Bimetallic synergistic degradation of chlorophenols by CuCoO_x–LDH catalyst in bicarbonate-activated hydrogen peroxide system

Abstract

Catalytic wastewater treatment is confronted by varied challenges like catalyst stability and efficiency in aqueous media due to the complex chemistry during organic compound degradation. Herein, we attempt to address this challenge by creating a synergistically stable and active bimetallic CuCoO_x–LDH catalyst via facile copper ion hydrothermal impregnation in a CoO_x–LDH catalyst. Different instrumental techniques like BET, XRD, FTIR, SEM, XPS and electrochemical studies etc. were conducted to investigate the properties of the catalyst before and after impregnation of copper ions. It was found that the changes in the electrochemistry and redox properties of the CuCoO_x–LDH catalyst based on cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and X-ray photoelectron spectroscopy (XPS) appeared in the form of enhanced activity and excellent stability. In the bicarbonate activation of hydrogen peroxide (BAP) system, the synthesized CuCoO_x–LDH catalyst can efficiently degrade 200 ppm 4-chlorophenol (4-CP) with 84% COD and 78% TOC removal in less than 40 minutes, and even 1000 ppm of 4-CP in hours, while the CoO_x–LDH and CuO_x–LDH catalysts or their physical mixtures are apparently sluggish. This catalyst can also effectively degrade various substituted phenols including 2,4-dichlorophenol (DCP), 2,4,4-trichlorophenol (TCP), 2-chlorophenol (2-CP), phenol, and chlorobenzene with significant COD removal. The findings from fluorescence, scavengers, electron paramagnetic resonance (EPR), XPS, and electrochemical studies suggest collectively the generation of ˙OH, ¹O₂, and ˙O₂⁻ species and that the regeneration of active sites may be part of the degradation process. This approach based on CV, EIS and XPS studies has provided novel knowledge about the intrinsic origins of synergetic acceleration of catalyst activity.

---

1. Introduction

Wastewater treatment is a hot topic due to the growing concerns of pollution control. The disposal of organic compounds or treatment of contaminated sites with available technologies is currently inadequate. From the versatile catalytic techniques for wastewater treatment, metal complexes or homogeneous metal catalysts contribute their own toxicity during treatment of wastewater. For example, during Fenton’s oxidation, the iron-containing sludge can lead to proliferation of algae, immobilized enzymes and genetically modified bacteria apart from a strong operating acidic pH.¹ To date, various modifications have been developed in the treatment of wastewater in order to minimize the limitations in the available systems. For example, the reactivity of the popular Fenton’s reagent is extended to a wide pH range with incorporated supports. Similarly, supported catalysts decrease the toxicity threats resulting from its direct use.^2–12 However, due to the complex chemistry of wastewater, the pH of the reaction medium drops during degradation which encourages catalyst leaching and deactivation, which still remains the biggest issue of heterogeneous catalysis.⁷ The application of bimetallic catalysts can sometimes minimize this problem. For example, the bimetallic catalysts Co–Fe/SBA, Cu–Mn/CeO₂, Ag@Cu₂O/NWs, and Fe–Cu/MCM-41 were found to be more active and stable than their corresponding monometallic catalysts during degradation of organic pollutants.^13–17 Moreover, the addition of a small amount of noble metal ions in heterogeneous catalysts was reported to promote their reducibility, and stabilize them against sintering and inhibit coke deposition as in water-gas shift reactions.¹⁷ In fact, in bimetallic heterogeneous catalysts, the second metal ion causes changes in the electronic and geometric structures of metal ions which may promote oxygen deficiencies, facile redox reactions, and novel active sites. An ideal catalyst would be inexpensive, react rapidly under ambient conditions and environmental pH, and would not leave any toxic residues during treatment of wastewater. Recently, bicarbonate was reported in minimizing catalyst leaching owing to its buffer pH apart from activating H₂O₂ and PS during the degradation of organic compounds.^7–9,12

The nature of the support is also very important in supported catalysts which can promote effective interactions between redox metal ions and the support, improve the dispersion of active components, and retain the properties of impregnated metals.¹⁸ Hydrotalcite-based materials show unique properties which makes them excellent materials for catalyst supports. For example, one can tune their properties by varying the metal proportion or replacing different metal ions within the layered structure.¹⁹ Most importantly, the metal ions in layered double hydroxides (LDHs) become more disperse and produce a highly diffuse mixture of metal oxides after thermal treatment making them an ideal material for catalyst supports.¹⁹ In such cases, acid–base pairs (M²⁺–O₂⁻) were produced which triggered catalyst activity after thermal treatment.²⁰

Here, we demonstrate an approach for promoting the stability and activity of a CoO_x–LDH-based catalyst upon facile hydrothermal doping of copper ions in a CuCoO_x–LDH catalyst. The structural, optical and especially the electrochemical properties of the CuCoO_x–LDH catalyst were studied in detail in order to understand the synergy responsible for the enhanced activity and stability. The higher redox potential of Co³⁺ (E_o = 1.18 V) than that of Cu²⁺ (E_o = 1.18 V) allows a feasible reaction thermodynamically in the CuCoO_x–LDH catalyst which promotes interfacial electron transfer as confirmed in XPS and electrochemical studies. Probably the synergetic effect developed from the reduction of Co³⁺ by Cu⁺ in the CuCoO_x–LDH catalyst during degradation may promote synergistically active sites, making it more reactive than its sluggish precursors such as CoO_x–LDH, CuO_x–LDH or their physical mixtures during the degradation of 4-CP. These interactions not only increase the catalyst activity but also promote stability as disclosed from the minimized catalyst leaching and prolonged reactivity.

2. Experimental section

2.1. Chemical and reagents

Co(NO₃)₂·6H₂O (≥99%), Cu(NO₃)₂·4H₂O (>99%), Mg(NO₃)₂·6H₂O (≥99%), Al(NO₃)₃·9H₂O (≥99%), Na₂CO₃ (≥99.8%), NaHCO₃ (99.5%), Na₂SO₄ (≥99%), NaNO₃ (≥99%), NaH₂PO₄ (≥99%), NaOH (99.8%), HCl (37.5%), chlorophenols (4-CP, 2,4-DCP, 2,4,6-TCP, 2-CP), phenol, nitrobenzene, H₂O₂ (30%), NaN₃ (99%), isopropyl alcohol (99.7%), benzoquinone (98.7%) and other chemicals were purchased from local Sinopharm Chemical Reagent, and were used as received without further purification. The EPR spin traps such as 5,5-dimethyl-1-pyrolene N-oxide (DMPO) (98%, Adamas Reagent Co. Ltd.) and N-tert-butyl-α-phenylnitrone (BPN) (98%, Alfa Aesar) were used for detecting both ˙OH and ˙O₂⁻ radicals.

2.2. Preparation of powdered catalyst for bench experiments

2.2.1 Preparation of MgAl as LDH support. First MgAl LDH was prepared by drop-wise addition of 15 mM Mg(NO₃)₂·6H₂O and 5 mM Al(NO₃)₃·9H₂O (250 mL aqueous solution) in a 90 mM basic solution of Na₂CO₃ at constant pH (10 ± 0.02) adjusted with 2 M NaOH and at 60 °C with continuous stirring. After the drop-wise addition, the white suspension was aged for 24 h at room temperature. The suspension was filtered, washed until it was at neutral pH with deionized water and dried overnight at 80 °C. The hard solid material after drying was crushed to a powder which acts as the support for catalysts and was named as LDH.

2.2.2 Preparation of CoO_x–LDH catalysts. In a typical procedure, trimetal-based LDH catalysts containing a fixed proportion of Mg(NO₃)₂·6H₂O and Al(NO₃)₃·9H₂O and a varied composition of Co(NO₃)₂·6H₂O as shown in Table S1 (ESI†) were prepared by the method mentioned above. After drying at 80 °C overnight, the crushed powder materials were calcined in air at 500 °C for 16 h and were named as the CoO_x–LDH catalyst. For comparison different other trimetal-based LDH catalysts such as CuO_x–LDH, MnO_x–LDH, FeO_x–LDH, ZnO_x–LDH, and NiO_x–LDH were also prepared on a common LDH support (MgAl) by the same procedure and their compositional details are also given in Table S1 (ESI†).

2.2.3 Preparation of CuCoO_x–LDH catalysts. To prepare the CuCoO_x–LDH catalyst, 0.5 g of the already prepared best CoO_x–LDH-1 catalyst was impregnated hydrothermally with different concentrations of Cu(II) salt as shown in Table 1, using a 100 mL autoclave at 80 °C for 4 h. The resulting suspension was transferred to a 100 mL beaker and heated on a hot plate with regular stirring until maximum water was evaporated. The resulting paste was dried at 80 °C overnight, and calcined at 500 °C for 5 h, and was named as the CuCoO_x–LDH catalyst.

Table 1 Improvement in efficiency and stability of CoO_x–LDH catalyst with impregnation of Cu^a

Catalyst	0.5 g CoOx–LDH-1 catalyst taken for impregnation				BET (m^2 g^−1)	% metal loading	pH	% CP removal	Leaching (ppm)
	Co^2+ (mM)	Mg^2+ (mM)	Al^3+ (mM)	Cu^2+ (mmol)					Co^2+	Cu^2+
a Conditions: 100 ppm 4-CP, 30 mM H2O2, 30 mM NaHCO3, temperature 40 °C, catalyst amount 1.5 g L^−1, reaction time 1 h.
CoOx–LDH-1	0.075	15	5		92	0.60	8.68	72	0.208	—
CuCoOx–LDH-1	—	—	—	0.50	123	3.9	8.63	83	0.180	0.461
CuCoOx–LDH-2	—	—	—	1.00	131	9.78	8.66	100	0.131	0.467
CuCoOx–LDH-3	—	—	—	1.50	119	13.38	8.65	96	0.112	0.512
CuCoOx–LDH-4	—	—	—	2.00	109	15.47	8.67	80	0.113	0.528
CuOx–LDH		—	—	1.00	84	7.4	8.53	42	—	1.192
MgAl as LDH		—	—		87	—	8.72	15	—	—

---

2.3. Catalyst characterization techniques

X-ray diffraction (XRD) patterns and Fourier transform infrared spectroscopy (FTIR) of the powdered materials were conducted with a PANalytical X’Pert PRO and Bruker Equinox 55, respectively. The specific surface area and thermal stability of the powdered catalysts were measured on a Micromeritics ASAP 2020 and STA 449C thermal analyzer (Netzsch, Germany), respectively. Scanning electron microscopy (SEM) and compositional analysis of the powdered materials were carried out with Virion 200 SEM and atomic emission spectroscopy (MP-AES 4100 Agilent), respectively. The redox properties of the prepared catalysts were investigated by means of photoelectron spectroscopy (XPS) with a V.G scientific ESCALAB mark II system. The cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) studies were performed on a PGSTAT-12 electrochemical analyzer (AUTO Lab BV Netherland). A conventional three-electrode system was adopted with glassy carbon (GC) as the working electrode, Pt as the counter electrode, and a standard calomel electrode (SCE) as the reference electrode. Electron paramagnetic resonance (EPR) for detecting ˙OH and ˙O₂⁻ radicals was conducted on a Bruker EMX 10/12 spectrometer at room temperature.

2.4. General procedure for degradation and analysis of CPs (chlorophenols) during bench experiments

A total of 20 mL of aqueous solution containing 100–300 ppm 4-CP, 25 mM H₂O₂, 30 mM NaHCO₃, and 10 mg catalyst was allowed to stir at room temperature for 1 h. The degradation of 4-CP was determined using high performance liquid chromatography (HPLC FL-2200) after collecting samples at specific time intervals (10–60 min). The HPLC equipment was equipped with a UV detector adjusted at 220 nm, and a C₁₈ column (250 × 4.6 mm), and the eluent composition for the mobile phase was 60% methanol with water (60 : 40 v/v). To assess the degree of mineralization, COD and TOC analyses were carried out on a portable (HACH-1010) COD and TOC analyzer (Analytikjena Multi N/C-3100), respectively. The amounts of leached Co and Cu ions during the reaction were determined by atomic emission spectroscopy (AES).

2.5. Procedure for monitoring the stability of the powdered catalyst

For investigating the stability of the powdered catalyst, 10 mL of stock 4-CP solution (5000 mg L⁻¹) was added to 40 mL of deionized water in a 100 mL round-bottom flask which already contained 10 mg of catalyst and 0.126 g of NaHCO₃. The resulting 4-CP solution (1000 mg L⁻¹) was stirred for 1 h to establish an adsorption equilibrium between the catalyst and 4-CP and then the reaction was initiated with 6 mM H₂O₂. The reaction was monitored for 9 h and each hour 3 mL of sample was withdrawn for 4-CP degradation and leaching studies. During this continuous reaction, 6 mM H₂O₂ was maintained throughout the reaction by adding 3 mL each hour from 100 mM H₂O₂ stock solution.

3. Results and discussion

3.1. Catalyst characterization

The XRD patterns of the prepared catalysts (CoO_x–LDH and CuO_x–LDH) are shown in Fig. 1A, and exhibit characteristic hydrotalcite-like peaks.⁹ The peaks at lower 2θ = 11°, 23° and 34° correspond to the 003, 006 and 009 basal planes, while the peaks at higher 2θ = 39°, 46°, 60° and 61° show the 015, 018, 110 and 113 non-basal planes of hydrotalcite. The composition of both Mg²⁺ and Al³⁺ in the CoO_x–LDH-1,2,3 catalysts remained constant, while the concentration of Co²⁺ increased in ascending order. The change in the concentration of Co²⁺ would affect the d-spacing of the LDH layers due to their different ionic radius (0.072 nm ionic radius for Mg²⁺, 0.054 nm for Al³⁺, and 0.065 nm for Co²⁺) and thus CoO_x–LDH-1,2,3 appeared with slightly different reflections in the XRD pattern. Here, the three peaks indexing 003, 006, and 009 decreased in intensity as the Co content increased, which indicates that the substitution of Mg²⁺/Al³⁺ by Co²⁺ resulted in poor crystallization.²¹ The XRD patterns of CoO_x–LDH-1, CuO_x–LDH and a series of CuCoO_x–LDH catalysts after calcination are displayed in Fig. 1B. The peaks at 2θ ≈ 43° and 63° demonstrate the MgO phase (JCPDS-079-0612) which also coincides with 2θ ≈ 42.5° and 61.8° for the Cu₂O phase (JCPDS 78-2076) in the CuCoO_x–LDH catalysts. The peaks at 2θ ≈ 35.6° and 38.8° which constantly appeared for all the CuCoO_x–LDH catalysts indicate the phases of pure CuO (JCPDS-041-0254). All the CuCoO_x–LDH catalysts show an almost similar pattern of peaks, indicating good crystallinity. In all the catalysts no separate peaks were detected for Co₃O₄ or other forms of cobalt due to the low Co loading (0.60%). However the spectral region 2θ ≈ 43° is reported to coincide with the MgO, CoO, Co₃O₄ and MgCo₂O₄ phases.²² This behaviour may be associated with the well-dispersed distribution and strong linkages of different phases producing a solid solution.

Fig. 1 (A) XRD pattern of LDH-based catalysts before calcination. (B) XRD patterns of a series of prepared catalysts after calcination (CuO (+), Cu₂O, MgO, CoO, Co₃O₄, MgCo₂O₄ (*)).

The BET surface area of the studied catalysts is shown in Tables 1 and S1 (ESI†). As shown, the CoO_x–LDH and CuO_x–LDH catalysts exhibited nearly the same surface area (84–94 m² g⁻¹), whereas in the Mn, Fe and Zn LDH-based catalysts, a smaller surface area was recorded. The surface area of CoO_x–LDH-1 increased during Cu impregnation by the hydrothermal method. With a high loading of Cu, the surface area slightly decreased which may be accounted for by the blocking of some channels or pores on the catalyst surface. In addition, the elemental analysis of the powdered catalysts is also shown in Tables 1 and S1 (ESI†).

The FTIR spectra of the prepared catalysts are shown in Fig. 2, demonstrating shifting of certain absorption bands in the wavelength region below 1500 cm⁻¹. For example, a common strong absorption band appears near 1385 cm⁻¹, corresponding to anti-symmetric ν₃ stretching of carbonate. However, a clear shift of this peak was observed in MgAl (lower wavelength), and CuCoO_x–LDH and CoO_x–LDH (higher wavelength). The presence of residual carbonate in the solid catalyst after calcination facilitates the reconstruction of a lamellar structure (memory effect) in the LDH-based catalysts after contacting an aqueous solution or moisture.²³ These results are in accordance with SEM analysis where a clear layered structure still exists after calcination. The lower region of FTIR (511–446 cm⁻¹) is very important, representing metal hydroxyl (M–OH) or metal–O–metal (M–O–M) linkages.²³ For example, a weak band around 446 cm⁻¹ in all the catalysts indicates a Mg/Al–OH transition band. Importantly, the transition bands around 511–473 cm⁻¹ appear only in the CuCoO_x–LDH catalyst, which may differentiate the metal–O–metal or metal–OH linkages from other catalysts.

Fig. 2 FTIR spectra of CuCoO_x–LDH-2 and the catalyst precursors.

Both the CoO_x–LDH-1 and CuCoO_x–LDH-2 catalysts were also characterized by SEM analysis to confirm the morphological changes during copper ion impregnation as shown in Fig. 3A and B. The as-synthesized CoO_x–LDH-1 catalyst demonstrates a clear layered structure where different layers stack over each other as shown in Fig. 3A. After copper ion impregnation as shown in Fig. 3B, the CuCoO_x–LDH-2 catalyst shows a completely different morphology in the form of small CuO or Cu₂O particles on the surface of the CoO_x–LDH-1 catalyst. These morphological changes were further confirmed by TEM analysis as shown in Fig. S2 (ESI†). The as-prepared materials indicate disperse coating of CuO_x over the surface of pristine CoO_x–LDH-1, making it very rough which can be visualized clearly at high resolution.

Fig. 3 (A) SEM images of CoO_x–LDH-1 and (B) SEM images of CuCoO_x–LDH-2 showing changes in morphology.

3.2. Performance of catalysts in the degradation of 4-CP

For searching for the best catalyst in the BAP system, a survey of MO_x–LDH catalysts was first performed using 4-CP as a model compound where M represents different metal ions like Co, Cu, Mn, Fe, Zn and Ni. In the studied MO_x–LDH catalysts, the CoO_x–LDH catalyst demonstrates the highest activity (Table S1 in the ESI†). Notably, the efficiency of the CoO_x–LDH catalyst decreases as the Co content increased which may be accounted for by Co fine dispersion and strong linkages in the LDH structure as confirmed through XRD analysis. Similarly, as shown in Table 1, Cu doping in the CoO_x–LDH-1 catalyst brings improvements both in efficiency and stability. In particular, all of the CuCoO_x–LDH catalyst series showed lower Co and Cu leaching than their precursors like CoO_x–LDH-1 and CuO_x–LDH, providing good stability and enhanced activity. These improved properties of the CuCoO_x–LDH catalyst series may arise from an increase in the surface area and a synergetic coupling effect of metal oxides as confirmed from BET, XPS, CV and EIS studies as below.

In order to address the intrinsic origins of the enhanced stability and activity in the CuCoO_x–LDH catalysts, XPS and electrochemical studies were performed on the CuCoO_x–LDH-2 catalyst. The XPS analysis provides information about the elementary states of metal ions in the solid catalyst. As shown in Fig. 4A (before reaction), the Cu 2p core spectrum appeared as 2p_3/2 and 2p_1/2 along with its satellite peaks. Copper oxide mainly exists in two semiconducting phases such as CuO or Cu₂O. The CuO phase is generally recognized by a 2p_3/2 peak (933.3–935.2 eV) along with a satellite peak.^24–26 Meanwhile Cu₂O is distinguished by a binding energy difference of 19.8 eV between the 2p_3/2 and 2p_1/2 peaks on the binding energy scale along with a 2p_3/2 peak at 931.2–932.8 eV.²⁴ Here, the 2p_3/2 peak after peak fitting gives peak A (932.4 eV) and peak B (933.7 eV) which were assigned to Cu₂O and CuO, respectively, in the CuCoO_x–LDH-2 catalyst before reaction.²⁴

Fig. 4 XPS analysis of the CuCoO_x–LDH-2 catalyst (A) before reaction and (B) after reaction.

The CuCoO_x–LDH-2 catalyst after reaction (Fig. 4B) displayed a spectrum with a core region (Cu 2p_2/3 and Cu 2p_1/2) along with satellite peaks. The Cu 2p_2/3 peak after peak fitting also split into peak A (933.7 eV) and peak B (935.2 eV). Both peaks were assigned to Cu²⁺ (CuO and Cu(OH)₂, respectively).^24,26 Thus from XPS analysis we confirmed that the oxidation state of Cu changes from Cu⁺ to Cu²⁺ during reaction. XPS can detect only signals on the upper 5 nm surface while XRD can detect it in a micro-order thickness surface.²⁷ SEM/TEM clearly visualized the morphological changes of CuO/Cu₂O on the surface of the catalyst. Probably, during impregnation, the copper encircles a maximum of cobalt, making it deeper and beyond the detection limit of XPS. As shown in Fig. 4B, no signal was detected for Co both before and after reaction. However, both Cu and Co in alloy shapes were reported to produce Co²⁺_1−xCu²⁺_x[Co³⁺]₂O₄ mixed phases after air calcination.²⁸

Electrochemical responses of the CuCoO_x–LDH-2 catalyst along with CoO_x–LDH-1 and CuO_x–LDH were also measured on oxidation of 4-aminophenol (4-AP) using cyclic voltammetry (CV) in 0.1 M phosphate buffer saline (PBS) at pH 7.4. Both CoO_x–LDH-1/GC and CuO_x–LDH/GC (Fig. 5A) demonstrate clear redox peaks during electrochemical oxidation of 4-AP. However in the case of CuCoO_x–LDH-2/GC, an obvious increase in peak current along with shifting of electrode potential was observed. However for the bare GC electrode under a similar experimental setup, no redox peaks were detected. The difference in redox potential among metals allows thermodynamically feasible reactions.^29,30 Probably in the CuCoO_x–LDH-2 catalyst, this property of metals is responsible for the enhanced peak current owing to the promotion of interfacial electron transfer.

Fig. 5 (A) Cyclic voltammetry (CV) studies of the CuCoO_x–LDH-2 catalyst and their precursors (CoO_x–LDH-1, CuO_x–LDH) in 0.1 mM 4-aminophenol and 0.1 M PBS solution at pH 7.4. (B) Nyquist plots of catalyst-modified electrodes in 0.1 M KCl containing 1.0 mM K₃Fe(CN)₆ and 1.0 mM K₄Fe(CN)₆ in the frequency range 0.1–10⁵ Hz.

To study the development of redox properties of catalysts more deeply, electrochemical impedance spectroscopy (EIS) was also conducted. The EIS studies offer significant clues about the probing of electron transfer kinetics occurring at catalyst-modified electrodes in 2 mM [Fe(CN)₆]³⁻/⁴⁻ solution. The Nyquist plots for various modified electrodes (MgAl as LDH, CoO_x–LDH-1, CuO_x–LDH and CuCoO_x–LDH-2) (Fig. 5B) present a semicircular portion, and the diameter (R_ct) represents the electron transfer limitations in the catalyst.³¹ Obviously, the R_ct value for CuCoO_x–LDH-2 (539 Ω) is less than that of the CoO_x–LDH-1 (650 Ω) and MgAl–LDH (723 Ω) catalysts. Likewise, the solution resistance (R_s) corresponding to the intersection of each curve at the real part Z′ is smaller for CuCoO_x–LDH-2 (1018 Ω) than for both CoO_x–LDH-1 (1185 Ω) and MgAl–LDH (1983 Ω).³¹ These results obviously indicate that Cu doping in the CuCoO_x–LDH-2 catalyst encourages the fast transfer of charge between the catalyst and electrolytic solution (2 mM [Fe(CN)₆]^3−/4−) and conforms well with the CV findings. From all these facts like a large active BET surface area, XPS findings (the change in the oxidation state of Cu) and enhanced electron transfer features, the excellent catalytic activities of the CuCoO_x–LDH-2 catalyst are favourably indicated.

To further evaluate the synergetic effect in a real practical environment, the best catalyst (CuCoO_x–LDH-2) was selected for detailed studies. The activities of the CuCoO_x–LDH-2 catalyst for 4-CP degradation (200 ppm), COD and TOC removal together with control experiments including CuO_x–LDH, CoO_x–LDH-1, their physical mixtures (CuO_x–LDH + CoO_x–LDH-1), MgAl–LDH, and leached Cu²⁺ and Co²⁺ ions as a homogeneous catalyst are shown in Fig. 6. It is clear that the triggered efficiency of the CuCoO_x–LDH-2 catalyst was better, with a 60% higher removal rate constant (5.93 × 10⁻² mg L⁻¹ min⁻¹) than that of the catalyst precursor like CoO_x–LDH-1 (8.3 × 10⁻³ mg L⁻¹ min⁻¹), and even much higher (80%) than that of CuO_x–LDH (4.0 × 10⁻³ mg L⁻¹ min⁻¹). As shown in Fig. 6A, the CuCoO_x–LDH-2 catalyst could completely degrade 200 ppm 4-CP in less than 40 min, while both the CoO_x–LDH-1 and CuO_x–LDH catalysts cause significantly lower degradation (less than 40% for CoO_x–LDH-1 and 20% for CuO_x–LDH) in 1 h. Using the physical mixtures (5 mg of each of CoO_x–LDH-1 and CuO_x–LDH) the degradation of 4-CP was even lower than that with the CoO_x–LDH-1 catalyst. These results clearly suggest that the enhanced activity does not originate from simple addition of metal ions in the solid catalyst, but that both Cu and Co ions in the CuCoO_x–LDH-2 catalyst synergistically promote the active sites. Moreover, the negligible activity of the catalyst support and leached Co and Cu ions further strengthens the synergetic concept in the CuCoO_x–LDH-2 catalyst.

Fig. 6 Synergy provided by Cu in the CuCoO_x–LDH-2 catalyst during degradation of 4-CP: (A) % 4-CP removal, and (B) % COD removal. Conditions: 200 ppm 4-CP, 25 mM H₂O₂, 30 mM NaHCO₃, catalyst amount 0.5 g L⁻¹, CuCoO_x–LDH reaction pH 8.62, CuO_x–LDH pH 8.50, CoO_x–LDH pH 8.65, temperature 30 °C, reaction time 1 h.

Likewise, both Cu and Co ions in the solid catalyst (CuCoO_x–LDH-2) have a key role in the deep oxidation of 4-CP as revealed from % COD and % TOC removal in Fig. 6B. Both the COD and TOC removals were much higher (84 and 78%) than those for CoO_x–LDH-1 (30 and 18%) and CuO_x–LDH (19 and 12%), respectively. A much lower COD removal (40%) was observed after 1 h reaction when cobalt and copper ions were used as a homogeneous catalyst (2 ppm Co²⁺, 4 ppm Cu²⁺). The deep oxidation (mineralization) with the CuCoO_x–LDH-2 catalyst was 64 and 76 times higher (% COD and TOC removal) than that with CoO_x–LDH-1 and even much higher than with the CuO_x–LDH catalyst. The enhanced degradation and much higher mineralization of 4-CP with CuO_xCo–LDH-2 than their precursors suggests that both Cu and Co ions appear with different properties which might be due to the changes in the electrochemical properties as well as the large BET surface area of the CuCoO_x–LDH-2 catalyst.

3.3. Catalyst stability

Almost all heterogeneous catalysts have a stability problem during the degradation of organic compounds, so it is important to check the stability of a catalyst in a real practical environment. For this purpose the CuCoO_x–LDH-2 catalyst was employed with a high dose of 4-CP (1000 ppm) for a long time service test as shown in Fig. 7. As shown the degradation continued until all of the 4-CP (1000 ppm) was removed completely. Similarly in the inset of Fig. 7, minor leaching for both Cu and Co ions (0.48 and 0.16 ppm, respectively) was detected in the first 1 h as the reaction was initiated by adding 6 mM solution of H₂O₂. However, the leaching profile then remained unchanged for both Cu and Co ions as the reaction continued further. These results clearly suggest good stability and enhanced activity of the powdered catalyst which might arise from regeneration of active sites through the redox cycle reaction and also the use of basic conditions.

Fig. 7 Stability profile of the CuCoO_x–LDH-2 catalyst during the degradation of 1000 ppm 4-CP solution. Conditions: 1000 ppm 4-CP, 6 mM H₂O₂, 30 mM NaHCO₃, amount of catalyst 0.5 g L⁻¹, temperature 30 °C, reaction time 9 h, pH 8.51.

The stability of the CuCoO_x–LDH-2 catalyst was further evaluated by its efficiency test after repeated use. The catalyst efficiency slowly dropped after consecutive runs as shown in Fig. S11 (ESI†). The efficiency of the catalyst was above 90% in the first two runs while after the fifth run, the efficiency decreased to nearly 50%. Furthermore, as shown in Fig. S12,† the XRD analysis provides us with very useful information about structural changes during the repeated use of the CuCoO_x–LDH-2 catalyst. It is apparent that the XRD analysis shows the complete stability of the catalyst during the first run, presenting the reflections as before the reaction. The appearance of the 001, 003 and 110 reflections at 2θ = 11°, 23° and 61° is due to the memory effect of the LDH structure. However after the fifth run, the reflections belonging to the CuO/Cu₂O or LDH structure become weak, indicating some active component loss. Interestingly, the LDH skeleton was still maintained even after the fifth run as apparent from the reflection at lower 2θ = 11° and 23°, and high 2θ = 110° which highlights the excellent stability of this material for a catalyst support.

3.4. Efficiency of catalyst in a BAP system

To evaluate the efficiency of the catalyst (CuCoO_x–LDH-2) in a BAP system, various control experiments, like (I) catalyst alone for adsorption study, (II) H₂O₂ + NaHCO₃, (III) catalyst + NaHCO₃, (IV) catalyst + H₂O₂, and (V) catalyst + H₂O₂ + NaHCO₃, were performed using 4-CP as a model compound. As shown in Fig. 8, the catalyst alone caused negligible degradation while H₂O₂ + NaHCO₃ showed 17% degradation. Similarly, 15% of 4-CP was removed with catalyst + NaHCO₃ that further increased up to 20% with catalyst + H₂O₂. The degradation of 4-CP remarkably increased to 100% in less than 10 min with catalyst + H₂O₂ + NaHCO₃, highlighting the bicarbonate activation of H₂O₂. The limited adsorption of 4-CP with the catalyst suggests that the degradation was mainly controlled by various active species. Moreover, in the BAP system, various reactive oxygen species (ROS) were produced during activation of H₂O₂, like ˙OH, ¹O₂, and ˙O₂⁻ along with popular ˙CO₃⁻ and HCO₄⁻.^7,32–34 Particularly, the BAP system is recognized with production of peroxy monocarbonate anions (HCO₄⁻), considered as a stronger oxidizing agent than H₂O₂ (HCO₄⁻/HCO₃⁻ = +1.8 V vs. NHE), which is capable of oxidizing different organic compounds alone.^35–37

Fig. 8 Activity of CuCoO_x–LDH-2 catalyst in the BAP system. Conditions: 100 ppm 4-CP, 25 mM H₂O₂, 30 mM NaHCO₃, amount of catalyst 0.5 g L⁻¹, temperature 30 °C, pH 8.66.

3.5. Efficiency of catalyst for other organic compounds

Wastewater is composed of mixtures of organic and inorganic compounds. Phenolic and benzene-like compounds are considered highly toxic and resistant to biodegradation. In the present study apart from 4-CP, the degradations of other organic compounds including 2-CP, 2,4-DCP (2,4-dichlorophenol), 2,4,6-TCP (2,4,6-trichlorophenol), phenol and nitrobenzene were also conducted on already optimized conditions and the results are summarized in Fig. 9. As shown, 2,4,6-TCP and 2,4-DCP were completely removed with 62% and 53% COD removal, respectively. Similarly, 90% of 2-CP and 58% of phenol were degraded with 50% and 41% COD removal, respectively. Nitrobenzene was found to be more resistive, and degradation followed the order of 4-CP > 2,4,6-TCP > 2,4-DCP > 2-CP > phenol > nitrobenzene. The presence of –Cl groups may have activated the ring due to their electron-donating and electron-withdrawing effect. That is why all the CPs are degraded faster than phenol. Regarding nitrobenzene, the –NO₂ group deactivates the ring due to its electron-withdrawing ability, thus making it difficult for catalytic degradation to occur.

Fig. 9 Activity of the CuCoO_x–LDH-2 catalyst for other organic compounds. Conditions: 100 ppm targeted compounds, 25 mM H₂O₂, 30 mM NaHCO₃, amount of catalyst 0.5 g L⁻¹, temperature 30 °C, pH 8.66, reaction time 1 h.

3.6. Effect of different parameters on the degradation of 4-CP

An ideal catalyst would be inexpensive, would react rapidly under ambient conditions and environmental pH, and would not leave any toxic residues during the treatment of wastewater. So it is important to check different parameters such as the effect of concentration, oxidant amount, pH and reaction temperature to understand its full use during the treatment of wastewater.

The effect of the 4-CP amount (100–300 ppm) was monitored and the results are summarized in Fig. 10. The CuCoO_x–LDH-2 catalyst can degrade 100, 150, 200, 250 and 300 ppm 4-CP with different degradation rates (6.99, 6.60, 5.93, 5.52 and 3.37 × 10⁻² mg L⁻¹ min⁻¹, respectively). The degradation rate decreased as the 4-CP concentration increased. For example, 100 ppm was degraded in less than 10 min with a degradation rate constant of 6.99 × 10⁻² mg L⁻¹ min⁻¹, while 300 ppm of 4-CP was completely removed in 60 min with a slower reaction rate (3.37 × 10⁻² mg L⁻¹ min⁻¹). The temperature plays a very important role as it affects the process efficiency as well as the cost on the process apart from controlling the activation energies required for degradation. As shown in Fig. S3,† the degradation rate of CP increased as the temperature increased. At lower temperature, i.e. 10 °C, the degradation rate was slow (3.37 × 10⁻² mg L⁻¹ min⁻¹) but increased to 5.93 × 10⁻² mg L⁻¹ min⁻¹ at 30 °C. Between 20 and 30 °C, the degradation rates were close to each other and led to complete degradation of 4-CP in less than 20 min. Just like other wastewater treatment systems, the activity of the present system is also strongly dependant on solution pH as shown in Fig. S4.† It is apparent that the catalyst shows the highest activity in the pH range 7–9. Probably at this pH, the bicarbonate is available in solution for activation of H₂O₂, however as the pH decreases, the bicarbonate will be converted to carbonic acid and no or less bicarbonate is available to activate H₂O₂, which results in the lower activity of the catalyst. Similarly at pH > 9, the activity of the catalyst also decreased. The other parameters like the effect of H₂O₂ and bicarbonate on the degradation of 4-CP were also studied and the results are summarized in Fig. S5 and S6 (ESI†). Hydrogen peroxide is an environmentally friendly mild oxidant and needs activation by different means including catalyst activation. An excessive amount of H₂O₂ was reported to play the role of a scavenger, thus affecting the catalyst efficiency. In the present system, the degradation smoothly increased and reached a maximum at 25 mM, and then decreased slightly by a further increase in the H₂O₂ concentration, which may be due to its scavenging action for various reactive species. Bicarbonate is a relatively non-toxic and abundantly available compound in nature in the range of 14.7–25 mM. Bicarbonate is a naturally occurring (soil, water, interfaces) complexation ligand that controls the bioavailability of essential trace metal ions in the ecosystem. In addition, bicarbonate is also reported to bring changes in the redox properties of metals during complexation which helps organisms with selective uptake, intracellular delivery and redox poisoning of metal ions on functional sites within macromolecules.³⁸ The role of bicarbonate as a H₂O₂ activator for different reactions including wastewater treatment has been very recently studied.^7–9,12 Apart from activating H₂O₂ in wastewater, bicarbonate also plays a very important role in minimizing or controlling catalyst leaching during degradation.^7–9 Here in Fig. S6 (ESI†), the degradation rate increased with the addition of bicarbonate. At 25 and 30 mM of bicarbonate, the difference in degradation becomes nearly negligible while this difference is more apparent at low concentrations.

Fig. 10 Effect of 4-CP concentration on the activity of the CuCoO_x–LDH-2 catalyst. Conditions: 100–300 ppm 4-CP, 25 mM H₂O₂, 30 mM NaHCO₃, amount of catalyst 0.5 g L⁻¹, temperature 30 °C, pH 8.60 ± 0.5 (changes with changing 4-CP concentration).

3.7. Mechanistic study

The formation of free radicals, high valent metal-oxo species or both has generally been reported in the literature during wet peroxide oxidative degradation of organic pollutants.³⁹ For example, the formation of a coloured peroxocobalt complex without ˙OH radicals was reported during the degradation of organic compounds with Co²⁺/Al₂O₃.⁴⁰ In another case, a peroxocomplex [Co²⁺(amm)₆(OOH)]⁺ was considered as an active intermediate species for the production of ˙OH radicals with a [Co²⁺(amm)₆]/SiO₂ catalyst.⁴¹ The reactivity of both monovalent Cu⁺ and divalent Cu²⁺ toward H₂O₂ is analogous to the Fe²⁺/H₂O₂ and Fe³⁺/H₂O₂ system as shown in eqn (1) and (2).⁴² In the BAP system, a caged Co²⁺ complex including the substrate and bicarbonate has been reported as the active intermediate during the production of ˙OH.^43,44 In contrast to these studies, singlet oxygen, superoxide, ˙OH and carbonate radicals as key active species with a supported cobalt catalyst were also proposed during the degradation of methylene blue and chlorophenols using the BAP system.^7–9 From these studies, it is clear that various mechanisms have already been reported under different circumstances.

Cu²⁺ + H₂O₂ → Cu⁺ + HO₂˙ + H⁺ (k = 4.6 × 10² M⁻¹ s⁻¹) (1)

Cu⁺ + H₂O₂ → Cu²⁺ + HO˙ + HO⁻ (k = 1.0 × 10⁴ M⁻¹ s⁻¹) (2)

Numerous scavengers have been reported in the literature to capture free radicals during reaction, and the existence or absence of these radicals could be reflected from the difference in the observed reaction rates.^7,8 Terephthalic acid (TA) is a widely employed reagent for detecting ˙OH radicals using photoluminescence technology.⁴³ As shown in Fig. 11, the fluorescence intensity increased sharply with the catalyst + H₂O₂ + NaHCO₃ and TA, implying that in situ ˙OH radicals were generated. The fluorescence intensity slightly decreased on the addition of 4-CP, suggesting the consumption of some ˙OH radicals during the oxidation of 4-CP. With sodium azide (NaN₃) and isopropyl alcohol (IPA), popular scavengers for ¹O₂ and ˙OH,^7,45 more than 75% and 90% of the fluorescence intensity decayed, respectively. However, both IPA and NaN₃ together kept almost all of the fluorescence intensity, which suggests the generation of ˙OH radicals and ¹O₂.

Fig. 11 Fluorescence studies showing the generation of a free radical environment. Conditions: 100 ppm 4-CP, 25 mM H₂O₂, 30 mM NaHCO₃, amount of catalyst 0.5 g L⁻¹, 1 M IPA, 0.134 mM TA, 2 mM BQ, 20 mM NaN₃, temperature 30 °C, pH 8.66.

To further justify the existence of ˙OH or singlet oxygen, NaN₃ and IPA were directly added in the reaction solution as shown in Fig. S7 and S8.† Obviously, the degradation rate was inhibited with both IPA and NaN₃, indicating the generation of both ˙OH and ¹O₂ in the reaction medium.^7,45 Similarly, the addition of benzoquinone (BQ) also decreased the degradation rate suggesting the generation of ˙O₂⁻ (Fig. S9†).^7,8,46 To further verify these active species, mixtures of two scavengers were directly added in the reaction medium as demonstrated in Fig. 12. As is apparent with the mixture of BQ and NaN₃, more than 80% degradation was inhibited, while these scavengers alone in Fig. S7 and S8† could scavenge only 47% and 38%, respectively. Similarly, a mixture of BQ + IPA or IPA + NaN₃ lowered 70% and 60% of the degradation, respectively, while IPA alone accounted for 30% of the degradation (Fig. S8†). All this information in Fig. 12 clearly suggests the generation of ˙OH, ¹O₂ and ˙O₂⁻ as active oxygen species, which were involved in the degradation process. EPR is a useful technique for detecting the paramagnetic species in the reaction medium. Here we applied DMPO and BPN as EPR spin traps for both ˙OH and ˙O₂⁻ radicals, respectively. As shown in Fig. S10,† characteristic quartet signals of DMPO–OH with g = 2.0065 for ˙OH radicals and triplet signals of BPN–O₂⁻ with g = 2.007 for ˙O₂⁻ were observed.^8,47 In summary, the information collected from fluorescence, scavengers and EPR studies confirmed the generation of ˙OH, ¹O₂ and ˙O₂⁻ as ROS involved in degradation. Previously different mechanisms were proposed in the BAP system. Recently, the published work either relied on the generation of free radicals or an in situ [M²⁺⋯(HCO₃⁻)]⁺-type complex before activating H₂O₂ or PS.^7,12,43,44 The data in Fig. 6 and 8 provide some important information about the possible reaction mechanism. For example, the negligible degradation with leached Cu²⁺ and Co²⁺ ions and improved the activity of the CuCoO_x–LDH-2 catalyst in Fig. 6 supports the heterogeneous surface activation in the BAP system. Similarly the typical Fenton’s reaction suggested in eqn (3)–(6) may be ruled out due to poor degradation as shown in Fig. 8.

≡Co³⁺ + H₂O₂ → S–Co²⁺ + ⁻˙O₂ + 2H⁺ (3)

≡Co²⁺ + H₂O₂ → S–Co³⁺ + ˙HO + OH⁻ (4)

≡Cu²⁺ + H₂O₂ → S–Cu⁺ + ˙OOH + H⁺ (5)

≡Cu⁺ + H₂O₂ → S–Cu²⁺ + ˙HO + OH⁻ (6)

H₂O₂ + HCO₃⁻ ↔ HCO₄⁻ + H₂O (7)

Fig. 12 Influence of mixtures of scavengers (BQ, NaN₃ and IPA) on the degradation of 4-CP. Conditions: 200 ppm 4-CP, 25 mM H₂O₂, 30 mM NaHCO₃, amount of catalyst 0.5 g L⁻¹, temperature 30 °C, pH 8.62.

Again, the mechanism based on bicarbonate activation as shown in eqn (7) cannot be considered, as such a system in Fig. 8 degrades only 17% 4-CP. From this clear evidence we can conclude that the mechanism does not rely on direct interaction of H₂O₂ or bicarbonate with the catalyst. In contrast it seems that the outer sphere interaction of the catalyst and bicarbonate makes H₂O₂ more active like those in the PS system.¹² Furthermore, bicarbonate has a structure like PMS, and we suspect that the in situ transient species [Cu^IICo^II⋯(HCO₃⁻)]³⁺ may initially be produced on the surface of the catalyst which would activate H₂O₂ more effectively. The formation of this complex may accelerate the electron transfer between the catalyst and H₂O₂. As evidence, bicarbonate was reported to accelerate the rate of electron transfer from Mn²⁺ to APO-PSII during photoactivation.³⁸ The results in Fig. S6† also provide evidence where the catalyst efficiency increased with the increase of bicarbonate concentration. Similarly the results in Fig. S4† show the highest activity in the pH range 7.0–9.0, also highlighting the importance of bicarbonate in the reaction. Particularly, the introduction of Cu in the CuCoO_x–LDH-2 catalyst may provide an opportunity to link a greater number of bicarbonate anions producing a more stable intermediate complex. This proposed complex will then activate H₂O₂ just as reported for CuO and CuFe₂O₄ to activate PMS producing surface ˙OH and ˙O₂⁻ radicals (eqn (8)–(11)).^29,30 The higher redox potential of Co³⁺ than Cu²⁺ (eqn (21) and (22)) allows the catalyst to regenerate the active sites (≡Cu²⁺, ≡Co²⁺) (eqn (20)) as also confirmed in XPS and electrochemical studies.^48,49 These radicals (˙HO and ˙O₂⁻) produced through eqn (8)–(11) then undergo a series of reactions (eqn (12)–(19)) to produce various reactive species including singlet oxygen and more selective carbonate radicals.^7,50 All these radicals as well the regeneration of ≡Cu⁺, ≡Cu²⁺ and ≡Co²⁺ may participate in the degradation process.

[≡Co²⁺⋯(HCO₃⁻)]⁺ + H₂O₂ → [≡Co³⁺(HCO₃⁻)]²⁺ + ˙HO + OH⁻ (8)

[≡Co³⁺⋯(HCO₃⁻)]²⁺ + H₂O₂ → [≡Co²⁺(HCO₃⁻)]⁺ + ˙O₂⁻ + 2H⁺ (9)

[≡Cu²⁺⋯(HCO₃⁻)]⁺ + H₂O₂ → [≡Cu⁺(HCO₃⁻)] + ˙O₂⁻ + 2H⁺ (10)

[≡Cu⁺⋯(HCO₃⁻)] + H₂O₂ → [≡Cu²⁺(HCO₃⁻)]⁺ + ˙HO + OH⁻ (11)

˙HO + H₂O₂ → ˙HO₂ + H₂O (12)

˙OOH ↔ ˙O₂⁻ + H⁺ (13)

˙HO + HCO₃⁻ → ˙CO₃⁻ + H₂O (14)

˙CO₃⁻ + H₂O₂ → HCO₃⁻ + ˙HO₂ (15)

˙O₂⁻ + ˙OH → ¹O₂ + OH⁻ (16)

˙O₂⁻ + ˙HO₂ → ¹O₂ + ⁻HO₂ (17)

˙HO₂ + ˙HO₂ → ¹O₂ + H₂O₂ (18)

˙CO₃⁻ + HCO₄⁻ → HCO₃⁻ + ˙CO₄⁻ (19)

≡Co³⁺ + ≡Cu⁺ → ≡Co²⁺ + ≡Cu²⁺ (20)

≡Cu²⁺ + e⁻ → ≡Cu⁺, E_o = 0.17 V (21)

≡Co³⁺ + e⁻ → ≡Co²⁺, E_o = 1.18 V (22)

4. Conclusions

The hydrothermal impregnation of Cu in CuCoO_x–LDH-2 triggered synergistically active sites which not only affects the catalyst efficiency but also promotes catalyst stability. The origin of the synergy was deeply studied based on chemical reactivity, stability and development of electrochemical properties. Contrary to the catalyst precursors like CoO_x–LDH-1, CuO_x–LDH or their physical mixture, the CuCoO_x–LDH-2 catalyst was found to be chemically more reactive. For instance, 200 ppm 4-CP was completely removed in 40 minutes with CuCoO_x–LDH-2 in contrast to the negligible degradation for CoO_x–LDH-1, CuO_x–LDH or their physical mixtures even after 1 h. Similarly, the CuCoO_x–LDH-2 catalyst was involved in deeper oxidation (84% COD and 78% TOC removal) compared to that of CoO_x–LDH-1 (30 and 18%) and CuO_x–LDH (19 and 12%), respectively. The electrochemical studies indicate the promotion of interfacial electron transfer during impregnation of Cu in CuCoO_x–LDH-2 which helped the catalyst regeneration. Probably this regeneration property provides both a good catalyst activity and prolongs the stability as supported by the minimum catalyst leaching. For this reason, the CuCoO_x–LDH-2 catalyst can degrade a high dose of 4-CP (1000 ppm) in hours without an incremental increase in the initial minor catalyst leaching.

Acknowledgements

This research was funded by Chutian Scholar Foundation from Hubei province, China and the National Natural Science Foundation of China (No. 21273086 and 21573082). The authors also thank the Analytical and Testing Center of Huazhong University of Science and Technology for help in XRD and XPS analysis.

References

1. A. Chanda, S. K. Khetan, D. Banerjee, A. Ghosh and T. J. Collins, J. Am. Chem. Soc., 2006, 128, 12058.
2. L. Hu, F. Yang, W. Lu, Y. Hao and H. Yuan, Appl. Catal., B, 2013, 134, 7.
3. P. Shukla, H. Sun, S. Wang, H. M. Ang and M. O. Tadé, Catal. Today, 2011, 175, 380.
4. Q. Yang, H. Choi, Y. Chen and D. D. Dionysiou, Appl. Catal., B, 2008, 77, 300.
5. T. Zhang, H. Zhu and J. P. Croué, Environ. Sci. Technol., 2013, 47, 2784.
6. F. Ji, C. Li and L. Deng, Chem. Eng. J., 2011, 178, 239.
7. L. Zhou, W. Song, Z. Chen and G. Yin, Environ. Sci. Technol., 2013, 47, 3833.
8. A. Jawad, X. Lu, Z. Chen and G. Yin, J. Phys. Chem. A, 2014, 118, 10028.
9. A. Jawad, Y. Li, X. Lu, Z. Chen, W. Liu and G. Yin, J. Hazard. Mater., 2015, 289, 165.
10. J. Madhavan, P. Maruthamuthu, S. Murugesan and S. Anandan, Appl. Catal., B, 2008, 83, 8.
11. P. Nfodzo and H. Choi, Chem. Eng. J., 2011, 174, 629.
12. Y. Lei, C. S. Chen, Y. J. Tu, Y. H. Huang and H. Zhang, Environ. Sci. Technol., 2015, 49, 6838.
13. Z. Han, Y. Dong and S. Dong, J. Hazard. Mater., 2011, 189, 241.
14. J. Fan, X. Jiang, H. Min, D. Li, X. Ran, L. Zou, Y. Sun, W. Li, J. Yang and W. Teng, J. Mater. Chem. A, 2014, 2, 10654.
15. H. Wu, G. Pantaleo, V. La Parola, A. M. Venezia, X. Collard, C. Aprile and L. F. Liotta, Appl. Catal., B, 2014, 156, 350.
16. J. Xiong, Z. Li, J. Chen, S. Zhang, L. Wang and S. Dou, ACS Appl. Mater. Interfaces, 2014, 6, 15716.
17. N. Al-Yassir and R. Le Van Mao, Appl. Catal., A, 2007, 317, 275.
18. A. Caballero, J. P. Holgado, V. M. Gonzalezdela, S. E. Habas, T. Herranz and M. Salmeron, Chem. Commun., 2010, 46, 1097.
19. L. Zhang, F. Li, D. G. Evans and X. Duan, Mater. Chem. Phys., 2004, 87, 402.
20. W. Shi, M. Wei, L. Jin and C. Li, J. Mol. Catal. B: Enzym., 2007, 47, 58.
21. R. Espinal, E. Taboada, E. Molins, R. J. Chimentao, F. Medina and J. L. lorca, RSC Adv., 2012, 2, 2946.
22. A. F. Lucredio and E. M. Assaf, J. Power Sources, 2006, 159, 667.
23. D. Hammoud, C. Gennequin, A. Aboukaïs and E. A. Aad, Int. J. Hydrogen Energy, 2015, 40, 1283.
24. A. Amri, X. Duan, C. Y. Yin, Z. T. Jiang, M. M. Rahman and T. Pryor, Appl. Surf. Sci., 2013, 275, 127–135.
25. G. Fierro, M. L. Jacono, M. Inversi, R. Dragone and P. Porta, Top. Catal., 2000, 10, 39.
26. T. Ghodselahi, M. A. Vesaghi, A. Shafiekhani, A. Baghizadeh and M. Lameii, Appl. Surf. Sci., 2008, 255, 2730–2734.
27. Z. Ai, L. Zhang, S. Lee and W. Ho, J. Phys. Chem. C, 2009, 113, 20896.
28. M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. Lau, A. R. Gerson and R. S. C. Smart, Appl. Surf. Sci., 2011, 257, 2717.
29. Y. Ding, L. Zhu, N. Wang and H. Tang, Appl. Catal., B, 2013, 129, 153.
30. Y. Yao, Z. Yang, D. Zhang, W. Peng, H. Sun and S. Wang, Ind. Eng. Chem. Res., 2012, 51, 6044.
31. F. Xiao, Y. Li, X. Zan, K. Liao, R. Xu and H. Duan, Adv. Funct. Mater., 2012, 22, 2487.
32. B. Balagam and D. E. Richardson, Inorg. Chem., 2008, 47, 1173.
33. J. M. Lin and M. Liu, J. Phys. Chem. B, 2008, 112, 7850.
34. M. Liu, L. Zhao and J. M. Lin, J. Phys. Chem. A, 2006, 110, 7509.
35. C. A. S. Regino and D. E. Richardson, Inorg. Chim. Acta, 2007, 360, 3971.
36. D. E. Richardson, H. Yao, K. M. Frank and D. A. Bennett, J. Am. Chem. Soc., 2000, 122, 1729.
37. H. Yao and D. E. Richardson, J. Am. Chem. Soc., 2003, 125, 6211.
38. J. Dasgupta, A. M. Tyryshkin, Y. N. Kozlov, V. V. Klimov and G. C. Dismukes, J. Phys. Chem. B, 2006, 110, 5099.
39. L. Cheng, M. Wei, L. Huang, F. Pan, D. Xia, X. Li and A. Xu, Ind. Eng. Chem. Res., 2014, 53, 3478.
40. I. A. Salem and M. S. El-Maazawi, Chemosphere, 2000, 41, 1173.
41. A. H. Gemeay, I. A. Mansour, R. G. El-Sharkawy and A. B. Zaki, J. Mol. Catal. A: Chem., 2003, 193, 109.
42. A. D. Bokare and W. Choi, J. Hazard. Mater., 2014, 275, 121.
43. A. Xu, X. Li, S. Ye, G. Yin and Q. Zeng, Appl. Catal., B, 2011, 102, 37.
44. L. Duan, Y. Chen, K. Zhang, H. Luo, J. Huang and A. Xu, RSC Adv., 2015, 5, 84303.
45. D. Xia, Z. Shen, G. Huang, W. Wang, J. C. Yu and P. K. Wong, Environ. Sci. Technol., 2015, 49, 6264.
46. M. Yin, Z. Li, J. Kou and Z. Zou, Environ. Sci. Technol., 2009, 43, 8361.
47. Z. Wang, W. Ma, C. Chen, H. Ji and J. Zhao, Chem. Eng. J., 2011, 170, 353–362.
48. Y. Wang, H. Zhao and G. Zhao, Appl. Catal., B, 2015, 164, 396.
49. C. Cai, H. Zhang, X. Zhong and L. Hou, J. Hazard. Mater., 2015, 283, 70.
50. C. Wu and K. G. Linden, Water Res., 2010, 44, 3585.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra10402a

---