In situ growth of Ag/Ag₂O nanoparticles on g-C₃N₄ by a natural carbon nanodot-assisted green method for synergistic photocatalytic activity

Abstract

Novel visible-light-driven Ag/Ag₂O@g-C₃N₄ (AAC) hybrid materials were synthesized successfully via a green, facile hydrothermal treatment approach by reducing AgNO₃ using carbon nanodots as reducing agent for the first time. Natural leaves were used as precursors to prepare carbon nanodots. By adjusting the feed mass ratio of AgNO₃ to carbon nanodots, the average diameter of Ag/Ag₂O can be modulated from 8 to 33 nm. The effects of Ag/Ag₂O deposition on the optical and photocatalytic performance of AAC in the degradation of rhodamine B (RhB) under visible light irradiation were systematically investigated. It was found that the Ag/Ag₂O content greatly influences the photocatalytic activity of AAC nanocomposites. A suitable amount of Ag/Ag₂O in the AAC system could obtain the best photocatalytic activity. The experimental results indicated that all the AAC nanocomposites showed higher photocatalytic activities compared with that of pure g-C₃N₄. The remarkable visible-light photocatalytic activity of AAC heterostructures could be attributed to their absorption in the visible region and low recombination rate of the electron–hole pairs because of the heterojunction formed between Ag₂O and g-C₃N₄. In addition, Ag nanoparticles are believed to play an essential role in affecting the photoreactivity because they are able to trap electrons, further separate the electron–hole pairs, and prolong the life of electron–hole pairs. In other words, it can be concluded that the increased photocatalytic activity of AAC was attributed to the synergic effect between g-C₃N₄, Ag₂O and Ag. It is believed that the present work could render useful information for steering the design and application of noble nanoparticle loaded g-C₃N₄-based heterojunction composites with high photocatalytic activity.

---

1. Introduction

The growing concerns about environmental and energy crises have attracted increasing attention for solar energy utilization. Heterogeneous photocatalysis has been recognized as a potential strategy for solar energy conversion and environmental remediation. Anatase TiO₂ is considered as one of the best photocatalysts, but the band gap energy of anatase TiO₂ (3.0–3.2 eV) requires UV light to be excited, and thus only a small portion of solar light is absorbed in the UV region.^1–4 Therefore, it becomes a challenge to find novel photocatalytic catalysts with visible-light response.

Graphitic carbon nitride (g-C₃N₄) polymer, a polymeric metal-free semiconductor with a band gap of about 2.70 eV, is a potentially viable candidate of the photocatalytic field, due to their peculiar thermal stability, low cost preparation and visible-light responsive property.⁵ Nevertheless, the fast recombination of photogenerated electron–hole pairs may still limit the enhancement of its photocatalytic activity.⁶ Transition metal modification and heterostructured composite fabrication are the two methods introduced for lowering the recombination rate.^6–8

Metal–semiconductor composite prepared by modifying metals, such as silver, gold and platinum onto a semiconductor can be significantly improve the photocatalytic activity.^9–13 Noble nanoparticles (NPs) can reduce the recombination rate of the photogenerated electron–hole pairs by transferring the electrons to themselves.¹⁴ Combined Ag and g-C₃N₄ can enhance the photocatalytic performance due to the substantial reduction in the rate of recombination of photogenerated charge carriers by the enhanced electron transport through Ag NPs and g-C₃N₄.¹⁵ Zhu¹⁶ and co-workers reported the enhanced photocatalytic activity for the photodegradation of methylene blue and hydrogen evolution reaction under visible light compared to the pure g-C₃N₄ sample. Chen et al.¹⁷ reported that modifying g-C₃N₄ with Ag significantly increases the photoelectric conversion performance. Ge et al.¹⁸ prepared Ag-modified g-C₃N₄ composite and found that the modification of Ag on the g-C₃N₄ surface significantly enhanced the visible-light photocatalytic capacity of g-C₃N₄.

In addition, coupling g-C₃N₄ with the other semiconductors to form heterostructures also provides a feasible route to inhibit the recombination of photogenerated electron–hole pairs, such as TiO₂@g-C₃N₄,^19,20 ZnO@g-C₃N₄,^21–23 TaON@g-C₃N₄,²⁴ SrTiO₃@g-C₃N₄ (ref. 25) and Fe₃O₄@g-C₃N₄.²⁶ Ag₂O is a fascinating p-type semiconductor with a narrow energy bandgap of ∼1.3 eV, which enables it to absorb visible light.²⁷ Under visible light, pure Ag₂O can absorb photons to generate electron–hole pairs. However, the fast recombination rate of charge carriers and its unstability limited its application. It has been found that Ag₂O can maintain its stability and enhance the photocatalytic activity when Ag₂O is coupled with other semiconductors.^28–35 There is one probable reason that the heterojunction structure is formed between Ag₂O and the other semiconductors (support materials) and thus the photogenerated electron–hole pairs can separate efficiently, leading to the enhanced photocatalytic performance under visible light irradiation. By comparing the energy levels of g-C₃N₄ with Ag₂O, it is fortunate to find that their well-matched overlapping band-structures are quite suitable to construct heterostructures that would bring an effective separation and transfer of photogenerated charges. Furthermore, Ag₂O NPs can be well dispersed on the surface of g-C₃N₄, which can provide more possible reaction sites for the catalytic reaction.

In this study, we have successfully developed a green, low-cost synthetic method to prepare Ag/Ag₂O NPs loaded g-C₃N₄ catalyst. In the preparation of the catalyst, Ag/Ag₂O was first synthesized using hydrothermal treatment approach by reducing AgNO₃ with carbon nanodots. A series of AAC-X heterostructures were prepared by mixing the as-prepared g-C₃N₄ with different amount of the precursor of Ag/Ag₂O. The synthesis route ensures not only the successful growth of Ag/Ag₂O NPs on g-C₃N₄ nanosheets but also the high dispersion of Ag/Ag₂O NPs on g-C₃N₄ without aggregation, and the size of Ag/Ag₂O can be controlled by adjusting the adding amount of AgNO₃. Photocatalytic activity of these Ag/Ag₂O-deposited g-C₃N₄ nanosheets was tested by photocatalytic degradation of RhB as a reference model representing pollutant. The formation of well-defined junctions among Ag, Ag₂O and g-C₃N₄ effectively facilitates charge transfer among Ag, Ag₂O and g-C₃N₄ and suppresses the recombination of photogenerated electrons and holes, resulting in extremely high activity and stability. In addition, possible photocatalytic mechanism of AAC-X nanosheets was investigated. The results present here could highlight the importance of designing ternary composite nanostructures using this facile, green, low-cost method for highly efficient photocatalysts.

2. Experimental

2.1. Materials

Absolute ethanol (C₂H₅OH), sodium borohydride (NaBH₄), silver nitrate (AgNO₃) and RhB are of analytical grade and obtained from Sinopharm Chemical Reagent Co., LTD. Terephthalic acid (TA) and polyvinylpyrrolidone (PVP) were bought from Aladdin Reagent Co., LTD. The leaves were obtained from Jiangsu University, China and washed by water for further use. All chemicals used in the experiments were used without further purification. Deionized (DI) water was used for all experiments.

2.2. Catalyst preparation

2.2.1. Preparation of carbon nanodots. In a typical process,³ leaves (2 g) were added into deionized water (10 mL). Next, the above mixture was transferred into a 25 mL Teflon-lined autoclave and heated at 180 °C for 3 h. The as-prepared samples were collected by removing impurity through filtration.

2.2.2. Preparation of g-C₃N₄ powder. Typically, the g-C₃N₄ was synthesized by thermal treatment of 40 g of urea in a crucible with a cover under ambient pressure in air. The precursor was heated to 600 °C in 1 h in a tube furnace and maintained at this temperature for 4 h. The resulted final light yellow powder was washed in sequence with water and anhydrous ethanol thoroughly and then collected by filtration and dried at room temperature.

2.2.3. Preparation of Ag/Ag₂O@g-C₃N₄ ternary hybrid materials. Typically, g-C₃N₄ (0.1 g) nanosheet was dispersed in absolute ethyl alcohol (20 mL) by ultrasonication for 20 min. After that, AgNO₃ (0.02 M) and PVP (1 g) were added to the above solution under vigorous stirring. The mixture was stirred for 30 min and impregnated for 12 h. At last, certain amount of carbon nanodots (1.5, 3, 6 and 9 mL) were added to the above mixture and stirred for 5 min. Then, the above mixture was transferred into a Teflon-lined stainless steel autoclave. The autoclave was kept at 180 °C for 3 h in an electric oven. After being cooled to room temperature, the samples were collected by centrifugation, washed with deionized water and absolute ethanol several times, and then dried in a vacuum for 48 h. The schematic illustration for the formation of the Ag/Ag₂O@g-C₃N₄ nanocomposite is shown in Scheme 1. The Ag/Ag₂O@g-C₃N₄ ternary hybrid materials were marked as AAC-X, X label as the added volume of AgNO₃ (X = 2.5, 5.0, 10.0 and 15.0 mL). Moreover, the molar ratio between Ag and Ag₂O is kept consistent (cal. 3.5 : 1).

Scheme 1 The schematic illustration for the formation of the AAC-X nanocomposites.

2.3. Characterization

The crystal structure and phase purity of the prepared samples were analyzed by X-ray diffraction (XRD) using D8 Advance X-ray diffraction (Bruker axs company, Germany) equipped with Cu-Kα radiation (λ = 1.5406 Å), employing a scanning rate of 0.02° s⁻¹ in the 2θ range from 20 to 80°. Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS) were performed on a scanning electron microscope (Hitachi S-4800 II, Japan) operated at an acceleration voltage of 10 kV to characterize the morphologies and the compositions of the prepared samples. Furthermore, the morphology and particle size of the products were also examined by transmission electron microscopy (TEM) which was recorded on a JEOL-JEM-2010 (JEOL, Japan) operating at 200 kV. UV-vis diffuse reflection spectroscopy (DRS) was performed on a Shimadzu UV2550 spectrophotometer using BaSO₄ as the reference. The N₂ physical absorption measurements were carried out at 77 K with a NOVA2000e analytical system made by Quantachrome Corporation (USA). The specific surface area was calculated by Brunauer–Emmett–Teller (BET) method. Pore size distribution and pore volume were calculated by Barrett–Joyner–Halenda (BJH) method. X-ray photo-electron spectroscopy (XPS) analysis was measured on an ESCALAB MK X-ray photoelectron spectrometer. The PL spectra of the samples were obtained with a QuantaMaster & TimeMaster Spectrofluorometer.

2.4. Photocatalytic activity for degradation of pollutant

The photocatalytic activities were evaluated by the decomposition of RhB under visible light irradiation. Typically, 80 mg of photocatalyst was dispersed in 80 mL of 10 mg L⁻¹ RhB aqueous solution in a reactor of double layer condensated by running water to keep the temperature (25 °C) unchanged. The pH value of the RhB solution was determined 6.2. Prior to irradiation, the suspension was magnetically stirred in dark for 30 min to ensure the establishment of an adsorption–desorption equilibrium between the photocatalyst and RhB. Then the suspension was irradiated under continuous stirring by using a 350 W xenon arc lamp with a UV-cutoff filter (420 nm) and was positioned 20 cm away from the reactor. Liquid samples were withdrawn at certain time intervals. The dye solution was separated from the catalyst using a centrifugal and the residual concentration of RhB was determined using a UV-vis spectrophotometer. For comparison, blank experiments without catalyst were also carried out. Additionally, the recycle experiments were performed for three consecutive cycles to test the durability.

2.5. Active species trapping experiments

For detecting the active species during the reaction, tert-butanol (t-BuOH), ammonium oxalate (AO) and 1,4-benzoquinone (BQ) were used as the hydroxyl radical (OH˙) scavenger, hole (h⁺) scavenger and superoxide radical (O₂˙⁻) scavenger, respectively. The process was parallel to the former photocatalytic activity experiment with the addition of quencher in the presence of RhB.

3. Results and discussion

3.1. TEM analysis

The detailed characterization of the morphologies and crystal structures of as-prepared samples were based on TEM. From the TEM images in Fig. 1, the smooth and flat layers in these series of g-C₃N₄ based samples could be clearly seen, which was consistent with the reported results.^25,36 After the introduction of Ag/Ag₂O, numerous dark Ag/Ag₂O NPs appear on the lamellar g-C₃N₄. All the Ag/Ag₂O NPs are attached on the surface of g-C₃N₄ strongly. Fig. 1A–F shows the TEM images of as-synthesized AAC-X prepared with different amount of AgNO₃. According to the TEM images, the as-prepared Ag/Ag₂O NPs are all well dispersed on the surface of g-C₃N₄ nanosheets. The TEM image for AAC-2.5 (Fig. 1A) sample shows that only a small amount of NPs with sizes approximately ranging from 4 to 13 nm (Fig. 2A) were deposited on the surface of g-C₃N₄ nanosheets. The average size of Ag/Ag₂O (AAC-2.5) was about 8 nm with an even distribution about this value. A series of experiments were performed by varying the amount of AgNO₃, while keeping other parameters constant. As shown in Fig. 1B–D, for AAC-5.0 (Fig. 1B) and AAC-10 (Fig. 1C), Ag/Ag₂O NPs with increasing amounts were uniformly distributed on g-C₃N₄ nanosheets, no apparent aggregation of the Ag–Ag₂O NPs was discerned which leads to the formation of interfaces between Ag–Ag₂O and g-C₃N₄ nanosheets. The size distribution of Ag/Ag₂O (AAC-5.0) ranged from 4 to 16 nm (Fig. 2B) and the average size was 11 nm. When the amount of AgNO₃ increased to 10 mL, the size of Ag/Ag₂O NPs only slightly increased (Fig. 2C). The resulting size of Ag/Ag₂O NPs (Fig. 1E) increases to 22–48 nm (Fig. 2D) with increasing the amount of AgNO₃ (15 mL). The diameter of Ag/Ag₂O was about 33 nm according to the size distribution. The NPs are still uniformly loaded on surface of g-C₃N₄ nanosheets without aggregation, offering high level exposure of the NP surface.

Fig. 1 TEM images of AAC-2.5 (A), AAC-5.0 (B), AAC-10.0 (C and D) and AAC-15.0 (E and F); wide-angle XRD pattern (G) of as-prepared samples.

Fig. 2 Particle size distribution of Ag/Ag₂O NPs ((A) AAC-2.5, (B) AAC-5.0, (C) AAC-10.0 and (D) AAC-15.0).

3.2. XRD analysis

XRD analysis is used to investigate the phase structure of AAC-X photocatalysts. The XRD patterns of as-prepared AAC-X composites, as well as the pure g-C₃N₄ and Ag were shown in Fig. 1I. Two distinct diffraction peaks at 13.1° and 27.5° in the C₃N₄ sample, corresponding to the in-plane structural packing motif and inter-layer stacking of aromatic segments, respectively, can be indexed as the (100) and (002) peaks for graphitic materials.^37,38 No significant diffraction peaks of any other phases or impurities can be detected in the composite samples, which indicates the introducing of Ag/Ag₂O does not affect the crystal structure of C₃N₄ based photocatalysts. In addition, these two peaks become weaker gradually with increasing Ag/Ag₂O contents, suggesting that Ag/Ag₂O NPs have been successfully deposited onto the surface of g-C₃N₄. Additionally, AAC-X catalysts were found to exhibit the diffraction peaks at 38.0, 44.4, 64.4 and 77.3°, which are assigned to the (111), (200), (220) and (311) planes of face centered cubic (FCC) Ag (JCPDS card no. 65-2871).³⁹ The peak intensity of Ag for AAC-X becomes stronger as the Ag contents increased, which may be explained by Ag plasmonic effects and high uniform dispersion in AAC-X samples.^16,40 In addition, the diffraction peaks at 33.7 and 38.2° are consistent with JCPDS no. 75-1532 of Ag₂O. Moreover, the weak peak of Ag₂O at 26.7° may be overlapped by the strong peak of g-C₃N₄ at 27.5° and cannot be observed in the XRD patterns.

3.3. FT-IR analysis

The composition of AAC-X samples were further confirmed by FT-IR analysis and the results were shown in Fig. 3. The pure g-C₃N₄ presents three characteristic absorption regions. The broad peak at 3100–3600 cm⁻¹ is ascribed to the stretching vibration of N–H and the stretching vibration of O–H of the physically adsorbed water.^41,42 The strong bands in the 1200–1650 cm⁻¹ regions were found in the spectrum, with the peaks at 1240, 1317, 1409, 1573 and 1636 cm⁻¹, which corresponded to the typical stretching vibration modes of C=N and C–N heterocycles.^22,43 The peak at 810 cm⁻¹ is related to characteristic breathing mode of triazine units.⁴⁴ It could also be clearly seen that the main characteristic peaks of g-C₃N₄ appeared in AAC-X photocatalysts, suggesting that no structural change of g-C₃N₄ occurs during the hybridization process. Furthermore, in the case of the AAC-X hybrid materials, the characteristic peaks of g-C₃N₄ did not move (red shift or blue shift) after the introduction of Ag/Ag₂O NPs.

Fig. 3 FT-IR spectra of g-C₃N₄ and AAC-X materials.

3.4. XPS measurements

To investigate the chemical composition and states of surface elements in the AAC-10.0 nanosheets, XPS measurements were carried out. Fig. 4A shows the XPS of ACC-10.0 in the C 1s binding energy regions and two peaks concerning at 288.1 and 284.6 are found. The peak centered at 284.6 eV can be ascribed to the C–C coordination of the surface adventitious carbon, whereas the peak at 288.1 eV corresponds to sp³-bonded C in C–N of g-C₃N₄.^8,44 In the N 1s spectrum (Fig. 4B), several binding energies can be separated. The main signal showed occurrence of C–N–C groups (398.4 eV), tertiary nitrogen N–(C)₃ groups (399.2 eV) and N–H groups (400.6 eV). The peak at 404.3 eV could be related to the charging effects.^45,46 Accordingly, the asymmetric O 1s peak shown in Fig. 4C can be split by using the XPS peak-fitting program. In the atomic O 1s XPS spectrum, the peak at 532.2 eV is attributed to the external –OH group or the water molecule adsorbed on the surface and the other O 1s peak located at around 531.1 eV was assigned to oxygen atoms in the lattice of Ag₂O.⁴⁷ Fig. 4D presents the XPS of AAC-10.0 in the Ag 3d binding energy regions. Two individual peaks at 367.6 and 373.5 eV, corresponding to Ag 3d5/2 and Ag 3d3/2 binding energies, respectively. These two peaks can be attributed to binding energy values of Ag⁰.³ In addition, the peaks located at about 368.4 and 374.1 eV can be attributed to Ag⁺, which are very close to the binding energy values of Ag⁺ in pure Ag₂O.⁴⁷ The results indicated the existence of both Ag and Ag₂O in the as-prepared composites. The binding energy values of Ag 3d and N 1s in the ternary materials are slightly higher than those of pure Ag₂O and pure g-C₃N₄.^6,27 The shift could be attributed to the surface coupling effect between Ag/Ag₂O and g-C₃N₄. Furthermore, results from XRD and HRTEM also revealed that the obtained composites contained Ag and Ag₂O. Evidently, from the all above the analysis, we can conclude that Ag/Ag₂O@g-C₃N₄ ternary hybrid materials have successfully prepared in this case.

Fig. 4 High-resolution XPS of AAC-10.0 nanocomposites in the (A) C 1s, (B) N 1s, (C) O 1s and (D) Ag 3d binding energy regions.

3.5. N₂ adsorption–desorption studies

It is well-known that photocatalysts with higher specific surface areas and bigger pore volumes are beneficial for the enhancement of photocatalytic activity due to there being more surface active sites for the adsorption of reactant molecules, ease of transportation of reactant molecules and products through the interconnected porous networks, and enhanced harvesting of light.⁴⁸ Fig. 5 shows the nitrogen adsorption–desorption isotherms at 77 K for pure g-C₃N₄ and a representative photocatalyst AAC-15.0. It can be seen that the nitrogen adsorption–desorption isotherms of these two catalysts are similar and both of them are type III with hysteresis loops according to the IUPAC classification.^49–51 The BET specific surface area and pore volume of samples are summarized in Table 1. As can been seen from Table 1, the BET specific surface area (S_BET) of the pure g-C₃N₄ is about 41.1 m² g⁻¹. When the Ag/Ag₂O NPs were loaded, the S_BET of the samples become smaller, which are slightly lower than pure g-C₃N₄. However, a further increase of Ag/Ag₂O NPs leaded to the severely decreased S_BET. It is easy to under stand that once Ag/Ag₂O NPs deposit on the surface of g-C₃N₄ nanosheets or embedded in the pores, they will reduce the specific surface area of the samples (Fig. 5).

Table 1 The BET specific surface area and pore volume of as-prepared samples

Samples	SBET (cm^2 g^−1)	Pore volume (cm^3 g^−1)
g-C3N4	41.1	0.28
AAC-2.5	39.8	0.25
AAC-5.0	38.3	0.19
AAC-10.0	37.4	0.18
AAC-15.0	31.1	0.11

---

Fig. 5 Typical nitrogen adsorption–desorption isotherm of g-C₃N₄ and AAC-15.0.

3.6. Optical absorption property

The light absorption properties of as-prepared g-C₃N₄ and typical AAC-10.0 composites are studied by UV-vis diffuse absorption spectra and the result is shown in Fig. 6. The as-prepared g-C₃N₄ nanosheets show a typical semiconductor absorption in the region of 200–450 nm, originating from charge transfer response of g-C₃N₄ from the VB populated by N 2p orbitals to the CB formed by C 2p orbitals. In addition, the sample colors shift from yellowish to dark brown with increasing Ag/Ag₂O loading. This phenomenon could be attributed to a charge-transfer transition among the Ag, Ag₂O species and the g-C₃N₄ sample.¹⁸ The band gap energy of a semiconductor can be determined by fitting the optical transition at the absorption edges using the Tauc/David–Mott model⁵² described by the equation:

(αhν)^1/n = A(hv − E_g) (1)

where α represents the absorption coefficient, h is Planck's constant, ν is the light frequency, E_g is the band gap and A is a constant. The value of the exponent n denotes the nature of the sample transition and is defined to be 0.5/2 for a directly/indirectly allowed transition. As previous literature reported, g-C₃N₄ is an indirectly allowed transition. After calculated experimental data,⁵³ E_g of pure g-C₃N₄ was determined from a plot of (αhν)^1/2 versus energy (hν) and the band gap energy of g-C₃N₄ is approximate 2.69 eV, which is consistent with the existing results.⁵⁴

Fig. 6 UV-vis diffuse reflection spectra of g-C₃N₄ and AAC-X.

3.7. Photocatalytic activity and photostability

RhB is widely used in many industrial fields such as textiles, plastics, paper, coatings, and rubber. The discharge of dyes into water has received global concern because of their overall environmental hazards. The removal of RhB from the environment become an important issue.³ To this end, the photocatalytic activities of as-prepared catalysts towards the decomposition of RhB under visible light were investigated in present study. Furthermore, commercially available Degussa P25 TiO₂ and pure g-C₃N₄ were used as references for comparison. According to the adsorption–desorption behavior results of AAC-X without light irradiation, after 30 minutes' adsorption, there is almost no change in the concentration of dye, which indicates that the adsorption–desorption behavior reaches an equilibrium. The photodegradation of RhB using AAC-10.0 was demonstrated by the help of plots between absorption and wavelength as shown in Fig. 7A. After 140 min of visible light exposure, the concentration of RhB was found to be less than 4.5%. Fig. 7B shows the photocatalytic evaluation of the AAC-X and P25 under visible light illumination. As can be seen from Fig. 7B, RhB can only be slightly degraded under visible light irradiation without catalysts, indicating that RhB is a stable molecule and that the photolysis mechanism can be ignored. As a reference material, the P25 sample displayed a low photocatalytic efficiency of 39% after irradiation for 140 min. Fig. 7B also indicates that the as-prepared samples exhibit good photocatalytic activities for the RhB degradation reaction. The catalytic activities of all AAC-X catalysts are better than pure g-C₃N₄ and Ag₂O. In addition, the presence of Ag/Ag₂O effectively enhances the photocatalytic activity of g-C₃N₄. Moreover, photocatalytic activity of Ag₂O@g-C₃N₄ is lower than those of AAC-X catalysts, suggesting the importance of Ag in degradation of RhB. When the optimum content of Ag/Ag₂O was obtained (AAC-10.0), the best photocatalytic activity of Ag/Ag₂O@g-C₃N₄ was obtained. With the Ag/Ag₂O contents increased, the photocatalytic activity of AAC-X first increased and then decreased. These changes in the photocatalytic activity could be understood through considering the synergetic effects of light absorption and heterojunction structure of AAC-X. When the Ag/Ag₂O content increased, the visible light absorption of Ag/Ag₂O@g-C₃N₄ strengthened and generated more electron–hole pairs and more Ag–Ag₂O@g-C₃N₄ heterojunction was formed, contributing to the separation of electron–hole pairs.⁴ However, when the Ag/Ag₂O content was above the optimal value, the high Ag/Ag₂O content may create some new e⁻/h⁺ pairs recombination centers, which could counteract its positive effect on the restraint of the recombination of e⁻/h⁺ pairs.⁵⁵ Consequently, the separation efficiency of electron–hole pairs decreased and further restrained the photocatalytic activity. In addition, the photocatalytic activity of catalyst is highly dependent on surface area because high surface area can not only provide more surface active sites for the efficient diffusion and transportation of the degradable organic molecules and active species in photochemical reaction, but also effectively promote the separation efficiency of the electron–hole pairs. Comparing to other three as-prepared catalysts, AAC-15.0 shows lower BET surface area as discussed above (Table 1). This reason may also lead to the decreasing of photocatalytic activity of AAC-15.0. Overall the above analysis, we can conclude that the photocatalytic activity of as-prepared catalysts could depend on the content ratio of Ag/Ag₂O@g-C₃N₄ and their surface area. Langmuir–Hinshelwood model has been widely applied for analysis of heterogeneous photocatalytic degradation kinetics of pollutants in the aqueous phase:where r₀ is the initial reaction rate (mg L⁻¹ min⁻¹), C is the concentration of pollutants (mg L⁻¹), t is the reaction time (min), k is the Langmuir–Hinshelwood reaction rate constant (mg L⁻¹ min⁻¹) and K is the Langmuir adsorption equilibrium constant (L mg⁻¹). At a dilute concentration of pollutants, pseudo-first-order kinetics model could be assumed as shown in eqn (3) and (4):where k_app is the apparent rate constant (min⁻¹), and C₀ is the initial concentration of pollutants (mg L⁻¹). The pseudo-first-order rate constants over P25, pure g-C₃N₄, Ag₂O, Ag₂O@g-C₃N₄, ACC-2.5, ACC-5.0, ACC-10.0 and ACC-15.0 are 0.0030, 0.0071, 0.0103, 0.0109, 0.0105, 0.0145, 0.0215 and 0.0167, respectively. These results imply that k values for all AAC-X are higher than those of P25, pure g-C₃N₄, Ag₂O and Ag₂O@g-C₃N₄. The photocatalytic decomposition rate of RhB over the AAC-10.0 catalyst is 3.0 and 7.0 times the activity over g-C₃N₄ (Ag₂O) and P25, respectively. The mineralization behavior of RhB over the AAC-10 photocatalyst was examined. About 82% of total organic carbon (TOC) was decreased with irradiation time of 140 min, suggesting most of the RhB was mineralized to inorganic molecules. All these results indicate that the formation of AAC-X could greatly enhance the photocatalytic efficiency of single g-C₃N₄.

Fig. 7 (A) UV-vis absorption spectra of RhB solutions in the presence of AAC-10.0 under visible light; (B) the relationship between C_t/C₀ and reaction time (t) in the photodegradation of RhB; (C) the relationship between ln(C_t/C₀) and reaction time (t) in the photodegradation of RhB; (D) the three run test of photocatalytic activity using AAC-10.0.

For practical photocatalytic applications in aqueous solution, the recycling of the AAC-X composite is an important but difficult task owing to the leakage of the individual components during the photocatalytic procedures. The photo-stability of AAC-10.0 was carried out by means of a recycling reaction. In recycling, the reacted catalysts were separated by high speed centrifuging (10 000 rpm). The separated catalysts were dried in a vacuum oven at a temperature of 80 °C for 12 h and used again. AAC-10.0 catalyst was repeatedly used for the RhB photodegradation reaction 6 times. The experimental results indicated that the catalytic activity of AAC-10.0 slightly decreased as compared to the fresh catalyst after 2 cycles, (Fig. 7D) which might because a part of Ag₂O NPs were reduced to Ag either by the reduction of photogenerated electron or the visible light irradiation. This decrease trend was essentially inhibited in the third cycle, revealing that the Ag/Ag₂O@g-C₃N₄ ternary photocatalyst has the good photostability. However, the catalytic activity further decreased in the fourth, fifth and sixth cycles because during the reusability of the photocatalyst, there would be a loss in the amount of photocatalyst as it is being centrifuged, washed and dried. To confirm the decreasing of photocatalytic activity after the third cycle was attribute to the loss in the amount of catalyst rather than the reduction of Ag₂O, the composition of the recyclable composite (after third and sixth cycles) was characterized by XRD. As shown in Fig. S1-A,† after third reaction, the diffraction peaks of Ag species become stronger which could be arised from the partial in situ photoreduction of Ag₂O. Previous research results demonstrated that, after the formation of a certain amount of metallic Ag, the obtained Ag₂O/Ag exhibits a stable structure.⁵⁵ In contrast to the three times used photocatalysts, the diffraction peaks of Ag species do not become stronger after sixth reaction. This result further proved that Ag/Ag₂O@g-C₃N₄ ternary photocatalyst has a good photostability. Moreover, the diffraction peak ascribed to the g-C₃N₄ is the same for the fresh and used photocatalysts, implying the high stability of g-C₃N₄. No obvious changes could be observed from the high resolution XPS spectra (Fig. S1-B†) of Ag 3d for the fresh and used photocatalysts further indicated that the ternary composites are stable. Moreover, TEM analysis (Fig. S1-C and D†) was also performed to investigate the stability of as-prepared catalyst. As can be observed from TEM image, no obvious agglomeration was found (fresh and six times used catalysts), further reflecting the stability of the ternary composite. As a result, the results demonstrate that the unique, Ag/Ag₂O embedded g-C₃N₄ nanostructures show good catalytic property in photo-stability.

3.8. PL emission spectra analysis

PL emission measurement has been widely used to investigate the fate of electron–hole pairs in semiconductor particles because PL emission is known to result from the recombination of excited electrons and holes for some semiconductors. The higher the PL emission intensity indicates the higher recombination efficiency of the photogenerated carriers and the lower the photocatalytic activity.⁵⁶ Fig. 8A shows the PL emission spectra of the pure g-C₃N₄ and AAC-X composites at room temperature. The excitation light source for pure g-C₃N₄ and AAC-X is a 360 nm He–Cd laser. The main emission peak for pure g-C₃N₄ is centered at about 460 nm, which is approximately equal to the band gap energy of pure g-C₃N₄. Compared with g-C₃N₄, the addition of Ag/Ag₂O does not alter the spectral position of the peaks, but reduces the relative intensity of PL spectral. In addition, the intensity of this emission peak depends on the Ag/Ag₂O concentration. AAC-10.0 composites give the weakest PL peak, while the intensity of the PL peak increases for the highest Ag/Ag₂O concentration (AAC-15.0). This result is corresponding to the photocatalysis activity. The lifetime of charge carriers in the g-C₃N₄ and AAC-10 were examined by using time-resolved transient PL spectroscopy, as shown in Fig. 8B. These spectra are both fitted by single-exponential decay models. The lifetime of AAC-10 (τ = 25.2 ns) is longer than the corresponding life time of g-C₃N₄, while maintaining similar lifetime intensities. This evidence indicates that the formation of Ag–Ag₂O/g-C₃N₄ heterostructures prolonged the lifetime, thereby reduces the recombination of photoexcited electrons and holes. This phenomenon was also investigated by the previous reported. Hou⁵⁷ and co-workers prepared g-C₃N₄/Bi₂MoO₆ heterojunctions, and it found that the lifetime is prolonged after coupling g-C₃N₄ with Bi₂MoO₆. The prolonged lifetime originates from the formation of the g-C₃N₄/Bi₂MoO₆ heterojunctions, which effectively reduces the recombination of photogenerated electrons and holes.

Fig. 8 Room temperature PL spectra (A) of g-C₃N₄ and AAC-X (X = 2.5, 5.0, 10.0 and 15.0) under the excitation wavelength of 360 nm, time-resolved fluorescence decay spectra (B) of g-C₃N₄ and AAC-10.0.

3.9. Detection of reactive species

A series of experiments were conducted to probe the mechanism responsible for this visible light-induced photocatalysis. Three different quenchers, t-BuOH, AO and BQ were used as the hydroxyl radical (OH˙) scavenger, hole (h⁺) scavenger and superoxide radical (O₂˙⁻) scavenger, respectively. As shown in Fig. 9, when 1 mmol of BQ as a scavenger for O₂˙⁻ radical species is added, no obvious decrease of degradation rate is observed, indicating the absence of O₂˙⁻ radical species. The experimental results (Fig. 9) also shown that the addition of 1 mmol t-BuOH or AO caused fast deactivation of the AAC-10.0 photocatalyst, reducing the photocatalytic activity for the degradation rate of RhB from 96% to 13% (t-BuOH) and 11% (AO) within 140 min. This result indicates that the OH˙ and h⁺ pathways have crucial roles in the process of RhB oxidation. In addition, it is well known that, holes can be adsorbed on the surface of the photocatalyst and can interact with hydroxyl radicals or water molecules to generate OH˙. The above results indicate that OH˙ is the most important oxidizing species during the RhB photocatalytic process, and that O₂˙⁻ in the solution is not the main active species. To further confirm hypothesis, we performed the photoactivity tests for OH˙ production. It is accepted that OH˙ radicals react with TA in solution and generate 2-hydroxylterephthalic acid (TAOH), which emits a unique fluorescence signal with a peak at 426 nm. Fluorescence spectra associated with TAOH were generated upon visible light irradiation of AAC-10.0, indicating that the surface structures are active in transferring photo-induced charge carriers to reactants, such as surface adsorbed water and hydroxyl groups. The nearly linear relationship between fluorescence intensity and irradiation time further confirms the stability of the as-prepared g-C₃N₄-based nanosheets (Fig. 10).

Fig. 9 Trapping experiment of active species during the photocatalytic degradation of RhB over AAC-10.0 under visible light irradiation.

Fig. 10 (A) Fluorescence spectra of AAC-10.0 nanosheets suspensions in 3 mM terephthalic acid irradiated by visible light at different irradiation times; (B) the time dependence of the fluorescence signal intensities at 426 nm on the irradiation time.

3.10. Possible photocatalytic mechanism of AAC-X nanosheets

In order to deeply understand the photocatalytic mechanism for the AAC-X nanosheets, the band edge positions of the valence band (VB) and conduction band (CB) of Ag₂O and g-C₃N₄ were investigated. The VB edge potentials of Ag₂O and g-C₃N₄ at the point of zero charge can be calculated by the following empirical equation:

E_VB = X − E^e + 0.5E_g (5)

where E_VB is the VB edge potential, X is the electronegativity of the semiconductor, which is the geometric mean of the electronegativity of the constituent atoms, E^e is the energy of free electrons on the hydrogen scale (approximately 4.5 eV), and E_g is the band gap energy of the semiconductor. The CB edge potential (E_CB) can be determined by E_CB = E_VB − E_g. The band gap of g-C₃N₄ was estimated as 2.69 eV according to the literatures⁵⁶ and the DRS analysis (Fig. 6). In the case of Ag₂O, the optical band gap was 1.2 eV, according to the reported literature.⁵ Based on the band gap positions, the CB and VB edge potentials of g-C₃N₄ were determined at −1.12 eV and +1.57 eV, respectively.⁷ The CB and VB edge potentials of Ag₂O were determined at +0.20 and +1.40 eV, respectively. The CB of g-C₃N₄ is more negative than that of Ag₂O and the VB of g-C₃N₄ is more positive than that of Ag₂O. Considering the inner electric field and energy band structure, it is accepted that transferring electrons between g-C₃N₄ and Ag₂O would be partly restricted, while the transfer of holes can be accelerated.⁵ However, the electrons can transfer from g-C₃N₄ and Ag directly. This causes an efficient separation of photogenerated electrons and holes to enhance the photocatalytic activity. On the basis of the above analysis, the process could be described as follows (Fig. 11): under the visible light irradiation, the electrons were promoted from the VB to the CB of the g-C₃N₄ and Ag₂O NPs, leaving the holes behind. When g-C₃N₄ and Ag₂O are in contact, the holes generated in the VB of Ag₂O NPs could migrate to the surface of g-C₃N₄, which promoted the effective separation of photoexcited electron–hole pairs and decreased the probability of electrons and holes recombination. The enriched holes on the VB of Ag₂O could not only be consumed by directly decomposing the organic molecules on the surface of as-prepared samples, but also captured by the OH⁻ ionized from water, which produced OH˙ free radicals with very strong oxidation capacities. The Ag₂O has been proved to be an effective sacrificial reagent to capture photo-induced electrons.⁵⁵ After the modification by Ag, the generated electrons on the CB of Ag₂O can effectively transfer to the surface of Ag, which can hinder the charge recombination in the electron-transfer processes, thus enhance the photocatalytic performance.⁵⁸ In addition, part of Ag may directly attach to the surface of g-C₃N₄, thus, the electrons on the CB of g-C₃N₄ can also transfer to the surface of Ag. Based on the above discussion, it can be accepted that the enhanced photocatalytic activity of the as-prepared AAC-X could be ascribed to the following reasons: (a) the improved optical absorption property arising from the heterojunction between g-C₃N₄ and Ag₂O, (b) the electrons transfer from g-C₃N₄ to Ag, noble-metal cocatalyst (Ag) acts as a reduction active site to promote the spatial separation of photogenerated electrons and holes, and the oxygen-reduction reaction on the noble-metal cocatalyst is expected to proceed by one electron transfer route,⁵⁹ (c) the synergetic effects of the inner electric field among Ag, g-C₃N₄ and Ag₂O, (d) the low recombination rate of the photogenerated electrons which brings from the matched energy band structure.

Fig. 11 Schematic diagrams for the possible photocatalytic mechanism of the AAC-X composites under visible light irradiation.

4. Conclusion

In summary, AAC-X nanocomposites have been prepared via a green and versatile method by directly reducing AgNO₃ using carbon nanodots for the first time. The average size of Ag/Ag₂O can be controlled from 8 to 33 nm by adjusting the adding amount of AgNO₃. The as-obtained AAC-X nanocomposites exhibited enhanced visible-light-driven photocatalytic activity and stability in the degradation of RhB in aqueous solution. The optimum photocatalytic activity of the AAC-10.0 nanocomposites for the degradation of RhB was almost 3.0 and 7.0 times higher than g-C₃N₄ and P25, respectively. The high activity and stability could contribute to the formation of well-defined junctions among Ag, Ag₂O and g-C₃N₄, which effectively facilitates charge transfer and suppresses the recombination of photogenerated electrons and holes. In addition, the possible radical species involved in the degradation of RhB were analyzed by means of PL. The results indicate that the photodegradation of RhB molecules is mainly attributed to holes and hydroxyl radicals. It is expected that our present work could provide useful information for the fabrication of other metal loaded g-C₃N₄-based heterojunction composites.

Acknowledgements

This work was partly supported by National Natural Science Foundation (21003065, 21306067), China Scholarship Council (201508320218), Industry High Technology Foundation of Jiangsu Province (BE2013090), Innovation Project for Graduate Student Research of Jiangsu Province (KYLX_1065/1291310024) and Science & Technology Foundation of Zhenjiang (SH2012011 and GY2012048), China.

References

1. G. Liu, H. G. Yang, X. Wang, L. Cheng, J. Pan, G. Q. Lu and H. M. Cheng, J. Am. Chem. Soc., 2009, 131, 12868–12869.
2. H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith, H. M. Cheng and G. Q. Lu, Nature, 2008, 453, 638–641.
3. Z. F. Jiang, X. M. Lv, D. L. Jiang, J. M. Xie and D. J. Mao, J. Mater. Chem. A, 2013, 1, 14963–14972.
4. Z. F. Jiang, J. J. Zhu, D. Liu, W. Wei, J. M. Xie and M. Chen, CrystEngComm, 2014, 16, 2384–2394.
5. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76–80.
6. X. C. Wang, X. F. Chen, A. Thomas, X. Z. Fu and M. Antonietti, Adv. Mater., 2009, 21, 1609–1612.
7. S. Ye, L. G. Qiu, Y. P. Yuan, Y. J. Zhu, J. Xia and J. F. Zhu, J. Mater. Chem. A, 2013, 1, 3008–3015.
8. Z. X Li, Y. Shen, Y. H. Guan, Y. H. Hu, Y. H. Lin and C. W. Nan, J. Mater. Chem. A, 2014, 2, 1967–1973.
9. H. Yan, J. Yang, G. Ma, G. Wu, X. Zong, Z. Lei, J. Shi and C. Li, J. Catal., 2009, 266, 165–168.
10. Y. Zheng, L. Zheng, Y. Zhan, X. Lin, Q. Zheng and K. Wei, Inorg. Chem., 2007, 46, 6980–6986.
11. Y. Di, X. Wang, A. Thomas and M. Antonietti, ChemCatChem, 2010, 2, 834–838.
12. M. Yoshida, A. Yamakata, K. Takanabe, J. Kubota, M. Osawa and K. Domen, J. Am. Chem. Soc., 2009, 131, 13218–13219.
13. J. Dong, M. Wang, X. Li, L. Chen, Y. He and L. Sun, ChemSusChem, 2012, 5, 2133–2138.
14. P. D. Cozzoil, R. Comparelli, E. Fanizza, M. L. Curri, A. Agostiano and D. Laub, J. Am. Chem. Soc., 2004, 126, 3868–3879.
15. Y. L. Meng, J. Shen, D. Chen and G. Xin, Rare Met. Mater. Eng., 2011, 30, 276–279.
16. X. Bai, R. Zong, C. Li, D. Liu, Y. Liu and Y. Zhu, Appl. Catal., B, 2014, 147, 82–91.
17. Y. Bu, Z. Chen and W. Li, Appl. Catal., B, 2014, 144, 622–630.
18. L. Ge, C. Han, J. Liu and Y. Li, Appl. Catal., A, 2011, 215, 409–410.
19. Z. F. Jiang, D. Liu, D. L. Jiang, W. Wei, K. Qian, M. Chen and J. M. Xie, Dalton Trans., 2014, 43, 13792–13802.
20. H. Yan and H. Yang, J. Alloys Compd., 2011, 509, L26–L29.
21. Y. Wang, R. Shi, J. Lin and Y. Zhu, Energy Environ. Sci., 2011, 4, 2922–2929.
22. D. Chen, K. Wang, D. Xiang, R. Zong, W. Yao and Y. Zhu, Appl. Catal., B, 2014, 147, 554–561.
23. J. Sun, Y. Yuan, L. Qiu, X. Jiang, A. Xie, Y. Shen and J. Zhu, Dalton Trans., 2012, 41, 6756–6763.
24. S. C. Yan, S. B. Lv, Z. S. Li and Z. G. Zou, Dalton Trans., 2010, 39, 1488–1491.
25. X. Xu, G. Liu, C. Randorn and J. T. S. Irvine, Int. J. Hydrogen Energy, 2011, 36, 13501–13507.
26. J. W. Lee, H. J. Jeon, H. J. Shin and J. K. Kang, Chem. Commun., 2012, 48, 422–424.
27. X. Wang, S. Li, H. Yu, J. Yu and S. Liu, Chem.–Eur. J., 2011, 17, 7777–7780.
28. Y. Li, H. Zhang, Z. Guo, J. Han, X. Zhao, Q. Zhao and S. J. Kim, Langmuir, 2008, 24, 8351–8357.
29. C. Hu, X. Hu, L. Wang, J. Qu and A. Wa, Environ. Sci. Technol., 2006, 40, 7903–7907.
30. Y. Bi, S. Ouyang, J. Cao and J. Ye, Phys. Chem. Chem. Phys., 2011, 13, 10071–10075.
31. J. F. Guo, B. W. Ma, A. Y. Yin, K. N. Fan and W. L. Dai, Appl. Catal., B, 2011, 101, 580–586.
32. J. Cao, B. Luo, H. Lin and S. Chen, J. Mol. Catal. A: Chem., 2011, 344, 138–144.
33. S. Rodrigues, S. Uma, I. N. Martyanov and K. J. Klabunde, J. Catal., 2005, 233, 405–410.
34. A. Pourahmad, S. Sohrabnezhad and E. Kashefian, Spectrochim. Acta, Part A, 2010, 77, 1108–1114.
35. Y. Zang, R. Farnood and J. Currie, Chem. Eng. Sci., 2009, 64, 2881–2886.
36. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2009, 25, 10397–10401.
37. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2010, 26, 3894–3910.
38. J. Fu, B. Chang, Y. Tian, F. Xi and X. Dong, J. Mater. Chem. A, 2013, 1, 3083–3090.
39. Y. Wen, H. Ding and Y. Shan, Nanoscale, 2011, 3, 4411–4417.
40. H. Sakai, T. Kanda, H. Shibata, T. Ohkubo and M. Abe, J. Am. Chem. Soc., 2006, 128, 4944–4945.
41. J. Fu, Y. L. Tian, B. B. Chang, F. N. Xi and X. P. Dong, J. Mater. Chem., 2012, 22, 21159–21166.
42. M. J. Bojdys, J. O. Muller, M. Antonietti and A. Thomas, Chem.–Eur. J., 2008, 14, 8177–8182.
43. Q. Xiang, J. Yu and M. Jaroniec, J. Phys. Chem. C, 2011, 115, 7355–7363.
44. H. Xu, J. Yan, Y. G. Xu, Y. H. Song, H. M. Li, J. X. Xia, C. J. Huang and H. L. Wan, Appl. Catal., B, 2013, 129, 182–193.
45. Y. S. Xu and W. D. Zhang, ChemCatChem, 2013, 5, 2343–2351.
46. Y. Q. Sun, C. Li, Y. X. Xu, H. Bai, Z. Y. Yao and G. Q. Shi, Chem. Commun., 2010, 46, 4740–4742.
47. L. Huang, H. Xu, Y. Li, H. Li, X. Cheng, J. Xia, Y. Xu and G. Cai, Dalton Trans., 2013, 42, 8606–8616.
48. D. Chandra, K. Saito, T. Yui and M. Yagi, Angew. Chem., Int. Ed., 2013, 52, 12606–12609.
49. Z. F. Jiang, J. M. Xie, D. L. Jiang, X. J. Wei and M. Chen, CrystEngComm, 2013, 15, 560–569.
50. Z. F. Jiang, J. M. Xie, D. L. Jiang, J. J. Jing and H. R. Qin, CrystEngComm, 2012, 14, 4601–4611.
51. Z. F. Jiang, J. M. Xie, D. L. Jiang, Z. X. Yan, J. J. Jing and D. Liu, Appl. Surf. Sci., 2014, 292, 301–310.
52. X. M. Li, H. W. Zhu, J. Q. Wei, K. L. Wang, E. Y. Xu, Z. Li and D. H. Wu, Appl. Phys. A: Mater. Sci. Process., 2009, 97, 341–344.
53. Z. D. Meng, T. Ghosh, L. Zhu, J. G. Choi, C. Y. Park and W. C. Oh, J. Mater. Chem., 2012, 22, 16127–16135.
54. J. S. Zhang, X. F. Chen, K. Takanabe, K. Maeda, K. Domen, J. D. Epping, X. Z. Fu, M. Antonietti and X. C. Wang, Angew. Chem., Int. Ed., 2010, 49, 441–444.
55. W. Yu, X. Liu, H. Chu, G. Zhu, J. Li, J. Liu, L. Niu, Z. Sun and L. Pan, J. Mol. Catal. A: Chem., 2015, 407, 25–31.
56. L. L. Chen, W. X. Zhang, C. Feng, Z. H. Yang and Y. M. Yang, Ind. Eng. Chem. Res., 2012, 51, 4208–4214.
57. H. Li, J. Liu, W. Hou, N. Du, R. Zhang and X. Tao, Appl. Catal., B, 2014, 160–161, 89–97.
58. X. Liu, J. Liu, H. Chu, J. Li, W. Yu, G. Zhu, L. Niu, Z. Sun, L. Pan and C. Q. Sun, Appl. Surf. Sci., 2015, 347, 269–274.
59. H. Yu, R. Liu, X. Wang, P. Wang and J. Yu, Appl. Catal., B, 2012, 111–112, 326–333.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra22176e

---