Heterogeneous degradation of tetracycline by magnetic Ag/AgCl/modified zeolite X–persulfate system under visible light

Abstract

Magnetic catalysts (Ag/AgCl/(x)FeX) were fabricated via a facile hydrothermal method followed by a precipitation–photo-reduction method. The Ag/AgCl/(x)FeX catalysts were applied to activate persulfate (PS) for tetracycline (TC) degradation. The degradation efficiencies of the Ag/AgCl/(x)FeX–PS system for TC removal were high due to the enhanced adsorptive capacity of (x)FeX, extended absorbance in the visible-light region and improved space separation of photo-induced charge carriers. The Ag/AgCl/(0.05)FeX catalyst exhibited better performance for TC removal compared with the other samples and complete degradation of TC could be achieved in 120 min by using 1 g L⁻¹ Ag/AgCl/(0.05)FeX and 5.0 mM PS at pH 3.5. The characterization of the catalysts and Electron Paramagnetic Resonance (EPR) studies indicated that TC was mainly degraded by surface-adsorbed radicals resulting from the reaction between PS and Fe(II) on the catalysts under visible light. The Ag/AgCl/(x)FeX–PS system provides new insights for contaminant removal from water.

---

1. Introduction

The widespread overuse of tetracycline (TC) has contaminated waters around the world via municipal and industrial wastewater effluent.¹ It has caused a series of environmental and human health issues.¹ Biological treatment processes are not suitable to remove antibiotics from water because antibiotics can be accumulated in microorganisms and inhibit the activities of microorganisms in bioreactors. Thus, physicochemical methods, such as adsorption and advanced oxidation processes, are suitable for removal of antibiotics from water because these processes are easily operated and give good performance.^2–4 However, some physicochemical processes involve the transfer of contaminants from one phase to another phase. Antibiotics are non-destructively removed from water to generate secondary contamination. Therefore, special attention has been given to advanced oxidation processes because organic pollutants can be mineralized by these processes.⁵ Avanced oxidation processes, such as ozone treatment and photocatalytic processes, rely on the generation of hydroxyl radicals (HO˙) for degradation of organics. Compared with HO˙, SO₄⁻˙ is more stable at room temperature and more non-selectively and widely reactive with environmental contaminants.^5,6 The sulfate radical (SO₄⁻˙) is a very strong one-electron oxidant and the SO₄⁻˙-based advanced oxidation process is one of the newest chemical oxidation technologies for the removal of organics from wastewater. The sulfate radical has a relatively longer lifetime than hydroxyl radicals (10⁻³ vs. 30–40 µs).⁶ In addition, the oxidizing power of SO₄⁻˙ is stronger than HO˙ in the pH range from 2.0 to 8.0.⁷ The SO₄⁻˙ can be generated through persulfate (PS, S₂O₈²⁻) and peroxymonosulfate (PMS, HSO₅⁻) activation by heat and UV, and chemical activation by transition metal ions.^8–10 However, intensive energy inputs and high oxidant costs represent practical constraints for PS/PMS pilot-scale application.⁸ Thus, it is necessary to develop a low-cost, high-efficiency SO₄⁻˙-based advanced oxidation process for rapid remediation of contaminated waters.

Activation under light irradiation is more attractive than thermal or chemical activation for producing SO₄⁻˙ with high efficiency.⁸ Semiconductor materials have played a vital role in SO₄⁻˙ activation under light irradiation. The Ag/AgCl composite has been widely applied to photocatalytic processes due to the surface plasmon resonance effect of noble metal nanoparticles under visible light irradiation.⁶ The major limitations of the Ag/AgCl composite are its difficult separation from water and low degradation efficiency of organics.⁷ The pre-concentration of organics, to enhance the degradation efficiency of organics, can be achieved by using a catalyst support and the catalysts can be easily separated from water while avoiding metal leaching. Thus, different materials, such as clays, polymers and molecular sieves, have been used as catalyst supports to obtain photocatalysts with various advanced properties.^7,8 According to the structure and pH-dependent property of TC, zeolites are the most versatile and chemically and thermally stable catalyst supports for photocatalytic TC degradation. This is because the Al–O–Si tetrahedral structure gives zeolites a cation exchange property due to charge imbalance. In addition, some foreign metal cations can be introduced into the frameworks of zeolites as catalytic sites.⁹

To date, various iron-containing catalysts have been developed to activate PS.^10–13 In particular, NZVI (Fe⁰) is non-toxic, cheap and easily obtained. It can not only be an alternative source of Fe²⁺, but can also recycle Fe³⁺ on its surface and reduce the precipitation of iron hydroxides during the reaction.¹³ Some studies were carried out to test the efficiencies for degradation of organics by a PS/Fe⁰ heterogeneous system under UV light.¹³ However, to the best of our knowledge, very limited information on iron-containing zeolite-based photocatalysts/PS oxidation of TC in water under visible light is available, though it is a likely option for the treatment of TC-induced water pollution.

In this study, zeolite X was impregnated with various weight percentages of NZVI to synthesize (x)FeX through a facile hydrothermal method. Crystalline Ag/AgCl nanoparticles were immobilized on the surface of (x)FeX by the deposition–photoreduction method. The objective of this study is to investigate the persulfate activation by Ag/AgCl/(x)FeX and the resultant degradation of TC. The effects of various parameters, such as pH and chloride ions, on TC degradation were investigated. Furthermore, the mechanisms for PS activation and TC degradation were explored by characterization of catalysts and measurement of free-radical formation by EPR spectroscopy.

2. Experimental

2.1 Materials and chemicals

The hydrochloride salt of tetracycline was purchased from the Sigma-Aldrich Corporation. Sodium silicate solution, sodium aluminate, sodium hydroxide, hydrochloric acid, silver sulfate, ammonia water, sodium chloride and NZVI were purchased from Fisher Scientific. The size of the NZVI was 50 nm. Sodium oxalate (Na₂C₂O₄), benzoquinone (BQ), ascorbic acid (AsA), dimethyl sulfoxide (DMSO), methanol and 5,5-dimethyl-1-pyrrolidine-N-oxide (DMPO) were purchased from Aladdin, China. All the chemicals were analytical grade and used without further purification. Milli-Q ultrapure water (18.2 MΩ cm) was used for all the experiments.

2.2 Synthesis of Ag/AgCl/(x)FeX

2.2.1 Synthesis of (x)FeX. The zeolite X was prepared with sodium silicate solution, sodium aluminate and sodium hydroxide. Firstly, the chemicals were mixed in deionized water and a white gel was formed. The molar ratio of Na₂O : Al₂O₃ : SiO₂ : H₂O equals 7.15 : 1 : 2.2 : 122. Then, the white gel was heated to boiling and was stirred for 1 h to form the precursors of zeolite X. Subsequently, various weight percentages of NZVI (0.03, 0.04, 0.05, and 0.06) were added to the precursors of zeolite X and a black gel was formed. The black gel was stirred for 1 h until it was homogenized. The black gel was crystallized in a Teflon-lined stainless steel autoclave at 343 K for 3 h and then at 368 K for 2 h. After crystallization, the autoclave was naturally cooled to room temperature. Finally, the product was washed with deionized water and dried at room temperature. The (x)FeX samples were obtained. The x denotes the weight percentage of NZVI and the values of x were 0.03, 0.04, 0.05 and 0.06. The zeolite X was also synthesized by the same procedure without the addition of NZVI.

2.2.2 Synthesis of Ag/AgCl/(x)FeX. Firstly, Ag₂SO₄ ammonia solution was prepared by dissolving Ag₂SO₄ (10 mmol) in 2 mol L⁻¹ ammonia water (50 mL). Then, (x)FeX (1 g) was mixed with the Ag₂SO₄ ammonia solution by vigorous mechanical agitation to give a Ag₂SO₄/(x)FeX solution. Saturated NaCl solution (50 mL) was also prepared in advance. Subsequently, the saturated NaCl solution was poured into the Ag₂SO₄/(x)FeX solution and the mixture was vigorously stirred to form a AgCl/(x)FeX solution. The AgCl/(x)FeX solution was illuminated using a 300 W Xe arc lamp with a UV filter (Newport, USA) for 30 min. Ag nanoparticles were generated on the surface of AgCl/(x)FeX by the photo-reduction method. Then, the product was centrifuged and washed with deionized water. Finally, the product was dried at room temperature. The Ag/AgCl/(x)FeX samples were obtained. For comparison, pure Ag/AgCl was also prepared by the same procedure without the addition of (x)FeX.

2.3 Characterization and analysis

X-ray diffraction (XRD) analysis of the Ag/AgCl/(x)FeX was carried out with a D/MAX-2500 diffractometer (Rigaku, Japan). The Brunauer–Emmett–Teller (BET) specific surface areas and the pore size distributions (PSDs) of the samples were measured in a BET analyzer (Gemini V; Micromeritics Instrument Corporation). Field emission scanning electron microscopy (FESEM) and transmission electron microscopy (FETEM) images were taken using a JEOL JSM-7600F FESEM and H-8100 FETEM, respectively. Energy dispersive X-ray spectroscopy (EDX) was carried out in conjunction with the FETEM studies. The diffuse reflectance spectroscopy (UV-vis DRS) measurements were recorded on a VARIAN CARY 100 spectrophotometer in the range of 200–800 nm. The freeze-dried samples were analyzed by X-ray photoelectron spectroscopy (XPS) (Shimadzu-Amics). The hysteresis loops of the samples at 293 K were measured using a vibrating sample magnetometer (VSM-7400, Lake Shore, USA). Electrochemical impedance spectroscopy (EIS) was carried out in a 0.1 M KCl solution containing 5 mM Fe(CN)₆^3−/4−. The electrochemical signals were recorded using an electrochemical analyzer (CHI 660B, Shanghai, China). The conventional three-electrode system consisted of a glassy carbon electrode as the working electrode, Ag/AgCl as the reference electrode and a platinum wire as the counter electrode. An electrochemical workstation was used with the frequency range from 0.01 Hz to 10 kHz at 0.23 V. To determine the isoelectric point (pH_zpc), zeta potential values of the samples were determined as follows: the pH of the solutions was adjusted by adding HCl or NaOH solution. The catalyst was added and stirred for 1 h at 300 rpm until a stable zeta potential was reached. The samples were collected and the zeta potential values of the samples were measured using a Nano-ZS90 instrument (Malvern, United Kingdom).

The TC concentration was determined by High Performance Liquid Chromatography (HPLC) (Agilent, USA) with a 4.6 × 150 mm Zorbax ODS column at 30 °C. The flow rate was 1 mL min⁻¹ and the wavelength of the ultraviolet (UV) detector was set at 355 nm. The mobile phase was 0.05 M phosphate buffer at pH 2.3 (80%) and acetonitrile (20%). The correlation coefficient of the standard curve (n = 10) was more than 0.999. The operational conditions were 3 kV capillary voltage, 553 K dissolution temperature, 403 K source temperature and 20 µL min⁻¹ flow velocity. All the TC concentrations were measured in triplicate and mean values were used. The iron concentration in the solution was measured by inductively coupled plasma mass spectrometry (ICP-MS) on an Agilent HP 4500.

2.4 (x)FeX adsorption studies

Various concentrations (100, 200, 300, 400 and 500 mg L⁻¹) of TC solutions were prepared. 1 g L⁻¹ of the catalyst was mixed with TC solution under constant stirring at 300 rpm. The solution samples were drawn periodically and filtered through 0.22 µm syringe filters. The TC concentrations of the supernatant were measured by HPLC analysis. Experiments were carried out in triplicate and mean values were used. The error was estimated as the standard deviation between triplicate runs.

2.5 TC degradation experiments

The TC degradation experiments were carried out using a photocatalytic reaction chamber with a 500 W Xe arc lamp with a UV filter (Newport, USA). In each experiment, 1 g L⁻¹ Ag/AgCl/(x)FeX sample was mixed with TC solution (100 mL) and 5 mM Na₂S₂O₈ solution. The pH of the TC solution was adjusted by adding one or two drops of 0.1 mol L⁻¹ HCl or NaOH solution. The solution samples were drawn periodically and filtered through 0.22 µm syringe filters. The TC concentrations of the supernatant were analyzed by HPLC analysis. The TC photolysis was measured in the absence of catalysts. The control experiment was carried out without the catalysts. All the experiments were carried out in triplicate and mean values were used. The error was estimated as the standard deviation between triplicate runs. The TC degradation efficiency W% was calculated using the following equation:
where C₀ and C are the initial TC concentration and the TC concentration at time t, respectively.

3. Results and discussion

3.1 Physicochemical characterization

3.1.1 XRD. As shown in Fig. 1B, the high crystallinity of the Ag/AgCl composite and zeolite X were still retained after NZVI impregnation. As the weight percentage of NZVI increased, the reflection intensity at 45° increased. In Fig. 1A, the reflection at 45° is the characteristic reflection of Fe⁰. This verified that some NZVI were immobilized on the surface of zeolite X. In addition, the diffraction peaks labeled as “α” are indexed to the reflections of AgCl crystals and the weak peaks marked with “β” are indexed to the reflections of the metallic Ag⁰ phase.⁶ This illustrated that the Ag/AgCl composite was formed on the surface of (x)FeX. The other reflections between 5° and 25° are the characteristic reflections of zeolite X. This indicated that there was no structural damage to zeolite X after NZVI impregnation.

Fig. 1 XRD patterns of (A) NZVI and (B) Ag/AgCl/(x)FeX samples.

3.1.2 FESEM and FETEM. From Fig. 2, the detailed morphological characteristics of the samples were examined by FESEM and FETEM. From the FESEM images of Ag/AgCl/(0.05)FeX, the size of the particle was about 4 µm and the particle was spherical. In addition, as shown by the magnified FETEM images, the surface of the particle was rough, jagged and uneven due to the NZVI and Ag/AgCl particles immobilized on the surface. This result is consistent with the XRD results. Fig. 2f shows the EDX image of Ag/AgCl/(0.05)FeX, from which the elements Fe, Ag, Cl, Si, Al, O and Na were detected.

Fig. 2 (a)–(c) FESEM images of Ag/AgCl/(0.05)FeX; (d) FETEM image of Ag/AgCl/(0.05)FeX particle and (e) the surface which the white circle shows; (f) the EDX image of Ag/AgCl/(0.05)FeX.

3.1.3 XPS. As shown in Fig. 3A, the XPS spectra confirmed the ingredient elements of zeolite X, Fe, Ag and Cl. This was consistent with the XRD results. In Fig. 3B, there are two bands at 368.2 and 374.2 eV, which were ascribed to Ag 3d5/2 and Ag 3d3/2 binding energies, respectively. The two bands can be further divided into four peaks, which are 367.5 and 368.4 eV, and 373.5 and 374.4 eV, respectively. The bands at 367.5 and 373.5 eV were ascribed to the Ag⁺ ion of AgCl. The bands at 368.4 and 374.4 eV were assigned to metallic silver (Ag⁰).¹¹ This indicated that the silver was present as the Ag⁺ ion of AgCl and metallic silver (Ag⁰). Through the peak differentiation simulation of the Ag 3d core level photoelectron spectra of Ag/AgCl/(0.05)FeX, the peak area ratio of the Ag⁺ to Ag⁰ was found to be 5.2. This also verified the photoreduction of some AgCl particles to metallic silver (Ag⁰) under light irradiation.¹²

Fig. 3 XPS spectra of Ag/AgCl/(0.05)FeX: (A) survey spectrum, (B) Ag 3d and (C) Cl 2p peaks.

As shown in Fig. 3C, the XPS spectra displayed binding energies of Cl 2p3 and Cl 2p1 at about 197.5 and 199.5 eV, respectively. This further verified the existence of AgCl particles on the surface of (0.05)FeX and was consistent with the XRD results.

3.1.4 Specific surface area and pore diameter. As shown in Table 1, the specific surface area of Ag/AgCl/(x)FeX increased slightly from 870.22 to 885.36 cm² g⁻¹ as the weight percentage of NZVI increased. The average pore diameters of the Ag/AgCl/(x)FeX samples were between 0.778 and 0.806 nm and the average pore diameter of zeolite X was 0.743 nm. The impregnation by NZVI had a slight influence on the specific surface areas and pore diameters of the samples. This indicated that most of the NZVI might have permeated into the framework of zeolite X.

Table 1 Specific surface areas, pore diameters and pore volumes of samples

Samples	S BET (cm^2 g^−1)	Pore diameter (nm)	Pore volume (cm^3 g^−1)
NZVI	10.06	0.822	0.003
Zeolite X	843.52	0.743	0.288
Ag/AgCl/zeolite X	864.31	0.772	0.274
Ag/AgCl/(0.03)FeX	870.22	0.778	0.268
Ag/AgCl/(0.04)FeX	875.56	0.784	0.259
Ag/AgCl/(0.05)FeX	880.26	0.793	0.248
Ag/AgCl/(0.06)FeX	885.36	0.806	0.237

---

3.1.5 UV-vis DRS spectra. The solar energy adsorption efficiencies and spectroscopic properties of Ag/AgCl/(x)FeX were evaluated from the UV-vis DRS spectra. As shown in Fig. 4, the samples exhibited strong absorption in the visible light region. The absorbance of Ag/AgCl/(x)FeX in the visible light region was enhanced as the weight percentage of NZVI increased. This was also reflected in the color change of the samples from gray to black.

Fig. 4 UV-vis diffuse reflectance spectra of Ag/AgCl/(x)FeX.

3.1.6 Electrochemical impedance spectroscopy (EIS). EIS was applied to study the charge transfer resistance and the separation efficiency between the photogenerated electrons and holes in the heterogeneous composites.¹⁴ Fig. 5 presents the typical EIS Nyquist plots of Ag/AgCl, Ag/AgCl/zeolite X and Ag/AgCl/(0.05)FeX. The x-axis and y-axis represent the real part of the impedance (Z′) and the negative number of the imaginary part of the impedance (−Z″), respectively. The Nyquist plot can be described by the Randles–Ershler circuit model. In the equivalent circuit, the semicircle radii R_ct are related to the charge transfer at the interface between the semiconductor and electrolyte.¹⁴

Fig. 5 EIS of Ag/AgCl, Ag/AgCl/Zeolite X and Ag/AgCl/(0.05)FeX.

The values of R_ct were 11 750, 8750 and 3500 ohm for Ag/AgCl, Ag/AgCl/zeolite X and Ag/AgCl/(0.05)FeX, respectively. The efficient charge transfer at the interface between the semiconductor and electrolyte suppresses charge recombination to enhance the photocatalytic efficiency.^15,16 If the semicircle radius in the EIS Nyquist plot is smaller, the charge transfer resistance will be lower.¹⁷ The EIS results verified that the impedance order was Ag/AgCl > Ag/AgCl/zeolite X > Ag/AgCl/(0.05)FeX. This indicated that Ag/AgCl/(0.05)FeX had better electron-transfer properties than the other samples. It also verified that Ag/AgCl/(0.05)FeX had a higher separation efficiency of the electron–hole pairs and lower recombination rate than Ag/AgCl and Ag/AgCl/zeolite X under visible light irradiation.

3.1.7 Photoluminescence (PL). As shown in Fig. 6, the PL spectrum intensity of the Ag/AgCl/(x)FeX samples decreased as the weight percentage of NZVI increased. This suggested that the recombination of the photo-electron and hole pairs of Ag/AgCl/(x)FeX was more inhibited as the weight percentage of NZVI increased. The iron ions generated by NZVI acted as a trap for the photo-electron and hole pairs and the separation of charge carriers was improved. There was a large decrease in the PL intensity of Ag/AgCl/(0.05)FeX and this illustrated that the optimum weight percentage of NZVI impregnation was 5%. However, when the weight percentage of NZVI was 6%, there was a slight increase in the PL intensity and this illustrated that the excess iron ions acted as recombination centers instead of a trap for photo-electron and hole pairs. Thus, Ag/AgCl/(0.05)FeX possesses much better optical and photoluminescent behaviors than the other samples.

Fig. 6 PL spectra of Ag/AgCl/(x)FeX.

3.2 Evaluation of TC degradation by various Ag/AgCl/(x)FeX samples

3.2.1 The effect of NZVI on TC adsorption by catalyst supports (x)FeX. Table 2 presents the maximum adsorption capacities (Q_max) of (x)FeX and the Langmuir adsorption constants (K_ads). The correlation coefficients R² of the linear form of the Langmuir model were close to 1 and it indicated the TC adsorption could be described by Langmuir model. This indicated that the TC adsorption process occurred on the homogeneous surface by monolayer adsorption.¹⁸ The Q_max of (x)FeX increased from 384.6 mg g⁻¹ to 476.2 mg g⁻¹ as the weight percentage of NZVI increased from 3% to 5%. However, the Q_max of (x)FeX decreased as the weight percentage of NZVI increased from 5% to 6%.

Table 2 The Langmuir isotherm model parameters of TC adsorption on (x)FeX (dosage of catalysts, 1 g L⁻¹; pH, 3.5)

Samples	Q max (mg g^−1)	K ads (L mg^−1)	R ^2
Zeolite X	384.6	1.32	0.934
(0.03)FeX	402.4	3.43	0.951
(0.04)FeX	434.8	4.61	0.959
(0.05)FeX	476.2	5.83	0.953
(0.06)FeX	427.4	3.83	0.987

---

As the pH is less than 5.5, the TCH³⁺ cation and TCH⁰₂ zwitterion are the main species of TC in the water. From Fig. 7A it can be seen that the weight percentage of aluminum decreased as the weight percentage of NZVI increased from 3% to 6%. This indicated that some aluminum in the framework of zeolite X was substituted by iron. As is shown in Fig. 7B, the intensity of the Na 1s spectrum of (0.05)FeX decreased after TC adsorption. This indicated that there was a cation exchange reaction between TC and sodium ions. As the weight percentage of NZVI increased from 5% to 6%, the Q_max of (x)FeX decreased because there was no extra aluminum in the framework of zeolite X to be substituted by iron to increase the number of sodium ions.

Fig. 7 (A) The weight percentage of aluminum and the TC adsorption capacity of (0.05)FeX as the weight percentage of NZVI increased (dosage of (0.05)FeX, 1 g L⁻¹; pH, 3.5). (B) Na 1s core level photoelectron spectra of (0.05)FeX before and after reaction with TC at pH 3.5. (C) The adsorption mechanism taking place between cationic TC and (x)FeX.

The atom–moles of the elements of zeolite X and (0.05)FeX before and after TC adsorption were determined by XPS. As shown in Table 3, the atom–moles of silicon and oxygen did not change after NZVI impregnation. However, the atom–moles of aluminum decreased and the atom–moles of sodium increased after NZVI impregnation. In addition, the increase in the atom–moles of sodium after NZVI impregnation was as much as the increase in the atom–moles of iron. This indicated that aluminum in the framework of zeolite X was substituted by iron, increasing the atom–moles of sodium. In addition, the cationic TC was removed by (0.05)FeX through the cation exchange reaction. Thus, the TC adsorption capacity of (0.05)FeX was increased after NZVI impregnation. Based on the discussion above, Fig. 7C presents the framework of (0.05)FeX and the adsorption mechanism happening between cationic TC and (x)FeX.

Table 3 The atom–moles of elements of zeolite X, and the unreacted and reacted (0.05)FeX

Samples (1 g)	Fe (mmol)	Na (mmol)	Al (mmol)	Si (mmol)	O (mmol)
Zeolite X	—	2.2	2.5	3.7	13.5
(0.05)FeX	2.3	4.5	0.2	3.7	13.5
(0.05)FeX after TC degradation	2.3	0.7	0.2	3.7	15.3

---

3.2.2 Comparison of TC removal efficiency by Ag/AgCl/(x)FeX–PS. As shown in Fig. 8A and B, the TC degradation through photolysis was negligible (<0.1% over 4 h) with and without PS at a pH of less than 5.5, because TC cannot be degraded under visible light due to the poor visible light absorption.¹⁸ The TC adsorption equilibrium on (0.05)FeX was achieved after 1 h and the Ag/AgCl on the surface of (x)FeX did not have a negative effect on the TC adsorption capacity. The order of the TC removal efficiencies by Ag/AgCl/(x)FeX–PS under visible light was Ag/AgCl/(0.05)FeX > Ag/AgCl/(0.04)FeX > Ag/AgCl/(0.06)FeX > Ag/AgCl/(0.03)FeX > Ag/AgCl/zeolite X.

Fig. 8 The efficiencies of TC removal by the catalysts under visible-light irradiation without PS (A) and with PS (B); photocatalytic degradation reaction kinetics for catalysts without PS (C) and with PS (D) (initial TC concentration, 100 mg L⁻¹; dosage of catalysts, 1 g L⁻¹; pH, 3.5).

The photocatalytic heterogeneous surface reaction could be described using the following Langmuir–Hinshelwood kinetic equation:

where k₂ is the Langmuir–Hinshelwood adsorption coefficient (L mg⁻¹) and k₁ is the reaction rate constant (mg L⁻¹ min⁻¹). The equation could be simplified to a pseudo-first order equation as the initial concentration was very low (<10⁻³ mol L⁻¹). The initial TC concentration was about 2.25 × 10⁻⁴ mol L⁻¹. Thus, the pseudo-first order equation could be used to describe the TC photocatalytic kinetics by the catalysts. The equation is as follows:
where k₃ denotes the pseudo-first order rate constant (min⁻¹) and it incorporates the reaction rate constant and the equilibrium adsorption constant. From Fig. 8C, based on the equation, the slopes of the fitted lines were the k₃ values. The k₃ values were used to quantitatively compare the photocatalytic activities of the catalysts for TC degradation under visible light. The k₃ values were calculated to be 0.208, 0.345, 0.654, 0.896 and 0.557 for Ag/AgCl/zeolite X, Ag/AgCl/(0.03)FeX, Ag/AgCl/(0.04)FeX, Ag/AgCl/(0.05)FeX and Ag/AgCl/(0.06)FeX, respectively, with R² values between 0.975 and 0.996. The photocatalytic activity of Ag/AgCl/(x)FeX increased as the weight percentage of NZVI increased.

From Fig. 8D, the k₃ values were calculated to be 0.391, 1.011, 1.179, 1.411 and 1.275 for Ag/AgCl/zeolite X, Ag/AgCl/(0.03)FeX, Ag/AgCl/(0.04)FeX, Ag/AgCl/(0.05)FeX and Ag/AgCl/(0.06)FeX, respectively, with R² values between 0.95 and 0.98. In addition, the TC photolysis with PS was negligible (<5% over 3 h). The TC removal efficiency of Ag/AgCl/(x)FeX–PS was more than that of Ag/AgCl/(x)FeX under visible light irradiation. This indicated that the PS was activated by Ag/AgCl/(x)FeX under visible light irradiation and the TC degradation efficiency of Ag/AgCl/(x)FeX–PS was increased as the weight percentage of NZVI increased. In addition, the k₃ value of Ag/AgCl/(0.05)FeX–PS under visible light was more than that of Ag/AgCl/(0.05)FeX–PS in the dark. This suggested that the PS activation was promoted under visible light.

3.2.3 The effect of pH. The Ag/AgCl/(0.05)FeX sample was used for the pH studies due to the TC degradation efficiency being more than the other samples. Fig. 9 shows the effect of pH on the TC degradation efficiency by Ag/AgCl/(0.05)FeX–PS. The TC degradation efficiency decreased as the pH increased from 3.5 to 7.5. This indicated that when the pH was lower, the breakdown of persulfate into sulfate free radicals could be further promoted and more SO₄⁻˙ radicals were generated, increasing the TC degradation efficiency. The reaction equations are as follows:

S₂O₈²⁻ + H⁺ → HS₂O₈⁻

HS₂O₈⁻ → SO₄²⁻ + H⁺ + SO₄⁻˙

Fig. 9 Effects of pH on the degradation of TC (TC concentration, 100 mg L⁻¹; dosage of catalyst, 1 g L⁻¹).

The Fe²⁺ ions were more easily generated from NZVI at lower pH. The persulfate could be activated by the homogeneous Fe²⁺ ions and the reaction was as follows:

Fe²⁺ + S₂O₈²⁻ → Fe³⁺ + SO₄⁻˙ + SO₄²⁻

However, the precipitation of Fe²⁺ and Fe³⁺ ions might be induced at pH 5.5–7.5.¹⁹ More sulfate radicals SO₄⁻˙ could be transformed to hydroxyl radicals OH˙ at higher pH through the following reaction:

SO₄⁻˙ + H₂O → SO₄²⁻ + OH˙ + H⁺

More sulfate radicals were produced at lower pH and more hydroxyl radicals were produced at higher pH. In addition, the reduction potential of the hydroxyl radicals was lower than that of the sulfate radicals at higher pH and their reduction potentials were similar at lower pH. Therefore, the TC degradation efficiency decreased as the pH increased.

3.2.4 The effect of chloride ions. Chloride ions often exist in wastewater, so it is necessary to study the influence of chloride ions on the TC degradation performance. As shown in Fig. 10, the TC degradation efficiency decreased as the concentration of chloride ions increased from 1 mM to 100 mM. This indicated that chloride ions had an inhibitory effect on the TC degradation efficiency of Ag/AgCl/(0.05)FeX–PS because the chloride ions could react with the oxidant radicals, such as SO₄⁻˙ and OH˙. Free chlorine atoms could be produced through reactions of Cl⁻ with SO₄⁻˙ and OH˙ (eqn (1)–(3)). The Cl˙, which is unstable in water, could complex with Cl⁻ to generate Cl₂⁻˙ when the Cl⁻ concentration is high (eqn (4)).²⁰ Cl₂⁻˙ might further react with Cl₂⁻˙ and Cl˙ to regenerate Cl⁻ in water (eqn (5) and (6)). Thus, the TC degradation performance was inhibited by Cl⁻ because Cl⁻ had scavenging effects on oxidant radicals.²¹

SO₄⁻˙ + Cl⁻ → SO₄²⁻ + Cl˙ (1)

OH˙ + Cl⁻ → ClOH⁻˙ (2)

ClOH⁻˙ + H⁺ ↔ Cl˙ + H₂O (3)

Cl˙ + Cl⁻ ↔ Cl₂⁻˙ (4)

Cl₂⁻˙ + Cl₂⁻˙ → Cl₂ + 2Cl⁻ (5)

Cl˙ + Cl₂⁻˙ → Cl₂ + Cl⁻ (6)

Fig. 10 Effects of chloride ions on the degradation of TC (TC concentration, 100 mg L⁻¹; dosage of catalyst, 1 g L⁻¹; pH, 3.5).

3.2.5 The stability of Ag/AgCl/(0.05)FeX. The iron ion concentration versus reaction time was detected during TC degradation. From Fig. 11, the maximum concentration of iron ion leaching was only 0.20 mg L⁻¹ and it was far below the European Union directives (2 mg L⁻¹). The iron ion leaching was negligible compared with the iron content of Ag/AgCl/(0.05)FeX (5%). Ag/AgCl/(0.05)FeX is a sustainable and environmentally friendly catalyst.

Fig. 11 Iron ion concentration in water during TC degradation by Ag/AgCl/(0.05)FeX.

3.3 Mechanism discussion

XPS is sensitive to the surface composition and could provide chemical state information. Based on the characterization results, some NZVI were coated on the surface of zeolite X. Fig. 12 shows the peak shape of Fe 2p (Fe 2p1/2 and Fe 2p3/2) and the binding energies of 710.33 and 710.59 eV denote the binding energies of Fe(II) and Fe(III), respectively. This indicates that the oxidation states of iron on the surface of (0.05)Fe-Z are +II and +III. Through the peak differentiation simulation of the Fe 2p core level photoelectron spectra of Ag/AgCl/(0.05)FeX before and after TC degradation, the peak area ratios of Fe(III) to Fe(II) before and after TC degradation were 1.89 and 0.86, respectively. This can be explained by the proportion of Fe(III) species being reduced compared with Fe(II) during the heterogeneous photocatalytic reaction.²² The Fe(III) could trap and transfer electrons and holes to inhibit the recombination of photo-excited holes and electrons.²² This indicated that an effective heterojunction electric field was formed on the interface between Fe(III) and Ag/AgCl, causing the photogenerated electrons and holes to be significantly strengthened. The reaction between Fe(II) and S₂O₈²⁻ ions was promoted. This would significantly strengthen the separation of the photo-electrons and holes. The SO₄⁻˙ formation efficiency increased, causing the photocatalytic efficiency to be enhanced.

Fig. 12 Fe 2p core level photoelectron spectra of Ag/AgCl/(0.05)FeX before (A) and after (B) TC degradation.

The Na₂C₂O₄, BQ, AsA, DMSO scavengers were employed to determine the type of reactive oxygen species (ROS) generated in the Ag/AgCl/(x)FeX–PS system. As shown in Fig. 13, there were various inhibitory effects of the different radical scavengers on TC degradation. Remarkable inhibitory effects on TC degradation were observed when BQ and oxalate were added. Moderate effects were also observed when AsA and DMSO were added. This indicated that ˙OH was one of the predominant oxidative species for oxidizing TC, generated through the reaction between ˙O₂⁻ and H₂O.²² The contribution of h⁺ in Cl⁰ generation for TC oxidation was supported by the observation of a moderate reduction in TC removal when an h⁺ scavenger (oxalate) was added.

Fig. 13 Effect of reactive radical scavengers on TC degradation by Ag/AgCl/(0.05)FeX–PS after 3 h (TC concentration, 100 mg L⁻¹; dosage of catalyst, 1 g L⁻¹; pH, 3.5).

It is worthy of discussion whether radicals are produced or not. In order to study this, radical quenching experiments were first carried out. As shown in Fig. 14A, the TC removal was only slightly decreased by 10% as 2 M methanol, which was 400 times the PS concentration, was added to the reaction. The radicals generated were detected by EPR experiments. As shown in Fig. 14B, the results clearly showed the existence of both SO₄⁻˙ and OH˙ radicals. In addition, as 2 M methanol was added to the TC solution, the intensity of DMPO adducts in the Ag/AgCl/(x)FeX–PS system decreased but the EPR signals were still observed. This indicated that methanol failed to capture the surface-adsorbed radicals completely and consequently the addition of 2 M methanol only scavenged about 10% of TC degradation. This illustrated that the surface radicals were probably responsible for the TC degradation.

Fig. 14 (A) Effects of methanol on TC degradation by the Ag/AgCl/(0.05)FeX–PS system ([PS] = 5.0 mM, [Ag/AgCl/(x)FeX] = 1 g L⁻¹, pH 3.5). (B) EPR spectra versus time for supernatant of Ag/AgCl/(x)FeX–PS–TC batch experiment ([PS] = 5.0 mM, [Ag/AgCl/(x)FeX] = 1 g L⁻¹, [DMPO] = 10 mM, pH 3.5).

In Fig. 15, a possible mechanism is proposed to explain the TC degradation process by Ag/AgCl/(x)FeX–PS. Firstly, cationic TC ions were adsorbed on the surface of (x)FeX through an ion exchange reaction. The adsorption capacity of (x)FeX was enhanced through impregnation with NZVI. Under visible light, Ag/AgCl nanoparticles on the surface of Ag/AgCl/(x)FeX could be excited to generate electron–hole (e⁻/h⁺) pairs (eqn (7)). In this process, the space separation efficiency of the photo-induced charge carriers was enhanced by the Fe(III)/Fe(II) cycle reaction because the impregnation with NZVI could lead to a more efficient utilization of solar energy for accelerating this reaction. After being excited by visible light, the conduction band (CB) electrons of Ag/AgCl shifted to Fe(III) on the surface of(x)FeX. The Fe(III) could act as an electron trap and resulted in the formation of Fe(II), which was less stable compared with Fe(III) due to the half-filled stable d5 configuration. Thus, the trapped charges could be easily released to form stable Fe³⁺ ions (eqn (8)). In this process, it led to the generation of ˙O₂⁻ and SO₄⁻˙ superoxide radicals for the degradation of organic pollutants (eqn (9) and (10)). The radicals ˙O₂⁻ and SO₄⁻˙ might react with H₂O to form the OH˙ radical (eqn (11) and (12)). Moreover, the photo-generated holes could be scavenged by Cl⁻ to produce Cl⁰ (eqn (14)). In addition, Cl⁰ is also an active species which can degrade organic pollutants. The Cl⁰ was reduced to Cl⁻ after the photocatalytic reaction. The TC molecule was oxidized by these superoxide species to intermediates and then was completely mineralized (eqn (13) and (15)). The reactions are schematically shown as follows:

Ag-NPs + hν → e⁻ + h⁺ (7)

Fe(III) + e⁻ → Fe(II) (8)

Fe(II) + O₂ → Fe(III) + ˙O₂⁻ (9)

Fe(II) + S₂O₈²⁻ → Fe(III) + SO₄⁻˙ + SO₄²⁻ (10)

SO₄⁻˙ + H₂O → SO₄²⁻ + OH˙ + H⁺ (11)

˙O₂⁻ + H₂O → OH⁻ + OH˙ (12)

TC + OH˙ → intermediates + CO₂ + H₂O + NO₃⁻ (13)

h⁺ + AgCl → Ag⁺ + Cl⁰ (14)

Cl⁰ + TC → intermediates + CO₂ + H₂O + Cl⁻ (15)

Fig. 15 Schematic diagram of mechanisms of TC degradation process by Ag/AgCl/(x)FeX–PS.

4. Conclusion

In conclusion, magnetic catalysts (Ag/AgCl/(x)FeX) were fabricated via a facile hydrothermal method followed by the precipitation–photo-reduction method. The Ag/AgCl/(0.05)FeX catalyst exhibited better performance for TC removal compared with other samples and complete degradation of TC could be achieved in 120 min by using 1 g L⁻¹ Ag/AgCl/(0.05)FeX and 5.0 mM PS at pH 3.5. The TC degradation efficiency decreased as the pH increased. The predominant reactive oxygen species for TC removal were the ˙O₂⁻ and SO₄⁻˙. The enhancement of the photocatalytic activity after NZVI impregnation was mainly attributed to the enhanced adsorption of cationic TC through a cation exchange reaction, the extended absorption in the visible light region and the efficient space separation of photo-induced charge carriers. This study may expand the further development of zeolite-based heterogeneous catalysts for the degradation of organic pollutants by activation of PS.

Acknowledgements

This work was supported by the Program of Shanghai Institute of Technology (No. (A06)-(10120K156075) YJ2015-33), the National Natural Science Foundation of China (21277093), Program for New Century Excellent Talents in University (NCET-13-0910), National Natural Science Foundation of China (No. 21007048), Key Projects in the National Science & Technology Pillar Program during the Twelfth Five-year Plan Period (2012BAJ25B06) and Twelfth Five-year Plan Period of Major Science and Technology Program for Water Pollution Control and Treatment (2012ZX07403-001).

References

1. H. Yu, C. J. Miller and A. I. Ohno, Photodegradation of contaminants using Ag@AgCl/rGO assemblages: possibilities and limitations, Catal. Today, 2014, 224, 122–131.
2. Y. M. Shin, S. K. Lee and S. H. Jang, Super hydrophobic properties of a hierarchical structure using a silicon micro-tip array decorated with ZnO nanowires, Curr. Appl. Phys., 2014, 14, 665–671.
3. M. M. Liu, B. D. Xi, L. A. Hou and S. L. Yu, Magnetic multi-functional nano-fly ash-derived zeolite composites for environmental applications, J. Mater. Chem. A, 2013, 1, 12617–12626.
4. J. T. Feng, J. J. Li and W. Lu, Synthesis of polypyrrole nano-fibers with hierarchical structure and its adsorption property of Acid Red G from aqueous solution, Synth. Met., 2014, 191, 66–73.
5. Z. D. Lin, S. J. Young and C. H. Hsiao, Adsorption sensitivity of Ag-decorated carbon nanotubes toward gas-phase compounds, Sens. Actuators, B, 2013, 188, 1230–1234.
6. Y. G. Xu, H. Xu and H. M. Li, Ionic liquid oxidation synthesis of Ag@AgCl core–shell structure for photocatalytic application under visible-light irradiation, Colloids Surf., A, 2013, 416, 80–85.
7. J. G. McEvoy, W. Q. Cui and Z. S. Zhang, Synthesis and characterization of Ag/Ag-activated carbon composites for enhanced visible light photocatalysis, Appl. Catal., B, 2014, 144, 702–712.
8. M. M. Liu, L. A. Hou, S. L. Yu and B. D. Xi, MCM-41 impregnated with A zeolite precursor: synthesis, characterization and tetracycline antibiotics removal from aqueous solution, Chem. Eng. J., 2013, 223, 678–687.
9. G. X. Du, Z. H. Li and L. B. Liao, Cr(VI) retention and transport through Fe(III)-coated natural zeolite, J. Hazard. Mater., 2012, 221, 118–123.
10. S. Sultanbayeva, R. Holze and R. M. Chernyakova, Removal of Fe²⁺-, Cu²⁺-, Al³⁺- and Pb²⁺-ions from phosphoric acid by sorption on carbonate-modified natural zeolite and its mixture with bentonite, Microporous Mesoporous Mater., 2013, 170, 173–180.
11. B. Jiang, Y. K. Liu and J. T. Zheng, Synergetic transformations of multiple pollutants driven by Cr(VI)–sulfite reactions, Environ. Sci. Technol., 2015, 49, 12363–12371.
12. X. N. Li, Z. H. Wang and B. Zhang, Fe_xCo_3−xO₄ nanocages derived from nanoscale metal–organic frameworks for removal of bisphenol A by activation of peroxymonosulfate, Appl. Catal., B, 2016, 181, 788–799.
13. Y. G. Guo, X. Y. Lou and C. L. Fang, Novel photo-sulfite system: toward simultaneous transformations of inorganic and organic pollutants, Environ. Sci. Technol., 2013, 47, 11174–11181.
14. X. Yao, X. H. Liu and D. Zhu, Synthesis of cube-like Ag/AgCl plasmonic photocatalyst with enhanced visible light photocatalytic activity, Catal. Commun., 2015, 59, 151–155.
15. H. Li, Y. J. Sun and B. Cai, Hierarchically Z-scheme photocatalyst of Ag@AgCl decorated on BiVO₄ (040) with enhancing photoelectron chemical and photocatalytic performance, Appl. Catal., B, 2015, 170–171, 206–214.
16. G. McEvoy and Z. S. Zhang, Synthesis and characterization of magnetically separable Ag/AgCl-magnetic activated carbon composites for visible light induced photocatalytic detoxification and disinfection, Appl. Catal., B, 2014, 160–161, 267–278.
17. J. Fukahori and T. Fujiwara, Modeling of sulfonamide antibiotic removal by TiO₂/high-silica zeolite HSZ-385 composite, J. Hazard. Mater., 2014, 272, 1–9.
18. M. M. Liu, B. D. Xi and L. A. Hou, Synthesis, characterization, and mercury adsorption properties of hybrid mesoporous aluminosilicate sieve prepared with flyash, Appl. Surf. Sci., 2013, 273, 706–716.
19. H. Song, Y. G. Li and Z. R. Lou, Synthesis of Fe-doped WO₃ nanostructures with high visible-light-driven photocatalytic activities, Appl. Catal., B, 2015, 166–167, 112–120.
20. Y. Ruixia, S. N. Ramjaun and W. Zhaohui, Effects of chloride ion on degradation of Acid Orange 7 by sulfate radical-based advanced oxidation process: implications for formation of chlorinated aromatic compounds, J. Hazard. Mater., 2011, 196, 173–179.
21. Y. Ruixia, W. Zhaohui and H. Yin, Probing the radical chemistry in UV/persulfate-based saline wastewater treatment: kinetics modeling and byproducts identification, Chemosphere, 2014, 109, 106–112.
22. Z. T. Hu, J. C. Liu and X. L. Yan, Low-temperature synthesis of graphene/Bi₂Fe₄O₉ composite for synergistic adsorption-photocatalytic degradation of hydrophobic pollutant under solar irradiation, Chem. Eng. J., 2015, 262, 1022–1032.

---