Introduction

Graphitic carbon nitride (g-C3N4) has been discovered very long ago, in the 1830s (Yew et al. 2016). Binary nitrogen and carbon materials are a new variety of carbonaceous materials doped with nitrogen. Since C and N are the most abundant elements in our planet, these covalent solids are environment friendly (Silva and Rojas, 2016). Graphite-like carbon nitride (g-C3N4) is the most stable allotrope of all the carbon nitride under ambient condition (Cheng et al. 2014). The g-C3N4 is a new two-dimensional material and has aroused a great deal of interest in recent decades (Liu et al., 2016a; Wang and Bi, 2015). It is environmentally friendly, with excellent biocompatibility and nontoxicity, earth abundant, and can be produced on a large scale at low cost (Cai et al. 2015; Xiong et al. 2016; Li et al. 2015).

The g-C3N4 has attracted many interests recently (Amiri et al. 2016). There are strong C–N covalent bonds within each layer of g-C3N4, instead of C–C covalent bonds present in graphite. The different layers are held together by weak van der Waals forces (Hatamie et al. 2018).

Graphitic carbon nitride, an analogue of graphene, has been of great interest due to the strong electron donor nature of nitrogen present in g-C3N4, which is absent in graphene (Gu et al. 2015). It is a defect rich, N-bridged poly (tri-s-triazine), whereby the defects and nitrogen atoms serve as active sites for electron conductivity (Lee et al. 2016). In addition, the presence of nitrogen in g-C3N4 itself plays an important role in improving the wettability of the electrode with the electrolytes and electron donor and provides more active reaction sites (Ansari and Cho, 2017). The tri-s-triazine ring structure and the high degree of condensation give rise to a medium band gap polymer, an indirect semiconductor (Liu et al., 2016b).

The conjugated aromatic tri-s-triazine polymer-like graphene tends to form p-conjugated planar layers (Wang et al. 2012). Due to its exceptional optical, thermal, electrical, mechanical, and chemical inert properties (Guan et al. 2015), g-C3N4 has drawn a huge attraction and is found in many applications, such as lithium-ion battery (Hou et al. 2014), water splitting (Ahmad et al. 2015; Xu et al. 2013; Gao et al. 2017), fuel cells (Mansor et al. 2016), degradation of pollutants (Cao et al. 2015; Zhu et al. 2014), conversion of carbon dioxide to fuels (Ye et al. 2015), NO decomposition (Sano et al. 2013), hydrogenation reactions (Wang et al. 2011), and fluorescent sensor (Rong et al. 2015).

Recently, application of g-C3N4 in electrochemical sensors has been reported by a few researchers (Medeiros et al. 2013; Liu et al. 2014; Zhang et al. 2014; Ma et al. 2015). Recent studies have shown that g-C3N4 can be used as an electrochemical sensing platform to study H2O2 (Liu et al., 2016a), mercuric ions (Sadhukhan and Barman, 2013), glucose (Tian et al. 2013), ascorbic acid, dopamine, uric acid, and adenine (Lee et al. 2016).

Synthetic colorants have been used to replace natural ones for many years in the food industry, because they show many advantages such as high stability to light, oxygen and pH, color uniformity, low microbiological contamination, and relative lower production costs (Gan et al., 2013a; Chao and Ma, 2015). Tartrazine is a synthetic organic food dye that can be found in common food products, such as beverages, candies, dairy products, and bakery products (Qiu et al. 2016; Taner Bişgin et al. 2015). However, tartrazine really has mutagenic and carcinogenic possibility due to the presence of azo group (–N=N–) (Wang and Zhao, 2015; Al-Degs 2009). So, high concentrations of tartrazine in our life can bring serious problems to human health and some studies show that tartrazine can cause the appearance of allergies, asthma, and childhood hyperactivity (Qiu et al. 2016; Wang and Bi, 2015; Yu et al. 2016).

Until now, different methods, such as high performance liquid chromatography (HPLC) (Al-Degs 2009; Hurst et al. 1981), spectrophotometry (Capitfin-Vallvey et al., 1997), high performance liquid chromatography–mass spectrometry (HPLC–MS) (Qiu et al. 2016), and capillary electrophoresis (Zhao et al. 2014), have been reported for the determination of tartrazine.

In recent years, electrochemical determination methods have also been proposed for the determination of tartrazine for their advantages of high sensitivity, simplicity, low cost, and convenience (Wang and Zhao, 2015; Arvand et al. 2016; Arvand et al. 2017). This method, such as graphene decorated with nickel nanoparticle-modified GCE (Gan et al., 2013a), graphene and mesoporous TiO2-modified carbon paste electrode (Gan et al., 2013b), graphene oxide and multi-walled carbon nanotube nanocomposite-modified GCE (Qiu et al. 2016), and nanogold-modified CPE (Ghoreishi et al. 2013), has been reported for electrochemical determination of tartrazine azo dye.

Photodegradation of azo dye with g-C3N4 was seen in many articles (Jiang et al. 2017; Tao et al. 2015; Yan et al. 2010), but until now, g-C3N4 without modification has not been used to identify this color by electrochemical method. In this work, for the first time, we applied g-C3N4 as a modifier for graphite pencil electrode (in this paper, graphite pencil is called graphite). Also, this electrode due to the existence of strong electron donor nitrogen group in g-C3N4 framework can increased the rate of the electron transfer for reduction reaction of tartrazine (Zhao et al. 2017) (Scheme 1). Therefore, this new electrode was used for electrochemical determination of tartrazine synthetic food dye in real samples.

Scheme 1
scheme 1

A general scheme for the reduction of tartrazine on the surface of g-C3N4/graphite in acidic medium

Full size image

Experimental

Materials

Melamine (Lobachemie Reagents company, India), tartrazine (Sigma-Aldrich), potassium chloride, potassium dihydrogen phosphate, and phosphoric acid (Merck) were purchased. All chemicals were of analytical grade and used without further purification. Phosphate buffer with pH 2.1 (0.1 mol L−1) was prepared using potassium dihydrogen phosphate and phosphoric acid. Tartrazine was dissolved into deionized water to prepare standard solutions and stored at 4 °C. Deionized water was used for the preparation of all solutions.

Apparatus

Electrochemical measurements were performed on a PGSTAT302 (Metrohm-Netherlands) electrochemical workstation. A platinum electrode, an Ag/AgCl electrode, and modified graphite electrode served as the auxiliary electrode, reference electrode, and working electrode, respectively. X-ray diffraction (XRD) patterns of the powder were collected with a scanning speed of 2°/min and a step of 0.02° (2θ) in the range from 10 to 80° on a Shimadzu (XRD-6000), X-ray diffractometer with Cu/Kα radiation (1.54 Å). The structural morphology of the g-C3N4 was investigated by using scanning electron microscopy (SEM) with Phenom ProX with a high performance. The Fourier transform infrared (FT-IR) spectroscopy was recorded using a Shimadzu IR Prestige-21 apparatus with the prepared powder diluted in KBr pellets.

Procedures

Synthesis of g-C3N4

The metal-free g-C3N4 powder was synthesized via direct pyrolysis of the melamine based on the report (Amiri et al. 2016). Typically, 5 g of melamine was put into an alumina crucible with a cover then heated in a muffle furnace for 4 h at 550 °C. After cooling down to room temperature, the pale yellow g-C3N4 were collected and milled into powder for further use. The g-C3N4 phase, morphology, composition, and structure were characterized by using SEM, XRD, and FT-IR spectroscopy.

Preparation of g-C3N4/Graphite Electrode

Firstly, the graphite electrode has been polished on a paper to obtain a shiny and smooth surface, and then rinsed with deionized water and acetone after polishing step. Then, the graphite electrode was dried at room temperature. Secondly, suspension (1 mg mL−1) of g-C3N4 was prepared from the dispersion of g-C3N4 powder in ethanol/water mixture (1:1), and sonication for 45 min. Finally, 5.0 μL of a 1 mg mL−1 of g-C3N4 suspension, as a modifier material, was deposited on to the graphite surface electrode and then allowed to dry at room temperature. The deposition step was repeated for ten times (50.0 μL) and uniform film of g-C3N4 material was formed on the surface of graphite electrode for electrochemical measurement.

Results and Discussion

Characterization of g-C3N4 Structure

Phase structure of the g-C3N4 was studied by XRD pattern, and the result along with its standard pattern is displayed in Fig. 1. Typically, g-C3N4 powder shows two diffraction peaks at about 27.3° corresponding to the stacking of the conjugated aromatic system, and a peak at about 13.1° corresponding to the interlayer structural packing (Amiri et al. 2016; Wang et al. 2009).

Fig. 1
figure 1

XRD pattern of g-C3N4

Full size image

The morphologies of as-synthesized g-C3N4 were characterized by SEM. The SEM images of the morphology and the microstructure of the as-prepared sample (Fig. 2) indicate the layered, stacked texture, and the smooth surface morphology of bulk g-C3N4 (Hong et al. 2015).

Fig. 2
figure 2

SEM image of g-C3N4

Full size image

In the FT-IR spectrum of pure g-C3N4 (Fig. 3), the broad band at 1200–1650 cm−1 is attributed to the stretching vibration modes of C−N and C=N in heterocycles (Vadivel et al. 2016). Furthermore, the band at 806 cm−1 is related to the breathing mode of the heptazine arrangement. The broad band at 3000–3300 cm−1 is due to the terminal NH2 or NH groups at the defect sites of g-C3N4 aromatic rings (Amiri et al. 2016).

Fig. 3
figure 3

FT-IR spectrum of g-C3N4

Full size image

Effect of pH

The electrochemical responses of 0.5 mmol L−1 tartrazine at g-C3N4/graphite electrode in different supporting electrolytes, with the same concentration (0.1 mol L−1), such as phosphoric acid and potassium di-hydrogen phosphate buffer solution, acetic acid and sodium acetate buffer solution, and hydrochloric acid solution, were examined by CV. The results show that phosphate buffer gives greater sensitivity to tartrazine. The solution of tartrazine was prepared in the referred supporting electrolyte and analyzed. Then, the influence of solution acidity on the reduction behaviors of 0.5 mmol L−1 tartrazine in 0.1 mol L−1 phosphate buffer in the range of pH 1–3 was tested. The results indicated that the reduction Ip of tartrazine increased with the pH increasing from 1.0 to 2.1, and then decreased when the pH was further increased up to 3 (Fig. 4a). So, the 0.1 mol L−1 phosphate buffer with a pH of 2.1 was chosen as the supporting electrolyte.

Fig. 4
figure 4

a Cyclic voltammogram of 0.5 mmol L−1 tartrazine in 0.1 mol L−1 phosphate buffer on the surface of g-C3N4/graphite in various pH in the range of 1.0–3.0. b Plot of the peak potential vs. pH

Full size image

As shown in Fig. 4b, by increasing pH, the reduction peak of tartrazine shifted toward a less positive potential, indicating that protons are involved in the reduction of tartrazine. The reduction Ep of tartrazine showed linear relationship with the solution pH. The regression equation for this relationship is obtained as:

$$ \mathrm{E},\mathrm{V}=\hbox{--} 0.055\mathrm{pH}\hbox{--} 0.229,\kern0.5em {\mathrm{R}}^2=0.9658 $$

The slope of peak potential curves versus pH for tartrazine was 55 mV/pH which indicates that equal numbers of electrons and protons are involved in the electro-reduction of tartrazine on the surface of the modified electrode.

Effect of Scan Rate

In Fig. 5a, the cyclic voltammogram of 0.5 mmol L−1 tartrazine has been showed in buffer solution with pH 2.1 on the surface of the g-C3N4/graphite electrode with various scan rates. In this figure, the maximum signal of reduction Ip was observed in 125 mV s−1, because at this voltage scan rates, the charge passed per unit time is greater and thus an increase in Ip is seen. The relation between peak currents and scan rates is approximately linear and the adsorption-controlled mechanism can be suggested for tartrazine reduction.

Fig. 5
figure 5

a Cyclic voltammogram of 0.5 mmol L−1 tartrazine in 0.1 mol L−1 phosphate buffer pH 2.1 on the surface of g-C3N4/graphite with various scan rates of 25, 35, 45, 55, 65, 75, 85, 95, 105, 115, and 125 mV s−1 (a → k). b Plot of Ip vs. scan rate

Full size image

Effect of Amount of g-C3N4

The reduction current of tartrazine at modified electrode can be affected by the amount of g-C3N4 on the electrode surface. This can be controlled by using the same volume (50.0 μL) of g-C3N4 suspensions with the different concentrations in the range of 0.4–1.4 mg mL−1 on the surface of graphite. Electrochemical behavior of 0.5 mmol L−1 tartrazine in 0.1 mol L−1 phosphate buffer with different concentrations of g-C3N4 in the range of 0.4, 0.6, 0.8, 1, 1.2, and 1.4 mg mL−1 were investigated by cyclic voltammetry. The results indicated that the reduction Ip of tartrazine enhanced with increase in the concentration of g-C3N4 up to 1.0 mg mL−1. Further increase, caused a decrease in the electrochemical response of electrode (Fig. 6). This behavior is related to the thickness of the film. If the film is thin, the amount of tartrazine adsorbed is small, and as a result, a small Ip is observed. When the film is thick, its conductivity gets reduced, and also, the g-C3N4 becomes unstable and could leave off the electrode surface. Therefore, 50.0 μL of 1.0 mg mL−1 of g-C3N4 suspension was selected as optimum amount for preparation of the modified electrode.

Fig. 6
figure 6

Effect of different amounts of g-C3N4 in the range of 0.4–1.4 mg mL−1 on the cyclic voltammogram of 0.5 mmol L−1 tartrazine in 0.1 mol L−1 phosphate buffer pH 2.1 with 125 mV s−1 scan rate

Full size image

Reproducibility and Stability of Graphite and g-C3N4/Graphite Electrode

The reproducibility and stability of graphite and g-C3N4/graphite were carried out in the solution of 0.5 mmol L−1 concentration level of tartrazine in buffer solution with pH 2.1 for 40 cycles. The relative standard deviation (RSD) of graphite and g-C3N4/graphite electrode after ten repetitions was 3.6 and 2.5%, respectively. These results show that g-C3N4/graphite electrode has a good stability and reproducibility for the determination of tartrazine (Fig. 7). The stability of g-C3N4/graphite electrode of its initial response from 2 months ago was investigated and RSD of electrode was measured 4.1%. The excellent long-term stability of g-C3N4 on the electrode surface and reproducibility of the g-C3N4/graphite make it attractive in the field of analytical applications.

Fig. 7
figure 7

Cyclic voltammogram of 0.5 mmol L−1 tartrazine in 0.1 mol L−1 phosphate buffer pH 2.1 on the surface of g-C3N4/graphite with 125 mV s−1 scan rate

Full size image

Cyclic Voltammetric Behaviors of Tartrazine at the Surface of Graphite and g-C3N4/Graphite

Regarding the chemical structure that is shown in Fig. 8, we clearly found that tartrazine contains azo group which is electrochemical active and can be reduced on g-C3N4/graphite electrode surface. Figure 9 indicates the electrochemical behavior of 0.5 mmol L−1 tartrazine on the surface of graphite (dashed line) and g-C3N4/graphite (solid line) electrodes by using cyclic voltammetry in 0.1 mol L−1 phosphate buffer with pH 2.1. On the surface of graphite, tartrazine exhibits a reduction Ep at approximately − 0.31 V and Ip − 71.8 μA. On the other hand, the reduction peak at a lower negative potential near − 0.34 V with Ip – 260 μA on the surface of g-C3N4/graphite is recorded and Ep at the g-C3N4/graphite shifted by about 30 mV toward negative values when compared with the unmodified graphite electrode. In comparison, at the g-C3N4/graphite, the voltammetric response of tartrazine was greatly improved. It can be correlated to decrease of overpotential at the surface of g-C3N4/graphite electrode. Also, this electrode, due to the existence of strong electron donor nitrogen group in g-C3N4 framework, can increased the rate of the electron transfer for reduction reaction of tartrazine and provide more electrochemical active site to improve the electrochemical activity of g-C3N4 (Zhao et al. 2017). Therefore, in this work, g-C3N4/graphite electrode was chosen for all electrochemical measurements, so that the tartrazine was measured in different concentrations on the surface of this electrode.

Fig. 8
figure 8

Chemical structure of tartrazine

Full size image
Fig. 9
figure 9

Cyclic voltammogram of 0.5 mmol L−1 tartrazine in 0.1 mol L−1 phosphate buffer pH 2.1 with 125 mV s−1 scan rate on the surface of a graphite (dashed line) and b g-C3N4/graphite (solid line)

Full size image

Calibration Curve and Detection Limit for Determination of Tartrazine

Under the optimized experimental conditions, differential pulse voltammetric (DPV) was used for quantitative determination of the tartrazine. In Fig. 10, a linear relationship is seen between reduction Ip and concentration of tartrazine in dynamic ranges of 1.0 × 10−7–1.0 × 10−5 mol L−1. The linear equation can be expressed as Ipc (μA) = 1.4961 C (μM) + 1.1959; (R2 = 0.9977).

Fig. 10
figure 10

a DPV of tartrazine in 0.1 mol L−1 phosphate buffer pH 2.1 on the surface of g-C3N4/graphite in concentrations of 1.0 × 10−7, 3.0 × 10−7, 5.0 × 10−7, 8.0 × 10−7, 1.0 × 10−6, 3.0 × 10−6, 5.0 × 10−6, 8.0 × 10−6, and 1.0 × 10−5 mol L−1. b Calibration plot of Ip vs. tartrazine concentration. DPV setting pulse amplitude, 100 mV; step potential, 15 mV; and modulation time, 50 ms.

Full size image

Figure 10 shows that the electrochemical Ip of tartrazine increased remarkably with the increase of tartrazine concentrations. The detection limit (LOD) was calculated as three times the standard deviations of the blank solution divided by the slope of the calibration curve (Sierra-Rosalesa et al. 2017). For tartrazine, the detection limit was found to be 0.21 μmol L−1. The result was compared with the reported electrodes which were shown in Table 1. It can be seen that this new sensing system exhibited lower detection limit for tartrazine than the electrodes reported in Table 1.

Table 1 Comparison of different electrodes in tartrazine detection
Full size table

The presented modifier exhibits the good analytical results, easy fabrication method, easy to use, and also, the preparation time and drying process is considerably short compare to the other electrodes. On the other hand, graphite electrode has many advantages such as low cost, user friendly, and ease in handling and disposability. The total of these properties has led to use of this electrode in the determination of tartrazine dye.

Real Sample Analysis

The electrochemical method is one of the identifying methods for determination of fraud in saffron. The proposed method in this paper was used to identify two samples of saffron fake powder without the standard logo of Iran that was purchased from a salesman in the city. For determination of saffron fake powder, 0.7 mg of it was introduced into a 25-mL volumetric flask and filled up with phosphate buffer 0.1 mol L−1 (pH 2.1) to desired volume and this process was repeated for another sample. Under the optimized conditions, the prepared test solution of saffron fake powder was detected on the surface of g-C3N4/graphite electrode by DPV. For each samples, during the potential sweep from − 0.1 to 0.35 V, the DPV curves were recorded and the good peak of tartrazine was observed. This indicates that the modified electrode can be used for practical application. The content of tartrazine was obtained by the standard addition method and recoveries of two samples were found 98.2 and 102.3% with the presented method for first and second saffron fake powder samples, respectively. This satisfactory recovery demonstrated that this method is very promising for detection of tartrazine in real samples, and as seen in Fig. 11, two saffron fake powder samples are content tartrazine dye.

Fig. 11
figure 11

Curves of two samples of saffron fake powder in the tartrazine plot a sample 1 (circle) and b sample 2 (star)

Full size image

Conclusions

The g-C3N4/graphite-sensitive sensor was conveniently created by facile processing route. The outstanding character of the proposed synthesis method lies in the green route and short reaction time. The maximum current at this working electrode depended significantly by the pH value, and therefore, we conclude that the g-C3N4/Tartrazine interaction is pH-dependent. It shows a good electrochemical activity toward the reduction of Tartrazine in an acidic medium at pH 2.1. In addition, this method is simple, environmental friendly, and economical for rapid and precision determination of trace amounts of Tartrazine in real samples. Also, the strong electron donor nature of nitrogen present in semiconductor-g-C3N4 framework provides more electrochemical active sites to improve the electrochemical activity of g-C3N4 and increased electron-transfer rate and therefore electrochemical response to dye was enhanced.