Cobalt modified mesoporous graphitic carbon nitride with enhanced visible-light photocatalytic activity

Abstract

Mesoporous graphitic carbon nitride (mpg-C₃N₄) was synthesized using silica nanoparticles as the hard template, and cobalt modified mpg-C₃N₄ (Co₃O₄/mpg-C₃N₄) was prepared via a facile impregnation method. UV-vis spectra showed that the mpg-C₃N₄ exhibited obvious absorption in the visible light range. The XPS results proved that the cobalt species deposited on mpg-C₃N₄ samples was Co₃O₄. The photocatalytic activity of Co₃O₄/mpg-C₃N₄ was evaluated in the photocatalytic degradation of bisphenol A (BPA) in aqueous solutions under visible light irradiation (λ > 420 nm). The photocatalytic degradation efficiency of BPA by Co₃O₄/mpg-C₃N₄ was significantly higher than that of mpg-C₃N₄ and the optimum content of Co₃O₄ loaded in the mpg-C₃N₄ was 1.5 wt%. Radical-trapping experiments certified that both hydroxyl radicals and superoxide radicals were the active species for the photocatalytic degradation of BPA. The total organic carbon (TOC) removal showed that BPA was mineralized by 1.5% Co₃O₄/mpg-C₃N₄ under visible light irradiation. The recycling tests indicated that 1.5% Co₃O₄/mpg-C₃N₄ was quite stable and there was no obvious decrease in photocatalytic activity after 5 cycles.

---

1. Introduction

Clean and sustainable solar energy has been extensively explored in order to overcome the serious energy and environmental challenges.^1–3 Some photocatalysts such as TiO₂ exhibit high photocatalytic activity under UV due to their wide band gap (3.2 eV).⁴ However, commercially available TiO₂ showed inefficient photocatalytic activity under visible light irradiation, which accounts for 43% of the solar spectrum. To make better use of the solar energy, visible light responsive photocatalysts have attracted more and more attention among researchers.

As a novel π-conjugated semiconductor (band gap = 2.7 eV), graphitic carbon nitride (g-C₃N₄) has a high thermal (stable in O₂ up to 550 °C) and chemical stability (insoluble in solution with pH from 0 to 14). Moreover, g-C₃N₄ is an excellent and sustainable metal-free photocatalyst. Wang et al.^5–7 first reported that g-C₃N₄ showed the ability to generate hydrogen from water under visible light irradiation. However, the application of g-C₃N₄ is limited by the low adsorption of solar light and high recombination of photogenerated electron–hole pairs. A lot of attention has been focused on improving its photocatalytic activity. It was reported that g-C₃N₄ doped with nonmetallic elements shows a prominent increase in photocatalytic activity. The efficiency of hydrogen generation from water by sulfur doped carbon nitride (g-CNS) was 12 times higher than that of g-C₃N₄.⁸ Introducing boron into the structural framework of g-C₃N₄ narrowed the band gap of g-C₃N₄ to absorb more visible light.⁹ The electric conductivity of P doped g-C₃N₄ is 4 orders of magnitude greater than g-C₃N₄.¹⁰ F-doped g-C₃N₄ presented superior performances in the oxidization of benzene to phenol under visible light irradiation.¹¹ Meanwhile, loading with noble metals (such as Ag,¹² Au,¹³ Pd,¹⁴ Pt¹⁵) can retard the recombination of electron–hole pairs to enhance the photocatalytic activity of the catalysts. Besides, integrating g-C₃N₄ with other semiconductors could improve the light response. The photocatalytic degradation efficiency of 4-aminobenzoic acid by g-C₃N₄/CdS was 41.6 and 2.7 times higher than those of g-C₃N₄ and CdS.¹⁶ Because of the enhancement of electron–hole separation on the interface of the semiconductors, the degradation efficiency of rhodamine B by DyVO₄/g-C₃N₄ was increased.¹⁷ In addition, NiS/g-C₃N₄,¹⁸ TiO₂/g-C₃N₄,¹⁹ BiOI/g-C₃N₄,²⁰ BiOBr/g-C₃N₄,^21,22 ZnO/g-C₃N₄,²³ Bi₂WO₆/g-C₃N₄²⁴ and ZnWO₄/g-C₃N₄²⁵ composites also showed higher photocatalytic activities than individual g-C₃N₄. As all known, the large surface area of catalyst can afford more active sites for photocatalysis. On the basis of this, the mesoporous graphitic carbon nitride (mpg-C₃N₄) which was synthesized using SiO₂ nanoparticles as hard template performed higher photocatalytic activity than g-C₃N₄.²⁶

As a typical p-type nano-structured semiconductor, Co₃O₄ has attracted considerable attention in recent years due to its potential applications in batteries,²⁷ supercapacitors,²⁸ electrocatalysis,²⁹ and sensing devices.³⁰ Co₃O₄-based catalyst was synthesized for photocatalysis. Phenol was degraded efficiently over Co₃O₄/BiVO₄ composite under visible light irradiation.³¹ Binitha et al.³² prepared Co₃O₄ nanoflakes for photodegradation of dyes. Xiao et al.³³ researched photodegradation of methylene blue by Co₃O₄/BiWO₆ under visible light and found that the composite with 0.2 wt% Co₃O₄ doping performed the highest photocatalytic activity. Pan et al.³⁴ prepared CuO/Co₃O₄ and investigated the mechanism of photocatalytic degradation of methyl orange on CuO/Co₃O₄ composite. Wang et al.⁶ synthesized Co₃O₄ deposited g-C₃N₄ and investigated the photocatalytic oxidation of water by Co₃O₄/g-C₃N₄. But the photocatalytic oxidation by Co₃O₄ loaded mpg-C₃N₄ has not been researched yet.

In this study, bulk g-C₃N₄, mpg-C₃N₄ and Co₃O₄/mpg-C₃N₄ with different loading of Co₃O₄ were prepared and their photocatalytic activities of BPA under visible light irradiation were investigated. The effect of Co₃O₄ in the photocatalytic reaction was researched and the photocatalytic mechanism was discussed.

2. Experimental

2.1. Preparation of bulk g-C₃N₄ and mpg-C₃N₄

All chemicals were of analytical grade and used without further purification. The bulk g-C₃N₄ was prepared by heating directly in ceramic crucible. Briefly, 5 g cyanamide was calcined for 4 h at 550 °C after heated at a rate of 2.3 °C min⁻¹ to 550 °C. The mpg-C₃N₄ was prepared as reported with a few diversifications.³⁵ 10 g cyanamide solution (50 wt%) was dissolved in 20 g solution with 40 wt% of 12 nm SiO₂ particles (Ludox HS40, Aldrich) used as hard template. The mixture was stirred for 2 h at 80 °C. The resulting powder was calcined at 550 °C with a rate of 2.3 °C min⁻¹ for 4 h in the air. Then yellow product was stirred in 4 mol L⁻¹ NH₄HF for 24 h to remove the silica template. At last, the product was centrifuged and washed with distilled water and ethanol, and dried at 70 °C overnight.

2.2. Preparation of Co₃O₄ loaded mesoporous graphitic carbon nitride (Co₃O₄/mpg-C₃N₄)

The Co₃O₄/mpg-C₃N₄ catalysts with different Co₃O₄ loaded amounts were prepared by the conventional impregnation method. 1 g mpg-C₃N₄ was impregnated in Co(NO₃)₂ solution with different concentrations and evaporated under stirring at 90 °C. The products were dried overnight and calcined at 300 °C for 2 h in the air. The products were donated as x% Co₃O₄/mpg-C₃N₄ where x standed for the mass of Co₃O₄/mpg-C₃N₄ mass percent.

2.3. Characterization of catalysts

X-ray diffraction (XRD) patterns were obtained in a Rigaku D/max-RA powder diffraction-meter using Cu Kα radiation. FT-IR spectra were obtained by a Nexus 870 FT-IR (Nicolet, USA) with KBr pellets. The transmission electron microscopy (TEM) images were obtained by JEOL 2100. The BET surface area analysis of the catalysts was accomplished at Micromeritics ASAP 2020. X-ray photoelectron spectroscopy (XPS) was recorded by a PHI 5000 VersaProbe. Photoluminescence spectra were carried out on a jobinYvon SPEX Fluorolog-3-P spectroscope. UV-vis diffuse reflectance spectroscopy (UV-vis-DRS) was performed with a Hitachi U-3010 UV-vis spectrometer. Photocurrent was obtained using a CHI 660B electrochemical workstation in a standard three-electrode system which used the products as the working electrodes. A 500 W xenon lamp worked as source of visible light irradiation.

2.4. Photocatalysis experiments

The photodegradation of BPA by catalysts was performed in a XPA-7 photochemical reactor (Xujiang Electromechanical plant Nanjing, China) and irradiated by a 500 W xenon lamp. Briefly, 0.08 g catalyst was dispersed in 200 mL of 15 mg L⁻¹ BPA solution in the quart tube reactor which was cooled by a thermostat. In order to reach adsorption–desorption equilibrium, the mixture was reacted in the dark for 60 min before photocatalytic reaction. 2 mL suspension was gained at certain time intervals. After BPA samples were centrifuged and filtered, the concentration of BPA was measured by High Performance Liquid Chromatography (Ultimate 3000, Dionex) equipped with a Dionex Accliam@120C18 column (4.6 nm × 250 nm) and emission wavelength of fluorescence detector was at 230 nm. The mobile-phase was 85% methanol and 15% water with a flow rate of 1 mL min⁻¹ and the column temperature was set at 30 °C.

3. Results and discussion

3.1. Catalyst characterization

Fig. 1a showed the XRD patterns of bulk g-C₃N₄ and Co₃O₄/mpg-C₃N₄ with different contents of Co₃O₄. For bulk g-C₃N₄, the strong peak at 27.4° indexed as (002) was a characteristic interplanar stacking peak of the conjugated aromatic systems with the interlayer distance of 0.326 nm. In addition, a pronounced peak at 13.1° corresponded to an in-plane structural packing motif, which was indexed as (100). The distance was determined to be 0.675 nm.^26,33 For mpg-C₃N₄ and Co₃O₄/mpg-C₃N₄, the overall intensities of (002) and (100) peaks weakened and the intra layer periodicity corresponding to 13.1° was broadened, which reflected the effect of geometric confinement. The previous studies reported that metal ion species restrained the polymeric condensation. The small amounts of the Co₃O₄ and good dispersion in the surface of Co₃O₄/mpg-C₃N₄ catalysts made no diffraction peaks of Co₃O₄ species observed in 1% Co₃O₄/mpg-C₃N₄, 1.5% Co₃O₄/mpg-C₃N₄, 2% Co₃O₄/mpg-C₃N₄ and 3% Co₃O₄/mpg-C₃N₄. For 5% Co₃O₄/mpg-C₃N₄, the diffraction peaks at 30.8° and 44.1° were distributed to the (220) and (400) planes of Co₃O₄.^36,37

Fig. 1 (a) XRD patterns of g-C₃N₄, mpg-C₃N₄ and Co₃O₄/mpg-C₃N₄ with different contents of Co₃O₄ and (b) FT-IR spectra of 1.5% Co₃O₄/mpg-C₃N₄.

Fig. 1b presented FT-IR spectra of 1.5% Co₃O₄/mpg-C₃N₄ between 400 and 4000 cm⁻¹. The features of the condensed C–N heterocycles were also found on the spectra of Co₃O₄/mpg-C₃N₄. The typical triazine ring mode was at 808 cm⁻¹ and its stretching mode was in the 1200–1650 cm⁻¹ regions. The peaks observed at 1575 and 1634 cm⁻¹ represented C=N stretching, while the three bands at 1245, 1321 and 1411 cm⁻¹ were typical for the stretching vibrations of aromatic C–N. A broad adsorption band at 3157 cm⁻¹ could be attributed to the stretching modes of terminal NH₂ or NH groups at the defect sites of the aromatic ring. Notably, an absorbance at 570 cm⁻¹ was accompanied with the cobalt incorporation on the spectra of Co₃O₄/mpg-C₃N₄, which was ascribed to the combination bands of Co–O fundamental vibrational modes. It was revealed that the original graphitic C–N network did not alter after the inclusion of Co, which was in good agreement with the results of XRD analyses.

The TEM images of g-C₃N₄, mpg-C₃N₄ and 1.5% Co₃O₄/mpg-C₃N₄ were observed in Fig. 2. As shown in Fig. 2a, the bulk g-C₃N₄ exerted interlayer-like structure. And in Fig. 2b, the holes of mpg-C₃N₄ copied the hard template of Ludox HS40 entirely. The dark spherical spots corresponding to well-distributed Co₃O₄ nanoparticles in the surface of mpg-C₃N₄ were observed in Fig. 2c. Moreover, the import of the Co₃O₄ nanoparticles had little influence on the morphology of the catalysts.

Fig. 2 TEM images of (a) g-C₃N₄, (b) mpg-C₃N₄ and (c), (d) 1.5% Co₃O₄/mpg-C₃N₄.

N₂ adsorption–desorption isotherms were observed in Fig. 3a. For mpg-C₃N₄ and Co₃O₄/mpg-C₃N₄, the physioadsorption isotherms could be classified as type IV in the IUPAC classification with a distinct hysteresis loop observed in the range of 0.5–1.0 P/P₀, which was characteristic of porous mater.³⁸ The BET specific surface of mpg-C₃N₄ was about 160.01 m² g⁻¹, which was slightly larger than that of Co₃O₄/mpg-C₃N₄ (149.62 m² g⁻¹). The pore size distribution was shown in Fig. 3b. The pores in catalysts were mainly distributed in the range 2–20 nm. The average pore size was about 12 nm attributed to the hard template of Ludox HS40. The introduction of Co₃O₄ slightly changed the BET specific surface and the mesoporous structure.

Fig. 3 (a) N₂ adsorption–desorption isotherms and (b) pore size distributions of mpg-C₃N₄ and 1.5 %Co₃O₄/mpg-C₃N₄.

The XPS indicated the core level chemical shifts and characterize the bonding nature of 1.5% Co₃O₄/mpg-C₃N₄. As shown in Fig. 4b, the peak at 284.6 eV corresponded to C–C coordination. The peak at 288.1 eV corresponding to the sp²-hybridized carbon in N=C–N₂ was confirmed as originating from carbon atoms that have one double and two single bonds with three neighboring N atoms. The peak at a binding energy of 293.9 eV might be ascribed to π-excitation.^39,40 The XPS of N1s separating into four binding energies was observed in Fig. 4c. The primary peak at 398.6 eV was attributed to electrons originated from sp₂-hybridized N atoms in N=C–N₂.⁵ The peak at 399.8 eV was ascribed to the tertiary nitrogen N–(C)₃ groups. The additional peak at 400.7 eV might be attributed to the amino functions carrying hydrogen (C–N–H), which corresponded to structural defects and incomplete condensation. The weak peak at 403.98 eV might be attributed to π-excitation.^39,40 The XPS of Co demonstrated a doublet corresponding to Co 2p_3/2 and Co 2p_1/2. Co 2p_3/2 peak consisted of two peaks at 780.9 eV and 785.5 eV, which were ascribed to Co³⁺ and Co²⁺. Co 2p_1/2 similarly consisted of peaks at 797.1 eV and 802.9 eV, which were attributed to Co³⁺ and Co²⁺.^6,31,37 The relative contents of Co were calculated by the areas of the fitted Gaussian components in Fig. 4d with 70.24% Co³⁺ and 29.76% Co²⁺. The result confirmed that Co in the form of the mixture (Co₃O₄) of Co³⁺ and Co²⁺ was loaded on the surface of mpg-C₃N₄.

Fig. 4 XPS spectra of 1.5% Co₃O₄/mpg-C₃N₄: (a) full scan, (b) C 1s, (c) N 1s and (d) Co 2p.

3.2. Photodegradation of BPA

It was shown in Fig. 5a that the photodegradation of BPA by g-C₃N₄, mpg-C₃N₄ and 1.5% Co₃O₄/mpg-C₃N₄. The results showed that BPA was hardly decomposed without catalysts under the visible light irradiation. The g-C₃N₄ represented very low photocatalytic activity and only 19.6% of BPA was degraded after irradiation for 180 min. Compared with g-C₃N₄, the photocatalytic efficiency of mpg-C₃N₄ was much higher, and 61.1% of BPA was degraded. Since the BET surface area of mpg-C₃N₄ was obviously larger than that of g-C₃N₄, mpg-C₃N₄ could furnish more catalytic active sites for reactions. The defects in the morphology of mpg-C₃N₄ improved the catalytic activity by the relocalization of electron in the graphitic layer. Furthermore, the appropriate pores were conducive to adsorption of the light waves deep inside the catalyst and the mobility of charges increased.^41,42 When 1.5% Co₃O₄ was presented on the surface of mpg-C₃N₄, the catalyst showed the highest photolytic efficiency that 93.6% of BPA was degraded under the visible light irradiation for 180 min. As a typical p-type semiconductor, Co₃O₄ could make great contributions to separate the electron–hole pairs. Thereby, the photocatalytic activity of mpg-C₃N₄ was enhanced by loading Co₃O₄.

Fig. 5 (a) Photocatalytic degradation of BPA under visible light irradiation in the presence of g-C₃N₄, mpg-C₃N₄, and 1.5% Co₃O₄/mpg-C₃N₄; (b) The pseudo-first-order rate constant (k_obs) for the photocatalytic degradation of BPA by Co₃O₄/mpg-C₃N₄ with different contents of Co₃O₄; (c) Influence of the initial pH on BPA degradation.

The first-order kinetic model was used to describe the kinetic of the photodegradation of BPA:

C_t = C₀ e^−k_obst

where C_t is the concentration of BPA solution at reaction time t (min); C₀ is the initial concentration of BPA solution (mg L⁻¹); k_obs is the rate constant (min⁻¹). The experimental data fitted the first-order reaction kinetic model very well (R² > 0.99). The k_obs of the reactions with Co₃O₄/mpg-C₃N₄ containing different contents of Co₃O₄ were observed in Fig. 5b. The k_obs of 1.5% Co₃O₄/mpg-C₃N₄ (0.014 min⁻¹) was higher than that of 1% Co₃O₄/mpg-C₃N₄ (0.010 min⁻¹). It might be caused by the higher interfacial electron transfer efficiency which enhanced the photocatalytic activity with more loaded Co₃O₄. However, with the content of Co₃O₄ increasing to 5%, the k_obs decreased to 0.007 min⁻¹. Since the over abundance of metal behaved as recombination centers of electrons and holes which inhibited the separation of electrons–holes pairs. On the other hand, excess Co₃O₄ formed overlapping agglomerates and covered the active sites on the surface of catalyst. Therefore, 1.5% Co₃O₄/mpg-C₃N₄ exhibited the most outstanding activity.

It was reported that the initial pH of the solution affect the surface charge of the semiconductor catalysts and the ionization of the organic pollutants in aqueous solution. Fig. 5c revealed the photocatalytic efficiency of BPA in solution with different initial pH. As the initial pH increased from 3.01 to 12.89, the photocatalytic degradation efficiency decreased from 98.19% to 28.98%. Wang et al.⁴³ had reported that the nitrogen functionalities on the surface might act as strong Lewis base sites, while the π-bonded planar layered configurations are expected to anchor phenol via special O–H⋯N or O–H⋯π interactions. Furthermore, O–H⋯N or O–H⋯π enhanced the adsorption of BPA on the surface of the catalyst. So the catalyst showed high BPA degradation efficiency at pH ranges from 3.01 to 10.05. The pK_a (acid dissociation constant) of BPA was reported as 9.6 to 11.3. When the value of pH was higher than pK_a of BPA, bisphenolate anion (–O–C₁₅H₁₄–O–) might be formed by (H–O–C₁₅H₁₄–O–H) due to deprotonation.^14,44,45 The interaction between BPA and the catalyst decreased which lead to less photodegradation efficiency at high pH > 11.00.

3.3. Photocatalytic mechanism

The band structure of the photocatalyst is responsible for the efficient generation and separation process of the electron–hole pairs. The theoretical calculated values of the conduction band (CB) and valence band (VB) potentials of the mpg-C₃N₄ material were −1.13 and 1.57 eV, respectively.^31,35 In the case of Co₃O₄, the values of the CB and VB potentials were 0.50 and 1.53 eV, respectively.²² Based on the experimental and theoretical results, the reaction mechanism diagram of the Co₃O₄/mpg-C₃N₄ is presented in Fig. 6. With the induction of light, both Co₃O₄ and mpg-C₃N₄ can engender photogenerated electron–hole pairs. Because the VB of mpg-C₃N₄ is more positive than that of Co₃O₄, the holes mpg-C₃N₄ would migrate to the surface of Co₃O₄. Therefore, the electron–hole pairs can be efficiently separated on the composite interface and the charge recombination could be easily suppressed. As a result, the Co₃O₄/mpg-C₃N₄ catalyst possessed an enhanced photocatalytic activity.

Fig. 6 The schematic diagram of photocatalytic degradation of BPA by 1.5% Co₃O₄/mpg-C₃N₄ catalyst under visible light irradiation.

The photoluminescence (PL) spectra of the photocatalysts indicated the migration, transfer, and recombination processes of the photogenerated electron–hole pairs in the semiconductors.⁴⁶ Fig. 7a presented the PL spectra of g-C₃N₄, mpg-C₃N₄, and 1.5% Co₃O₄/mpg-C₃N₄ at the excitation wavelength of 400 nm. It could be seen that the emissions of g-C₃N₄, mpg-C₃N₄, and 1.5% Co₃O₄/mpg-C₃N₄ are centered at 460 nm, corresponding to the recombination process of self-trapped excitations. Compare to g-C₃N₄, the emission intensity of the mpg-C₃N₄ decreased remarkably, reflecting of the improved separation efficiency of electron–hole pairs. The emission intensity of the 1.5% Co₃O₄/mpg-C₃N₄ decreased further, suggesting that the presence of Co₃O₄ suppressed the recombination of the electron–hole pairs and consequently performed much higher photocatalytic activity.

Fig. 7 (a) Photoluminescence spectra, (b) Photocurrent measurements, (c) UV-vis diffuse reflectance spectra of g-C₃N₄, mpg-C₃N₄ and 1.5% Co₃O₄/mpg-C₃N₄.

The photocurrent measurements were carried out for mpg-C₃N₄ and Co₃O₄/mpg-C₃N₄ to investigate the synergistic effect between Co₃O₄ and mpg-C₃N₄. As observed in Fig. 7b, the photocurrent intensity of Co₃O₄/mpg-C₃N₄ (0.15 uA) was much higher than that generated by mpg-C₃N₄ (0.04 uA), demonstrating that the synergistic effect could improve markedly the separation efficiency of photoinduced electrons and holes.

The UV-vis diffuse reflectance spectra represented in Fig. 7c. The bulk g-C₃N₄ showed absorbance in a wide range of wavelength from UV to visible light and the wavelength of the absorption edge is 460 nm. Compared with the bulk g-C₃N₄, the absorption of light in mpg-C₃N₄ was enhanced. The pores were made for the deep adsorption of light waves on the catalyst as described previously. However, the light absorption of Co₃O₄/mpg-C₃N₄ increased markedly, which showed multiple bands at 550 nm and 760 nm that originated from O²⁻ → Co²⁺ and O²⁻ → Co³⁺ transitions.⁴⁷ Co₃O₄ enhanced the light absorption capacity of mpg-C₃N₄ in the range of near infrared and visible light, which made contributions to produce more electron–hole pairs. Therefore, the photocatalytic activity was enhanced by introducing Co₃O₄ nanoparticles on the surface of mpg-C₃N₄.

It was important to verify the main active oxidant in the photocatalytic reaction process. The active oxidants generated in the photocatalytic reaction were measured through trapped by t-BuOH (hydroxyl radicals scavenger), potassium iodide (photogenerated holes scavenger) and benzoquinone (superoxide radicals scavenger).¹⁴ As shown in Fig. 8a, the photodegardation of BPA by the catalyst was restrained with the addition of t-BuOH. With presence of 1 mmol L⁻¹ and 5 mmol L⁻¹ t-BuOH, 89.5% and 80.4% of BPA were degraded. With the concentration of t-BuOH further increased to 10 mmol L⁻¹, the photocatalytic efficiency decreased slightly. It indicated that hydroxyl radicals played an important role in the photocatalytic degradation of BPA. Since the photodegradation of BPA was restrained with the addition of t-BuOH. However, the photocatalytic reaction of BPA was promoted with the presence of KI (Fig. 8b). It might be attributed to that scavenging h⁺ improved the electron–hole separation and thus more free electrons generated more hydroxyl radicals.⁴⁸ Therefore, KI in the solution improved the photocatalytic efficiency. The photocatalysis was inhibited obviously with the degradation efficiency reduced to 34.9% as the concentration of benzoquinone was 1 mmol L⁻¹. With the concentration of benzoquinone up to 5 mmol L⁻¹, the suppression was not further intensified. It was reported that the photogenerated electrons in the CB of graphitic carbon nitride could react with O² to generate ˙O₂⁻. Further, the superoxide radicals were momentous for the ring cleavage of aromatic compounds. Besides, the single-electron oxidation of phenoxyl radicals generated by phenols and the presence of ˙O₂, phenoxyl radicals promoted the cleavage reactions of aromatic rings.^48–50 Hence, ˙O₂⁻ acted as major active radicals in the photocatalytic reaction. In conclusion, both hydroxyl radicals and superoxide radicals played important role in the photocatalytic reaction.

Fig. 8 Photocatalytic degradation of BPA by 1.5% Co₃O₄/mpg-C₃N₄ with the addition of (a) t-BuOH, (b) KI and (c) benzoquinone.

The TOC removal ratio was used to evaluate the mineralization efficiency of BPA by 1.5% Co₃O₄/mpg-C₃N₄ under visible light irradiation. As observed in Fig. 9, the TOC decreased gradually and the efficiency of TOC removal was 80.4%. It was noted that the efficiency of TOC removal was lower than that of photodegradation due to the incomplete mineralization of the BPA and generation of low molecular weight organic.

Fig. 9 Change in TOC abatements during photocatalytic degradation of BPA by 1.5% Co₃O₄/mpg-C₃N₄.

3.4. Stability of photocatalyst

Besides the activity of the photocatalyst, the stability is another significant effect on the assessment of the photocatalyst. Once the photocatalytic reaction of a testing cycle was completed, the suspension was filtered. The separated catalysts were washed with deionized water and dried at 70 °C before next testing cycle. As shown in Fig. 10, the cycles of the 1.5% Co₃O₄/mpg-C₃N₄ were performed to evaluate the stability of the photocatalyst. After 5 times cycles, the catalytic efficiency had slight decrease. Therefore, the 1.5% Co₃O₄/mpg-C₃N₄ showed significant stability, facilitating the application of the photocatalyst.

Fig. 10 Cycling runs for the photocatalytic degradation of BPA in the presence of 1.5% Co₃O₄/mpg-C₃N₄.

4. Conclusions

Co₃O₄ loaded mpg-C₃N₄ catalyst was prepared by the conventional impregnation method and the effects of contents of Co₃O₄ loaded on the photocatalytic activity were investigated. The highest degradation efficiency of BPA was achieved 93.6% by 1.5% Co₃O₄/mpg-C₃N₄. The catalyst showed high BPA degradation efficiency at pH ranges from 3.01 to 10.05. The benzoquinone and t-BuOH inhibited the photocatalytic reaction, while KI enhanced the reaction, indicating the hydroxyl radicals and the superoxide radicals were the major oxidative species in photocatalytic reaction. BPA was mineralized by Co₃O₄/mpg-C₃N₄ under the visible light irradiation. Besides the 1.5% Co₃O₄/mpg-C₃N₄ showed significant stability.

Acknowledgements

The financial supports from the Natural Science Foundation of China (no. 51178223 and 51208257), the Natural Science Foundation of Jiangsu Province (SBK201240759), China Postdoctoral Science Foundation (no. 2013M541677), the Jiangsu Planned Projects for Postdoctoral Research Funds and the project from Environmental Protection Department of Jiangsu Province of China (no. 2012008) are gratefully acknowledged.

References

1. D. Chatterjee and S. Dasgupta, J. Photochem. Photobiol., C, 2005, 6, 186.
2. A. Fujishima and K. Honda, Nature, 1972, 238, 37.
3. Y. Mu, H. Q. Yu, J. C. Zheng and S. J. Zhang, J. Photochem. Photobiol., A, 2004, 163, 311.
4. F. Jiang, S. Zheng, L. An and H. Chen, Appl. Surf. Sci., 2012, 258, 7188.
5. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2008, 8, 76.
6. J. Zhang, M. Grzelczak, Y. Hou, K. Maeda, K. Domen, X. Fu, M. Antonietti and X. Wang, Chem. Sci., 2012, 3, 443.
7. Y. Wang, X. Wang and M. Antonietti, Angew. Chem., Int. Ed., 2012, 51, 68.
8. J. Zhang, J. Sun, K. Maeda, K. Domen, P. Liu, M. Antonietti, X. Fu and X. Wang, Energy. Environ. Sci., 2011, 4, 675.
9. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2010, 26, 3894.
10. Y. Zhang and M. Antonietti, J. Am. Chem. Soc., 2010, 132, 6294.
11. Y. Wang, Y. Di, M. Antonietti, H. Li, X. Chen and X. Wang, Chem. Mater., 2010, 22, 5119.
12. Y. Meng, J. Shen, D. Chen and G. Xin, Rare Met., 2011, 30, 276.
13. N. Cheng, J. Tian, Q. Liu, C. Ge, A. H. Qusti, A. M. Asiri, A. O. Al-Youbi and X. Sun, ACS Appl. Mater. Interfaces, 2013, 5, 6815.
14. C. Chang, Y. Fu, M. Hu, C. Wang, G. Shan and L. Zhu, Appl. Catal., B, 2013, 142–143, 553.
15. K. Ghosh, M. Kumar, H. Wang, T. Maruyama and Y. Ando, J. Phys. Chem. C, 2010, 114, 5107.
16. J. Fu, B. Chang, Y. Tian, F. Xi and X. Dong, J. Mater. Chem. A, 2013, 1, 3083.
17. Y. He, J. Cai, T. Li, Y. Wu, Y. Yi, M. Luo and L. Zhao, Ind. Eng. Chem. Res., 2012, 51, 14729.
18. J. Hong, Y. Wang, Y. Wang, W. Zhang and R. Xu, ChemSusChem, 2013, 6, 2263.
19. X. Lu, Q. Wang and D. Cui, J. Mater. Sci. Technol., 2010, 26, 925.
20. C. Chang, L. Zhu, S. Wang, X. Chu and L. Yue, ACS Appl. Mater. Interfaces, 2014, 6, 5083.
21. J. Fu, Y. Tian, B. Chang, F. Xi and X. Dong, J. Mater. Chem., 2012, 22, 21159.
22. J. Di, J. Xia, S. Yin, H. Xu, M. He, H. Li, L. Xu and Y. Jiang, RSC Adv., 2013, 3, 19624.
23. J. X. Sun, Y. P. Yuan, L. G. Qiu, X. Jiang, A. J. Xie, Y. H. Shen and J. F. Zhu, Dalton Trans., 2012, 41, 6756.
24. L. Ge, C. Han and J. Liu, Appl. Catal., B, 2011, 108–109, 100.
25. L. Sun, X. Zhao, C. J. Jia, Y. Zhou, X. Cheng, P. Li, L. Liu and W. Fan, J. Mater. Chem., 2012, 22, 23428.
26. F. Goettmann, A. Fischer, M. Antonietti and A. Thomas, Angew. Chem., Int. Ed., 2006, 45, 4467.
27. S. W. Hwang, A. Umar, S. H. Kim, S. A. Al-Sayari, M. Abaker, A. Al-Hajry and A. M. Stephan, Electrochim. Acta, 2011, 56, 8534.
28. X. Dong, H. Xu, X. Wang, Y. Huang, M. B. Chan-Park, H. Zhang, L. Wang, W. Huang and P. Chen, ACS Nano, 2012, 6, 3206.
29. D. Zhang, Q. Xie, A. Chen, M. Wang, S. Li, X. Zhang, G. Han, A. Ying, J. Gong and Z. Tong, Solid State Ionics, 2010, 181, 1462.
30. D. Barreca, D. Bekermann, E. Comini, A. Devi, R. A. Fischer, A. Gasparotto, M. Gavagnin, C. Maccato, C. Sada, G. Sberveglieri and E. Tondello, Sens. Actuators, B, 2011, 160, 79.
31. M. Long, W. Cai, J. Cai, B. Zhou, X. Chai and Y. Wu, J. Phys. Chem. B, 2006, 110, 20211.
32. N. N. Binitha, P. V. Suraja, Z. Yaakob, M. R. Resmi and P. P. Silija, J. Sol-Gel Sci. Technol., 2009, 53, 466.
33. Q. Xiao, J. Zhang, C. Xiao and X. Tan, Catal. Commun., 2008, 9, 1247.
34. L. Pan and Z. Zhang, J. Mater. Sci.: Mater. Electron., 2010, 21, 1262.
35. A. Thomas, A. Fischer, F. Goettmann, M. Antonietti, J. O. Müller, R. Schlögl and J. M. Carlsson, J. Mater. Chem., 2008, 18, 4893.
36. Y. Q. Liang, Z. D. Cui, S. L. Zhu, Z. Y. Li, X. J. Yang, Y. J. Chen and J. M. Ma, Nanoscale, 2013, 5, 10916.
37. Z. Ding, X. Chen, M. Antonietti and X. Wang, ChemSusChem, 2010, 4, 274.
38. Y. Huang, Z. Ai, W. Ho, M. Chen and S. Lee, J. Phys. Chem. C, 2010, 114, 6342.
39. Y. Li, J. Zhang, Q. Wang, Y. Jin, D. Huang, Q. Cui and G. Zou, J. Phys. Chem. B, 2010, 114, 9429.
40. Q. Guo, Y. Xie, X. Wang, S. Zhang, T. Hou and S. Lv, Chem. Commun., 2004, 26.
41. X. Wang, J. Yu, C. Ho, Y. Hou and X. Fu, Langmuir, 2005, 21, 2552.
42. L. Zhang and J. C. Yu, Chem. Commun., 2003, 2078.
43. Y. Wang, J. Yao, H. Li, D. Su and M. Antonietti, J. Am. Chem. Soc., 2011, 133, 2362.
44. C. A. Staples, P. B. Dorn, G. M. Klecka, S. T. O'Block and L. R. Harris, Chemosphere, 1998, 36, 2149.
45. C. Wang, H. Zhang, F. Li and L. Zhu, Environ. Sci. Technol., 2010, 44, 6843.
46. J. Pal and P. Chauhan, Mater. Charact., 2010, 61, 575.
47. Y. Fu, H. Chen, X. Sun and X. Wang, Appl. Catal., B, 2012, 111–112, 280.
48. J. Ng, X. Wang and D. D. Sun, Appl. Catal., B, 2011, 110, 260.
49. P. Raja, A. Bozzi, H. Mansilla and J. Kiwi, J. Photochem. Photobiol., A, 2005, 169, 271.
50. X. Wang, W. Yang, F. Li, Y. Xue, R. Liu and Y. Hao, Ind. Eng. Chem. Res., 2013, 52, 17140.

---