxygen, respectively (Fig. 9b) [71]. It is clearly seen from Fig. 9 that the lattice oxygen was relatively weakened and physically adsorbed oxygen was detected after the catalyst reactions. As illustrated in Fig. S13, it is also clearly found that PMS concentration was decreased in the S-BFO/PMS system. It further confirms that the generated 1O2 in the degradation of LVF in the S-BFO/PMS system comes from the lattice oxygen and PMS (although weaker in the control test). Based on the characterization of S-BFO, EPR investigation and the results of the XPS analysis, the mechanism of PMS activation by S-BFO for LVF degradation are proposed as follows. Fig. 10. Mechanism on the heterogeneous activation of peroxymonosulfate by S-BFO.

135

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

4. Conclusions [15]

The sillenite Bi25FeO40 was synthesized via a hydrothermal process, and was first used as an activator for PMS on the degradation of levofloxacin (LVF). The increase of the PMS concentration (0.0675–1.35 mM), catalyst dosages (100–2000 mg/L), reaction temperature (15–40 °C) and pH (3.0–9.0) promoted the LVF degradation in the coupled S-BFO/PMS system, despite of the inhibition at the extreme alkaline condition. When NaN3 and FFA were added, the degradation rate of LVF got decreased (by 27.2–41.6%) compared to TBA, IP and the control test. The probes compounds for 1O2 (phenol or nitrobenzene) showed the significant inhibition performance. Singlet oxygen was further identified using EPR, HPLC-MS and DO monitoring experiments with DMPO and TEMP. Two generation pathways were proposed through XPS analysis for 1O2, including the PMS activation by Bi5+ (direct formation) and the oxygen vacancy/O∗ production from the replacement of Bi5+ to Bi3+ (indirect formation). As a green oxidation, this study provides a potential application to degrade organic contaminants in water using S-BFO with PMS.

[16]

[17]

[18]

[19]

[20]

[21]

[22]

Acknowledgements

[23]

This work was financially supported by the National Natural Science Foundation of China (51508354) and the Project of Science & Technology Bureau of Chengdu (2015-HM01-00502-SF). This work was also partly supported by the Fundamental Research Funds for the Central Universities (No. 2012017yjsy166).The authors are thankful to all the anonymous reviewers for their insightful comments and suggestions.

[24]

[25]

[26]

Appendix A. Supplementary data

[27]

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2018.02.125.

[28]

[29]

References [30]

[1] M. Humayun, A. Zada, Z.J. Li, M.Z. Xie, X.L. Zhang, Y. Qu, F. Raziq, L.Q. Jing, Enhanced visible-light activities of porous BiFeO3 by coupling with nanocrystalline TiO2 and mechanism, Appl. Catal. B-Environ. 180 (2016) 219–226. [2] Z.T. Hu, Z. Chen, R. Goei, W. Wu, T.T. Lim, Magnetically recyclable Bi/Fe-based hierarchical nanostructures via self-assembly for environmental decontamination, Nanoscale 8 (2016) 12736–12746. [3] T.Y. Tan, W. Xie, G.J. Zhu, J. Shan, P.F. Xu, L.N. Li, J.W. Wang, Fabrication and photocatalysis of BiFeO3 with inverse opal structure, J. Porous Mater. 22 (2015) 659–663. [4] J. Silva, A. Reyes, H. Esparza, H. Camacho, L. Fuentes, BiFeO3: a review on synthesis, doping and crystal structure, Integr. Ferroelectr. 126 (2011) 47–59. [5] A.W. Sun, H. Chen, C.Y. Song, F. Jiang, X. Wang, Y.S. Fu, Magnetic Bi25FeO40graphene catalyst and its high visible-light photocatalytic performance, RSC Adv. 3 (2013) 4332–4340. [6] A.M. Lopes, J.P. Araujo, S. Ferdov, Room temperature synthesis of Bi25FeO39 and hydrothermal kinetic relations between sillenite- and distorted perovskite-type bismuth ferrites, Dalton Trans. 43 (2014) 18010–18016. [7] L.W. Matzek, K.E. Carter, Activated persulfate for organic chemical degradation: a review, Chemosphere 151 (2016) 178–188. [8] M.G. Antoniou, A.A. de la Cruz, D.D. Dionysiou, Degradation of microcystin-LR using sulfate radicals generated through photolysis, thermolysis and e- transfer mechanisms, Appl. Catal. B-Environ. 96 (2010) 290–298. [9] X.M. Xiong, B. Sun, J. Zhang, N.Y. Gao, J.M. Shen, J.L. Li, X.H. Guan, Activating persulfate by Fe0 coupling with weak magnetic field: performance and mechanism, Water Res. 62 (2014) 53–62. [10] S. Wang, N. Zhou, Removal of carbamazepine from aqueous solution using sonoactivated persulfate process, Ultrason. Sonochem. 29 (2016) 156–162. [11] O.S. Furman, A.L. Teel, R.J. Watts, Mechanism of base activation of persulfate, Environ. Sci. Technol. 44 (2010) 6423–6428. [12] X.X. He, A.A. de la Cruz, D.D. Dionysiou, Destruction of cyanobacterial toxin cylindrospermopsin by hydroxyl radicals and sulfate radicals using UV-254 nm activation of hydrogen peroxide, persulfate and peroxymonosulfate, J. Photochem. Photobiol. A 251 (2013) 160–166. [13] S.S. Abu Amr, H.A. Aziz, M.N. Adlan, Optimization of stabilized leachate treatment using ozone/persulfate in the advanced oxidation process, Waste Manage. (Oxford) 33 (2013) 1434–1441. [14] I. Epold, N. Dulova, Oxidative degradation of levofloxacin in aqueous solution by

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40] [41]

[42]

136

S2O82-/Fe2+, S2O82-/H2O2 and S2O82-/OH- processes: a comparative study, J. Environ. Chem. Eng. 3 (2015) 1207–1214. W.-D. Oh, Z. Dong, T.-T. Lim, Generation of sulfate radical through heterogeneous catalysis for organic contaminants removal: current development, challenges and prospects, Appl. Catal. B-Environ. 194 (2016) 169–201. W.D. Oh, Z.L. Dong, T.T. Lim, Hierarchically-structured Co-CuBi2O4 and CuCuBi2O4 for sulfanilamide removal via peroxymonosulfate activation, Catal. Today 280 (2017) 2–7. Y. Wang, H. Sun, H.M. Ang, M.O. Tadé, S. Wang, 3D-hierarchically structured MnO2 for catalytic oxidation of phenol solutions by activation of peroxymonosulfate: structure dependence and mechanism, Appl. Catal. B-Environ. 164 (2015) 159–167. W.D. Oh, Z.L. Dong, Z.T. Hu, T.T. Lim, A novel quasi-cubic CuFe2O4-Fe2O3 catalyst prepared at low temperature for enhanced oxidation of bisphenol A via peroxymonosulfate activation, J. Mater. Chem. A 3 (2015) 22208–22217. W.D. Oh, S.K. Lua, Z.L. Dong, T.T. Lim, A novel three-dimensional spherical CuBi2O4 consisting of nanocolumn arrays with persulfate and peroxymonosulfate activation functionalities for 1H-benzotriazole removal, Nanoscale 7 (2015) 8149–8158. X. Duan, Z. Ao, L. Zhou, H. Sun, G. Wang, S. Wang, Occurrence of radical and nonradical pathways from carbocatalysts for aqueous and nonaqueous catalytic oxidation, Appl. Catal. B-Environ. 188 (2016) 98–105. F. Ji, C. Li, X. Wei, J. Yu, Efficient performance of porous Fe2O3 in heterogeneous activation of peroxymonosulfate for decolorization of Rhodamine B, Chem. Eng. J. 231 (2013) 434–440. C. Tan, N. Gao, Y. Deng, J. Deng, S. Zhou, J. Li, X. Xin, Radical induced degradation of acetaminophen with Fe3O4 magnetic nanoparticles as heterogeneous activator of peroxymonosulfate, J. Hazard. Mater. 276 (2014) 452–460. S.-Y. Oh, S.-G. Kang, D.-W. Kim, P.C. Chiu, Degradation of 2,4-dinitrotoluene by persulfate activated with iron sulfides, Chem. Eng. J. 172 (2011) 641–646. Y. Ren, L. Lin, J. Ma, J. Yang, J. Feng, Z. Fan, Sulfate radicals induced from peroxymonosulfate by magnetic ferrospinel MFe2O4 (M=Co, Cu, Mn, and Zn) as heterogeneous catalysts in the water, Appl. Catal. B-Environ. 165 (2015) 572–578. Y.B. Ding, L.H. Zhu, N. Wang, H.Q. Tang, Sulfate radicals induced degradation of tetrabromobisphenol A with nanoscaled magnetic CuFe2O4 as a heterogeneous catalyst of peroxymonosulfate, Appl. Catal. B-Environ. 129 (2013) 153–162. W.D. Oh, Z. Dong, G. Ronn, T.T. Lim, Surface-active bismuth ferrite as superior peroxymonosulfate activator for aqueous sulfamethoxazole removal: performance, mechanism and quantification of sulfate radical, J. Hazard. Mater. 325 (2017) 71–81. W. Luo, L.H. Zhu, N. Wang, H.Q. Tang, M.J. Cao, Y.B. She, Efficient removal of organic pollutants with magnetic nanoscaled BiFeO3 as a reusable heterogeneous fenton-like catalyst, Environ. Sci. Technol. 44 (2010) 1786–1791. J.J. An, L.H. Zhu, N. Wang, Z. Song, Z.Y. Yang, D.Y. Du, H.Q. Tang, Photo-Fenton like degradation of tetrabromobisphenol A with graphene-BiFeO3 composite as a catalyst, Chem. Eng. J. 219 (2013) 225–237. J.J. An, L.H. Zhu, Y.Y. Zhang, H.Q. Tang, Efficient visible light photo-Fenton-like degradation of organic pollutants using in situ surface-modified BiFeO3 as a catalyst, J. Environ. Sci.-China 25 (2013) 1213–1225. H. Lee, H.J. Lee, J. Jeong, J. Lee, N.B. Park, C. Lee, Activation of persulfates by carbon nanotubes: oxidation of organic compounds by nonradical mechanism, Chem. Eng. J. 266 (2015) 28–33. J. Kang, X.G. Duan, L. Zhou, H.Q. Sun, M.O. Tade, S.B. Wang, Carbocatalytic activation of persulfate for removal of antibiotics in water solutions, Chem. Eng. J. 288 (2016) 399–405. T. Zhang, Y. Chen, Y.R. Wang, J. Le Roux, Y. Yang, J.P. Croue, Efficient peroxydisulfate activation process not relying on sulfate radical generation for water pollutant degradation, Environ. Sci. Technol. 48 (2014) 5868–5875. Y. Zhou, J. Jiang, Y. Gao, J. Ma, S.Y. Pang, J. Li, X.T. Lu, L.P. Yuan, Activation of peroxymonosulfate by benzoquinone: a novel nonradical oxidation process, Environ. Sci. Technol. 49 (2015) 12941–12950. X. Cheng, H. Guo, Y. Zhang, X. Wu, Y. Liu, Non-photochemical production of singlet oxygen via activation of persulfate by carbon nanotubes, Water Res. 113 (2017) 80–88. M. Sturini, A. Speltini, F. Maraschi, L. Pretali, A. Profumo, E. Fasani, A. Albini, R. Migliavacca, E. Nucleo, Photodegradation of fluoroquinolones in surface water and antimicrobial activity of the photoproducts, Water Res. 46 (2012) 5575–5582. J.M. Conley, S.J. Symes, M.S. Schorr, S.M. Richards, Spatial and temporal analysis of pharmaceutical concentrations in the upper Tennessee River basin, Chemosphere 73 (2008) 1178–1187. H.C. Su, G.G. Ying, R. Tao, R.Q. Zhang, J.L. Zhao, Y.S. Liu, Class 1 and 2 integrons, sul resistance genes and antibiotic resistance in Escherichia coli isolated from Dongjiang River, South China, Environ. Pollut. 169 (2012) 42–49. S.K. Kansal, P. Kundu, S. Sood, R. Lamba, A. Umar, S.K. Mehta, Photocatalytic degradation of the antibiotic levofloxacin using highly crystalline TiO2 nanoparticles, New J. Chem. 38 (2014) 3220. I. Epold, M. Trapido, N. Dulova, Degradation of levofloxacin in aqueous solutions by Fenton, ferrous ion-activated persulfate and combined Fenton/persulfate systems, Chem. Eng. J. 279 (2015) 452–462. T. Liu, Y. Xu, J. Zhao, Low-temperature synthesis of BiFeO3 via PVA sol-gel route, J. Am. Ceram. Soc. 93 (2010) 3637–3641. W. Ji, M. Li, G. Zhang, P. Wang, Controlled synthesis of Bi25FeO40 with different morphologies: growth mechanism and enhanced photo-Fenton catalytic properties, Dalton Trans. 46 (2017) 10586–10593. R. Köferstein, T. Buttlar, S.G. Ebbinghaus, Investigations on Bi25FeO40 powders synthesized by hydrothermal and combustion-like processes, J. Solid State Chem. 217 (2014) 50–56.

Chemical Engineering Journal 343 (2018) 128–137

Y. Liu et al.

radical mechanism, J. Hazard. Mater. 320 (2016) 571–580. [60] Q. Wang, X.H. Lu, Y. Cao, J. Ma, J. Jiang, X.F. Bai, T. Hu, Degradation of Bisphenol S by heat activated persulfate: kinetics study, transformation pathways and influences of co-existing chemicals, Chem. Eng. J. 328 (2017) 236–245. [61] A. Outsiou, Z. Frontistis, R.S. Ribeiro, M. Antonopoulou, I.K. Konstantinou, A.M.T. Silva, J.L. Faria, H.T. Gomes, D. Mantzavinos, Activation of sodium persulfate by magnetic carbon xerogels (CX/CoFe) for the oxidation of bisphenol A: process variables effects, matrix effects and reaction pathways, Water Res. 124 (2017) 97–107. [62] Y. Liu, Y. Zhang, H. Guo, X. Cheng, H. Liu, W. Tang, Persulfate-assisted photodegradation of diethylstilbestrol using monoclinic BiVO4 under visible-light irradiation, Environ. Sci. Pollut. Res. Int. 24 (2017) 3739–3747. [63] W. Tian, H. Zhang, Z. Qian, T. Ouyang, H. Sun, J. Qin, M.O. Tadé, S. Wang, Breadmaking synthesis of hierarchically Co@C nanoarchitecture in heteroatom doped porous carbons for oxidative degradation of emerging contaminants, Appl. Catal. BEnviron. 225 (2018) 76–83. [64] M.A.J. Rodgers, Solvent-induced deactivation of singlet oxygen – additivity relationships in non-aromatic solvents, J. Am. Chem. Soc. 105 (1983) 6201–6205. [65] D. Wu, S.T. Yue, W. Wang, T.C. An, G.Y. Li, H.Y. Yip, H.J. Zhao, P.K. Wong, Boron doped BiOBr nanosheets with enhanced photocatalytic inactivation of Escherichia coli, Appl. Catal. B-Environ. 192 (2016) 35–45. [66] Y.X. Wang, Z.M. Ao, H.Q. Sun, X.G. Duan, S.B. Wang, Activation of peroxymonosulfate by carbonaceous oxygen groups: experimental and density functional theory calculations, Appl. Catal. B-Environ. 198 (2016) 295–302. [67] Y.H. Guan, J. Ma, X.C. Li, J.Y. Fang, L.W. Chen, Influence of pH on the formation of sulfate and hydroxyl radicals in the UV/peroxymonosulfate system, Environ. Sci. Technol. 45 (2011) 9308–9314. [68] D.F. Evans, M.W. Upton, Studies on singlet oxygen in aqueous-solution. 3. The decomposition of peroxy-acids, J. Chem. Soc. Dalton (1985) 1151–1153. [69] S. Miyamoto, G.R. Martinez, M.H.G. Medeiros, P. Di Mascio, Singlet molecular oxygen generated from lipid hydroperoxides by the Russell mechanism: studies using O-18-labeled linoleic acid hydroperoxide and monomol light emission measurements, J. Am. Chem. Soc. 125 (2003) 6172–6179. [70] X.W. Liu, T.Q. Zhang, Y.C. Zhou, L. Fang, Y. Shao, Degradation of atenolol by UV/ peroxymonosulfate: kinetics, effect of operational, parameters and mechanism, Chemosphere 93 (2013) 2717–2724. [71] Y.W. Gao, Z.Y. Zhang, S.M. Li, J. Liu, L.Y. Yao, Y.X. Li, H. Zhang, Insights into the mechanism of heterogeneous activation of persulfate with a clay/iron-based catalyst under visible LED light irradiation, Appl. Catal. B-Environ. 185 (2016) 22–30. [72] B.T. Zhang, L.X. Zhao, J.M. Lin, Determination of folic acid by chemiluminescence based on peroxomonosulfate-cobalt(II) system, Talanta 74 (2008) 1154–1159. [73] X.W. Liu, K.B. Zhou, L. Wang, B.Y. Wang, Y.D. Li, Oxygen vacancy clusters promoting reducibility and activity of ceria nanorods, J. Am. Chem. Soc. 131 (2009) 3140–3141. [74] Y.B. Ding, G.L. Zhang, X.R. Wang, L.H. Zhu, H.Q. Tang, Chemical and photocatalytic oxidative degradation of carbamazepine by using metastable Bi3+ selfdoped NaBiO3 nanosheets as a bifunctional material, Appl. Catal. B-Environ. 202 (2017) 528–538. [75] T. Zhang, Y.B. Ding, H.Q. Tang, Generation of singlet oxygen over Bi(V)/Bi(III) composite and its use for oxidative degradation of organic pollutants, Chem. Eng. J. 264 (2015) 681–689.

[43] L. Zhang, X. Zhang, Y. Zou, Y.-H. Xu, C.-L. Pan, J.-S. Hu, C.-M. Hou, Hydrothermal synthesis, influencing factors and excellent photocatalytic performance of novel nanoparticle-assembled Bi25FeO40 tetrahedrons, CrystEngComm 17 (2015) 6527–6537. [44] Y. Tang, Z. Jiang, J. Deng, D. Gong, Y. Lai, H.T. Tay, I.T. Joo, T.H. Lau, Z. Dong, Z. Chen, Synthesis of nanostructured silver/silver halides on titanate surfaces and their visible-light photocatalytic performance, ACS Appl. Mater. Interface 4 (2012) 438–446. [45] S. Chauhan, M. Kumar, S. Chhoker, S.C. Katyal, A comparative study on structural, vibrational, dielectric and magnetic properties of microcrystalline BiFeO3, nanocrystalline BiFeO3 and coreeshell structured BiFeO3@SiO2 nanoparticles, J. Alloy. Compd. 666 (2016) 454–467. [46] L. Zhang, Y. Zou, J. Song, C.-L. Pan, S.-D. Sheng, C.-M. Hou, Enhanced photocatalytic activity of Bi25FeO40–Bi2WO6 heterostructures based on the rational design of the heterojunction interface, RSC Adv. 6 (2016) 26038–26044. [47] Y.X. Wang, H.Q. Sun, X.G. Duan, H.M. Ang, M.O. Tade, S.B. Wang, A new magnetic nano zero-valent iron encapsulated in carbon spheres for oxidative degradation of phenol, Appl. Catal. B-Environ. 172 (2015) 73–81. [48] Y. Wu, H.J. Luo, X.L. Jiang, H. Wang, J.J. Geng, Facile synthesis of magnetic Bi25FeO40/rGO catalyst with efficient photocatalytic performance for phenolic compounds under visible light, RSC Adv. 5 (2015) 4905–4908. [49] S. Li, G.S. Zhang, H.S. Zheng, Y.J. Zheng, P. Wang, Stability of BiFeO3 nanoparticles via microwave-assisted hydrothermal synthesis in Fenton-like process, Environ. Sci. Pollut. Res. 24 (2017) 24400–24408. [50] D.C. Craig, N.C. Stephenson, Structural studies of some body-centered cubic phases of mixed oxides involving Bi2O3 – structures of Bi25FeO40 and Bi38ZnO60, J. Solid State Chem. 15 (1975) 1–8. [51] S. Farhadi, N. Rashidi, Preparation and characterization of pure single-phase BiFeO3 nanoparticles through thermal decomposition of the heteronuclear Bi[Fe(CN)6] center dot 5H2O complex, Polyhedron 29 (2010) 2959–2965. [52] R.G. Zepp, B.C. Faust, J. Hoigne, Hydroxyl radical formation in aqueous reactions (pH 3–8) of iron(II) with hydrogen-peroxide – the photo-fenton reaction, Environ. Sci. Technol. 26 (1992) 313–319. [53] G.P. Anipsitakis, D.D. Dionysiou, Radical generation by the interaction of transition metals with common oxidants, Environ. Sci. Technol. 38 (2004) 3705–3712. [54] E. Saputra, S. Muhammad, H.Q. Sun, H.M. Ang, M.O. Tade, S.B. Wang, Manganese oxides at different oxidation states for heterogeneous activation of peroxymonosulfate for phenol degradation in aqueous solutions, Appl. Catal. B-Environ. 142 (2013) 729–735. [55] J.A.O. Gonzalez, M.C. Mochon, F.J.B. de la Rosa, Spectrofluorimetric determination of levofloxacin in tablets, human urine and serum, Talanta 52 (2000) 1149–1156. [56] W. Guo, Y. Shi, H. Wang, H. Yang, G. Zhang, Intensification of sonochemical degradation of antibiotics levofloxacin using carbon tetrachloride, Ultrason. Sonochem. 17 (2010) 680–684. [57] H. Kim, H.Y. Yoo, S. Hong, S. Lee, S. Lee, B.S. Park, H. Park, C. Lee, J. Lee, Effects of inorganic oxidants on kinetics and mechanisms of WO3-mediated photocatalytic degradation, Appl. Catal. B-Environ. 162 (2015) 515–523. [58] E. Saputra, H. Zhang, Q. Liu, H. Sun, S. Wang, Egg-shaped core/shell alphaMn2O3@alpha-MnO2 as heterogeneous catalysts for decomposition of phenolics in aqueous solutions, Chemosphere 159 (2016) 351–358. [59] J.B. Chen, L.M. Zhang, T.Y. Huang, W.W. Li, Y. Wang, Z.M. Wang, Decolorization of azo dye by peroxymonosulfate activated by carbon nanotube: radical versus non-

137