Ultra small gold nanoparticles–graphitic carbon nitride composite: an efficient catalyst for ultrafast reduction of 4-nitrophenol and removal of organic dyes from water

Abstract

Synthesis of supported ultra small gold particles is important for their unusual properties and catalytic applications. We report a facile method for synthesis of ultra small gold nanoparticles supported on carbon nitride sheets (Au–CN_x). The ultrasonication was used to form Au–CN_x composite from carbon nitride quantum dots and HAuCl₄ without any aid of external reducing agent. The Au–CN_x composite was well characterized by tunneling electron microscopy, selected area electron diffraction, energy dispersive X-ray spectroscopy, powder X-ray diffraction, X-ray photo electron spectroscopy methods. Electron microscopic measurements confirm that thin graphitic carbon nitride sheets provided two dimensional supports to stabilize ultra small, 1–3 nm sized gold nanoparticles (AuNPs). The Au–CN_x composite showed excellent catalytic activity towards the reduction of 4-nitrophenol in aqueous medium in presence of sodium borohydride (NaBH₄) with very fast kinetics and good stability. The Au–CN_x catalyst can be used as an absorbent for the removal of organic dye [Rhodamine B (RhB), methylene blue (MB), and methyl red (MR)] from water. It showed excellent adsorption capacity for RhB and MB due to electrostatic interaction between anionic catalyst and cationic organic dye molecules. The catalyst can easily be reused after removing adsorbed dye from the catalyst simply by changing pH of the solution. In addition, Au–CN_x composite exhibited superior photo catalytic degradation of various dyes (RhB, MB and MR) on irradiation of UV, visible light and natural sunlight. Excellent photodegradation rate constants for RhB (0.024 min⁻¹), MB (0.024 min⁻¹) and MR (0.02 min⁻¹) were observed although high concentration of dyes were used for degradation. Au–CN_x is one of the best adsorbent for RhB adsorption with maximum adsorption capacity of 400 mg g⁻¹ and catalyst for nitrophenol reduction. We hope that this Au–CN_x composite will find its application as an effective catalyst for water purification.

---

Introduction

In recent years, noble metal nanoparticles have attracted tremendous attention due to their potential applications in various fields such as catalysis,¹ sensors,^2–6 optics,⁶ fuel cell.⁷ For example, gold nanoparticles⁸ are being used for a long time in drug delivery, different biological application and catalysis. Silver and platinum nanoparticles show promising capabilities in biosensors⁹ and in fuel cell.¹⁰ Very small metal NPs has an interesting size and shape dependent reactivity¹¹ due to the increase of volume to surface ratio. The reduction in size of metal nanoparticles causes a significant change in reduction potential of metal nanoparticles as compared to bulk materials because their Fermi potential becomes more negative. This typical important property helps them in electron transfer in various catalysis reactions. But synthesis of ultra small metal nanoparticle is a very challenging task due its high agglomeration property, arises from its high surface energy. In order to avoid aggregation, nanoparticles should be stabilized by using organic molecules or polymer as a capping agent. But their catalytic activity greatly reduced due to the presence of organic or polymer molecules around nanoparticles. Another alternative available method is to stabilize them on the surface of a solid support.¹² Two dimensional (2D) graphene, due to its unique properties¹³ such as high specific surface area, excellent electrical conductivity, high charge carrier mobility, high mechanical strength, became a promising candidate as a support for different nanoparticles. The graphene containing metal or semiconductor nanoparticles have confirmed to be effective nano-composite materials for photovoltaic,^14,15 catalysis and biosensor applications.¹⁶ In recent year another two dimensional material graphitic carbon nitride (g-C₃N₄), has attracted a significant attention due to its moderate band gap which leads to a potential application in the photocatalysis,^17–19 organic reactions^20,21 and fuel cell.^17,22,23 g-C₃N₄ consists of repeating triazine units which has large number of binding sites to stabilize the MNPs making this materials as a promising supports for noble MNPs. But, most of the reports^24,25 for g-C₃N₄ as supported materials are mesoporous due to the better diffusion and strong binding of metal NPs through the surface pore of the g-C₃N₄. The synthetic procedure of mesoporous compound needs of hazardous materials and also high temperature which is not environmental friendly. Some time it was difficult to remove the template from the products which demises the products purity. Thus, 2D sheet of carbon nitride can be considered as feasible alternative for stabilizing metal nanoparticles for their unusual catalytic properties.

In 21^st century, environmental pollution is increasing due to the fast growth of chemical industries and the release of various types of industrial contaminant into water.²⁶ The different industrial contaminants are chlorinated organic compounds, volatile organic compounds, nitro aromatics and fertilizers, dyes etc. Among these toxic threats, dyes and nitro aromatics are more concerned to us because of their high chemical and biological stability which makes difficult to removes from water. In chemical industries 4-NP is generally used to prepare different synthetic dyes, medicines and pesticides. The remediation of aromatic nitro compounds is important, because aromatic nitro moiety is one the most characteristic of anthropogenic contaminants. 4-Nitro phenol (4-NP) is toxic pollutant and hazardous compound, declared by U.S. Environmental Protection Agency. 4-NP can causes disorder of our central nervous system, and can damage kidney, liver of humans as well as animals. So removal of 4-NP is very much important. A conventional water purification treatment is not effective for removal of 4-NP due to its high stability and solubility in water. So reduction of 4-NP to 4-AP is the best idea as 4-AP is a starting material in various industrial synthesis like pharmaceutical, photographic, and corrosion inhibitor. Therefore, recently scientific community has given a lot of attention for inventing new catalyst for the reduction of 4-NP to 4-AP.^27,28

The effluents from the pesticides, dying, petrochemical, and pharmaceutical industries contains highly water pollutant dyes which release in environment and cause various problem such as some skin diseases in human health, and also aquatic biota are also adversely affected.²⁹ So removal of these type environmental threats is very important. Different methods are used to remove dyes from the environment such as coagulation–flocculation,³⁰ photodegradation,^31,32 electrochemical oxidation³³ and adsorption.³⁴ Among them, sorption and degradation process are most effective treatment for cleaning up water. The adsorption property depends on the surface structure of adsorbent where as photocatalytic activity control by the band gap of the semiconductor material or hetero-junction structure of the metal–semiconductor composite. The adsorption procedure used frequently to remove the dye molecules from water due the simplicity of operation procedure, low cost, and also easy recycling of the adsorbent. Scientific community has given more attention on the porous molecule³⁵ as a good adsorbent due to easy diffusion of the dye molecule through the pore of the adsorbent. Different natural polymers were found to be good adsorbent. A. J. Hunt and his co-workers developed polysaccharide based polymer for removal of dyes from water.³⁶ The main limitation of the polymer based catalyst is lower thermal stability, poor hydrodynamic volume as well as low surfaces area. Some time the polymer molecules shows lack of separation selectivity, low adsorbing capacity and recycling property. Polymeric g-C₃N₄ doesn't have good adsorption property. The adsorption property of C₃N₄ was improved by doping boron into the carbon nitride framework.³⁵ Recently Y. Chen and his co-workers synthesized porous BN sheet for effective water cleaning agent.³⁵ However there was no report on adsorption of dye using metal nanoparticles deposited carbon nitride as an adsorbent.

Photocatalyst based on heterogeneous semiconductor has created a lot of attention due to its potential application on reducing of global water pollution which was well accepted strategy to solve the environmental problem.³⁷ Among different photocatalytic system semiconductor³⁸ TiO₂ explored more but unfortunately TiO₂ is photo catalytically active only under UV-light irradiation³⁹ due to its wide band gap. The quantum yield of this type semiconductor catalyst is very low under solar light, since only 5% of solar energy is UV light. This is the major limitation of their applications. The photocatalysts, sensitive to visible light, are promising candidate for degradation of dyes, since visible light present in of solar energy is ∼40%. Therefore, both visible light as well as UV light driven photocatalysis are significantly important for practical applications. There are several reports on the different photocatalytic system based on their different photophysical properties such as band gap and band position of conduction and valence band, surfaces area, size, shape and morphology of the catalyst. But last few years scientific community are mostly focused on the metal oxide, sulfide and metal nitride based photocatalytic systems which are mostly toxic and costly materials. The enormous efforts has been given to develop the environmental friendly less toxic and cheap materials.⁴⁰ Among them, carbon, nitrogen based semiconductor, graphitic carbon nitride (g-C₃N₄) has created more attention in last few years due to their extreme stability under harsh thermal and photochemical condition in compared to other polymeric system and potential photocatalytic activity under visible light irradiation.⁴¹ But the low surface area, high recombination rate of photo excited charge carriers electron–hole, and low quantum efficiency is limiting their practical applications.⁴¹ Several methods have been applied to improve the charge separation efficiency of g-C₃N₄, such as designing the porous structure,⁴² coupled with different n-type semiconductor⁴⁰ and making hetero-junction composite with noble metals.^37,43–48 Among these processes, hetero-junction composites are most widely used.^49,50 X. Hu and his co-workers reported³⁷ Ag/AgCl/g-C₃N₄ composite for MB degradation under visible light. X. Wang⁵¹ and his co-worker also reported highly efficient visible light induced photocatalytic degradation of various dyes by Ag/AgCl/g-C₃N₄ composite. But there is very few reported on AuNPs deposited g-C₃N₄ based photocatalysis.⁴³

Herein, we demonstrated ultrasonication mediated synthesis of ultra small AuNPs supported on g-C₃N₄ sheets and their superior catalytic properties. Nearly monodispersed ultra small gold nanoparticles with diameter of 1–3 nm were formed on carbon nitride sheets (CN_x) by ultrasound mediated reduction of HAuCl₄ in presence of carbon nitride quantum dots. The Au–CN_x composite exhibited superior catalytic activity toward the reduction of 4-NP in presence of NaBH₄. It can reduce 0.1 mM 4-NP solution in 15 s with NaBH₄. The Au–CN_x composite showed superior adsorption capacity for various dyes. The surface charge of the Au–CN_x sheets is responsible superior adsorption of cationic dyes (RhB, MB) at neutral medium and anionic dye (MR) in acidic medium. In addition, this catalyst showed superior catalytic activity towards the degradation organic dye such as RhB, MB, MR under UV/visible/sun light. Possible mechanism for degradation of dyes upon Au–CN_x catalyst under UV and visible light irradiation were proposed. The photo-degradation of RhB on the Au–CN_x surface takes place via N-de-ethylation processes and formation of N-de-ethylated products was confirmed by UV-visible and ESI-MS spectroscopy. The Au–CN_x catalyst is very efficient for removal of cationic as well as anionic organic dye from aqueous environment by its dual actions with adsorption and photo catalytic degradation of dye.

Experimental section

Preparation of g-C₃N₄

Formamide used as precursor for synthesized carbon nitride quantum dots by microwave mediated method.^52,53 Typically 30 ml of formamide was heated using microwave synthesizer for 2 h at 180 °C. Resulting brown coloured solution was evaporated at 180 °C in a rotary-evaporator to get bulk amount of black product. This product was washed with water, filtered and vacuum dried to obtained dry, solid g-CNQDs.

Preparation of Au–CN_x composite

5 mg of as prepared g-CNQDs was dispersed in 3 ml of water by sonication for 10 min. In another vial 0.032 g of HAuCl₄, xH₂O was taken in 5 ml of water. Next, the dispersed g-CNQDs and gold solution were mixed together and again sonicated for 15 min. Finally, the mixture was ultra sounded with ∼28 kHz frequency for 3 hours at 400 watt. After ultrasound the solid product was separated by centrifugation. After washing this solid product with water and vacuum dried, was taken for further characterization and others experimental procedure.

Catalytic reduction method of 4-nitrophenol

In this typical reaction procedure, 3.0 ml of 0.1 mM 4-NP was taken in a glass-vial followed by addition of 0.3 mmol of NaBH₄. Here NaBH₄ has taken in excess amount to avoid the concentration change throughout the reaction. As soon as NaBH₄ was added the color of the reaction solution changed to dark yellow from light yellow. Then 0.005 g of the nanocomposite catalyst was added to it. Within few second the colour of the solution was changed from dark yellow to colourless. After completion of the reaction, a little amount of ammonium chloride was added to neutralize the excess NaBH₄ and the catalyst was recovered after doing centrifugation for 20 min at 16 000 rpm. Here we have also studied the regeneration of catalytic activity of our separated catalyst.

Adsorption efficiency measurement

Adsorption activity of the prepared catalyst was measured on two cationic dye [Rhodamine B (RB), methylene blue (MB)] and anionic azo dye [methyl red (MR)]. Adsorption study of RB, MB was done in neutral medium and MR was done in acidic medium (pH 3). Typically 1 ml catalyst was taken and kept in oven at 80 °C for half an hour. Then desired concentration of dye was added and kept in a dark place with stirring for 12 hours. After reached in adsorption–desorption equilibrium solution was centrifuged at 1600 rpm and concentration of remaining dye was determined by UV-vis spectrometer. The percentage of adsorption was calculated using eqn (1)

And equilibrium uptake was calculated using eqn (2)

where C₀ is initial concentration of dye solution (mg L⁻¹) C_e is the equilibrium concentration of dye solution (mg L⁻¹) q_e is the equilibrium capacity of dye on absorbent (mg g⁻¹) V is volume of dye solution (L) and W is the weight of adsorbent (g) used.

Photocatalytic activity measurement

For degradation of RB and MR, UV light was illuminated with 84 watt of 250 nm wavelength light, for visible light 135 watt xenon lamp with cut off filter 3 (λ ≥ 400 nm) and all dye degradation in sunlight was done between 10 am to 3 pm under a clear sky. All photo catalysis experiment was done after complete adsorption of dye and pH was adjusted to 3 by adding 0.1 (M) H₂SO₄. The degradation of dye was monitored by UV-visible spectroscopy by withdrawn centrifuged (10 000 rpm, 5 min) dye solution from reaction mixture.

Result and discussion

The powder X-ray diffraction pattern of g-C₃N₄ and Au–CN_x nanocomposite is shown in Fig. 1. In both p-XRD pattern the peak at 2θ value of 27.3° corresponds to (002) plane of g-C₃N₄ with interlayer d-spacing 3.27 Å.⁵² The additional four well resolved peaks of Au–CN_x composite located at 2θ values 38.2°, 44.4°, 64.5°, 77.54° and 81.8° can be indexed as (111), (200), (220), (311) and (222) reflection planes respectively for pure f.c.c. metallic Au nanoparticles⁴³ (JCPDS no. 04-0784). This suggests that the formation of AuNPs on CN_x sheets by ultrasonic treatment. TEM (Fig. 2a and b) and AFM (Fig. S1a and b ESI†) measurements were extensively used to study morphology of carbon nitride sheets. It shows formation of two dimensional CN_x sheet during evaporation of aqueous solution of g-CNQDs. Recently we have reported the formation of 2-dimensional sheets on a solid substrate by evaporation induced self assembly and condensation of carbon nitride quantum dots.⁵² Highly ordered assembly of silver nanoparticles on carbon nitride sheets and their application in non-enzymatic detection of glucose/H₂O₂ was also reported by our group recently.³ TEM samples were prepared by evaporation of aqueous solution of Au–CN_x composite on TEM grid. Fig. 2c–e show TEM images of very small gold nanoparticles, highly dispersed on the carbon nitride sheets indicating very high loading of AuNPs on CN_x sheets. The size distribution of AuNPs as shown in Fig. 2f and the size of dispersed AuNPs in CN_x sheets are in the range 1–3 nm with mean particle size of 1.5 nm. In addition few larger gold particles of 6–8 nm size are also visible in TEM images. The SAED image of AuNPs, taken from Fig. 2d is shown in Fig. 2g. The interlayer spacing calculated from SAED image are 2.36, 2.03, 1.22, 0.946 Å which corresponds to (111), (200), (220), (311) and (331) planes of face centered cubic AuNPs respectively, based on powder X-ray data base (JCPDS no. 04-0784). This is in agreement with p-XRD data (Fig. 1) of Au–CN_x composites. The EDS spectrum (Fig. 2h) has taken from Au NPs–carbon nitride sheets in the image of Fig. 2c shows the presence of carbon, nitrogen and gold, also confirming deposition of AuNPs on carbon nitride sheets. In addition, EDS spectrum of Au–CN_x was also done on silicon wafer in order to calculate the amount of Au present in Au–CN_x sheets (Fig. S2, ESI†) and it was found ∼40 weight% Au present in the composite. The exact amount of gold loading in Au–CN_x composite was determined by Thermo Gravimetric Analysis (TGA) method. TGA curves of Au–CN_x and CN_x were shown in Fig. S3 (ESI†). It shows that 46 wt% gold present in this Au–CN_x composite. The Au–CN_x composite with different amount of gold loading were prepared by ultrasonic treatment of CN_x sheets and different amount of AuCl₄ salts. The maximum loading of gold in CN_x sheets we could achieve is 50 wt%. X-ray photoelectron spectroscopy (XPS) measurements were carried out to know the oxidation state of gold particle and chemical environment around carbon and nitrogen atoms in Au–CN_x. Fig. 3a shows the survey scan of the XPS spectrum which clearly indicates the presence of gold, carbon, nitrogen and oxygen atoms in Au–CN_x composite. The binding energy spectra of Au 4f are very sensitive to the chemical environment around the gold surface and size of gold nanoparticles. It is well known that the binding energy spectra of Au 4f of metallic gold NPs is appeared as doublet with binding energy 83.9 eV for 4f_7/2 and 87.3 for 4f_5/2 peak.⁵⁴ It was reported^55,56 by various groups that binding energy of Au 4f electron for very small gold cluster shifts towards higher binding energy in compared to the position of bulk as the size of the nanoparticles is reduced. Recently Mark Turner et al. reported⁵⁷ binding energy of Au 4f_7/2 electron for very small (1–2 nm) gold nanoparticles was 1.1 eV higher than that of bulk Au. This typical shift in binding energy for ultra small metal nanoparticles on various supports materials was also reported.^55,56 The reduced core-hole screening in small metal nanoparticles and temporal charging^58,59 of nanoparticles during photo emission process is responsible for this shift. The electronic properties are significantly different for these very small particles and unusual catalytic properties were observed due to size dependent alteration of electronic structure. David P. Anderson et al. reported⁶⁰ Au 4f_7/2 peak for Au₈, Au₉, Au₁₁ clusters dispersed on titanium oxide were appeared in the region 85–86.2 eV whereas for Au₁₀₁ cluster appeared at 83.9 eV. Fig. 3b display the Au 4f spectrum for Au–CN_x and two peaks appeared at 85.6 and 88.9 eV. Thus, peaks at 85.6 and 88.9 eV can be assigned to au 4f_7/2 and 4f_5/2 photoelectron. The binding energy of Au 4f electron shifts towards higher value due to very small Au NPs dispersed on CN_x. It is reported⁵² for C 1s XPS spectra of CN_x that peaks due to C–N bonding are generally appeared in the range 285–287 eV and the peaks in the range 288–290 eV are attributed to CO_x. Fig. 3c and d show the XPS spectrum in the carbon 1s and nitrogen 1s peaks, which was de-convoluted to three and two main Gaussian peaks respectively. The binding energies of 285.3 eV and 286.45 eV are due to the presence of (sp²) N–C]–N (sp²) (carbon bonded to two nitrogen atoms) and C(–N)3 (planar trigonal carbon geometry) respectively, whereas the peak at 288.8 eV is attributed to CO.⁵² The peak at 400.1 eV refers to the presence of (sp²) C–N1]C (sp²), nitrogen bonded to two carbon atoms or pyridone moiety whereas the second peak in this region at 401.6 eV corresponds to quaternary nitrogen which is bonded to three sp² carbon atoms, known as graphitic nitrogen.^52,61 Based on p-XRD, TEM, SAED, EDS and XPS measurements it can be concluded that ultra small (1–3 nm), naked gold nanoparticles are highly dispersed on carbon nitride sheets.

Fig. 1 p-XRD of g-C₃N₄ and Au–CN_x composite.

Fig. 2 (a and b) TEM images of g-carbon nitride. (c, d and e) TEM images of Au–CN_x composite shows ultra small Au NPs are highly dispersed on the CN_x sheets. (f) Size distribution plot of Au NPs taken from (d). (g and h) SEAD and EDS spectra of Au–CN_x composite respectively, has taken from (c).

Fig. 3 (a) XPS spectra of 4f orbital of Au. (b and c) The de-convoluted spectra of 1s carbon (C) and nitrogen (N) of Au–CN_x respectively.

Fig. 4 (a) UV-visible spectra for the reduction 4-NP catalysed by Au–CN_x composite. 4-NP is represented by green colour curve, red color curve is in presence of only NaBH₄ without catalyst and blue color curve represent reductive product 4-AP. (b) The stability of the Au–CN_x catalyst with 5 successive cycles keeping the conditions same at every cycles.

Catalytic reduction of 4-nitrophenol

The catalytic activity of ultra small AuNPs supported on CN_x sheets was first tested for reduction of 4-NP was monitored by observing the change in UV-vis absorption spectra as well as colour change of aqueous solution. UV-vis absorption spectrum of 4-NP solution shows in Fig. 4a. Aqueous solution of 4-NP shows an absorption maxima at 317 (Fig. 4a green line) but after addition of NaBH₄ the peak shifted to 400 nm with increasing the intensity (Fig. 4a red curve) due to the formation of phenolate ion which is fully conjugated with benzene moiety. The yellow color of 4-nitrophenolate ion immediately bleached to colorless solution on addition of Au–CN_x catalyst and absorption maxima at 400 nm diminished with appearance of a new peak at 298 nm (Fig. 4a blue curve). This confirmed the reduction of 4-NP and formation of 4-AP.²⁷ Here, Au–CN_x composite reduced 4-NP completely (100%) in 15 s which is comparable to other reported values, given in Table T1 (ESI†). Recently M. Antonietti and his co-worker reported⁶² reduction of 4-NP with AuNPs doped mesoporous carbon nitride (m-CNR–Au), only 96% conversion was happened with NaBH₄ in 300 s. But the Au–CN_x catalyst take only 15 s for full conversion of 4-NP. The superior catalytic activity of this Au–CN_x composite is probably due to the presence of ultra small, naked gold particle on CN_x sheet. The stability of the catalyst was studied by doing the measurement repeatedly at the same condition. Catalyst can be successfully recycled by centrifugation, followed by washing with water and dried in vacuum. This catalyst can be used for at least 5 successive cycles with high conversion efficiency (Fig. 4b). This proves that catalyst has good catalytic efficiency and stability.

Adsorption of organic dyes

The cationic dyes (RhB, MB) consider as primary toxic pollutant in water resource. The Au–CN_x can be used as adsorbent to remove these dyes. Change of UV-vis absorption spectra was used to study the adsorption process on surface of Au–CN_x. Fig. 5a shows the UV-vis adsorption of 70 ppm RhB in aqueous solution. The absorption maximum of RhB at 552 nm was monitored for adsorption process on the Au–CN_x. The corresponding decay plot (C/C₀ vs. t) of RhB is shown in Fig. 5b. It suggests that the Au–CN_x effectively removes 91% RhB dye within 380 min. Another cationic dye, MB also shows good adsorption on the Au–CN_x surface. Fig. 5d shows the change UV-vis adsorption spectra of MB solution with adsorption time in presence of Au–CN_x catalyst. This Au–CN_x catalyst can effectively remove 90% dye within 420 minutes. The versatility of the catalyst has been evaluated by measuring the adsorption capacity of textile an anionic dye (MR). MR has poor adsorption on Au–CN_x composite at neutral pH. However, the adsorption of MR over Au–CN_x was increased significantly at lower pH (pH 3). The change of UV-visible adsorption spectra of MR in presence of Au–CN_x catalyst with different time is shown in Fig. S4a (ESI†) and the corresponding decay plot (C/C₀ vs. t) is also shown in Fig. S4b (ESI†). Adsorption isotherm demonstrated the actual distribution of dye in equilibrium between liquid and solid adsorbent. The adsorption isotherm model of different dye was established by batch experiment where a known amount of Au–CN_x catalyst was dispersed in different known concentration of dye in dark with contentious stirring for certain time to establish the equilibration between adsorbent and adsorbed. The interactive behaviour of dyes and Au–CN_x was described by the different appropriate adsorption model including Freundlich, Langmuir and Sips isotherms,⁶³ as described below.

Freundlich isotherm q_e = K_FC_e^1/n (a)

Langmuir isotherm q_e = q_mbC_e/(1 + K_LC_e) (b)

Sips isotherm q_e = q_mbC_e^1/n/(1 + K_sC_e^1/n) (c)

Freundlich isotherm model described the ratio of concentration of solute adsorbed on surface and solute in liquid which is in contact with adsorbent (eqn (a)). Langmuir model describes the chemisorptions of solute on the surface of adsorbent and only monolayer of solute was formed with maximum limit (eqn (b)). The Sips model is an improved model of Freundlich and Langmuir. According to sips model solute follow the Freundlich model at lower concentration and Langmuir model at higher concentration (eqn (c)). The experimental data for adsorption of RhB and MB on Au–CN_x composite were fitted with different adsorption isotherm models (Fig. 5c and f). The fitting parameters of RhB, MB and MR are shown in Table 1, T2 and T3 ESI† respectively. On the basis of R² value, the experimental data of RhB, MB and MR follow the sips isotherm suggesting multilayered adsorption of RhB, MB and MR. The maximum adsorption capacities for RhB, MB and MR were found to be 400 mg g⁻¹, 250 mg g⁻¹ and 130 mg g⁻¹ respectively. There were few reports on the adsorption of RhB dye in literature.^64,65 Recently K. S. Kim reported⁶⁶ reduced graphene oxide based adsorbent for RhB adsorption with maximum capacity 30 mg g⁻¹. But our Au–CN_x show comparable adsorption capacity with other reported value. Generally dye molecules are adsorbed on the adsorbent by two ways (a) by electrostatic interaction between solute and adsorbent and (b) π–π interaction of aromatic ring of dye molecule and the conjugated domain of the adsorbent. Graphitic carbon nitride doesn't have good adsorption property where as Au–CN_x shows superior dye adsorption properties. To understand the enhanced adsorption property of Au–CN_x compared to CN_x, zeta potential measurements of aqueous solution of C₃N₄ and Au–CN_x were done as a function of pH, shown in Fig. 6a and b. The average zeta potential value for CN_x sheets was +30 mV at neutral pH whereas value for Au–CN_x was −45 mV. RhB and MB are cationic dye at neutral pH and Au–CN_x catalyst has negatively charged at neutral pH. Therefore, Au–CN_x can adsorb RhB and MB at neutral pH due to electrostatic attraction between dye and the catalyst. But g-C₃N₄ has positive charge at neutral pH and these two dyes is also positively charged. The CN_x sheets show poor adsorption due to electrostatic repulsion of same chare. However MR is negatively charged at neutral pH. Thus Au–CN_x doesn't show any adsorption at neutral pH. But at higher pH (pH = 3) MR gets positive charge and Au–CN_x is still negatively charged. So it shows significant adsorption at pH 3. Hence, here dye adsorption by Au–CN_x composite was mainly monitor by the electrostatic interaction.

Fig. 5 (a and d) UV-visible absorption spectra of aqueous solution of RhB and MB respectively at different time interval in presence of Au–CN_x catalyst. (b and e) Plot of C/C₀ vs. time for RhB and MB dyes respectively where C₀ is initial concentration of dyes (70 ppm) and C is the concentration at different time interval. (c and f) Fitting of different isotherm model with adsorption of RhB and MB on Au–CN_x surfaces respectively. All the experiment was done with 1 ml of 70 ppm dyes with 1 mg Au–CN_x catalyst.

Table 1 Parameters of different isotherm model of RhB dye

RhB dye	Isotherm models
Parameter	Freundlich	Langmuir	Sips
K (mg g^−1)	31	—	—
N	2.5	—	1.63
qmax (mg g^−1)	—	250	400
B	—	0.02	0.04
R^2	0.966	0.950	0.971

---

Fig. 6 (a and b) Change of zeta potential of g-CN_x and Au–CN_x composite with different pH respectively. (c and d) The adsorption–desorption cycle of RhB and MB respectively.

Desorption of adsorbed dyes

Recycling of the catalyst is one of the important properties of a good adsorbent in the economical point of view. In literature, there are several methods are available for desorption of dyes. Q. Wang and his co-workers⁶⁷ reported a method for desorption by annealing the adsorbed dye-BN hallow sphere at 300 °C for 2 h. Expensive solvent such as ethylene glycol was used for washing the catalyst to remove the dye by K. S. Kim⁶⁶ and his co-workers. Most of the reported desorption process need long time as well as hazardous materials and methods was also not eco-friendly. Recently H. Schönherr and his co-workers reported³⁴ an easy method for desorption of dye by changing the pH of the solution. The Au–CN_x catalyst can be reused after removing adsorbed dye from the catalyst simply by changing the pH of the solution with a mild sonication. Fig. 6c and d show the plot of percentage of adsorption–desorption at different cycle number for RhB and MB respectively. After complete adsorption at pH 6, when pH of the solution is increased to 12 by adding 1 (M) NaOH solution, the adsorbed dye gradually desorbed and maximum desorption occurred after a mild sonication. The 89% and 87% of maximum desorption was achieved for RhB and MB at pH 12 in the first cycle. After 4th consecutive absorption–desorption processes the desorption efficiency of RhB reduced to 81% and the small decrease of desorption efficiency (80%) of MB is also observed after 4th cycle. The de-adsorption process of dyes at high pH can easily be explained by the zeta potential of Au–CN_x and the charge of the dyes. At pH 12, the Au–CN_x surfaces is negatively charged with zeta potential of −55 mV and at the same time these two dyes has also negative charge. The electrostatic repulsion between negatively charged dye molecules and negatively charged Au–CN_x surfaces is responsible for superior desorption of the dyes. Thus it can be concluded that this Au–CN_x catalyst has an efficient recycling ability for the removal of these two cationic dyes.

Photocatalytic degradation of organic dyes

The photo catalytic activity of the Au–CN_x catalyst was evaluated by degradation of Rhodamine B (RhB), methylene blue (MB) and methyl red (MR) on irradiation of the visible light, UV light and natural sunlight. In typical photo catalytic reaction catalyst was dispersed in the dye solution and then kept in dark with continuous stirring for 1.5 hours before degradation. 130 watt visible light, 84 watt UV light and natural sunlight were used for degradation of organic dyes. A UV-visible absorption maximum at 544 nm was monitor for RhB degradation where as 644 nm and 591 nm were used for MB and MR degradation. Since the Au–CN_x composite showed good adsorption of dye, all the photo-catalytic dye degradation experiment was performed after 1.5 hours adsorption of the dyes on Au–CN_x catalyst in dark. Initial concentrations of dyes were 70 ppm. After 1.5 h of adsorption, the remaining dye in solution was considered as initial concentration (C₀) for photocatalysis degradation. Fig. 7a shows the change of UV-visible absorption spectrum of ∼30 ppm RhB under visible light irradiation. The corresponding changes of ratio, C/C₀ with time on irradiation of UV, visible and natural sunlight are given in Fig. 7b and the change of C/C₀ in absence of light (control experiment) is also given for comparison. Similar to the dye degradation the concentration of dye after 1.5 h adsorption was taken as C₀ for the control experiment (absence of light). It shows that ∼95% of dye was removed in presence of light in 180 minutes whereas only 15% of dye was removed in absence of light due to adsorption. It confirms that enhanced dye removal efficiency of Au–CN_x composite under light is due to photo-catalytic degradation of dye. The 98%, 90% and 82% of RhB dye were degraded in 120 minutes under visible, natural sunlight and UV light. Fig. 7b also contains the change of C/C₀ of dye under visible light in presence of CN_x. This indicates the very good photo stability of dye and poor degradation ability of CN_x. Fig. 7d and g shows the change of UV-visible absorption spectra of 35 ppm MB and 35 ppm MR under visible light respectively. The C/C₀ vs. time plot for under different conditions for MB and MR are shown in Fig. 7e and h. MB and MR also shows excellent photodegradation with Au–CN_x on light irradiation. The 81%, 96% and 90% of MB dye were degraded in 120 minutes under UV, visible and natural sunlight where as 82%, 96% and 92% of MR dye were removed in 150 minutes. The Au–CN_x composite with different amount of gold loading were prepared by ultrasonic treatment of CN_x sheets and different amount of HAuCl₄. The maximum loading of gold in CN_x sheets we could achieve is 80 wt%. The catalytic activity of AuCN_x composite is better when gold loading in Au–CN_x composite is 40–50%. All the measurements were done with 40 wt% Au–CN_x composite.

Fig. 7 Degradation of 30 ppm RhB, 35 ppm MB, and 35 ppm MR dye by Au–CN_x under irradiation of visible light, natural sunlight, and UV light. (a, d and g) The change of UV-visible absorption spectra of RhB, MB, MR over Au–CN_x catalyst on irradiation of visible light respectively. (b, e and h) Change of the concentration (C) of RhB, MB, and MR relative to their initial values (C₀) with time on irradiation of different lights. (c, f and i) The pseudo first order kinetics plots of RhB, MB, MR degradation under different UV, visible and natural sunlight respectively.

The rate of the photo oxidative degradation of different dye was measured by the Langmuir–Hinshelwood model in equation.

where r is reaction rate, K is the absorption coefficient, k is reaction rate constant, and C is the reactant concentration. If the reactant concentration is low these equation reduced to
where k is the pseudo-first order rate constant, C₀ and C are the initial concentration and concentration after time t respectively. The plot of ln(C₀/C) vs. t for RhB as shown in Fig. 7c gives a straight line going through origin with a positive slope (k). The rate constant for degradation of RhB are 0.024, 0.016 and 0.011 min⁻¹ in visible light, natural sunlight and UV light respectively. Fig. 7f and i shows the plot of ln(C₀/C) vs. t for MB and MR dye respectively. The rate constant for degradation of MB visible light, natural sunlight and UV light are 0.024, 0.017 and 0.014 min⁻¹ respectively whereas rate constant for MR dye is 0.02, 0.014 and 0.011 min⁻¹ under natural sunlight, visible light and UV light respectively. This confirmed that Au–CN_x is efficient catalyst for degradation of dye under irradiation of visible as well as UV light. The comparative study for degradation rate of different dye with Au–CN_x catalyst with different reported catalyst is given in Table T4 (ESI†). The photocatalytic rate constant of this Au–CN_x catalyst is comparable to other reported photocatalysts, although high concentration, 30 to 35 ppm of dyes was used in this case. The effect of concentration of dye on degradation rate constant was also studied. Fig. S5 (ESI†) display the plot of ln(C₀/C) vs. time for different concentration of RhB. The rate constant for 15, 30 and 45 ppm RhB are 0.036 min⁻¹, 0.024 min⁻¹, and 0.015 min⁻¹. As the concentration of dye increased, rate constant is decreased. The decreases of the rate constant with concentration is due to the less percentage of light properly contact with catalyst as because high concentration of dye hindrance the light. One of the most important properties of the catalyst was the photo stability and reusability of the catalyst. The reusability of the catalyst for degradation of 30 ppm RhB is shown in Fig. 8. The Au–CN_x catalyst easily degraded 95% RhB dye in 120 min upto 3^rd cycle but it take little more time after 4^th cycle. The catalyst has high photostability upto four cycles with high degradation ability. The recycled catalyst was well characterized by p-XRD. The p-XRD pattern (Fig. S8, ESI†) of recycled Au–CN_x catalyst confirms the presence of Au nano-particles on CN_x sheets. Au–CN_x catalyst photodegradation of dyes are highly sensitive to pH. The plot of degradation percentage vs. pH of 30 ppm RhB has shown in Fig. S6 (ESI†). Au–CN_x catalyst shows poor degradation efficiency in higher pH compared to neutral pH. But the degradation efficiency of RhB at lower pH (pH 3) little higher than neutral pH. So all the degradation experiment was done in acidic pH (pH 3). Since the catalyst have significant adsorption we have tested the photo degradation of dye of very high concentration of 250 ppm RhB after keeping 12 hours in dark for full adsorption of dye. Fig. S7a and b (ESI†) shows the change of UV-visible spectra and corresponding C/C₀ vs. time plot. The Au–CN_x showed excellent degradation in much higher concentration and it can easily degraded 85% within 300 min. Thus, this photocatalyst is effective for removal of low and very high concentration of organic dye by using its dual (adsorbent and photocatalyst) properties. The much higher degradation efficiency of Au–CN_x in compared to CN_x is probably due the hetero junction, made from AuNPs and CN_x sheets.

Fig. 8 (a) Reusability of Au–CN_x catalyst for degradation of 30 ppm RhB dye where conditions of each cycle are kept same.

In order to examine the effect of presence of ultra small AuNPs on CN_x sheets, fluorescence measurements were done to investigate separation efficiency of photo-generated electrons and holes. As fluorescence originated from the recombination of the photoinduced electron and hole in valence band, so fluorescence spectra was investigated to understand the separation efficiency of photo excited charge carriers of the photocatalyst. The fluorescence spectra of CN_x and Au–CN_x composite was shown in Fig. 9a. The highly blue fluorescence of the graphitic carbon nitride was quenched in Au–CN_x due to loading of AuNPs on the graphitic carbon nitride sheets. The actual relationship is that higher the fluorescence intensity higher is the recombination rate of the electron–hole and lower is the efficiency of photoexcited charge separation. The fluorescence emission spectra of CN_x at ∼435 nm represent the band gap of emission light energy equal to the ban gape of pure CN_x and this emission was quenched fully after formation of Au–CN_x composite. The less intensity of fluorescence emission spectra of Au–CN_x confirmed the formation of heterostructure where the photoexcited electron traps by the AuNPs and stabilized the photoinduced charge separation. To evaluate active species involved in this photocatalytic dye degradation we are studied the RhB degradation under visible light irradiation in presence of different scavenger. EDTA–Na complex is known for hole (h⁺) scavenger, isopropanol is OH radical (˙OH) scavenger and benzoquinone (BQ) is a superoxide radical scavenger. Fig. 9b shows the effect of scavenger on the degradation rate. It can easily be seen that rate of RhB degradations remain unchanged on addition of the EDTA–Na complex. But the rate of the photodegradation was decreased on addition of isopropanol and benzoquinone. It proved that the hydroxyl and superoxide radical are the reactive oxygen species (ROS) responsible for the photodegradation of the dye. Photodegradation mechanism of RhB dye was studied by UV-visible spectroscopy and LC-MS/MS spectroscopy. The photodegradation of RhB generally occurs by two ways⁶⁸ (a) the complete cleavage of chromophores, (b) N-de-ethylation of N-substituted groups on chromophores. The change of UV-visible spectra of RhB under visible light and their normalized spectra are shown in Fig. 10a and b respectively. It shows that the adsorption maxima of RhB showed hyposochromic shift from 544 nm to 499 nm with increasing the irradiation time. This hyposochromic behavior of absorption maxima is due to the consequence of N-de-ethylation process.⁶⁹ As shown in Fig. 10b, the different adsorption maxima for different de-ethylated intermediates. The 539 nm peak is due to N,N,N′-tri-ethylated Rhodamine; 522 nm for N,N′-di-ethylated Rhodamine, where as 510 nm and 500 nm are for N-ethylated Rhodamine and Rhodamine respectively. The formation of different intermediate of the de-ethylated products was also confirmed by LC-MS mass spectroscopy. Fig. 11a–e show the mass spectra of different intermediates which was formed during the photo degradation of RhB. Initially RhB shows only one peak at m/z 443 which corresponds to dechlorinated RhB. New peaks were appeared with m/z value at 415, 387, 359, and 331 with increasing the irradiation time. The differences between two consecutive values are 28 m/z that corresponds to mass of ethylene molecule. Based on the UV-vis and mass spectra it can be concluded that degradation of RhB on Au–CN_x catalyst occurred via de-ethylation process. If the conjugate ring structure is not destroyed then fully de-ethylated RhB should have a maximum at 498 nm with molar extinction co-efficient (Є_max) 70% of the initial extinction co-efficient (Є_max) of RhB. But intensity of the absorption peak at 498 nm of 30 ppm RhB after visible light irradiation of 120 minutes is reduced to 5% of the initial intensity of RhB. So it confirms that the degradation of RhB on Au–CN_x occurred via the N-de-ethylation process followed by complete mineralization of the chromophores. The schematic diagram of the proposed mechanism for de-ethylation process is shown in Scheme 1. RhB moiety was easily adsorbed on the surfaces of Au–CN_x through the N-alkylated site due to the electrostatic interaction. During the photo irradiation the surface adsorbed oxygen is reduced to form hydroxyl radicals which can easily take the aliphatic proton from N-ethyl to form ethylene and N,N,N′-tri-ethylated Rhodamine. Some of the N,N,N′-tri-ethylated Rhodamine can desorbed from the catalyst surface and then again adsorbed on AuNPs surfaces. In next step de-ethylation of N,N,N′-tri-ethylated Rhodamine occurs in the same way by the hydroxyl radical to form N,N′-di-ethylated Rhodamine and then released in solution. In the sequential de-ethylation processes Rhodamine B converted to de-ethylated Rhodamine. Finally, Rhodamine was degraded to form UV inactive products.

Fig. 9 (a) Fluorescence emission spectra of g-CN_x and Au–CN_x composite. (b) Scavenger effect on rate of 30 ppm RhB dye degradation.

Fig. 10 (a) UV-vis absorption spectra for degradation of RhB under irradiation of visible light. (b) Normalized absorption spectra of different N-de-ethylated intermediate products taken from (a).

Fig. 11 (a–e) Represent LC-MS mass spectra of different N-alkylated intermediate, formed during degradation of RhB.

Scheme 1 Schematic diagram of de-ethylation process of RhB on the surface of Au–CN_x composite under irradiation of light.

Based on the above experimental observations we proposed a mechanism for degradation of organic dye over Au–CN_x catalyst on irradiation of visible light as shown in Fig. 12a. The dye was first excited to dye* and then excited electron transfer to conduction band of CN_x to from cationic dye radical. At the same time on irradiation of visible light, CN_x also form exciton i.e. electron excited to conduction band from valence band leading to electron–hole charge separation. Now, the injected electron from dye and excited electron transport to AuNPs by interfacial electron transports. Then, surfaces adsorbed O₂ molecule reduced on the Au surface to form superoxide radicals which can easily reduced to hydroxyl radical in acidic medium. Finally, hydroxyl radical can easily degrade the dyes easily shown in Fig. 12a. It is well known that Au and Ag NPs show considerable absorption of UV light due to inter band transition (transition of 5d or 4d electrons to 6sp or 5sp band).⁷⁰ The degradation of dye over Au–CN_x catalyst under UV light irradiation can be explained by the UV light absorption property of the gold nanoparticles through inter band electron transition. Recently H. Zhu and his co-worker established the mechanism of phenol degradation by the Ag NPs supported on metal oxide under UV light irradiation through 4d to 5sp inter band transition³¹ and D. Wang and his co-workers also established the actual mechanism of UV light adsorption by the Au NPs supported on the metal oxide via inter band transition of 5d to 6sp.⁷¹ A proposed mechanism for dye degradation under UV light was depicted in Fig. 12b. On irradiation of UV light 5d electrons of gold nanoparticles of Au–CN_x composite was excited to 6sp band (inter band transition). Among those excited electrons some of the electrons are in the higher energy state than the energy (red line) of the reduction potential for O₂/O₂⁻. These electrons can easily able to reduce the adsorbed oxygen molecule into superoxide radicals which in acidic medium form hydroxyl radicals. The hydroxyl radicals are then available for degradation of dyes. The slight greater the rate constant value under visible light may be due to the synergetic effect of electron transport of supported g-carbon nitride and dye molecule where under UV light only AuNPs responsible for electron transport. The adsorption and photocatalytic degradation of organic dye (RhB) with other noble metal–CN_x composites such as Ag–CN_x were also studied. The Ag–CN_x didn't show any significant absorption and photocatalytic degradation of RhB. The zeta potential of Ag–CN_x was found to be +31 mV whereas Au–CN_x was −25 mV at neutral pH. The adsorption and photo-catalytic behavior of Ag–CN_x composite are similar to pure CN_x compound since both have positive surface charge. The negative surface charge on the catalyst in solution is important for strong adsorption and photocatalytic degradation of positively charged organic dye such as RhB and MB.

Fig. 12 A proposed mechanism of photocatalytic degradation of dyes (a) under sunlight or visible light and (b) under irradiation of UV light in presence of Au–CN_x catalyst.

Conclusion

In conclusion, we have demonstrated a facile synthetic route for the formation and very high loading of ultra small gold nano-particles on two dimensional carbon nitride sheets. This composite was found to be superior catalyst for the reduction of nitro phenol. We have also showed that the Au–CN_x composite has high adsorption capacity for cationic dye (RhB, MB) in neutral medium and anionic dye (MR) in acidic medium with good reusability and easy desorption process. The Au–CN_x composite is not only good adsorbent but also a very efficient heterogeneous photocatalyst towards the degradation of various dyes. The degradation of RhB, MB and MR over Au–CN_x catalyst was performed on irradiation of UV/visible light/natural sunlight. The controlled experiment proves that hydroxyl radical is main active species towards photocatalytic degradation of organic dye. The catalyst is also very efficient for removal of high concentration, 250 ppm dye due to its superior photo catalytic activity. This superior catalytic property of Au–CN_x is probably due to naked, ultra small AuNPs on CN_x sheets. Moreover, this Au–CN_x composite can be reused several times for photo-catalytic degradation and as well as for dye adsorption. Thus, this Au–CN_x composite is a suitable catalyst for water cleaning.

Acknowledgements

SB thanks DST, Govt. of India for financial support (project number: SR/S1/PC-10/2011). We would like to thank Prof. Shikha Varma (Institute of Physics, Bhubaneswar, India) for XPS measurement.

References

1. D. Astruc, F. Lu and J. R. Aranzaes, Angew. Chem., Int. Ed., 2005, 44, 7852–7872.
2. P. K. Jain, X. Huang, I. H. El-Sayed and M. A. El-Sayed, Acc. Chem. Res., 2008, 41, 1578–1586.
3. M. K. Kundu, M. Sadhukhan and S. Barman, J. Mater. Chem. B, 2015, 3, 1289–1300.
4. J. Tian, Q. Liu, A. M. Asiri, A. H. Qusti, A. O. Al-Youbi and X. Sun, Nanoscale, 2013, 5, 11604–11609.
5. J. Tian, Q. Liu, C. Ge, Z. Xing, A. M. Asiri, A. O. Al-Youbi and X. Sun, Nanoscale, 2013, 5, 8921–8924.
6. A. D. McFarland and R. P. Van Duyne, Nano Lett., 2003, 3, 1057–1062.
7. Y. Zhang, G. Chang, S. Liu, J. Tian, L. Wang, W. Lu, X. Qin and X. Sun, Catal. Sci. Technol., 2011, 1, 1636–1640.
8. M.-C. Daniel and D. Astruc, Chem. Rev., 2003, 104, 293–346.
9. J. Lin, C. He, Y. Zhao and S. Zhang, Sens. Actuators, B, 2009, 137, 768–773.
10. J.-D. Qiu, G.-C. Wang, R.-P. Liang, X.-H. Xia and H.-W. Yu, J. Phys. Chem. C, 2011, 115, 15639–15645.
11. B. Yoon, H. Häkkinen, U. Landman, A. S. Wörz, J.-M. Antonietti, S. Abbet, K. Judai and U. Heiz, Science, 2005, 307, 403–407.
12. J. M. Campelo, D. Luna, R. Luque, J. M. Marinas and A. A. Romero, ChemSusChem, 2009, 2, 18–45.
13. A. K. Geim, Science, 2009, 324, 1530–1534.
14. W.-J. Ong, L.-L. Tan, S.-P. Chai and S.-T. Yong, Chem. Commun., 2015, 51, 858–861.
15. T.-F. Yeh, S.-J. Chen and H. Teng, Nano Energy, 2015, 12, 476–485.
16. S. Guo and S. Dong, Chem. Soc. Rev., 2011, 40, 2644–2672.
17. Y. Zheng, J. Liu, J. Liang, M. Jaroniec and S. Z. Qiao, Energy Environ. Sci., 2012, 5, 6717–6731.
18. H. Gao, S. Yan, J. Wang and Z. Zou, Dalton Trans., 2014, 8178–8183.
19. W.-J. Ong, L.-L. Tan, S.-P. Chai, S.-T. Yong and A. R. Mohamed, Nano Energy DOI:10.1016/j.nanoen.2015.03.014.
20. Y. Wang, X. Wang and M. Antonietti, Angew. Chem., Int. Ed., 2012, 51, 68–89.
21. K. Maeda, R. Kuriki, M. Zhang, X. Wang and O. Ishitani, J. Mater. Chem. A, 2014, 2, 15146–15151.
22. J. Tian, R. Ning, Q. Liu, A. M. Asiri, A. O. Al-Youbi and X. Sun, ACS Appl. Mater. Interfaces, 2013, 6, 1011–1017.
23. Z. Wu, H. Gao, S. Yan and Z. Zou, Dalton Trans., 2014, 12013–12017.
24. W.-J. Ong, L.-L. Tan, S.-P. Chai and S.-T. Yong, Dalton Trans., 2015, 1249–1257.
25. Z.-F. Huang, J. Song, L. Pan, Z. Wang, X. Zhang, J.-J. Zou, W. Mi, X. Zhang and L. Wang, Nano Energy, 2015, 12, 646–656.
26. J. McCann and B. N. Ames, Proc. Natl. Acad. Sci. U. S. A., 1976, 73, 950–954.
27. N. Pradhan, A. Pal and T. Pal, Colloids Surf., A, 2002, 196, 247–257.
28. G. Chang, Y. Luo, W. Lu, X. Qin, A. M. Asiri, A. O. Al-Youbi and X. Sun, Catal. Sci. Technol., 2012, 2, 800–806.
29. M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann, Chem. Rev., 1995, 95, 69–96.
30. B. H. Tan, T. T. Teng and A. K. M. Omar, Water Res., 2000, 34, 597–601.
31. X. Chen, Z. Zheng, X. Ke, E. Jaatinen, T. Xie, D. Wang, C. Guo, J. Zhao and H. Zhu, Green Chem., 2010, 12, 414–419.
32. Z. Xiong, L. L. Zhang, J. Ma and X. S. Zhao, Chem. Commun., 2010, 46, 6099–6101.
33. S. H. Lin and C. F. Peng, Water Res., 1996, 30, 587–592.
34. S. Ghorai, A. Sarkar, M. Raoufi, A. B. Panda, H. Schönherr and S. Pal, ACS Appl. Mater. Interfaces, 2014, 6, 4766–4777.
35. W. Lei, D. Portehault, D. Liu, S. Qin and Y. Chen, Nat. Commun., 2013, 4, 1777.
36. H. L. Parker, A. J. Hunt, V. L. Budarin, P. S. Shuttleworth, K. L. Miller and J. H. Clark, RSC Adv., 2012, 2, 8992–8997.
37. X. Yao, X. Liu and X. Hu, ChemCatChem, 2014, 6, 3409–3418.
38. H. Sun, S. Liu, S. Liu and S. Wang, Appl. Catal., B, 2014, 146, 162–168.
39. R. Thapa, S. Maiti, T. H. Rana, U. N. Maiti and K. K. Chattopadhyay, J. Mol. Catal. A: Chem., 2012, 363–364, 223–229.
40. J. Zhang, M. Zhang, R.-Q. Sun and X. Wang, Angew. Chem., Int. Ed., 2012, 51, 10145–10149.
41. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2009, 25, 10397–10401.
42. Y. Wang, X. Wang, M. Antonietti and Y. Zhang, ChemSusChem, 2010, 3, 435–439.
43. 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, 6815–6819.
44. L. Ge, C. Han, J. Liu and Y. Li, Appl. Catal., A, 2011, 409–410, 215–222.
45. H. Xu, J. Yan, Y. Xu, Y. Song, H. Li, J. Xia, C. Huang and H. Wan, Appl. Catal., B, 2013, 129, 182–193.
46. X. Lü, J. Shen, Z. Wu, J. Wang and J. Xie, J. Mater. Res., 2014, 29, 2170–2178.
47. H. Li, Y. Liu, X. Gao, C. Fu and X. Wang, ChemSusChem, 2015, 8, 1189–1196.
48. X. Yang, H. Tang, J. Xu, M. Antonietti and M. Shalom, ChemSusChem, 2015 DOI:10.1002/cssc.201403168.
49. S. Zhang, J. Li, M. Zeng, G. Zhao, J. Xu, W. Hu and X. Wang, ACS Appl. Mater. Interfaces, 2013, 5, 12735–12743.
50. S. Zhang, J. Li, M. Zeng, J. Li, J. Xu and X. Wang, Chem.–Eur. J., 2014, 20, 9805–9812.
51. S. Zhang, J. Li, X. Wang, Y. Huang, M. Zeng and J. Xu, ACS Appl. Mater. Interfaces, 2014, 6, 22116–22125.
52. S. Barman and M. Sadhukhan, J. Mater. Chem., 2012, 22, 21832–21837.
53. M. Sadhukhan and S. Barman, J. Mater. Chem. A, 2013, 1, 2752–2756.
54. C. D. Wagner and G. E. Muilenberg, Handbook of X-ray photoelectron spectroscopy: a reference book of standard data for use in X-ray photoelectron spectroscopy, Physical Electronics Division, Perkin-Elmer Corp, 1979.
55. G. K. Wertheim and S. B. DiCenzo, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 844–847.
56. K. Takahiro, S. Oizumi, A. Terai, K. Kawatsura, B. Tsuchiya, S. Nagata, S. Yamamoto, H. Naramoto, K. Narumi and M. Sasase, J. Appl. Phys., 2006, 100, 084325.
57. M. Turner, V. B. Golovko, O. P. H. Vaughan, P. Abdulkin, A. Berenguer-Murcia, M. S. Tikhov, B. F. G. Johnson and R. M. Lambert, Nature, 2008, 454, 981–983.
58. H. Hövel, B. Grimm, M. Pollmann and B. Reihl, Phys. Rev. Lett., 1998, 81, 4608–4611.
59. G. K. Wertheim, S. B. DiCenzo and S. E. Youngquist, Phys. Rev. Lett., 1983, 51, 2310–2313.
60. D. P. Anderson, J. F. Alvino, A. Gentleman, H. A. Qahtani, L. Thomsen, M. I. J. Polson, G. F. Metha, V. B. Golovko and G. G. Andersson, Phys. Chem. Chem. Phys., 2013, 15, 3917–3929.
61. R. Arrigo, M. Havecker, R. Schlogl and D. S. Su, Chem. Commun., 2008, 4891–4893,  10.1039/b812769g.
62. X.-H. Li, X. Wang and M. Antonietti, Chem. Sci., 2012, 3, 2170–2174.
63. S. G. Muntean, M. E. Radulescu-Grad and P. Sfarloaga, RSC Adv., 2014, 4, 27354–27362.
64. Y. Dong, H. Lin, Q. Jin, L. Li, D. Wang, D. Zhou and F. Qu, J. Mater. Chem. A, 2013, 1, 7391–7398.
65. S. Zhang, X. Wang, J. Li, T. Wen, J. Xu and X. Wang, RSC Adv., 2014, 4, 63110–63117.
66. J. N. Tiwari, K. Mahesh, N. H. Le, K. C. Kemp, R. Timilsina, R. N. Tiwari and K. S. Kim, Carbon, 2013, 56, 173–182.
67. G. Lian, X. Zhang, S. Zhang, D. Liu, D. Cui and Q. Wang, Energy Environ. Sci., 2012, 5, 7072–7080.
68. T. Wu, G. Liu, J. Zhao, H. Hidaka and N. Serpone, J. Phys. Chem. B, 1998, 102, 5845–5851.
69. K. Yu, S. Yang, H. He, C. Sun, C. Gu and Y. Ju, J. Phys. Chem. A, 2009, 113, 10024–10032.
70. C. Voisin, N. Del Fatti, D. Christofilos and F. Vallée, J. Phys. Chem. B, 2001, 105, 2264–2280.
71. H. Zhu, X. Chen, Z. Zheng, X. Ke, E. Jaatinen, J. Zhao, C. Guo, T. Xie and D. Wang, Chem. Commun., 2009, 7524–7526.

---

Footnotes

† Electronic supplementary information (ESI) available: Materials details, and additional data. See DOI: 10.1039/c5ra04913j

‡ Tanmay Bhowmik and Manas Kumar Kundu equally contributed to this manuscript.

---