Preparation of luminescent graphitic C₃N₄ NS and their composites with RGO for property controlling

Abstract

Thin graphitic C₃N₄ nanosheets (g-C₃N₄ NS) with bright blue photoluminescence were prepared through an acid and alkali corrosion and ultrasonic-assisted method. g-C₃N₄ NS/reduced graphene oxide (RGO) sample which was prepared through a hydrothermal synthesis offered high separation and transfer efficiency of photogenerated carriers because of the thin g-C₃N₄ NS attached closely to the RGO. The introduction of RGO sheets provided prominent advantages for enhanced light absorption and promoted the transfer rate of photogenerated carriers from g-C₃N₄ NS. A g-C₃N₄ NS/RGO (3 : 2) composite not only functions as a sensor to detect Pb²⁺ and Cd²⁺ in an aqueous solution but also exhibits efficient visible-light photocatalytic performance. The application in a sensor is ascribed to the intercalation impact between thin g-C₃N₄ NS and heavy metal ions. Owing to the presence of more active functional groups (e.g., carboxyl and amino groups) on the surface of the g-C₃N₄ NS compared with the bulk form, the ions can be easily adsorbed onto g-C₃N₄ NS. The g-C₃N₄ NS/RGO composites are highly sensitive and selective in ion sensing, as well as having superior photocatalytic activity.

---

Introduction

In recent years, many events of heavy metal poisoning pollution and water pollution occur frequently, attracting the attention of society, which can be ascribed to several industrial activities. Lead and cadmium ions are detrimental to human and other living beings when it is accumulated in biological bodies, which have been widely studied and represent one of the main public health concerns.¹ Heavy metal poisoning damages the nervous system, alimentary system, hemopoietic system and so on. In order to reduce the presence of heavy metals in the environment, it is important to find a rapid, effective and “green” approach to detect them. Among the heavy metal ions, Pb²⁺ and Cd²⁺ are highly dangerous even at very low concentration as a result of bio-accumulation and bio-amplification.² Up to now, many methods have been developed for detecting Pb²⁺ and Cd²⁺, such as high performance liquid chromatography (HPLC), fluorescence spectrometry and electro-chemiluminescent immunoassay.^3–5 However, some of them demand expensive instrumentation and multi-step preparation. Comparatively, sensor analysis surpasses other analytical methods because of its low cost, easy operation and great visibility, which has been widely recognized as an ideal technique for lead and cadmium ion detection, owing to its remarkable sensitivity that allows the determination of heavy metal ions at trace and ultra-trace levels. Alternative sensors have been developed for monitoring cadmium and lead residues and pollutions in recent works.⁶ Researchers have focused on not only sensors but also efficient visible-light-driven photocatalysts for environmental remediation. It is necessary to develop active and stable visible-light responsive photocatalysts. Among the semiconductor photocatalysts employed in water purification, heterostructure composite photocatalysts can show high performance due to strong interfacial interaction and efficient light harvesting ability.⁷ However, there are few reports about a kind of heterostructure material which could not only degrade organic dyes under visible light irradiation but also detect heavy metal ions in solution.

Graphitic carbon nitride (g-C₃N₄) has attracted much attention owing to its chemical and thermal stability with an intermediate band gap (2.4–2.8 eV). As an organic semiconductor consisting of sp²-bonded carbon and nitrogen atoms with layered structure, g-C₃N₄ exhibits structural similarity to graphene but with different properties.⁸ Besides, the exceptional chemical stability, low cost, green precursors and in particular the metal-free feature endow g-C₃N₄ with great promise in photocatalysis. However, the photocatalytic efficiency of bulk g-C₃N₄ is limited because of the high recombination rate of photogenerated electron–hole pairs. g-C₃N₄-based heterostructure photocatalysts with improved physicochemical properties and high photocatalytic activities are increasingly required, such as g-C₃N₄/TiO₂, g-C₃N₄/BiOBr and so on. g-C₃N₄-based two-dimensional (2D) heterojunction photocatalysts have been new research areas in recent years, which exhibit enhanced photocatalytic activity, such as ultrathin nanosheet/g-C₃N₄ 2D–2D heterostructures and SnNb₂O₆ nanosheet/g-C₃N₄ 2D composites.^9,10 In particular, there is a great interest in combining g-C₃N₄ with graphene to improve its conductivity and catalytic performance. As a 2D macromolecular sheet of carbon atoms with a honeycomb structure, graphene has attracted much attention because of its outstanding thermal, optical, and electrical properties and wide range of applications in bio-sensing, capacitors and catalysis. The reduction of graphene oxide (GO) can also serve as a support material for various graphene-based composites to improve the performance of optoelectronic and energy conversion devices.

It is worth noting that usual g-C₃N₄-based composites are still limited in their application both in the degradation of organic dyes and the detection of heavy metal ions in solution. In g-C₃N₄ structure the interlayer spacing is 0.325 nm and the cavity is 0.713 nm.¹¹ It is likely that metal ions could intercalate into g-C₃N₄ by coordination with some N-atoms or insert into the interlayers in theory since the size of metal ions is smaller than the spacing between layers of g-C₃N₄.¹¹ The interactions between g-C₃N₄ and metal ions such as Pb²⁺, Mg²⁺, Al³⁺ and Pb²⁺ have been reported.^12–14 Zhang et al. reported a sensitive sensor for the electrochemical determination of trace Hg²⁺ by employing ultrathin g-C₃N₄ nanosheets as an enhanced sensing platform.¹⁵ Cheng et al. employed g-C₃N₄ to fabricate an electrogenerated chemiluminescence sensor which shows high selectivity to Cu²⁺ determination.¹⁶ Apart from the application of carbon nitride as a metal-free catalyst for photocatalysis, graphitic carbon nitride quantum dots (g-C₃N₄ NS) have been explored as two-photon absorption and ion probes for environmental and biological detection.¹⁷ g-C₃N₄ NS have the advantages of bright fluorescence, good stability, water solubility and excellent biocompatibility, making them good candidates for replacing traditional quantum dots. After coupling with graphene, electronic transmission capacity can be strongly enhanced due to the good conductivity of graphene. Fine dispersibility and similar structure could provide convenience for ion embedding. From the above discussion, g-C₃N₄ NS/graphene composites will be a growing research hotspot and their broad application potentials in the fields of materials will attract more attention.

In this study, we prepare graphitic-C₃N₄ nanosheets by an acid and alkali corrosion and ultrasonic-assisted method. Building from the above ideas, a sensitive Pb²⁺ and Cd²⁺ sensor has been constructed based on ultrathin g-C₃N₄ nanosheets/reduced graphene oxide (RGO) as a sensing medium in aqueous solution. Meanwhile, g-C₃N₄ NS/RGO composites exhibit efficient photocatalytic activity under visible-light irradiation. The introduction of RGO sheets provided prominent advantages for enhanced light absorption and promoted separation and transfer rate of photogenerated carriers from g-C₃N₄ NS, thus leading to effective separation of electron–hole pairs and consequently an improvement in photocurrent intensity and photocatalytic performance. The g-C₃N₄ NS/RGO composites can be easily modified on the surface of ITO electrodes. It was found that the g-C₃N₄ NS/RGO electrode demonstrates great affinity towards Pb²⁺ and Cd²⁺ and achieved a detection limit of 10⁻⁷ mol L⁻¹. This g-C₃N₄ NS/RGO composite system may become a novel and competitive sensor or/and photocatalyst for a broad range of applications.

Experimental

Chemicals

Polyethylene(diallyldimethylammonium chloride) (PDDA) solution, cadmium chloride hemi-pentahydrate (CdCl₂·2.5H₂O), and graphite powder were obtained from Aladdin. Ammonia solution (NH₃·H₂O) was purchased from Shanghai Chemical Reagent Company. Cobalt chloride hexahydrate (CoCl₂·6H₂O), lead chloride (PbCl₂), manganese dichloride tetrahydrate (MnCl₂·4H₂O), ferrous chloride tetrahydrate (FeCl₂·4H₂O), iron nitrate nonahydrate (Fe(NO₃)₃·9H₂O), magnesium chloride hexahydrate (MgCl₂·6H₂O), calcium nitrate tetrahydrate (Ca(NO₃)₂·4H₂O), copper sulfate pentahydrate (CuSO₄·5H₂O), aluminum nitrate nonahydrate (Al(NO₃)₃·9H₂O), and potassium permanganate (KMnO₄) were purchased from Tianjin Chemical Reagent Company. Nickel chloride hexahydrate (NiCl₂·6H₂O), copper sulfate pentahydrate (CuSO₄·5H₂O), zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O), chromium nitrate nonahydrate (Cr(NO₃)₂·9H₂O) and melamine were purchased from Sinopharm Chemical Reagent Co. Ltd, China. Hydrogen peroxide (H₂O₂), concentrated sulfuric acid (H₂SO₄), hydrochloric acid (HCl) and orthophosphoric acid (H₃PO₄) were obtained from Laiyang Chemical Reagent Company. All chemicals and solvents were used directly with no additional purification.

Synthesis of samples

Typically, bulk g-C₃N₄ powders were prepared by heating 3 g of melamine at 550 °C in a tube furnace and then keeping at this temperature for 2 h. Then, 0.2 g of bulk g-C₃N₄ was treated in a mixture of 15 mL of concentrated sulfuric acid (H₂SO₄) and 15 mL of nitric acid (HNO₃) for 4 h at room temperature. The mixture was then washed with deionized water several times, the as-obtained white product being porous g-C₃N₄. Second, 0.02 g of the porous g-C₃N₄ was dispersed in 30 mL concentrated NH₃·H₂O, and the mixed solution was stirred for about 30 min and then transferred into a Teflon-lined stainless steel autoclave, heated at 160 °C and sustained for 4 h. Finally, the porous g-C₃N₄ was peeled into porous nanosheets. Upon cooling down to room temperature, the precipitate was washed with water several times to remove adsorbed NH₃ molecules. Third, 10 mg of porous g-C₃N₄ nanosheets was dispersed in 50 mL water, and then treated with ultrasound for about 8 h. The as-obtained aqueous suspension was then centrifuged at 8000 rpm, and dialysis method was used with a dialysis bag to remove large-sized particles.

GO was synthesized using graphite powder by a modified Hummers method.¹⁸ Firstly, 1 mg mL⁻¹ g-C₃N₄ micro-sized particle solution was ultrasonically mixed with 2 mg mL⁻¹ GO solution for 2 h and then stirred for 2 h. GO nanosheets served as the surfactant to disperse the g-C₃N₄ nanosheets, leading to the formation of g-C₃N₄/GO hybrid nanosheets. Subsequently, the resultant stable suspension was put into a 50 mL Teflon vessel and then sealed in an autoclave and heated at 180 °C for 6 h. Furthermore, the color of the solution turned from dark brown into black, accompanied by outgassing. The mixture was washed with water several times. The resulting black solid was freeze-dried. A series of g-C₃N₄ NS/RGO composites with different weight ratios of RGO and g-C₃N₄ were prepared by changing the amounts of g-C₃N₄ NS (m(g-C₃N₄ NS) : m(GO) = 2, 1.5, 1.25, 1, 0.75 and pure g-C₃N₄ NS), which are denoted as CNG-2, CNG-1.5, CNG-1.25, CNG-1, CNG-0.75 and CN, respectively.

Instrumentation

The crystal structures of samples were identified using X-ray diffraction (XRD) (Bruker D8-Advance, Germany). Photocatalytic activity was assessed by the degradation of rhodamine B (RhB) under 300 W Xe lamp irradiation with a cutoff filter (λ ≥ 420 nm). UV-visible absorption spectra were obtained through the diffuse reflection method using a spectrometer (Hitachi U-4100, Japan). The morphologies of the samples were investigated by using scanning electron microscopy (SEM) (QUANTA 250 FEG) and transmission electron microscopy (TEM) with a JEOL-2100 microscope under an acceleration voltage of 200 kV. The photocurrent intensity and photoelectrochemical sensing process were carried out with a CHI 660D electrochemical workstation (Shanghai Chenhua Limited, China). The photoluminescence (PL) spectra of samples were detected with a Hitachi F-4600 spectrometer using an excitation wavelength of 325 nm.

Photoelectrochemical measurements for Pb²⁺ and Cd²⁺ sensing

The photoelectrochemical measurements were carried out using a CHI 660D electrochemical workstation. A conventional three-electrode cell using Ag/AgCl as reference electrode and Pt wire as the counter electrode was used here. The electrolyte was 0.1 M Na₂SO₄ aqueous solution without additive. This solution was degassed with highly pure nitrogen for 10 min before experiments. Each stock solution of Pb²⁺, Mn²⁺, Zn²⁺, K⁺, Fe²⁺, Cd²⁺, Ni²⁺, Fe³⁺, Cu²⁺, Mg²⁺, Ca²⁺, Al³⁺, and Cr²⁺ was prepared by dissolving appropriate amounts of the compounds PbCl₂, MnCl₂, Zn(NO₃)₂, KCl, FeCl₂, CdCl₂, NiCl₂, Fe(NO₃)₃, CuSO₄, MgCl₂, Ca(NO₃)₂, Al(NO₃)₃, and Cr(NO₃)₂ into deionized water, respectively. The stock solutions of Pb²⁺ and Cd²⁺ were further diluted whenever necessary.

Photocatalytic investigation

The photocatalytic performance of the catalyst was tested by degradation of RhB solution. The RhB solution of 10 mg L⁻¹ was first prepared. In each test, 0.01 g of catalyst was added to ∼25 mL of the RhB solution and sonicated for 10 min. Then, the suspension was magnetically stirred for one hour to achieve an adsorption–desorption balance. The degradation process was performed under visible light illumination. 2 mL of solution was taken out at regular intervals. Then, the supernatant was separated from the solution by centrifuging. The concentration of RhB was determined by monitoring the changes in the maximal absorbance at approximately λ = 554 nm characterized by a UV-visible spectrometer.

Results and discussion

Fig. 1a shows the UV-visible spectrum of thin g-C₃N₄ NS with an apparent absorption band at about 313 nm and the PL spectrum which indicates that the PL emission is a recombination process of g-C₃N₄ after the charge separation of the electron (in the conduction band) and the hole (in the valence band) upon photoexcitation. The g-C₃N₄ NS emit strong blue fluorescence at about 430 nm via the one-photon excitation mechanism.¹⁹ The morphology and microstructure of g-C₃N₄ NS were investigated by TEM and HRTEM observations. As shown in Fig. 1b, the diameter of the g-C₃N₄ NS ranges from 20 to 45 nm. The suitable size of g-C₃N₄ NS endows them with the possibility to interlaminate with GO nanosheets. The TEM images in Fig. 1b and 2a indicate that as-prepared g-C₃N₄ NS and GO nanosheets have similar laminar structure.

Fig. 1 UV-visible absorption and PL spectra of thin g-C₃N₄ NS (a) and TEM image of sample CN (b).

Fig. 2 TEM images of GO (a) and CNG-1.5 (b). The inset in (b) shows the HRTEM image of the corresponding sample.

The TEM image in Fig. 2b shows that the g-C₃N₄ NS/RGO sample has a wrinkled, porous and loose structure stacked by some nanosheets. From the HRTEM image (see the inset in Fig. 2b), nanosheets forming a layer-by-layer assembly framework were crinkly and crossing. By measuring the lattice parameters, the lattice fringes were observed to have a d spacing of about 0.33 nm, which corresponded to the typical (002) interlayer stacking distance of g-C₃N₄. Meanwhile, g-C₃N₄ nanosheets are tightly inserted in layers of RGO nanosheets, constituting a layer-by-layer structure. It is observed that a strong and coherent interface exists between g-C₃N₄ NS and RGO. The heterostructure formation can improve the transfer of electrons, which benefits the electrochemical sensing and photocatalytic activity.

XRD patterns were obtained to characterize the crystal structure of as-prepared pure g-C₃N₄, GO, and as-prepared CNG-1 and CNG-1.5 samples. Fig. 3a shows the XRD pattern of pure g-C₃N₄ NS which reveals a peak at 27.3° with high intensity, corresponding to the result of HRTEM image (inset in Fig. 2b), reflecting the interlayer stacking of atomic segments, which is indexed as (002) plane observed for graphitic materials (JCPDS 87-1526).²⁰ The peak at 13.1° can be indexed as the (100) plane related to in-plane structural packing motif.²⁰ No apparent peaks of graphene were found in the XRD pattern of sample CNG-1.5. This could be because the regular stacking of GO was destroyed during the reduction process and could be resolved by XRD. A sharp peak appears at about 11° in the pattern of pure GO indexed to the (002) plane of GO. With increasing amount of g-C₃N₄ NS, the characteristic peak of GO is gradually weakened for sample CNG-1.5. These results indicate that GO is partially translated into RGO during the hydrothermal process, further indicating the evolved graphitic structures. Fig. 3b shows the FTIR spectra of pure g-C₃N₄ NS and CNG-1.5. The FTIR spectrum of pure g-C₃N₄ presents strong characteristic absorption bands in the 1200–1650 cm⁻¹ region, which are attributed to the stretching modes of heterocycles.²¹ The breathing mode of triazine units at 806 cm⁻¹ was also observed. The peaks at about 1240, 1406, and 1631 cm⁻¹ correspond to the typical stretching modes of CN heterocycles.¹⁹ It is worth noting that the above characteristic peaks present in the spectrum of sample g-C₃N₄ NS/RGO demonstrate the successful hybridization of g-C₃N₄ with RGO. Particularly, the peak of the amide I band at 1675 cm⁻¹ implies the formation of covalent bonds between g-C₃N₄ and RGO,²² in favor of the stability of composites. Significantly, a new peak emerges at 1557 cm⁻¹, which is attributed to the skeletal vibration of the graphene sheets,²³ indicating the presence of these sheets in the g-C₃N₄/graphene composite.

Fig. 3 XRD patterns of pure g-C₃N₄, GO, and as-prepared CNG-1 and CNG-1.5 samples (a) and FTIR spectra of pure g-C₃N₄ and as-prepared CNG-1.5 sample (b).

A schematic illustration of the process for the preparation of g-C₃N₄ NS/RGO heterostructure nanocomposites is presented in Scheme 1. An alternative method for the preparation of thin g-C₃N₄ NS, it could be mainly divided into three steps. First, porous g-C₃N₄ was formed by acid corrosion. Then, the porous g-C₃N₄ was exfoliated into ultrathin nanosheets via alkali treatment and ultrasound into thin g-C₃N₄ NS in water. The bulk g-C₃N₄ can be broken into thin nanosheets finally. Subsequently, the as-prepared GO solution was added to g-C₃N₄ NS aqueous solution. The g-C₃N₄ NS/GO composites were synthesized via hydrothermal co-assembly. During the reduction process, g-C₃N₄ NS/RGO hybrid nanosheets were formed, which is equivalent to the formation of GO sheets via the self-assembly of GO nanosheets. It can be inferred that the π-stacking interactions and the hydrogen-bonding interactions between GO nanosheets and g-C₃N₄ NS are responsible for the formation of the g-C₃N₄ NS/RGO framework.

Scheme 1 Schematic illustration of preparation of g-C₃N₄ NS/RGO.

In Fig. 4a, after combination with GO, CNG-1.5 exhibits two peaks at about 230 nm and 310 nm, respectively, corresponding to the absorption bands of GO and g-C₃N₄ NS, which exhibits direct differences from pure g-C₃N₄ NS. The PL spectra of CN and CNG-1.5 samples are shown in Fig. 4b. Pure g-C₃N₄ NS (sample CN) has an absorption edge at about 460 nm, originating from its band gap of 2.7 eV. CNG-1.5 exhibits a similar absorption edge to CN. After the introduction of GO nanosheets, the optical absorption of CNG-1.5 composites in the visible-light region increases. Also, CNG-1.5 shows a higher absorption capacity for visible light than the CN sample indicating that GO nanosheets are conducive to enhance light absorption. It could be concluded that layer-by-layer assembly structure could contribute to enhance light absorption, leading to higher separation and transfer efficiency of photogenerated electron–hole pairs. A g-C₃N₄ nanosheet is tightly connected to a GO nanosheet, forming a sheet-on-sheet structure. A compact and interacted interface exists between GO and g-C₃N₄ suggesting heterostructure formation. This heterostructure can accelerate the electron transfer from g-C₃N₄ to GO, thus leading to development of photocatalytic activity. The electron–hole recombination of g-C₃N₄ can be restrained by the good electrical conductivity of GO, the visible-light utilization may be strengthened via light multi-reflection across the connected open framework, and the catalytic surface reactive sites can be increased through the high adsorption capability.

Fig. 4 UV-visible absorption spectra of CN and CNG-1.5 in solution (a) and UV-visible diffuse absorption spectra of CN and CNG-1.5 composite samples (b).

XPS spectra were measured to explore the elemental composition of CNG-1.5 composite and chemical states of C, N, and O elements in the sample. Fig. 5a shows the XPS survey spectrum for CNG-1.5 composite and proves the presence of C, N, and O elements. In Fig. 5b, The C 1s curve of CNG-1.5 can be divided into three peaks. The peaks at 284.8 eV and 287.8 eV correspond to the sp² C–C bond and sp²-bonded carbon in N-containing aromatic rings (N–C=N), respectively. The peak at 286.28 eV for the CNG-1.5 sample is newly present compared to pure g-C₃N₄, which could be attributed to the residual C–O bond from GO after the hydrothermal treatment²⁴ and the bonding formation of N atoms in g-C₃N₄ with the defect sp²-C and sp³-C atoms in graphene.²⁵ Besides, the main peak in the N 1s spectrum could be fitted to three different peaks at 398.47, 399.86, and 400.54 eV (Fig. 5c), corresponding to sp² aromatic N bonded to carbon atoms (C=N–C), tertiary N groups (N–(C)₃), and amino functional groups (C–N–H), respectively. The position of the peaks in the N 1s spectrum shifted to lower binding energies compared to pure g-C₃N₄, which proved the chemical environment of N element had changed.²⁶ Fig. 5d shows a peak at 531.5 eV in the O 1s XPS spectrum further confirming the existence of the C–O bond.

Fig. 5 (a) XPS survey spectrum of CNG-1.5, and high-resolution (b) C 1s, (c) N 1s and (d) O 1s spectra of CNG-1.5.

To further prove the efficient separation of photogenerated electrons and holes in g-C₃N₄ NS/RGO composites, the transient photocurrent responses of pure g-C₃N₄ (sample CN) and g-C₃N₄ NS/RGO composite (samples CNG-0.75, CNG-1, CNG-1.25, CNG-1.5, CNG-2) electrodes were recorded for several on–off cycles of irradiation. As shown in Fig. 6a, when the UV irradiation turns off, the photocurrent value rapidly decreased; conversely, when turned on, the value increased. This shows that the photogenerated electrons transferred to the back contact across the samples to produce photocurrent under UV irradiation. Among all the composites, CNG-1.5 exhibits the highest photocurrent response value, which means that CNG-1.5 possesses the fewest electron–hole pairs at the interface between GO and g-C₃N₄ NS. At the same time, photogenerated holes transfer to the NS surface where they are trapped by reduced species in the solution. However, the photocurrents decay in pure g-C₃N_4, CNG-0.75, CNG-1, CNG-1.25 and CNG-2 samples, which indicates that recombination processes are occurring. In previous reports, PL analysis was used to clarify the efficiency of photogenerated carrier trapping, transfer and separation and to explore the destiny of photogenerated electrons and holes in semiconductors,²⁷ since the PL emission stems from the PL measurement for g-C₃N₄ NS (sample CN) and g-C₃N₄ NS/RGO (CN, CNG-1.5 and CNG-2), as shown in Fig. 6b. Upon photoexcitation at 365 nm, the g-C₃N₄ exhibits a PL band at 460 nm. The PL band of CNG composites appears at about 460 nm as well as pure g-C₃N₄ NS, which is consistent with 2.7 eV. This strong peak was owing to a band–band PL phenomenon. This PL signal is attributed to exciton emission, which mainly results from the n–π* electronic transitions consisting of lone pairs of nitrogen atoms in g-C₃N₄.²⁸ The intensity of PL for CNG-1.5 is the lowest among the three composites, which indicates that the CNG-1.5 and CNG-2 composites have lower recombination rate of photogenerated electrons and holes under UV irradiation. The photogenerated electrons are excited from the valence band to the conduction band, holding back the direct recombination of charge carriers. We could conclude that GO nanosheets in g-C₃N₄ NS/RGO heterostructure composites could accelerate the separation and transfer rates of photogenerated electrons and holes. These graphene sheets provided electronic transmission channels and were considered as fine electron accepters due to the π-conjugation structure.²⁹

Fig. 6 Transient photocurrent responses of ITO electrodes made of CN, CNG-0.75, CNG-1, CNG-1.25, CNG-1.5, and CNG-2 composites in 0.1 M Na₂SO₄ solution under UV irradiation (a) and PL spectra of samples CN, CNG-1.5 and CNG-2 (b).

Recently, nanomaterials especially carbon materials have been designed as promising sensitive nanosensors based on an optical or electrical signal due to the interaction between the target analyte and nanomaterial. Photo-electrochemical sensing is a newly emerged analytical method which is regarded as an ideal substitute for electrochemical and fluorescence sensing. In Fig. 7a, it is seen that the CNG-1.5 composite has a strong photocurrent response and is considered to be a superior candidate for sensor applications. The performance in sensing of CNG-1.5 composites was tested by the sensing of Cd²⁺ and Pb²⁺. Herein, CNG-1.5 composites were prepared by depositing g-C₃N₄ NS on graphene nanosheets, and electrostatically connected to an ITO electrode by using positively charged PDDA. For the sake of exploring the selectivities to Cd²⁺ and Pb²⁺, Fig. 7a displays the effects of different metal ions (Pb²⁺, Ni²⁺, Fe³⁺, Cr²⁺, Cd²⁺, Zn²⁺, Mn²⁺, K⁺, Ca²⁺, Al³⁺, Mg²⁺, Fe²⁺ and Cu²⁺) with concentration of 50 μM on the photocurrent intensity of CNG-1.5 composites. Apparently, among the metal ions, Cd²⁺ and Pb²⁺ ions could greatly increase the photocurrent intensity of CNG-1.5 composites. Herein, the effects of Cd²⁺ and Pb²⁺ ion concentrations on the photocurrent intensity of CNG-1.5 composites were investigated, as shown in Fig. 7b and c, respectively. In Fig. 7b, a good linear relationship (R = 0.9917) can be seen between the value of photocurrent decrease (I − I₀)/I₀ and the logarithm of the concentration of Cd²⁺ over the range from 0.5 to 50 μM.³⁰ In Fig. 7c, a good linear relationship (R = 0.99664) can be seen when the logarithm of the concentration of Pb²⁺ ranges from 0.3 to 50 μM. I and I₀ are the photocurrent intensity of CNG-1.5 composites in the presence and absence of Cd²⁺ or Pb²⁺, respectively.

Fig. 7 Effects of various metal ions on photocurrent intensity of ITO/(PDDA/CNG-1.5) electrode in 0.1 M Na₂SO₄ solution. All the metal ions have concentrations of 50 μM (a). Plots of photocurrent increase [(I − I₀)/I₀] of ITO/(PDDA/CNG-1.5) versus log[Cd²⁺] (b) and log[Pb²⁺] (c).

Similar to graphite, g-C₃N₄ has a layered structure involving weak van der Waals interactions between the adjacent C–N layers.¹¹ g-C₃N₄ NS could provide more surface active sites. The g-C₃N₄ planes constructed via highly ordered triazine (C₃N₃) or tri-s-triazine (C₆N₇) units contain many coordination sites, which can interact with metal ions through the lone-pair electrons of nitrogen. Few-layer structure due to the short distance and thin nanosheets contributed to carrier fast transfer from the inside to the surface. Furthermore, periodic cavities exist in the layers and the conjugate diameter of middle cavity structure is about 0.325 nm.³¹ In theory, most metal ions could be inserted into g-C₃N₄ layers or filled into cavities by complexation with nitrogen atoms so that metal ions could interact with g-C₃N₄. Compared with other metal ions, Pb²⁺ and Cd²⁺ could easily be inserted in the layered structure, due to the fact that a large amount of free electrons existing in g-C₃N₄ could adsorb more Pb²⁺ and Cd²⁺ ions, thus increasing the photocurrent response intensity. On the one hand, the reason that CNG composites can be used to detect Cd²⁺ and Pb²⁺ ions is that ionic radius of Cd²⁺ and Pb²⁺ are much larger than those of other ions. Cd²⁺ and Pb²⁺ ions could be tightly attached to the tri-s-triazine units and occupy cavities, which provide more linked electron transportation channels. On the other hand, compared with the significant responses of Cd²⁺ and Pb²⁺, the responses of the hybrid sensor to other ions were much weaker, due to the fact that the amidogen groups of CNG favor binding with Cd²⁺ and Pb²⁺, affecting the specific interaction between Pb²⁺ and CNG. The increasing change can be attributed to the decreasing hole concentration in the RGO sheets.³² Graphene nanosheets could improve the electron transmission of g-C₃N₄ and assist in improving the response current in the detection process.

The photocatalytic performance of as-prepared CN and CNG samples with various weight ratios was tested by RhB degradation under visible-light irradiation as shown in Fig. 8a. A mixed solution of RhB and photocatalyst was stirred for 1 h to reach adsorption equilibrium before visible irradiation. Fig. 8a shows that the removal of RhB dye was nearly 29% by g-C₃N₄ NS (sample CN) under visible-light irradiation for 2 h due to the rapid recombination of photogenerated carriers. Comparing with pure g-C₃N₄ NS, all CNG composites exhibit higher degradation efficiency for RhB under the same conditions, indicating the existence of a synergistic effect between g-C₃N₄ NS and RGO. Among the CNG samples, CNG-1.5 reaches the highest degradation efficiency of about 80% in 2 h. In the CNG system, g-C₃N₄ NS act as the photocatalyst which could generate electron–hole pairs under visible light irradiation; and RGO sheets could improve the separation and transfer of photogenerated carriers. According to the tendency of degradation, to quantitatively determine the reaction kinetics of RhB dye photodegradation by the as-prepared samples, the experimental data of CNG CN and CNG composites were fitted with the pseudo-first-order model, ln(C₀/C) = kt, where k is the apparent first-order rate constant, C₀ is the initial concentration of RhB solution, C is the concentration of RhB solution at time t, and k is the kinetic constant.³³ The corresponding kinetic constants (k) were calculated and are plotted in Fig. 8b. The reaction rate constant of CNG-1.5 is 0.0105 min⁻¹, which is about 2.84 times higher than that of pure g-C₃N₄ NS. When the weight ratio of g-C₃N₄ NS to GO is less than 1.5, the photocatalytic activity of CNG gradually increases with an increase of g-C₃N₄ NS content. On further increasing the g-C₃N₄ content, the photocatalytic activity of CNG slightly decreases. Results of recycling degradation experiments using CNG-1.5 under visible light irradiation are shown in Fig. 9. After cycling four times for 2 h, the photocatalytic degradation efficiency of RhB is about 58.8%. The as-prepared CNG-1.5 shows a good catalytic stability, maintaining a similar level of reactivity under visible light irradiation. The slight decrease could originate from the inescapable loss of catalyst during the recycling process.

Fig. 8 Photocatalytic degradation of RhB (C₀ = 10 mg mL⁻¹) under visible-light irradiation (a) and corresponding kinetic simulation curves (b).

Fig. 9 Recycling runs in photodegradation of RhB for CNG-1.5 under visible light irradiation.

On the basis of the results above, a possible photocatalytic mechanism for as-prepared CNG composites under visible light irradiation is illustrated in Scheme 2. When visible light irradiates the CNG catalyst, the light could be directly absorbed by CNG and also scattered by its porous framework. Under visible-light irradiation, electrons (e⁻) are excited from the valence band (VB) populated by N 2p orbitals to the conduction band (CB) formed by C 2p orbitals of g-C₃N₄, creating holes (h⁺) in the VB. Subsequently, the photoexcited electrons transfer to the GO surface for O₂ reduction, due to the high conductivity and lower Fermi level of GO (−0.08 eV vs. NHE).²¹ g-C₃N₄ NS are immobilized on the surface of graphene sheets to form the layered composites, and the photogenerated electrons in the CB of g-C₃N₄ NS tend to transfer to graphene sheets due to their excellent electronic conductivity, resulting in hole–electron separation. The holes remaining in the VB of g-C₃N₄ can directly degrade the RhB molecules or react with the surface-adsorbed OH⁻ to generate ·OH reactive radicals for the degradation. However, when the content of g-C₃N₄ NS is low, photocatalytic active sites are reduced and the excess GO sheets stack together hindering the electronic transmission channels. In contrast, when the g-C₃N₄ NS content is greatly increased, there are not enough GO nanosheets to disperse leading to worse photocatalytic activity. Therefore, we conclude that the excellent photocatalytic activity of CNG could be attributed to three factors as follows. First, the large coherent interface between g-C₃N₄ NS and GO nanosheets can inhibit the recombination of photogenerated carriers. Second, layer-by-layer assembly structure improves visible-light absorption. Lastly, the GO nanosheets provide a large number of electronic transport channels for photogenerated electrons, promoting the separation and transfer rates of electron–hole pairs.

Scheme 2 Schematic diagram of photodegradation processes for CNG under visible-light irradiation.

Conclusions

In summary, thin g-C₃N₄ NS were prepared by an acid and alkali corrosion and ultrasonic-assisted method. A sensitive Pb²⁺ and Cd²⁺ heavy metal ion sensor has been constructed based on the ultrathin g-C₃N₄ NS/RGO as a sensing medium in aqueous solution. Meanwhile, g-C₃N₄ NS/RGO composites also exhibit efficient photocatalytic activity under visible-light irradiation. There exist covalent bonds between g-C₃N₄ and RGO. The RGO sheets act as conductive channels to provide prominent advantages for enhanced light absorption and promote separation and transfer rate of photogenerated carriers from g-C₃N₄ NS, thus leading to improvement in photocurrent intensity and photocatalytic performance. Among all the CNG samples, CNG-1.5 exhibits the highest photocatalytic performance for RhB degradation and sensing for Pb²⁺ and Cd²⁺. This novel g-C₃N₄ NS/RGO composite system not only opens the field of development of electrochemical sensors but also sheds new light on the optical properties of g-C₃N₄ NS.

Acknowledgements

This work was supported in part by the program for Taishan Scholars, the projects from National Natural Science Foundation of China (grant no. 51572109, 51501071, 51302106, 51402123, and 51402124).

References

1. J. Godt, F. Scheidig, C. Grosse-Siestrup, V. Esche, P. Brandenburg, A. Reich and D. A. Groneberg, J. Occup. Med. Toxicol., 2006, 1, 1.
2. D. Neagu, F. Arduini, J. C. Quintana, P. Di Cori, C. Forni and D. Moscone, Environ. Sci. Technol., 2014, 48, 7477.
3. H. Y. Cheng, C. L. Wu, L. H. Shen, J. H. Liu and Z. G. Xu, Anal. Chim. Acta, 2014, 828, 9.
4. Z. Li, Y. Wang, Y. N. Ni and S. Kokot, Sens. Actuators, B, 2015, 207, 490.
5. F. Cai, Q. Zhu, K. Zhao, A. Deng and J. Li, Environ. Sci. Technol., 2015, 49, 5013.
6. E. Nagles, V. Arancibia, C. Rojas and R. Segura, Talanta, 2012, 99, 119.
7. G. L. Li, Q. W. Chen and J. Lan, ACS Appl. Mater. Interfaces, 2014, 6, 22561.
8. J. Zhang, X. Chen, K. Takanabe, K. Maeda, K. Domen, J. D. Epping, X. Fu, M. Antonietti and X. Wang, Angew. Chem., Int. Ed., 2010, 49, 441.
9. D. Jiang, T. Wang, Q. Xu, D. Li, S. Meng and M. Chen, Appl. Catal., B, 2017, 201, 617.
10. Z. Zhang, D. Jiang, D. Li, M. He and M. Chen, Appl. Catal., B, 2016, 183, 113.
11. X. Wang, X. Chen, A. Thomas, X. Fu and M. Antonietti, Adv. Mater., 2009, 21, 1609.
12. X. Lu, H. Wang, Y. Yang and T. Liu, J. Mater. Sci. Technol., 2011, 27, 245.
13. Y. Wang, J. Yao, H. Li, D. Su and M. Antonietti, J. Am. Chem. Soc., 2011, 133, 2362.
14. J. Liu and Y. Lu, Chem. Mater., 2004, 16, 3231.
15. J. Zhang, Z. Zhu, J. Di, Y. Long, W. Li and Y. Tu, Electrochim. Acta, 2015, 186, 192.
16. C. Cheng, Y. Huang, X. Tian, B. Zheng, Y. Li, H. Yuan, S. Xie and M. F. Choi, Anal. Chem., 2012, 84, 4754.
17. J. H. Lee, C. S. Lim, Y. S. Tian, J. H. Han and B. R. Cho, J. Am. Chem. Soc., 2010, 132, 1216.
18. D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L. B. Alemany, W. Lu and J. M. Tour, ACS Nano, 2010, 4, 4806.
19. X. Zhang, H. Wang, H. Wang, Q. Zhang, J. Xie, Y. Tian, J. Wang and Y. Xie, Adv. Mater., 2014, 26, 4438.
20. B. Yao, Y. Chen, X. Zhang, W. Zhang and Z. Yang, Appl. Phys. Lett., 2003, 82, 280.
21. S. Yan, Z. Li and Z. Zou, Langmuir, 2009, 25, 10397.
22. Z. Tong, D. Yang, J. Shi, Y. Nan, Y. Sun and Z. Jiang, ACS Appl. Mater. Interfaces, 2015, 7, 25693.
23. C. Nethravathi and M. Rajamathi, Carbon, 2008, 46, 1994.
24. Z. Y. Zhang, F. Xiao, Y. L. Guo, S. Wang and Y. Q. Liu, ACS Appl. Mater. Interfaces, 2013, 5, 2227.
25. Z. H. Sheng, L. Shao, J. J. Chen, W. J. Bao, F. B. Wang and X. H. Xia, ACS Nano, 2011, 5, 4350.
26. H. Huang, S. Yang, R. Vajtai, R. Wang and P. M. Ajayan, Adv. Mater., 2014, 26, 5160.
27. H. Liu, Z. Jin and Z. Xu, Dalton Trans., 2015, 44, 14368.
28. V. N. Khabashesku, J. L. Zimmerman and J. L. Margrave, Chem. Mater., 2000, 12, 3264.
29. A. K. Geim, Science, 2009, 324, 1530.
30. E. Chow, E. L. Wong, O. Pascoe, D. B. Hibbert and J. J. Gooding, Anal. Bioanal. Chem., 2007, 387, 1489.
31. X. Wang, X. Chen, A. Thomas, X. Fu and M. Antonietti, Adv. Mater., 2009, 21, 1609.
32. J. D. Fowler, M. J. Allen, V. C. Tung, Y. Yang, R. B. Kaner and B. H. Weiller, ACS Nano, 2009, 3, 301.
33. Y. Liu, J. Wang, P. Yang and K. Matras-Postolek, RSC Adv., 2015, 5, 61657.

---