- Email: [email protected]
Zn-Fe-CNTs catalytic in situ generation of H2O2 for Fenton-like degradation of sulfamethoxazole
Journal of Hazardous Materials 342 (2018) 166–176
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Zn-Fe-CNTs catalytic in situ generation of H2 O2 for Fenton-like degradation of sulfamethoxazole Yong Liu a,b , Qin Fan a , Jianlong Wang b,c,∗ a b c
College of Chemistry and Materials Science, Sichuan Normal University, Chengdu 610066, PR China Collaborative Innovation Center for Advanced Nuclear Energy Technology, INET, Tsinghua University, Beijing 100084, PR China Beijing Key Laboratory of Radioactive Wastes Treatment, Tsinghua University, Beijing 100084, PR China
h i g h l i g h t s • • • •
• Zn-Fe-CNTs was capable of converting O2 to H2 O2 and further to OH. The Fenton-like degradation of sulfamethoxazole (SMX) using Zn-Fe-CNTs as catalyst was studied. The removal efficiency of SMX and TOC was 100% and 51.3%, respectively. The possible reaction mechanism of Zn-Fe-CNTs/O2 process was tentatively proposed.
a r t i c l e
i n f o
Article history: Received 24 May 2017 Received in revised form 5 August 2017 Accepted 7 August 2017 Available online 9 August 2017 Keywords: Fenton-like In situ generation of H2 O2 Micro-electrolysis PPCPs Sulfamethoxazole
a b s t r a c t A novel Fenton-like catalyst (Zn-Fe-CNTs) capable of converting O2 to H2 O2 and further to • OH was prepared through infiltration fusion method followed by chemical replacement in argon atmosphere. The catalyst was characterized by SEM, EDS, TEM, XRD and XPS. The reaction between Zn-Fe-CNTs and O2 in aqueous solution could generate H2 O2 in situ, which was further transferred to • OH. The Fentonlike degradation of sulfamethoxazole (SMX) using Zn-Fe-CNTs as catalyst was evaluated. The results indicated that Zn-Fe-CNTs had a coral porous structure with a BET area of 51.67 m2 /g, exhibiting excellent adsorption capacity for SMX, which enhanced its degradation. The particles of Zn0 and Fe0 /Fe2 O3 were observed on the surface of Zn-Fe-CNTs. The mixture of Zn0 and CNTs could reduce O2 into H2 O2 by microelectrolysis and Fe0 /Fe2 O3 could catalyze in-situ generation of H2 O2 to produce • OH through Fenton-like process. When initial pH = 1.5, T = 25 ◦ C, O2 flow rate = 400 mL/min, Zn-Fe-CNTs = 0.6 g/L, SMX = 25 mg/L and reaction time = 10 min, the removal efficiency of SMX and TOC was 100% and 51.3%, respectively. The intermediates were detected and the possible pathway of SMX degradation and the mechanism of Zn-Fe-CNTs/O2 process were tentatively proposed. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Advanced oxidation processes (AOPs) have been widely used for the degradation of toxic organic contaminants in water and wastewater in the past decades [1]. The classic Fenton- and related reactions are convenient, economical, and green ways to generate • OH for the removal of non-biodegradable and/or toxic compounds [2]. Fenton-like process, using solid iron compounds/complexes as heterogeneous catalysts [3,4], has received increasing atten-
∗ Corresponding author at: Energy Science Building, INET, Tsinghua University, Beijing 100084, PR China. E-mail addresses: [email protected], [email protected] (J. Wang). http://dx.doi.org/10.1016/j.jhazmat.2017.08.016 0304-3894/© 2017 Elsevier B.V. All rights reserved.
tion in recent years. It is also attractive to develop Fenton process which can spontaneously generate H2 O2 in situ, such as electroFenton system and photo-Fenton process [5,6]. However, the low and hardly regulated concentration of H2 O2 generated insitu decreased the degradation efficiency of organic contaminants [7–9]. Therefore, it is necessary to increase the concentration of H2 O2 in solid iron compounds/O2 system. The generation of H2 O2 in-situ by the reaction between Zn0 and O2 was investigated, but the productivity was low [10–12]. According to the theory of electrochemical corrosion, the rate of oxidation-reduction reaction on the surface of metal can be accelerated in the galvanic-type corrosion cell without external power supply. Therefore, the in-situ generation of H2 O2 can be improved by the electrode reaction in the corrosion cell using Zn0 as anode. Cathodic H2 O2 could be produced on the surface of carbon materi-
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
als through the electrochemical process [13,14]. Carbon nanotubes (CNTs) may offer significant advantages for the two-electron reduction of O2 due to their good electrical conductivity, high surface activity and mechanical strength [15]. Thus, high concentration of H2 O2 can be possibly produced by the reduction of O2 in the micro-electrolysis system of Zn0 -CNTs, which formed numerous corrosion cells between the particles of Zn0 and CNTs in aqueous solution. Moreover, the redox potential of Fe2+ /Fe0 (−0.44 V) is higher than that of Zn2+ /Zn0 (−0.76 V), the Fe0 can be deposited on Zn0 /CNTs via chemical replacement. Therefore, when the composite of Zn0 , CNTs and Fe0 reacts with O2 , H2 O2 can be continuously supplied through the micro-electrolysis reaction between O2 and Zn0 -Fe0 /CNTs, which can catalyze the generated H2 O2 to produce • OH. In this system, Fe0 or Fe2+ can be regenerated either by a direct reduction with Zn0 or the reaction with H2 O2 . The release of antibiotics into the aquatic environment caused by extensive use of antibiotics is receiving increasing attention in recent years [16–18]. Sulfamethoxazole (SMX), an antibiotic used worldwide for treatment of urinary infection, has been identified in many wastewater treatment plants (WWTPs) effluents [19]. It is required to develop effective approaches for SMX removal. In this study, Zn-Fe-CNTs composite was synthesized, characterized and used as catalyst for Fenton-like degradation of SMX. The effect of initial pH, Zn-Fe-CNTs dosage, and initial SMX concentration on the removal efficiency of SMX was determined. The intermediates were detected by LC–MS and IC, and the possible degradation pathway of SMX was tentatively proposed. 2. Materials and methods 2.1. Materials and chemicals Sulfamethoxazole (SMX) (C10 H11 N3 O3 S) was a sigma Aldrich product (≥99.0% purity). Multi-walled CNTs (d < 8 nm, l = 10–30 m) were produced by Beijing Deke Daojing NanoCompany, China. Zinc powders (size: 120 meshes, specific surface area: 0.72 m2 /g, purity: >99.8%) were obtained from Shandong Xiya Corporation Ltd., China. Commercial iron powders (≥98.0% purity) with specific surface area of 1.26 m2 /g were purchased from Tianjin Jinke Fine Chemical Industry Research Institute., China. H2 SO4 , NaOH, FeSO4 ·7H2 O and H2 O2 were all of analytical grade. De-ionized water was used in all experiments.
167
Scanning electron microscopy (SEM) and energy dispersive spectrometer (EDS) analyses were performed on emission scanning electron microscope (SEM, SU8010, Hitachi). Transmission electron microscopy (TEM) images were recorded on a transmission electron microscope (HRTEM, JEM 2100 and JEOL). BET-surface areas were measured using a NOVA 3200e bath at 77 K and degassing at 373 K. The pore size distribution was determined by the BJH method. Powder X-ray diffraction (XRD) measurement was performed with Cu K␣ radiation (MiniFlex 600, Rigaku, 40 kV and 15 mA) at a scanning rate of 0.5◦ /min. X-ray photoelectron spectroscopy (XPS) was measured by an AXIS-Ultra instrument (Kratos Analytical, UK). 2.3. Fenton-like degradation of SMX Experiments were conducted at room temperature in a glass bottle (500 mL) containing a 250 mL solution. The real effluent of wastewater was collected at a local municipal wastewater treatment plant. The sample had an initial content of TOC of 9.77 ± 0.36 mg/L and pH of 7.0 ± 0.2. 2.4. Analytical methods H2 O2 concentration was measured by a photometric method on an UV–vis spectrometer (PerkinElmer Lambda 25) at 385 nm using potassium titanium oxalate as chromogenic reagent. SMX concentration was determined by HPLC (Agilent 1200) equipped with a diode array detector (DAD) and a C18 reversed-phase column (4.6 mm × 150 mm). The mobile phase used for SMX was a mixture of distilled water and acetonitrile (55:45 (v/v)) at a flow rate of 1.0 mL/min with a column temperature of 30 ◦ C, and the analytical wavelength was 255 nm. SMX had a retention time of 3.0 min under these conditions. HPLC–MS equipped with the above-mentioned column coupled to a Shimadzu 2010EV mass spectrometer with ESI ion source (LC–MS 2010, Columbia, USA) was used to analyze organic intermediates. It equipped with a photo diode array (PDA) detector, and operated in a negative mode. The aforementioned solvent conditions were used for the analysis of the intermediates. The injected volume was 30 L. Total organic carbon (TOC) content was measured by a Multi TOC/TN Analyzer (2100, Analytik Jena. Germany). The concentration of dissolved zinc and iron generated from the reaction process was determined with a flame atomic adsorption spectroscopy (AAS, HITACHI ZA-3000, Japan).
2.2. Preparation and characterization of Zn-Fe-CNTs
3. Results and discussion
The Zn-Fe-CNTs preparation was as follows. Firstly, 1.5 g Zinc powder was mixed with 0.5 g multi-walled CNTs. 2 mL 40% w polyethyleneglycol 4000 was dropped into the mixture, stirred for 30 min, then dried at room temperature, sintered in the tube furnace at 550 ◦ C for 120 min with a N2 flow speed of 60 mL/min and cooled into room temperature to obtain Zn-CNTs. Secondly, FeSO4 ·7H2 O (0.5 g) were dissolved in 200 mL of aqueous sulfuric acid solution (pH = 3), and Zn-CNTs (1 g) were added in argon atmosphere. After 1 h, Zn-Fe-CNTs composites were obtained. The formation of Zn-Fe-CNTs was based on the following principles: polyethyleneglycol 4000 was used as binder and to bind zinc powder and multi-walled CNTs. When the mixture of zinc powder, multi-walled CNTs and polyethyleneglycol 4000 was heated into 550 ◦ C, zinc powder with melting points of 419.5 ◦ C was melted into liquid and polyethyleneglycol 4000 was decomposed into volatile gases. When the heated mixture was cooled into room temperature, liquid zinc was solidified on the surface of multi-walled CNTs and form Zn-CNTs. When as-prepared Zn-CNTs were put into the Fe2+ solution, the chemical replacement between Fe2+ and Zn0 in Zn-CNTs resulted in the formation of Zn-Fe-CNTs composites.
3.1. Characterization of Zn-Fe-CNTs The N2 adsorption-desorption isotherms and pore size distributions of Zn-Fe-CNTs and Zn-CNTs were given in Fig. 1. The Zn-Fe-CNTs sample provided the IV type isotherm, which are typical for the highly ordering mesoporous material with narrow size distribution. The isotherm of Zn-CNTs was identified as type an III/IV mixed type isotherm, which was characteristic of co-existence of microporous and mesoporous. Zn-Fe-CNTs had a larger BET specific surface area (51.67 m2 /g) than that of Zn-CNTs (26.99 m2 /g). The pore size distribution was calculated using the Barrett-JoynerHalenda (BJH) method (Fig. 4b). The result showed that the average pore sizes of Zn-Fe-CNTs and Zn-CNTs were 16.89 nm and 13.52 nm, respectively. TEM image and SEM micrograph of Zn-Fe-CNTs were shown in Figs. 2 and 3. The samples exhibited coral porous structure and large dispersive metal nano-particles were adhered to the surface of CNTs in the Zn-Fe-CNTs. In order to confirm the element distribution and relative element content in Zn-Fe-CNTs, EDS spectra and element
168
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 1. N2 adsorption/desorption isotherm and pore size distribution (the inset) of Zn-Fe-CNTs (a) and Zn-CNTs(b).
species typically seen in Fe2 O3 [24,26]. Fe 2p1/2 spectra at 723.5 eV was the satellite peaks of Fe3 O4 [27]. XPS results indicated that Fe0 and Zn0 existed on the surface of Zn-Fe-CNTs. However, a few atomic layers of Fe (III) and Zn (II) species existed on Zn-Fe-CNTs surface, probably due to the exposure of Zn-Fe-CNTs in air and aqueous solution. Compared with newly Zn-Fe-CNTs, the variety of the valence of iron was found in the used Zn-Fe-CNTs, indicating the possible oxidation of Fe0 and reduction of Fe (III) by Zn0 in the Zn-Fe-CNTs/O2 system. 3.2. The synergetic effect of Zn-Fe-CNTs/O2 process
Fig. 2. TEM image of Zn-Fe-CNTs.
mapping were recorded in Fig. 3. The element peaks of C and O were observed at 0.25 keV and 0.5 keV, respectively. The peaks of Fe were at 0.7 keV and 6.4 keV and the peaks of Zn were at 1.0 keV and 8.6 keV. The atom percentage of Zn, O, C and Fe was 40.3, 29.1, 23.3 and 7.1, respectively. The XRD pattern of the Zn-Fe-CNTs was shown in Fig. 4. The peaks at 26.77◦ and 59.75◦ was corresponded to the (002) and (103) reflections of carbon (JCPDS 99-0057). The peaks at 36.54◦ , 39.33◦ , 43.47◦ , 54.55◦ and 70.24◦ were assigned to the (002), (100), (101), (102) and (103) reflections of Zn0 (JCPDS 99-0110), respectively. The peaks at 32.03◦ , 34.66◦ , 47.85◦ , 56.82◦ ,63.08◦ and 68.15◦ were assigned to the (100), (002), (102), (110), (103) and (112) reflections of ZnO (JCPDS 99-0111), respectively. The peaks at 32.95◦ was attributed to the (104) reflections of Fe2 O3 (JCPDS 99-0060). The weak peaks at 44.67◦ was correspond to the (110) reflections of Fe0 (JCPDS 99-0064) XPS was used to examine the chemical compositions of Zn-FeCNTs (Fig. 5). Fig. 5a clearly illustrated the existence of Zn, O, Fe and C. Fig. 5b exhibited the Auger KE of Zn [20], and the peak at 987.7 eV could be assigned to ZnO [21]. This was in consistent with the XRD analysis in Fig. 4. In Fig. 5b, a broad peak around 717.4 eV assigned to Fe 2p1/2 was detected, indicating the presence of Fe0 on the surface of Zn-Fe-CNTs [22]. However, binding energies at 710.7 eV and 724.3 eV assigned to Fe 2p3/2 and Fe 2p1/2 were characteristic of Fe2 O3 [23,24], which was also a good catalyzer for the Fenton oxidation [25]. Fig. 5c exhibited the Fe 2p spectra of Zn-Fe-CNTs after reaction. Typically, it featured a strong Fe3+ 2p1/2 peak at 725.6 eV. The peaks at 711.6 eV and 719.1 eV were characteristic for the Fe3+
To evaluate the superiority and synergetic effect of the Zn-Fe-CNTs/O2 process, four control experiments were performed, including: (a) Zn-Fe-CNTs/O2 ; (b) Zn-CNTs + Fe0 /O2 ; (c) Zn-CNTs/O2 ; and (d) Fe0 /O2 ; to treat the SMX aqueous solution under the same conditions (i.e., initial pH of 1.5, O2 gas flow rate of 400 mL/min, Zn-Fe-CNTs, Zn-CNTs and Fe0 dosage of 0.6 g/L, 0.54 g/L and 0.06 g/L, respectively). As shown in Fig. 6a, high SMX removal efficiency (e.g., 100% after 10 min treatment) was obtained by the Zn-CNTs +Fe0 /O2 process, while the low SMX removal efficiency was achieved by Zn-CNTs/O2 alone (e.g., 26.1% after 10 min treatment) and Fe0 /O2 alone (e.g., 35.3% after 10 min treatment), suggesting that the synergetic effect between Zn-CNTs and Fe0 are responsible for the degradation of SMX. The removal efficiency of SMX by Zn-Fe-CNTs/O2 process (e.g., 95.3% after 2 min treatment) was higher than that of Zn-CNTs +Fe0 /O2 process (e.g., 80.2% after 2 min treatment). It may be due to the BET surface area and the average pore sizes of Zn-Fe-CNTs were higher than that of Zn-CNTs (Fig. 1), which improved the mass transfer and active sites. The other reason for the better performance of Zn-Fe-CNTs might be that nano-iron particles had higher catalytic activity for the Fentonlike oxidation [28]. According to Fig. 6b, TOC removal efficiency obtained by the ZnCNTs +Fe0 /O2 process (e.g., 31.8% after 10 min treatment) was about 1.22 and 1.87 times higher than that of the Zn-CNTs/O2 process (e.g., 17.2% after 10 min treatment) and Fe0 /O2 process (e.g., 11.1% after 10 min treatment), respectively. The TOC removal efficiency by ZnFe-CNTs/O2 reached about 51.3% after 10 min treatment, which was much higher than that of the Zn-CNTs +Fe0 /O2 process, indicating that the synergetic effect between Zn-CNTs and iron in Zn-FeCNTs composite played a vital role in the further mineralization of the intermediates to obtain high TOC removal efficiency. The SMX removal efficiency was quite similar with the TOC removal efficiency in Zn-CNTs/O2 process, suggesting that the removal of SMX in Zn-CNTs/O2 process was due to the adsorption of SMX by ZnCNTs. In Fe0 /O2 process, the reaction of Fe0 (s) with oxygen could lead to the formation of reactive oxygen species that were capable of oxidizing contaminants [29]. In our work, the relatively low removal efficiency of SMX was due to the heavy corrosion of Fe0 ,
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 3. SEM-EDS analysis of Zn-Fe-CNTs. (a) SEM image; (b) EDS spectrum; (c) EDS mapping pictures of O, Zn, Fe, C elements.
169
170
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
which limits the efficiency of this process and the weak oxidant of ferryl ion (Fe(IV)) produced from the corrosion of Fe0 by oxygen cannot effectively degrade pollutants, which is in consistent with many literatures [30,31].
3.3. Involved active oxidation species
Fig. 4. The XRD pattern of Zn-Fe-CNTs.
To evaluate the performance of active oxidation species generated by Zn-Fe-CNTs/O2 process, the accumulated H2 O2 concentration in different systems was measured (Fig. 7). The dosage of both Zn-Fe-CNTs and Zn-CNTs were 0.6 g/L. Fig. 7 indicated that the concentration of H2 O2 gradually increased with the reaction time and the H2 O2 accumulation was dependent on initial pH in Zn-CNTs/O2 system. In Zn-CNT/O2 system, the H2 O2 concentration was 5.26 and 22.2 mg/L at 10 min, respectively when the initial pH was 1.5 and 7.0, respectively. It confirmed that H2 O2 could be generated by the reduction of oxygen on the surface of Zn-CNTs due to micro-electrolysis. The H2 O2 generation via the reaction of Zn0 and O2 was evaluated by Zhang and Wen [11,12]. However, the
Fig. 5. XPS full survey spectra (a), Zn Auger KE spectra (b), and Fe 2p spectra (c) of Zn-Fe-CNTs, Fe 2p spectra of Zn-Fe-CNTs after reaction (d).
Fig. 6. The degradation (a) and mineralization (b) of SMX in different processes.
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 7. Variation of H2 O2 concentration with reaction time in different systems.
171
of SMX. There was no obvious difference in the removal efficiency of SMX between t-BuOH addition and KI addition, suggesting that the degradation of SMX was primarily attributed to the action of the surface bounded • OH. In addition, the degradation efficiency of Zn-Fe-CNTs/O2 and Zn-Fe-CNTs/N2 was compared to confirm the role of dissolved oxygen (DO) and the adsorption capacity of Zn-Fe-CNTs for SMX. As shown in Fig. 8, SMX removal (e.g., 100% after 10 min) by the ZnFe-CNTs/O2 process was about five times of that (e.g., 20.1% after 10 min) obtained by Zn-Fe-CNTs/N2 , demonstrating that the generation of H2 O2 and • OH was caused by the reduction of O2 on the surface of Zn-Fe-CNTs and Zn-Fe-CNTs had high adsorption capacity for SMX. 19%-21% of removal was still achieved with addition of t-BuOH or Iodide ion, which can be attributed to the adsorption of SMX by Zn-Fe-CNTs. Fe0 can react with O2 to produce H2 O2 through a two-electron transfer or a series of one-electron transfers via the reaction ferrous iron produced by Fe0 oxidation with oxygen [29–31]. However, the low and hardly regulated concentration of H2 O2 generation was found in Fe0 /O2 system because the action of Fe0 might present in two aspects, i.e. a source of Fe2+ and an electron donor to generate H2 O2 intermediates, which decreased the degradation of organics. In Zn-Fe-CNTs/O2 system, many active intermediate species H2 O2 can be generated and the content of H2 O2 can be adjusted by the component content in Zn-Fe-CNTs material, which is in favor of the generation of • OH and the degradation of organics. 3.4. Factors influencing SMX degradation
Fig. 8. The degradation of SMX by Zn-Fe-CNTs under different condition.
concentration of generated H2 O2 in the Zn0 /O2 system was below 3 mg/L with a stoichiometry of same dosage at neutral solution. In Zn-CNT/O2 system, the high concentration of in situ H2 O2 generation was due to the following reasons. The Zn0 was attached to the surface of CNTs and the galvanic-type corrosion cell was formed between Zn0 and CNTs. The CNTs could improve the two-electron reduction of oxygen due to its good electrical conductivity and high surface activity, which was proved by the in situ H2 O2 generation in electric catalysis process [15,32]. Compared with Zn-CNT/O2 system, at initial pH 1.5 and 7, the concentration of accumulated H2 O2 was rather low (<13.5 mg/L), particularly under acid condition (<1.6 mg/L), indicating that the iron in Zn-Fe-CNTs improved the decomposition of H2 O2 and high decomposition rate of H2 O2 under acid condition was obtained. The influence of different scavengers on the degradation of SMX at initial pH 1.5 was determined to investigate involved active radicals in the process (Fig. 8). Tertiary butanol (t-BuOH) is known as hydroxyl radicals scavengers, which are used to examine the role of hydroxyl radicals [33]. As shown in Fig. 8, the removal of SMX was strongly inhibited with the presence of excess t-BuOH (200.0 mM), indicating that SMX was mainly decomposed by the attack of • OH radicals (including surface-bounded • OH and free • OH). Iodide ion is used to scavenger hydroxyl radicals produced at the surface of Zn-Fe-CNTs [28]. As shown in Fig. 8, addition of excess KI (20 mM) resulted in a considerable decrease in the extent of SMX degradation from 100% (in 10 min in absence of KI) to 21%, indicating that surface-bounded • OH played a significant role in the degradation
3.4.1. Effect of initial pH The pH value of solution played an important role in degradation of organic pollutants by Zn-Fe-CNTs/O2 process, owing to its effect on in-situ generation of H2 O2 from the reduction of O2 by Zn0 CNTs and the in-situ generation of • OH from the decomposition of H2 O2 by Fe0 /Fe(II). On one hand, it is well-known that Fenton oxidation in water is strongly depended on pH value [34]. A lower initial pH value benefits the generation of • OH and results in the higher removal efficiency of pollutant [35]. On the other hand, insitu generation of H2 O2 from the reduction of O2 by Zn-CNTs was largely affected by pH value of solution, as shown in Fig. 7. The effect of pH on SMX removal efficiency was investigated. Since the acidic solution benefit the Fenton oxidation, the initial pH was selected in the range of 1–3 in this study. As shown in Fig. 9a, the SMX removal efficiency decreased with the initial pH increasing from 1.5 to 3. The final removal efficiency of SMX was 100% after 20 min at initial pH of 1.5, but the removal efficiency of SMX was 73.1% and 14.9% at initial pH of 2.0 and 3.0, respectively. The removal efficiency of SMX was 90% after 20 min at the pH of 1, which was lower than at the initial pH of 1.5. Therefore, the initial pH of 1.5 was chosen for the degradation of SMX by Zn-Fe-CNTs/O2 process. In order to explain the effect of initial pH on the removal efficiency of SMX, the variation of pH during the Zn-Fe-CNTs/O2 process was determined and shown in Fig. 9b. It can be seen that the pH value increased gradually with reaction time. The final pH reached 2.1, 2.8, 6.3 and 6.8, respectively when initial pH was 1.0, 1.5, 2.0 and 3.0, respectively. This was attributed to the reduction of H+ and O2 on the surface of Zn-Fe-CNTs. Fenton oxidation of organic pollutants occurred faster with a decrease of pH and better removal efficiency can be achieved at pH of 2–3 [5,36]. In addition, Zn(II) was slowly released into the solution during the degradation of SMX. The concentration of Zn(II) in solution at 20 min reached 169.6, 104.8, 14, and 2.6 mg/L, respectively when initial pH was 1.0, 1.5, 2.0 and 3.0, respectively. The generated Zn(II) possibly led to another environmental problem, but it can be recovered in the formed of Zn(OH)2 when pH rose to neutral with the
172
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 9. Effect of operating parameters on SMX degradation in the Zn-Fe-CNTs/O2 system: (a) initial pH; (b) the variation of the solution pH; (c) initial SMX concentration; (d) Zn-Fe-CNTs dosage.
increase of reaction time or with addition of NaOH to adjust pH because the Ksp of Zn(OH)2 was very low. Meanwhile, there was an inevitable leaching of iron at acidic solution, and the dissolved iron concentration was 4.6 mg/L at initial pH 1.0. However, the dissolved iron concentration was low due to higher pH value and the iron ion could be reduced to Fe0 by Zn0 . When the solution pH rose to neutral, no dissolved iron was detected in this system.
3.4.2. Effect of initial concentration of SMX Fig. 9c shows the effect of initial SMX concentration on SMX degradation with initial pH 1.5, Zn-Fe-CNTs dosage 0.6 g/L and O2 flow rate 400 mL/min. The results showed that the removal efficiency of SMX was 100%, 85% and 65% when SMX concentration was 25, 50 and 75 mg/L, respectively after 10 min, and the apparent rate constant was calculated to be 1.07, 1.95 and 5.38.
3.4.3. Effect of Zn-Fe-CNTs/O2 dosage Fig. 9d shows that when the Zn-Fe-CNTs dosage increased from 0.2 g/L to 0.6 g/L, an obvious increase in the removal efficiency of SMX was observed. For example, after 20 min, the SMX removal efficiency was 57.9%, 88.3%, and 100% respectively. This was mainly due to the increasing amount of active sites for the O2 reduction to produce more reactive oxidants such as H2 O2 and • OH. Nevertheless, the degradation degree was only slightly enhanced when the Zn-Fe-CNTs dosage increased from 0.6 to 1.0 g/L, which was likely attributed to the percentage of • OH scavenged by Fe(II) through undesirable reaction. Thus, the optimum Zn-Fe-CNTs dosage was 0.6 g/L in this experiment.
Fig. 10. The degradation and mineralization of SMX in wastewater by Zn-FeCNTs/O2 process.
3.5. Degradation of SMX in real wastewater The effectiveness of the Zn-Fe-CNTs/O2 process in treating WWTP effluent containing SMX (around12.5 mg/L and 25 mg/L) was evaluated at initial pH 1.5, Zn-Fe-CNTs 0.6 g/L and O2 flow rate 400 mL/min. Fig. 10 shows the variation in SMX concentration and TOC during the treatment process. It can be seen that the Zn-FeCNTs/O2 process was very effective in removing SMX from WWTP
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 11. Evolution of small organic acids and inorganic ions.
effluent. SMX was completely removed after 10 min when its initial concentration was 12.5 mg/L and 25 mg/L. The initial TOC concentration of WWTP effluent containing 12.5 mg/L SMX and 25 mg/L SMX were 15.1 mg/L and 20.3 mg/L, respectively. The removal efficiency of TOC after 10 min was 51.1% and 17.3% when initial concentration of SMX was 12.5 mg/L and 25 mg/L, respectively. 3.6. The intermediate and pathway of SMX degradation 3.6.1. The formation of small organic acids and inorganic ions Short-chain carboxylic acids generated during the degradation of SMX process were detected and shown in Fig. 11. These chromatograms exhibited the peaks related to propanoic (t = 3.53 min), formic (t = 3.67 min) and acetic (t = 3.31 min) acids. The concentration of propanoic and acetic acids was firstly increased and then decreased. The max accumulation of propanoic acid and acetic acid was 6.09 mg/L and 4.85 mg/L respectively, and then decreased to 4.54 mg/L and 3.58 mg/L after 20 min, indicating that propanoic and acetic acids was degraded by • OH. The concentration of formic acid increased continuously up to 3.52 mg/L after 20 min, because the oxidation of formic acid was the final step for the mineralization of organic matters. The evolution of inorganic ions (NH4 + , NO3 − ) during the degradation of SMX was also shown in Fig. 11. In this study, SO4 2− ions were not monitored because H2 SO4 was employed as the pH regulator in the solution. SO4 2− was released during the degradation of SMX by • OH [37]. In Fig. 11, stable NH4 + concentration was observed during SMX degradation process and 2.36 mg/L of NH4 + was generated after 20 min. The concentration of NO3 − was firstly increased and then decreased because the reduction of NO3 − by low valence zinc and iron, 1.52 mg/L of NO3 − was generated after 20 min. The nitrogen content of SMX was mainly converted into ammonium. 3.6.2. Evolution of aromatic intermediates and SMX degradation pathway The transformation of SMX by Fenton process under different conditions has been reported. However, the SMX degradation pathway by Zn-Fe-CNTs/O2 was not clear. The intermediates formed during SMX degradation in the Zn-Fe-CNTs/O2 process were identified by LC–MS in order to determine a probable reaction pathway. The molecular structure of SMX and possible molecular structures of six intermediates and the corresponding mass spectrum were presented in Fig. 12.
173
Several intermediates were found during the Zn-Fe-CNTs/O2 process. These intermediates were produced at different oxidation reaction time. In the mass spectrometry, the peaks of SMX appeared at [M+H]+ 254 with the retention time of 5.8 min. The molecular ion A of the peak with the retention time of 2.1 min was found at [M+H]+ 270, which can be attributed to the addition of 16 mass units to the parent peak [1]. This formula was in consistent with the addition of an • OH to the SMX structure in different positions to yield mono-hydroxylated derivative. Another peak was observed at [M+H]+ 256 with the retention time of 1.6 min, which corresponds to intermediate B. Intermediate B mass units can be attributed to the substitution of the methyl group by • OH radical attack on the isoxazole ring, resulting in the formation of a hydroxylated structure [38]. The intermediate C with the retention time of 1.2 min at [M+H]+ 232 can be attributed to the substitution aromatic ring of hydroxyl group without abstraction of amino group. In addition to steric effect of the molecule, the sulfo- group in the meta-position (towards the sulfo- group) and in the ortho-position towards the amino group was an electron donating substituent [39], which was easily substituted by • OH and generated the intermediate C. Structure of sulphonamide bond was stable in basic solution, but during Zn-Fe-CNTs/O2 reaction, an attack of hydroxyl radicals occurred in the acidic medium, which caused the degradation highly probable. As a result, intermediate D with the retention time of 1.4 min at [M+H]+ 191 was already detected as oxidation products of SMX. The molecular ions E and F for the two peaks were observed at [M+H]+ 237 and [M+H]+ 143, respectively, which was in consistent with the addition of • OH to the intermediate E structure in different positions. The fragment [M+H]+ 173 and 99 were missing, indicating that the initial degradation of SMX was due to the attack of • OH radicals on the benzene ring or isoxazole ring. Base on above analyses, a reaction pathway for the complete mineralization of SMX with • OH as the main oxidant was proposed in Fig. 13. The first oxidation step of SMX was the attack by • OH radicals at different sites of the SMX molecule to form hydroxylated derivatives. These hydroxylated derivatives were further oxidized by • OH into short-chain carboxylic acids accompanied by the release of SO4 2− , NH4 + and NO3 − ions. These short-chain carboxylic acids can be further mineralized into CO2 and H2 O by • OH radicals. 3.7. Proposed reaction mechanism of Zn-Fe-CNTs/O2 process The SEM, EDS, XRD, XPS results demonstrated that nano Zn0 particles and nano Fe0 /Fe2 O3 particles were successfully dispersed into the CNTs skeletal structure and formed Zn-Fe-CNTs composite. TEM results demonstrated that metal particles contacted directly with CNTs, which was benefit for the formation of corrosion cell with Zn0 as anode due to the potential difference of Zn, Fe, CNTs and Zn with lowest potential among them. The control experiments and involved active oxidation species analysis confirmed the occurrence of the in situ generation of H2 O2 via the reduction of O2 by Zn-CNTs and the in situ generation of • OH from H2 O2 by Fe0 /Fe2 O3 catalysis during the reaction between Zn-Fe-CNTs and O2 . The reduction of O2 by Zn0 was prior to Fe0 /Fe2 O3 due to the lower potential of Zn, which favored the catalysis of Fe0 /Fe2 O3 to H2 O2 . According to these analysis data and the present literatures, the proposed reaction mechanisms abridged general view describing of Zn-Fe-CNTs/O2 process are summarized and presented in Fig. 14. The Zn-Fe-CNTs/O2 processes for the oxidation of pollutants can be divided into three stages: (a) The in situ generation of H2 O2 : Base on corrosion electrochemistry theory, the direct contact of Zn0 with CNTs could form zinc-carbon galvanic cells, when Zn-Fe-CNTs con-
174
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 12. The mass spectrum and corresponding possible molecular structures of the detected intermediates during the Zn-Fe-CNTs/O2 process.
tacted with O2 in solution, Zn0 lost electrons and the formed Zn2+ entered into solution due to hydration. The electrons were transferred to the surface of CNTs and transmitted to the adsorbed O2
on the surface of CNTs, to generate H2 O2 . (b) The in situ generation of • OH by Fenton-like reaction: H2 O2 generated in situ was decomposed into • OH by Fe0 /Fe2 O3 . (c) Oxidation of pollutants by
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
175
Fig. 13. Proposed degradation pathway of SMX by Zn-Fe-CNTs/O2.
oxidation for pollutants by • OH. In this study, experiments were conducted to investigate the influence of operational parameters on the degradation of SMX. Complete degradation was achieved within 10 min with initial pH of 1.5, O2 flow rate of 400 mL/min, ZnFe-CNTs of 0.6 g/L, SMX of 25 mg/L, with 51.3% of TOC was removed after 10 min reaction. This system could be used to degrade and mineralize toxic organic pollutants in water, with no need for addition of H2 O2 . The high efficiency of this system makes it a promising alternative for wastewater treatment. Acknowledgements Fig. 14. The proposed reaction scheme diagram of Zn-Fe-CNTs/O2 process.
The generated • OH could oxidize the adsorbed pollutants in situ, or diffuse into the solution phase to oxidize the non-adsorbed pollutants. Moreover, the Zn2+ in solution could be converted to Zn(OH)2 via the combination with generated OH− due to the reduction of H+ and O2 (Fig. 5). The solubility product constant (ksp ) of Zn(OH)2 was only 1.2 × 10−17 at pH 7 and 25 ◦ C. Therefore, the secondary pollution of Zn2+ could be avoided by solid-liquid separation in neutral condition. During Zn-Fe-CNTs/O2 process, the generation of H2 O2 in situ was due to the sacrifice of Zn0 . Therefore, the consumption of Zn0 would influence the stability of the catalyst. However, the catalyst was relatively stable due to its relatively slow consumption rate. How to further increase the stability of the catalyst will be studied in our future work.
This research was supported by the National Natural Science Foundation of China (51338005) and the Program for Changjiang Scholars and Innovative Research Team in University (IRT-13026).
• OH:
4. Conclusions In summary, a novel Fenton-like system that spontaneously generated • OH and H2 O2 by the reaction between Zn-Fe-CNTs and O2 was developed. The Zn-Fe-CNTs/O2 process for the oxidation of pollutants involved in the in situ generation of H2 O2 , the in situ generation of • OH via the decomposition H2 O2 by Fe0 /Fe2 O3 and the
References [1] J.L. Wang, L.J. Xu, Advanced oxidation processes for wastewater treatment: formation of hydroxyl radical and application, Crit. Rev. Environ. Sci. Technol. 42 (2012) 251–325. [2] E. Neyens, J. Baeyens, A review of classic Fenton’s peroxidation as an advanced oxidation technique, J. Hazard. Mater. 98 (2003) 33–50. [3] Z. Wan, J. Hu, J.L. Wang, Removal of sulfamethazine antibiotics using Ce-Fe-graphene nanocomposite as catalyst by Fenton-like process, J. Environ. Manage. 182 (2016) 284–291. [4] N. Wang, T. Zheng, G. Zhang, P. Wang, A review on Fenton-like processes for organic wastewater treatment, Chem. Eng. J. 4 (2016) 762–787. [5] H. Zhang, X. Ran, X. Wu, Electro-Fenton treatment of mature landfill leachate in a continuous flow reactor, J. Hazard. Mater. 241–242 (2012) 259–266. [6] W. Liu, H. Liu, Z. Ai, In-situ generated H2 O2 induced efficient visible light photo-electrochemical catalytic oxidation of PCP-Na with TiO2 , J. Hazard. Mater. 288 (2015) 97–103. [7] W. Wang, Y. Qu, B. Yang, X. Liu, W. Su, Lactate oxidation in pyrite suspension: a Fenton-like process in situ generating H2 O2 , Chemosphere 86 (2012) 376–382. [8] C. Annabi, F. Fourcade, I. Soutrel, F. Geneste, D. Floner, N. Bellakhal, A. Amrane, Degradation of enoxacin antibiotic by the electro-fenton process: optimization, biodegradability improvement and degradation mechanism, J. Environ. Manage. 165 (2016) 96–105. [9] A. Dirany, S. Efremova Aaron, N. Oturan, I. Sirés, M.A. Oturan, J.J. Aaron, Study of the toxicity of sulfamethoxazole and its degradation products in water by a bioluminescence method during application of the electro-Fenton treatment, Anal. Bioanal. Chem. 400 (2011) 353–360.
176
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
[10] Y. Li, L. Yang, C. Chen, Y. Lan, Zn(0)-catalyzed ozonation degradation of acid orange 7 (AO7) in aqueous solution, Water Air Soil Pollut. 227 (2016) 364–382. [11] J. Zhang, Y. Wu, L. Liu, Y. Lan, Rapid removal of p-chloronitrobenzene from aqueous solution by a combination of ozone with zero-valent zinc, Sep. Purif. Technol. 151 (2015) 318–323. [12] G. Wen, S. Wang, J. Ma, T. Huang, Z. Liu, L. Zhao, J. Su, Enhanced ozonation degradation of di-n-butyl phthalate by zero-valent zinc in aqueous solution: performance and mechanism, J. Hazard. Mater. 265 (2014) 69–78. [13] G.R. Agladze, G.S. Tsurtsumia, B.I. Jung, J.S. Kim, G. Gorelishvili, The In-cell and Ex-cell Fenton treatment of phenol, 4-chlorophenol and aniline, J. Appl. Electrochem. 37 (2007) 385–393. [14] L. Fu, S. You, G. Zhang, F. Yang, X. Fang, Degradation of azo dyes using in-situ Fenton reaction incorporated into H2 O2 -producing microbial fuel cell, Chem. Eng. J. 160 (2010) 164–169. [15] Z. Ai, T. Mei, J. Liu, J. Li, F. Jia, L. Zhang, J. Qiu, Fe@Fe2 O3 core-shell nanowires as an iron reagent. 3. Their combination with CNTs as an effective oxygen-fed gas diffusion electrode in a neutral electro-Fenton system, J. Phys. Chem. C 111 (2007) 14799–14803. [16] J.L. Wang, Z.Y. Bai, Fe-based catalysts for heterogeneous catalytic ozonation of emerging contaminants in water and wastewater, Chem. Eng. J. 312 (2017) 79–98. [17] J.L. Wang, S.Z. Wang, Removal of pharmaceuticals and personal care products (PPCPs) from wastewater: a review, J. Environ. Manage. 182 (2016) 620–640. [18] J.L. Wang, L.B. Chu, Irradiation treatment of pharmaceutical and personal care products (PPCPs) in water and wastewater: an overview, Radiat. Phys. Chem. 125 (2016) (2016) 56–64. [19] N. Klamerth, L. Rizzo, S. Malato, M.I. Maldonado, A. Agüera, A.R. Fernández-Alba, Degradation of fifteen emerging contaminants at (g L−1 initial concentrations by mild solar photo-Fenton in MWTP effluents, Water Res. 44 (2010) 545–554. [20] W.L. Dai, Q. Sun, J.F. Deng, D. Wu, Y.H. Sun, XPS studies of Cu/ZnO/Al2 O3 ultra-fine catalysts derived by a novel gel oxalate co-precipitation for methanol synthesis by CO2 +H2 , Appl. Surf. Sci. 177 (2001) 172–179. [21] A.C. Hegde, K. Venkatakrishna, N. Eliaz, Electrodeposition of Zn-Ni, Zn-Fe and Zn-Ni-Fe alloys, Surf. Coat. Technol. 205 (2010) 2031–2041. [22] X. Song, Q. Shi, H. Wang, S. Liu, C. Tai, Z. Bian, Preparation of Pd-Fe/graphene catalysts by photocatalytic reduction with enhanced electrochemical oxidation-reduction properties for chlorophenols, Appl. Catal. B: Environ. 203 (2017) 442–451. [23] Y. Lu, F. Yu, J. Hu, J. Liu, Catalytic conversion of syngas to mixed alcohols over Zn-Mn promoted Cu-Fe based catalyst, Appl. Catal. A: Gen. 429–430 (2012) 48–58. [24] X. Weng, M. Guo, F. Luo, Z. Chen, One-step green synthesis of bimetallic Fe/Ni nanoparticles by eucalyptus leaf extract: biomolecules identification, characterization and catalytic activity, Chem. Eng. J. 308 (2017) 904–911.
[25] Y. Liu, W. Jin, Y. Zhao, G. Zhang, W. Zhang, Enhanced catalytic degradation of methylene blue by ␣-Fe2 O3 /graphene oxide via heterogeneous photo-Fenton reactions, Appl. Catal. B: Environ. 206 (2017) 642–652. [26] X. Deng, J. Lee, C. Matranga, Preparation and characterization of Fe3 O4 (111) nanoparticles and thin films on Au (111), Surf. Sci. 604 (2010) 627–632. [27] K. Li, S. Tao, J. Li, X. Wang, Controllable Fe introduction into ordered mesoporous carbon with interconnected small pores for investigating Fe doping effect on hydrogen adsorption, Int. J. Hydrogen Energy 42 (2017) 4733–4740. [28] L.J. Xu, J.L. Wang, A heterogeneous Fenton-like system with nanoparticulate zero-valent iron for removal of 4-chloro-3-methyl phenol, J. Hazard. Mater. 186 (2011) 256–264. [29] C.R. Keenan, D.L. Sedlak, Factors affecting the yield of oxidants from the reaction of nanoparticulate zero-valent iron and oxygen, Environ. Sci. Technol. 42 (2008) 1262–1267. [30] S.E. Mylon, Q. Sun, T.D. Waite, Process optimization in use of zero valent iron nanoparticles for oxidative transformations, Chemosphere 81 (2010) 127–131. [31] S. Kang, W. Choi, Oxidative degradation of organic compounds using zero-valent iron in the presence of natural organic matter serving as an electron shuttle, Environ. Sci. Technol. 43 (2009) 878–883. [32] S. Zhang, D. Wang, L. Zhou, X. Zhang, P. Fan, X. Quan, Intensified internal electrolysis for degradation of methylene blue as model compound induced by a novel hybrid material: multi-walled carbon nanotubes immobilized on zero-valent iron plates (Fe0 -CNTs), Chem. Eng. J. 217 (2013) 99–107. [33] H. Niu, D. Zhang, S. Zhang, X. Zhang, Z. Meng, Y. Cai, Humic acid coated Fe3 O4 magnetic nanoparticles as highly efficient Fenton-like catalyst for complete mineralization of sulfathiazole, J. Hazard. Mater. 190 (2011) 559–565. [34] J.J. Pignatello, E. Oiveros, A. MacKay, Advanced oxidation processes for organic contaminant destruction based on the Fenton reaction and related chemistry, Crit. Rev. Env. Sci. Technol. 36 (2006) 1–84. [35] M. Wang, G. Fang, P. Liu, D. Zhou, C. Ma, D. Zhang, J. Zhan, Fe3 O4 @-CD nanocomposite as heterogeneous Fenton-like catalyst for enhanced degradation of 4-chlorophenol (4-CP), Appl. Catal. B: Environ. 188 (2016) 113–122. [36] G. Xia, Y. Lu, H. Xu, Electrogeneration of hydrogen peroxide for electro-Fenton via oxygen reduction using polyacrylonitrile-based carbon fiber brush cathode, Electrochim. Acta 158 (2015) 390–396. [37] P. Calza, E. Pelizzetti, C. Minero, The fate of organic nitrogen in photocatalysis: an overview, J. Appl. Electrochem. 35 (2005) 665–673. [38] A.G. Trovó, R.F.P. Nogueira, A. Agüera, A.R. Fernandez-Alba, C. Sirtori, S. Malato, Degradation of sulfamethoxazole in water by solar photo-Fenton. Chemical and toxicological evaluation, Water Res. 43 (2009) 3922–3931. [39] D. Marciocha, J. Kalka, J. Turek-Szytow, J. Wiszniowski, J. Surmacz-Górska, Oxidation of sulfamethoxazole by UVA radiation and modified Fenton reagent: toxicity and biodegradability of by-products, Water Sci. Technol. 60 (2009) 2555–2562.
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Zn-Fe-CNTs catalytic in situ generation of H2 O2 for Fenton-like degradation of sulfamethoxazole Yong Liu a,b , Qin Fan a , Jianlong Wang b,c,∗ a b c
College of Chemistry and Materials Science, Sichuan Normal University, Chengdu 610066, PR China Collaborative Innovation Center for Advanced Nuclear Energy Technology, INET, Tsinghua University, Beijing 100084, PR China Beijing Key Laboratory of Radioactive Wastes Treatment, Tsinghua University, Beijing 100084, PR China
h i g h l i g h t s • • • •
• Zn-Fe-CNTs was capable of converting O2 to H2 O2 and further to OH. The Fenton-like degradation of sulfamethoxazole (SMX) using Zn-Fe-CNTs as catalyst was studied. The removal efficiency of SMX and TOC was 100% and 51.3%, respectively. The possible reaction mechanism of Zn-Fe-CNTs/O2 process was tentatively proposed.
a r t i c l e
i n f o
Article history: Received 24 May 2017 Received in revised form 5 August 2017 Accepted 7 August 2017 Available online 9 August 2017 Keywords: Fenton-like In situ generation of H2 O2 Micro-electrolysis PPCPs Sulfamethoxazole
a b s t r a c t A novel Fenton-like catalyst (Zn-Fe-CNTs) capable of converting O2 to H2 O2 and further to • OH was prepared through infiltration fusion method followed by chemical replacement in argon atmosphere. The catalyst was characterized by SEM, EDS, TEM, XRD and XPS. The reaction between Zn-Fe-CNTs and O2 in aqueous solution could generate H2 O2 in situ, which was further transferred to • OH. The Fentonlike degradation of sulfamethoxazole (SMX) using Zn-Fe-CNTs as catalyst was evaluated. The results indicated that Zn-Fe-CNTs had a coral porous structure with a BET area of 51.67 m2 /g, exhibiting excellent adsorption capacity for SMX, which enhanced its degradation. The particles of Zn0 and Fe0 /Fe2 O3 were observed on the surface of Zn-Fe-CNTs. The mixture of Zn0 and CNTs could reduce O2 into H2 O2 by microelectrolysis and Fe0 /Fe2 O3 could catalyze in-situ generation of H2 O2 to produce • OH through Fenton-like process. When initial pH = 1.5, T = 25 ◦ C, O2 flow rate = 400 mL/min, Zn-Fe-CNTs = 0.6 g/L, SMX = 25 mg/L and reaction time = 10 min, the removal efficiency of SMX and TOC was 100% and 51.3%, respectively. The intermediates were detected and the possible pathway of SMX degradation and the mechanism of Zn-Fe-CNTs/O2 process were tentatively proposed. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Advanced oxidation processes (AOPs) have been widely used for the degradation of toxic organic contaminants in water and wastewater in the past decades [1]. The classic Fenton- and related reactions are convenient, economical, and green ways to generate • OH for the removal of non-biodegradable and/or toxic compounds [2]. Fenton-like process, using solid iron compounds/complexes as heterogeneous catalysts [3,4], has received increasing atten-
∗ Corresponding author at: Energy Science Building, INET, Tsinghua University, Beijing 100084, PR China. E-mail addresses: [email protected], [email protected] (J. Wang). http://dx.doi.org/10.1016/j.jhazmat.2017.08.016 0304-3894/© 2017 Elsevier B.V. All rights reserved.
tion in recent years. It is also attractive to develop Fenton process which can spontaneously generate H2 O2 in situ, such as electroFenton system and photo-Fenton process [5,6]. However, the low and hardly regulated concentration of H2 O2 generated insitu decreased the degradation efficiency of organic contaminants [7–9]. Therefore, it is necessary to increase the concentration of H2 O2 in solid iron compounds/O2 system. The generation of H2 O2 in-situ by the reaction between Zn0 and O2 was investigated, but the productivity was low [10–12]. According to the theory of electrochemical corrosion, the rate of oxidation-reduction reaction on the surface of metal can be accelerated in the galvanic-type corrosion cell without external power supply. Therefore, the in-situ generation of H2 O2 can be improved by the electrode reaction in the corrosion cell using Zn0 as anode. Cathodic H2 O2 could be produced on the surface of carbon materi-
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
als through the electrochemical process [13,14]. Carbon nanotubes (CNTs) may offer significant advantages for the two-electron reduction of O2 due to their good electrical conductivity, high surface activity and mechanical strength [15]. Thus, high concentration of H2 O2 can be possibly produced by the reduction of O2 in the micro-electrolysis system of Zn0 -CNTs, which formed numerous corrosion cells between the particles of Zn0 and CNTs in aqueous solution. Moreover, the redox potential of Fe2+ /Fe0 (−0.44 V) is higher than that of Zn2+ /Zn0 (−0.76 V), the Fe0 can be deposited on Zn0 /CNTs via chemical replacement. Therefore, when the composite of Zn0 , CNTs and Fe0 reacts with O2 , H2 O2 can be continuously supplied through the micro-electrolysis reaction between O2 and Zn0 -Fe0 /CNTs, which can catalyze the generated H2 O2 to produce • OH. In this system, Fe0 or Fe2+ can be regenerated either by a direct reduction with Zn0 or the reaction with H2 O2 . The release of antibiotics into the aquatic environment caused by extensive use of antibiotics is receiving increasing attention in recent years [16–18]. Sulfamethoxazole (SMX), an antibiotic used worldwide for treatment of urinary infection, has been identified in many wastewater treatment plants (WWTPs) effluents [19]. It is required to develop effective approaches for SMX removal. In this study, Zn-Fe-CNTs composite was synthesized, characterized and used as catalyst for Fenton-like degradation of SMX. The effect of initial pH, Zn-Fe-CNTs dosage, and initial SMX concentration on the removal efficiency of SMX was determined. The intermediates were detected by LC–MS and IC, and the possible degradation pathway of SMX was tentatively proposed. 2. Materials and methods 2.1. Materials and chemicals Sulfamethoxazole (SMX) (C10 H11 N3 O3 S) was a sigma Aldrich product (≥99.0% purity). Multi-walled CNTs (d < 8 nm, l = 10–30 m) were produced by Beijing Deke Daojing NanoCompany, China. Zinc powders (size: 120 meshes, specific surface area: 0.72 m2 /g, purity: >99.8%) were obtained from Shandong Xiya Corporation Ltd., China. Commercial iron powders (≥98.0% purity) with specific surface area of 1.26 m2 /g were purchased from Tianjin Jinke Fine Chemical Industry Research Institute., China. H2 SO4 , NaOH, FeSO4 ·7H2 O and H2 O2 were all of analytical grade. De-ionized water was used in all experiments.
167
Scanning electron microscopy (SEM) and energy dispersive spectrometer (EDS) analyses were performed on emission scanning electron microscope (SEM, SU8010, Hitachi). Transmission electron microscopy (TEM) images were recorded on a transmission electron microscope (HRTEM, JEM 2100 and JEOL). BET-surface areas were measured using a NOVA 3200e bath at 77 K and degassing at 373 K. The pore size distribution was determined by the BJH method. Powder X-ray diffraction (XRD) measurement was performed with Cu K␣ radiation (MiniFlex 600, Rigaku, 40 kV and 15 mA) at a scanning rate of 0.5◦ /min. X-ray photoelectron spectroscopy (XPS) was measured by an AXIS-Ultra instrument (Kratos Analytical, UK). 2.3. Fenton-like degradation of SMX Experiments were conducted at room temperature in a glass bottle (500 mL) containing a 250 mL solution. The real effluent of wastewater was collected at a local municipal wastewater treatment plant. The sample had an initial content of TOC of 9.77 ± 0.36 mg/L and pH of 7.0 ± 0.2. 2.4. Analytical methods H2 O2 concentration was measured by a photometric method on an UV–vis spectrometer (PerkinElmer Lambda 25) at 385 nm using potassium titanium oxalate as chromogenic reagent. SMX concentration was determined by HPLC (Agilent 1200) equipped with a diode array detector (DAD) and a C18 reversed-phase column (4.6 mm × 150 mm). The mobile phase used for SMX was a mixture of distilled water and acetonitrile (55:45 (v/v)) at a flow rate of 1.0 mL/min with a column temperature of 30 ◦ C, and the analytical wavelength was 255 nm. SMX had a retention time of 3.0 min under these conditions. HPLC–MS equipped with the above-mentioned column coupled to a Shimadzu 2010EV mass spectrometer with ESI ion source (LC–MS 2010, Columbia, USA) was used to analyze organic intermediates. It equipped with a photo diode array (PDA) detector, and operated in a negative mode. The aforementioned solvent conditions were used for the analysis of the intermediates. The injected volume was 30 L. Total organic carbon (TOC) content was measured by a Multi TOC/TN Analyzer (2100, Analytik Jena. Germany). The concentration of dissolved zinc and iron generated from the reaction process was determined with a flame atomic adsorption spectroscopy (AAS, HITACHI ZA-3000, Japan).
2.2. Preparation and characterization of Zn-Fe-CNTs
3. Results and discussion
The Zn-Fe-CNTs preparation was as follows. Firstly, 1.5 g Zinc powder was mixed with 0.5 g multi-walled CNTs. 2 mL 40% w polyethyleneglycol 4000 was dropped into the mixture, stirred for 30 min, then dried at room temperature, sintered in the tube furnace at 550 ◦ C for 120 min with a N2 flow speed of 60 mL/min and cooled into room temperature to obtain Zn-CNTs. Secondly, FeSO4 ·7H2 O (0.5 g) were dissolved in 200 mL of aqueous sulfuric acid solution (pH = 3), and Zn-CNTs (1 g) were added in argon atmosphere. After 1 h, Zn-Fe-CNTs composites were obtained. The formation of Zn-Fe-CNTs was based on the following principles: polyethyleneglycol 4000 was used as binder and to bind zinc powder and multi-walled CNTs. When the mixture of zinc powder, multi-walled CNTs and polyethyleneglycol 4000 was heated into 550 ◦ C, zinc powder with melting points of 419.5 ◦ C was melted into liquid and polyethyleneglycol 4000 was decomposed into volatile gases. When the heated mixture was cooled into room temperature, liquid zinc was solidified on the surface of multi-walled CNTs and form Zn-CNTs. When as-prepared Zn-CNTs were put into the Fe2+ solution, the chemical replacement between Fe2+ and Zn0 in Zn-CNTs resulted in the formation of Zn-Fe-CNTs composites.
3.1. Characterization of Zn-Fe-CNTs The N2 adsorption-desorption isotherms and pore size distributions of Zn-Fe-CNTs and Zn-CNTs were given in Fig. 1. The Zn-Fe-CNTs sample provided the IV type isotherm, which are typical for the highly ordering mesoporous material with narrow size distribution. The isotherm of Zn-CNTs was identified as type an III/IV mixed type isotherm, which was characteristic of co-existence of microporous and mesoporous. Zn-Fe-CNTs had a larger BET specific surface area (51.67 m2 /g) than that of Zn-CNTs (26.99 m2 /g). The pore size distribution was calculated using the Barrett-JoynerHalenda (BJH) method (Fig. 4b). The result showed that the average pore sizes of Zn-Fe-CNTs and Zn-CNTs were 16.89 nm and 13.52 nm, respectively. TEM image and SEM micrograph of Zn-Fe-CNTs were shown in Figs. 2 and 3. The samples exhibited coral porous structure and large dispersive metal nano-particles were adhered to the surface of CNTs in the Zn-Fe-CNTs. In order to confirm the element distribution and relative element content in Zn-Fe-CNTs, EDS spectra and element
168
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 1. N2 adsorption/desorption isotherm and pore size distribution (the inset) of Zn-Fe-CNTs (a) and Zn-CNTs(b).
species typically seen in Fe2 O3 [24,26]. Fe 2p1/2 spectra at 723.5 eV was the satellite peaks of Fe3 O4 [27]. XPS results indicated that Fe0 and Zn0 existed on the surface of Zn-Fe-CNTs. However, a few atomic layers of Fe (III) and Zn (II) species existed on Zn-Fe-CNTs surface, probably due to the exposure of Zn-Fe-CNTs in air and aqueous solution. Compared with newly Zn-Fe-CNTs, the variety of the valence of iron was found in the used Zn-Fe-CNTs, indicating the possible oxidation of Fe0 and reduction of Fe (III) by Zn0 in the Zn-Fe-CNTs/O2 system. 3.2. The synergetic effect of Zn-Fe-CNTs/O2 process
Fig. 2. TEM image of Zn-Fe-CNTs.
mapping were recorded in Fig. 3. The element peaks of C and O were observed at 0.25 keV and 0.5 keV, respectively. The peaks of Fe were at 0.7 keV and 6.4 keV and the peaks of Zn were at 1.0 keV and 8.6 keV. The atom percentage of Zn, O, C and Fe was 40.3, 29.1, 23.3 and 7.1, respectively. The XRD pattern of the Zn-Fe-CNTs was shown in Fig. 4. The peaks at 26.77◦ and 59.75◦ was corresponded to the (002) and (103) reflections of carbon (JCPDS 99-0057). The peaks at 36.54◦ , 39.33◦ , 43.47◦ , 54.55◦ and 70.24◦ were assigned to the (002), (100), (101), (102) and (103) reflections of Zn0 (JCPDS 99-0110), respectively. The peaks at 32.03◦ , 34.66◦ , 47.85◦ , 56.82◦ ,63.08◦ and 68.15◦ were assigned to the (100), (002), (102), (110), (103) and (112) reflections of ZnO (JCPDS 99-0111), respectively. The peaks at 32.95◦ was attributed to the (104) reflections of Fe2 O3 (JCPDS 99-0060). The weak peaks at 44.67◦ was correspond to the (110) reflections of Fe0 (JCPDS 99-0064) XPS was used to examine the chemical compositions of Zn-FeCNTs (Fig. 5). Fig. 5a clearly illustrated the existence of Zn, O, Fe and C. Fig. 5b exhibited the Auger KE of Zn [20], and the peak at 987.7 eV could be assigned to ZnO [21]. This was in consistent with the XRD analysis in Fig. 4. In Fig. 5b, a broad peak around 717.4 eV assigned to Fe 2p1/2 was detected, indicating the presence of Fe0 on the surface of Zn-Fe-CNTs [22]. However, binding energies at 710.7 eV and 724.3 eV assigned to Fe 2p3/2 and Fe 2p1/2 were characteristic of Fe2 O3 [23,24], which was also a good catalyzer for the Fenton oxidation [25]. Fig. 5c exhibited the Fe 2p spectra of Zn-Fe-CNTs after reaction. Typically, it featured a strong Fe3+ 2p1/2 peak at 725.6 eV. The peaks at 711.6 eV and 719.1 eV were characteristic for the Fe3+
To evaluate the superiority and synergetic effect of the Zn-Fe-CNTs/O2 process, four control experiments were performed, including: (a) Zn-Fe-CNTs/O2 ; (b) Zn-CNTs + Fe0 /O2 ; (c) Zn-CNTs/O2 ; and (d) Fe0 /O2 ; to treat the SMX aqueous solution under the same conditions (i.e., initial pH of 1.5, O2 gas flow rate of 400 mL/min, Zn-Fe-CNTs, Zn-CNTs and Fe0 dosage of 0.6 g/L, 0.54 g/L and 0.06 g/L, respectively). As shown in Fig. 6a, high SMX removal efficiency (e.g., 100% after 10 min treatment) was obtained by the Zn-CNTs +Fe0 /O2 process, while the low SMX removal efficiency was achieved by Zn-CNTs/O2 alone (e.g., 26.1% after 10 min treatment) and Fe0 /O2 alone (e.g., 35.3% after 10 min treatment), suggesting that the synergetic effect between Zn-CNTs and Fe0 are responsible for the degradation of SMX. The removal efficiency of SMX by Zn-Fe-CNTs/O2 process (e.g., 95.3% after 2 min treatment) was higher than that of Zn-CNTs +Fe0 /O2 process (e.g., 80.2% after 2 min treatment). It may be due to the BET surface area and the average pore sizes of Zn-Fe-CNTs were higher than that of Zn-CNTs (Fig. 1), which improved the mass transfer and active sites. The other reason for the better performance of Zn-Fe-CNTs might be that nano-iron particles had higher catalytic activity for the Fentonlike oxidation [28]. According to Fig. 6b, TOC removal efficiency obtained by the ZnCNTs +Fe0 /O2 process (e.g., 31.8% after 10 min treatment) was about 1.22 and 1.87 times higher than that of the Zn-CNTs/O2 process (e.g., 17.2% after 10 min treatment) and Fe0 /O2 process (e.g., 11.1% after 10 min treatment), respectively. The TOC removal efficiency by ZnFe-CNTs/O2 reached about 51.3% after 10 min treatment, which was much higher than that of the Zn-CNTs +Fe0 /O2 process, indicating that the synergetic effect between Zn-CNTs and iron in Zn-FeCNTs composite played a vital role in the further mineralization of the intermediates to obtain high TOC removal efficiency. The SMX removal efficiency was quite similar with the TOC removal efficiency in Zn-CNTs/O2 process, suggesting that the removal of SMX in Zn-CNTs/O2 process was due to the adsorption of SMX by ZnCNTs. In Fe0 /O2 process, the reaction of Fe0 (s) with oxygen could lead to the formation of reactive oxygen species that were capable of oxidizing contaminants [29]. In our work, the relatively low removal efficiency of SMX was due to the heavy corrosion of Fe0 ,
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 3. SEM-EDS analysis of Zn-Fe-CNTs. (a) SEM image; (b) EDS spectrum; (c) EDS mapping pictures of O, Zn, Fe, C elements.
169
170
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
which limits the efficiency of this process and the weak oxidant of ferryl ion (Fe(IV)) produced from the corrosion of Fe0 by oxygen cannot effectively degrade pollutants, which is in consistent with many literatures [30,31].
3.3. Involved active oxidation species
Fig. 4. The XRD pattern of Zn-Fe-CNTs.
To evaluate the performance of active oxidation species generated by Zn-Fe-CNTs/O2 process, the accumulated H2 O2 concentration in different systems was measured (Fig. 7). The dosage of both Zn-Fe-CNTs and Zn-CNTs were 0.6 g/L. Fig. 7 indicated that the concentration of H2 O2 gradually increased with the reaction time and the H2 O2 accumulation was dependent on initial pH in Zn-CNTs/O2 system. In Zn-CNT/O2 system, the H2 O2 concentration was 5.26 and 22.2 mg/L at 10 min, respectively when the initial pH was 1.5 and 7.0, respectively. It confirmed that H2 O2 could be generated by the reduction of oxygen on the surface of Zn-CNTs due to micro-electrolysis. The H2 O2 generation via the reaction of Zn0 and O2 was evaluated by Zhang and Wen [11,12]. However, the
Fig. 5. XPS full survey spectra (a), Zn Auger KE spectra (b), and Fe 2p spectra (c) of Zn-Fe-CNTs, Fe 2p spectra of Zn-Fe-CNTs after reaction (d).
Fig. 6. The degradation (a) and mineralization (b) of SMX in different processes.
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 7. Variation of H2 O2 concentration with reaction time in different systems.
171
of SMX. There was no obvious difference in the removal efficiency of SMX between t-BuOH addition and KI addition, suggesting that the degradation of SMX was primarily attributed to the action of the surface bounded • OH. In addition, the degradation efficiency of Zn-Fe-CNTs/O2 and Zn-Fe-CNTs/N2 was compared to confirm the role of dissolved oxygen (DO) and the adsorption capacity of Zn-Fe-CNTs for SMX. As shown in Fig. 8, SMX removal (e.g., 100% after 10 min) by the ZnFe-CNTs/O2 process was about five times of that (e.g., 20.1% after 10 min) obtained by Zn-Fe-CNTs/N2 , demonstrating that the generation of H2 O2 and • OH was caused by the reduction of O2 on the surface of Zn-Fe-CNTs and Zn-Fe-CNTs had high adsorption capacity for SMX. 19%-21% of removal was still achieved with addition of t-BuOH or Iodide ion, which can be attributed to the adsorption of SMX by Zn-Fe-CNTs. Fe0 can react with O2 to produce H2 O2 through a two-electron transfer or a series of one-electron transfers via the reaction ferrous iron produced by Fe0 oxidation with oxygen [29–31]. However, the low and hardly regulated concentration of H2 O2 generation was found in Fe0 /O2 system because the action of Fe0 might present in two aspects, i.e. a source of Fe2+ and an electron donor to generate H2 O2 intermediates, which decreased the degradation of organics. In Zn-Fe-CNTs/O2 system, many active intermediate species H2 O2 can be generated and the content of H2 O2 can be adjusted by the component content in Zn-Fe-CNTs material, which is in favor of the generation of • OH and the degradation of organics. 3.4. Factors influencing SMX degradation
Fig. 8. The degradation of SMX by Zn-Fe-CNTs under different condition.
concentration of generated H2 O2 in the Zn0 /O2 system was below 3 mg/L with a stoichiometry of same dosage at neutral solution. In Zn-CNT/O2 system, the high concentration of in situ H2 O2 generation was due to the following reasons. The Zn0 was attached to the surface of CNTs and the galvanic-type corrosion cell was formed between Zn0 and CNTs. The CNTs could improve the two-electron reduction of oxygen due to its good electrical conductivity and high surface activity, which was proved by the in situ H2 O2 generation in electric catalysis process [15,32]. Compared with Zn-CNT/O2 system, at initial pH 1.5 and 7, the concentration of accumulated H2 O2 was rather low (<13.5 mg/L), particularly under acid condition (<1.6 mg/L), indicating that the iron in Zn-Fe-CNTs improved the decomposition of H2 O2 and high decomposition rate of H2 O2 under acid condition was obtained. The influence of different scavengers on the degradation of SMX at initial pH 1.5 was determined to investigate involved active radicals in the process (Fig. 8). Tertiary butanol (t-BuOH) is known as hydroxyl radicals scavengers, which are used to examine the role of hydroxyl radicals [33]. As shown in Fig. 8, the removal of SMX was strongly inhibited with the presence of excess t-BuOH (200.0 mM), indicating that SMX was mainly decomposed by the attack of • OH radicals (including surface-bounded • OH and free • OH). Iodide ion is used to scavenger hydroxyl radicals produced at the surface of Zn-Fe-CNTs [28]. As shown in Fig. 8, addition of excess KI (20 mM) resulted in a considerable decrease in the extent of SMX degradation from 100% (in 10 min in absence of KI) to 21%, indicating that surface-bounded • OH played a significant role in the degradation
3.4.1. Effect of initial pH The pH value of solution played an important role in degradation of organic pollutants by Zn-Fe-CNTs/O2 process, owing to its effect on in-situ generation of H2 O2 from the reduction of O2 by Zn0 CNTs and the in-situ generation of • OH from the decomposition of H2 O2 by Fe0 /Fe(II). On one hand, it is well-known that Fenton oxidation in water is strongly depended on pH value [34]. A lower initial pH value benefits the generation of • OH and results in the higher removal efficiency of pollutant [35]. On the other hand, insitu generation of H2 O2 from the reduction of O2 by Zn-CNTs was largely affected by pH value of solution, as shown in Fig. 7. The effect of pH on SMX removal efficiency was investigated. Since the acidic solution benefit the Fenton oxidation, the initial pH was selected in the range of 1–3 in this study. As shown in Fig. 9a, the SMX removal efficiency decreased with the initial pH increasing from 1.5 to 3. The final removal efficiency of SMX was 100% after 20 min at initial pH of 1.5, but the removal efficiency of SMX was 73.1% and 14.9% at initial pH of 2.0 and 3.0, respectively. The removal efficiency of SMX was 90% after 20 min at the pH of 1, which was lower than at the initial pH of 1.5. Therefore, the initial pH of 1.5 was chosen for the degradation of SMX by Zn-Fe-CNTs/O2 process. In order to explain the effect of initial pH on the removal efficiency of SMX, the variation of pH during the Zn-Fe-CNTs/O2 process was determined and shown in Fig. 9b. It can be seen that the pH value increased gradually with reaction time. The final pH reached 2.1, 2.8, 6.3 and 6.8, respectively when initial pH was 1.0, 1.5, 2.0 and 3.0, respectively. This was attributed to the reduction of H+ and O2 on the surface of Zn-Fe-CNTs. Fenton oxidation of organic pollutants occurred faster with a decrease of pH and better removal efficiency can be achieved at pH of 2–3 [5,36]. In addition, Zn(II) was slowly released into the solution during the degradation of SMX. The concentration of Zn(II) in solution at 20 min reached 169.6, 104.8, 14, and 2.6 mg/L, respectively when initial pH was 1.0, 1.5, 2.0 and 3.0, respectively. The generated Zn(II) possibly led to another environmental problem, but it can be recovered in the formed of Zn(OH)2 when pH rose to neutral with the
172
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 9. Effect of operating parameters on SMX degradation in the Zn-Fe-CNTs/O2 system: (a) initial pH; (b) the variation of the solution pH; (c) initial SMX concentration; (d) Zn-Fe-CNTs dosage.
increase of reaction time or with addition of NaOH to adjust pH because the Ksp of Zn(OH)2 was very low. Meanwhile, there was an inevitable leaching of iron at acidic solution, and the dissolved iron concentration was 4.6 mg/L at initial pH 1.0. However, the dissolved iron concentration was low due to higher pH value and the iron ion could be reduced to Fe0 by Zn0 . When the solution pH rose to neutral, no dissolved iron was detected in this system.
3.4.2. Effect of initial concentration of SMX Fig. 9c shows the effect of initial SMX concentration on SMX degradation with initial pH 1.5, Zn-Fe-CNTs dosage 0.6 g/L and O2 flow rate 400 mL/min. The results showed that the removal efficiency of SMX was 100%, 85% and 65% when SMX concentration was 25, 50 and 75 mg/L, respectively after 10 min, and the apparent rate constant was calculated to be 1.07, 1.95 and 5.38.
3.4.3. Effect of Zn-Fe-CNTs/O2 dosage Fig. 9d shows that when the Zn-Fe-CNTs dosage increased from 0.2 g/L to 0.6 g/L, an obvious increase in the removal efficiency of SMX was observed. For example, after 20 min, the SMX removal efficiency was 57.9%, 88.3%, and 100% respectively. This was mainly due to the increasing amount of active sites for the O2 reduction to produce more reactive oxidants such as H2 O2 and • OH. Nevertheless, the degradation degree was only slightly enhanced when the Zn-Fe-CNTs dosage increased from 0.6 to 1.0 g/L, which was likely attributed to the percentage of • OH scavenged by Fe(II) through undesirable reaction. Thus, the optimum Zn-Fe-CNTs dosage was 0.6 g/L in this experiment.
Fig. 10. The degradation and mineralization of SMX in wastewater by Zn-FeCNTs/O2 process.
3.5. Degradation of SMX in real wastewater The effectiveness of the Zn-Fe-CNTs/O2 process in treating WWTP effluent containing SMX (around12.5 mg/L and 25 mg/L) was evaluated at initial pH 1.5, Zn-Fe-CNTs 0.6 g/L and O2 flow rate 400 mL/min. Fig. 10 shows the variation in SMX concentration and TOC during the treatment process. It can be seen that the Zn-FeCNTs/O2 process was very effective in removing SMX from WWTP
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 11. Evolution of small organic acids and inorganic ions.
effluent. SMX was completely removed after 10 min when its initial concentration was 12.5 mg/L and 25 mg/L. The initial TOC concentration of WWTP effluent containing 12.5 mg/L SMX and 25 mg/L SMX were 15.1 mg/L and 20.3 mg/L, respectively. The removal efficiency of TOC after 10 min was 51.1% and 17.3% when initial concentration of SMX was 12.5 mg/L and 25 mg/L, respectively. 3.6. The intermediate and pathway of SMX degradation 3.6.1. The formation of small organic acids and inorganic ions Short-chain carboxylic acids generated during the degradation of SMX process were detected and shown in Fig. 11. These chromatograms exhibited the peaks related to propanoic (t = 3.53 min), formic (t = 3.67 min) and acetic (t = 3.31 min) acids. The concentration of propanoic and acetic acids was firstly increased and then decreased. The max accumulation of propanoic acid and acetic acid was 6.09 mg/L and 4.85 mg/L respectively, and then decreased to 4.54 mg/L and 3.58 mg/L after 20 min, indicating that propanoic and acetic acids was degraded by • OH. The concentration of formic acid increased continuously up to 3.52 mg/L after 20 min, because the oxidation of formic acid was the final step for the mineralization of organic matters. The evolution of inorganic ions (NH4 + , NO3 − ) during the degradation of SMX was also shown in Fig. 11. In this study, SO4 2− ions were not monitored because H2 SO4 was employed as the pH regulator in the solution. SO4 2− was released during the degradation of SMX by • OH [37]. In Fig. 11, stable NH4 + concentration was observed during SMX degradation process and 2.36 mg/L of NH4 + was generated after 20 min. The concentration of NO3 − was firstly increased and then decreased because the reduction of NO3 − by low valence zinc and iron, 1.52 mg/L of NO3 − was generated after 20 min. The nitrogen content of SMX was mainly converted into ammonium. 3.6.2. Evolution of aromatic intermediates and SMX degradation pathway The transformation of SMX by Fenton process under different conditions has been reported. However, the SMX degradation pathway by Zn-Fe-CNTs/O2 was not clear. The intermediates formed during SMX degradation in the Zn-Fe-CNTs/O2 process were identified by LC–MS in order to determine a probable reaction pathway. The molecular structure of SMX and possible molecular structures of six intermediates and the corresponding mass spectrum were presented in Fig. 12.
173
Several intermediates were found during the Zn-Fe-CNTs/O2 process. These intermediates were produced at different oxidation reaction time. In the mass spectrometry, the peaks of SMX appeared at [M+H]+ 254 with the retention time of 5.8 min. The molecular ion A of the peak with the retention time of 2.1 min was found at [M+H]+ 270, which can be attributed to the addition of 16 mass units to the parent peak [1]. This formula was in consistent with the addition of an • OH to the SMX structure in different positions to yield mono-hydroxylated derivative. Another peak was observed at [M+H]+ 256 with the retention time of 1.6 min, which corresponds to intermediate B. Intermediate B mass units can be attributed to the substitution of the methyl group by • OH radical attack on the isoxazole ring, resulting in the formation of a hydroxylated structure [38]. The intermediate C with the retention time of 1.2 min at [M+H]+ 232 can be attributed to the substitution aromatic ring of hydroxyl group without abstraction of amino group. In addition to steric effect of the molecule, the sulfo- group in the meta-position (towards the sulfo- group) and in the ortho-position towards the amino group was an electron donating substituent [39], which was easily substituted by • OH and generated the intermediate C. Structure of sulphonamide bond was stable in basic solution, but during Zn-Fe-CNTs/O2 reaction, an attack of hydroxyl radicals occurred in the acidic medium, which caused the degradation highly probable. As a result, intermediate D with the retention time of 1.4 min at [M+H]+ 191 was already detected as oxidation products of SMX. The molecular ions E and F for the two peaks were observed at [M+H]+ 237 and [M+H]+ 143, respectively, which was in consistent with the addition of • OH to the intermediate E structure in different positions. The fragment [M+H]+ 173 and 99 were missing, indicating that the initial degradation of SMX was due to the attack of • OH radicals on the benzene ring or isoxazole ring. Base on above analyses, a reaction pathway for the complete mineralization of SMX with • OH as the main oxidant was proposed in Fig. 13. The first oxidation step of SMX was the attack by • OH radicals at different sites of the SMX molecule to form hydroxylated derivatives. These hydroxylated derivatives were further oxidized by • OH into short-chain carboxylic acids accompanied by the release of SO4 2− , NH4 + and NO3 − ions. These short-chain carboxylic acids can be further mineralized into CO2 and H2 O by • OH radicals. 3.7. Proposed reaction mechanism of Zn-Fe-CNTs/O2 process The SEM, EDS, XRD, XPS results demonstrated that nano Zn0 particles and nano Fe0 /Fe2 O3 particles were successfully dispersed into the CNTs skeletal structure and formed Zn-Fe-CNTs composite. TEM results demonstrated that metal particles contacted directly with CNTs, which was benefit for the formation of corrosion cell with Zn0 as anode due to the potential difference of Zn, Fe, CNTs and Zn with lowest potential among them. The control experiments and involved active oxidation species analysis confirmed the occurrence of the in situ generation of H2 O2 via the reduction of O2 by Zn-CNTs and the in situ generation of • OH from H2 O2 by Fe0 /Fe2 O3 catalysis during the reaction between Zn-Fe-CNTs and O2 . The reduction of O2 by Zn0 was prior to Fe0 /Fe2 O3 due to the lower potential of Zn, which favored the catalysis of Fe0 /Fe2 O3 to H2 O2 . According to these analysis data and the present literatures, the proposed reaction mechanisms abridged general view describing of Zn-Fe-CNTs/O2 process are summarized and presented in Fig. 14. The Zn-Fe-CNTs/O2 processes for the oxidation of pollutants can be divided into three stages: (a) The in situ generation of H2 O2 : Base on corrosion electrochemistry theory, the direct contact of Zn0 with CNTs could form zinc-carbon galvanic cells, when Zn-Fe-CNTs con-
174
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
Fig. 12. The mass spectrum and corresponding possible molecular structures of the detected intermediates during the Zn-Fe-CNTs/O2 process.
tacted with O2 in solution, Zn0 lost electrons and the formed Zn2+ entered into solution due to hydration. The electrons were transferred to the surface of CNTs and transmitted to the adsorbed O2
on the surface of CNTs, to generate H2 O2 . (b) The in situ generation of • OH by Fenton-like reaction: H2 O2 generated in situ was decomposed into • OH by Fe0 /Fe2 O3 . (c) Oxidation of pollutants by
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
175
Fig. 13. Proposed degradation pathway of SMX by Zn-Fe-CNTs/O2.
oxidation for pollutants by • OH. In this study, experiments were conducted to investigate the influence of operational parameters on the degradation of SMX. Complete degradation was achieved within 10 min with initial pH of 1.5, O2 flow rate of 400 mL/min, ZnFe-CNTs of 0.6 g/L, SMX of 25 mg/L, with 51.3% of TOC was removed after 10 min reaction. This system could be used to degrade and mineralize toxic organic pollutants in water, with no need for addition of H2 O2 . The high efficiency of this system makes it a promising alternative for wastewater treatment. Acknowledgements Fig. 14. The proposed reaction scheme diagram of Zn-Fe-CNTs/O2 process.
The generated • OH could oxidize the adsorbed pollutants in situ, or diffuse into the solution phase to oxidize the non-adsorbed pollutants. Moreover, the Zn2+ in solution could be converted to Zn(OH)2 via the combination with generated OH− due to the reduction of H+ and O2 (Fig. 5). The solubility product constant (ksp ) of Zn(OH)2 was only 1.2 × 10−17 at pH 7 and 25 ◦ C. Therefore, the secondary pollution of Zn2+ could be avoided by solid-liquid separation in neutral condition. During Zn-Fe-CNTs/O2 process, the generation of H2 O2 in situ was due to the sacrifice of Zn0 . Therefore, the consumption of Zn0 would influence the stability of the catalyst. However, the catalyst was relatively stable due to its relatively slow consumption rate. How to further increase the stability of the catalyst will be studied in our future work.
This research was supported by the National Natural Science Foundation of China (51338005) and the Program for Changjiang Scholars and Innovative Research Team in University (IRT-13026).
• OH:
4. Conclusions In summary, a novel Fenton-like system that spontaneously generated • OH and H2 O2 by the reaction between Zn-Fe-CNTs and O2 was developed. The Zn-Fe-CNTs/O2 process for the oxidation of pollutants involved in the in situ generation of H2 O2 , the in situ generation of • OH via the decomposition H2 O2 by Fe0 /Fe2 O3 and the
References [1] J.L. Wang, L.J. Xu, Advanced oxidation processes for wastewater treatment: formation of hydroxyl radical and application, Crit. Rev. Environ. Sci. Technol. 42 (2012) 251–325. [2] E. Neyens, J. Baeyens, A review of classic Fenton’s peroxidation as an advanced oxidation technique, J. Hazard. Mater. 98 (2003) 33–50. [3] Z. Wan, J. Hu, J.L. Wang, Removal of sulfamethazine antibiotics using Ce-Fe-graphene nanocomposite as catalyst by Fenton-like process, J. Environ. Manage. 182 (2016) 284–291. [4] N. Wang, T. Zheng, G. Zhang, P. Wang, A review on Fenton-like processes for organic wastewater treatment, Chem. Eng. J. 4 (2016) 762–787. [5] H. Zhang, X. Ran, X. Wu, Electro-Fenton treatment of mature landfill leachate in a continuous flow reactor, J. Hazard. Mater. 241–242 (2012) 259–266. [6] W. Liu, H. Liu, Z. Ai, In-situ generated H2 O2 induced efficient visible light photo-electrochemical catalytic oxidation of PCP-Na with TiO2 , J. Hazard. Mater. 288 (2015) 97–103. [7] W. Wang, Y. Qu, B. Yang, X. Liu, W. Su, Lactate oxidation in pyrite suspension: a Fenton-like process in situ generating H2 O2 , Chemosphere 86 (2012) 376–382. [8] C. Annabi, F. Fourcade, I. Soutrel, F. Geneste, D. Floner, N. Bellakhal, A. Amrane, Degradation of enoxacin antibiotic by the electro-fenton process: optimization, biodegradability improvement and degradation mechanism, J. Environ. Manage. 165 (2016) 96–105. [9] A. Dirany, S. Efremova Aaron, N. Oturan, I. Sirés, M.A. Oturan, J.J. Aaron, Study of the toxicity of sulfamethoxazole and its degradation products in water by a bioluminescence method during application of the electro-Fenton treatment, Anal. Bioanal. Chem. 400 (2011) 353–360.
176
Y. Liu et al. / Journal of Hazardous Materials 342 (2018) 166–176
[10] Y. Li, L. Yang, C. Chen, Y. Lan, Zn(0)-catalyzed ozonation degradation of acid orange 7 (AO7) in aqueous solution, Water Air Soil Pollut. 227 (2016) 364–382. [11] J. Zhang, Y. Wu, L. Liu, Y. Lan, Rapid removal of p-chloronitrobenzene from aqueous solution by a combination of ozone with zero-valent zinc, Sep. Purif. Technol. 151 (2015) 318–323. [12] G. Wen, S. Wang, J. Ma, T. Huang, Z. Liu, L. Zhao, J. Su, Enhanced ozonation degradation of di-n-butyl phthalate by zero-valent zinc in aqueous solution: performance and mechanism, J. Hazard. Mater. 265 (2014) 69–78. [13] G.R. Agladze, G.S. Tsurtsumia, B.I. Jung, J.S. Kim, G. Gorelishvili, The In-cell and Ex-cell Fenton treatment of phenol, 4-chlorophenol and aniline, J. Appl. Electrochem. 37 (2007) 385–393. [14] L. Fu, S. You, G. Zhang, F. Yang, X. Fang, Degradation of azo dyes using in-situ Fenton reaction incorporated into H2 O2 -producing microbial fuel cell, Chem. Eng. J. 160 (2010) 164–169. [15] Z. Ai, T. Mei, J. Liu, J. Li, F. Jia, L. Zhang, J. Qiu, Fe@Fe2 O3 core-shell nanowires as an iron reagent. 3. Their combination with CNTs as an effective oxygen-fed gas diffusion electrode in a neutral electro-Fenton system, J. Phys. Chem. C 111 (2007) 14799–14803. [16] J.L. Wang, Z.Y. Bai, Fe-based catalysts for heterogeneous catalytic ozonation of emerging contaminants in water and wastewater, Chem. Eng. J. 312 (2017) 79–98. [17] J.L. Wang, S.Z. Wang, Removal of pharmaceuticals and personal care products (PPCPs) from wastewater: a review, J. Environ. Manage. 182 (2016) 620–640. [18] J.L. Wang, L.B. Chu, Irradiation treatment of pharmaceutical and personal care products (PPCPs) in water and wastewater: an overview, Radiat. Phys. Chem. 125 (2016) (2016) 56–64. [19] N. Klamerth, L. Rizzo, S. Malato, M.I. Maldonado, A. Agüera, A.R. Fernández-Alba, Degradation of fifteen emerging contaminants at (g L−1 initial concentrations by mild solar photo-Fenton in MWTP effluents, Water Res. 44 (2010) 545–554. [20] W.L. Dai, Q. Sun, J.F. Deng, D. Wu, Y.H. Sun, XPS studies of Cu/ZnO/Al2 O3 ultra-fine catalysts derived by a novel gel oxalate co-precipitation for methanol synthesis by CO2 +H2 , Appl. Surf. Sci. 177 (2001) 172–179. [21] A.C. Hegde, K. Venkatakrishna, N. Eliaz, Electrodeposition of Zn-Ni, Zn-Fe and Zn-Ni-Fe alloys, Surf. Coat. Technol. 205 (2010) 2031–2041. [22] X. Song, Q. Shi, H. Wang, S. Liu, C. Tai, Z. Bian, Preparation of Pd-Fe/graphene catalysts by photocatalytic reduction with enhanced electrochemical oxidation-reduction properties for chlorophenols, Appl. Catal. B: Environ. 203 (2017) 442–451. [23] Y. Lu, F. Yu, J. Hu, J. Liu, Catalytic conversion of syngas to mixed alcohols over Zn-Mn promoted Cu-Fe based catalyst, Appl. Catal. A: Gen. 429–430 (2012) 48–58. [24] X. Weng, M. Guo, F. Luo, Z. Chen, One-step green synthesis of bimetallic Fe/Ni nanoparticles by eucalyptus leaf extract: biomolecules identification, characterization and catalytic activity, Chem. Eng. J. 308 (2017) 904–911.
[25] Y. Liu, W. Jin, Y. Zhao, G. Zhang, W. Zhang, Enhanced catalytic degradation of methylene blue by ␣-Fe2 O3 /graphene oxide via heterogeneous photo-Fenton reactions, Appl. Catal. B: Environ. 206 (2017) 642–652. [26] X. Deng, J. Lee, C. Matranga, Preparation and characterization of Fe3 O4 (111) nanoparticles and thin films on Au (111), Surf. Sci. 604 (2010) 627–632. [27] K. Li, S. Tao, J. Li, X. Wang, Controllable Fe introduction into ordered mesoporous carbon with interconnected small pores for investigating Fe doping effect on hydrogen adsorption, Int. J. Hydrogen Energy 42 (2017) 4733–4740. [28] L.J. Xu, J.L. Wang, A heterogeneous Fenton-like system with nanoparticulate zero-valent iron for removal of 4-chloro-3-methyl phenol, J. Hazard. Mater. 186 (2011) 256–264. [29] C.R. Keenan, D.L. Sedlak, Factors affecting the yield of oxidants from the reaction of nanoparticulate zero-valent iron and oxygen, Environ. Sci. Technol. 42 (2008) 1262–1267. [30] S.E. Mylon, Q. Sun, T.D. Waite, Process optimization in use of zero valent iron nanoparticles for oxidative transformations, Chemosphere 81 (2010) 127–131. [31] S. Kang, W. Choi, Oxidative degradation of organic compounds using zero-valent iron in the presence of natural organic matter serving as an electron shuttle, Environ. Sci. Technol. 43 (2009) 878–883. [32] S. Zhang, D. Wang, L. Zhou, X. Zhang, P. Fan, X. Quan, Intensified internal electrolysis for degradation of methylene blue as model compound induced by a novel hybrid material: multi-walled carbon nanotubes immobilized on zero-valent iron plates (Fe0 -CNTs), Chem. Eng. J. 217 (2013) 99–107. [33] H. Niu, D. Zhang, S. Zhang, X. Zhang, Z. Meng, Y. Cai, Humic acid coated Fe3 O4 magnetic nanoparticles as highly efficient Fenton-like catalyst for complete mineralization of sulfathiazole, J. Hazard. Mater. 190 (2011) 559–565. [34] J.J. Pignatello, E. Oiveros, A. MacKay, Advanced oxidation processes for organic contaminant destruction based on the Fenton reaction and related chemistry, Crit. Rev. Env. Sci. Technol. 36 (2006) 1–84. [35] M. Wang, G. Fang, P. Liu, D. Zhou, C. Ma, D. Zhang, J. Zhan, Fe3 O4 @-CD nanocomposite as heterogeneous Fenton-like catalyst for enhanced degradation of 4-chlorophenol (4-CP), Appl. Catal. B: Environ. 188 (2016) 113–122. [36] G. Xia, Y. Lu, H. Xu, Electrogeneration of hydrogen peroxide for electro-Fenton via oxygen reduction using polyacrylonitrile-based carbon fiber brush cathode, Electrochim. Acta 158 (2015) 390–396. [37] P. Calza, E. Pelizzetti, C. Minero, The fate of organic nitrogen in photocatalysis: an overview, J. Appl. Electrochem. 35 (2005) 665–673. [38] A.G. Trovó, R.F.P. Nogueira, A. Agüera, A.R. Fernandez-Alba, C. Sirtori, S. Malato, Degradation of sulfamethoxazole in water by solar photo-Fenton. Chemical and toxicological evaluation, Water Res. 43 (2009) 3922–3931. [39] D. Marciocha, J. Kalka, J. Turek-Szytow, J. Wiszniowski, J. Surmacz-Górska, Oxidation of sulfamethoxazole by UVA radiation and modified Fenton reagent: toxicity and biodegradability of by-products, Water Sci. Technol. 60 (2009) 2555–2562.