An Electrochemical Sensor of Poly(EDOT-pyridine-EDOT)/Graphitic Carbon Nitride Composite for Simultaneous Detection of Cd2+ and Pb2+

In this study, poly(2,5-bis(3,4-ethylenedioxythienyl)pyridine)/graphitic carbon nitride composites (poly(BPE)/g-C3N4) were prepared by an in situ chemical polymerization method. Composites were characterized by using Fourier transform infrared spectroscopy (FT-IR), ultraviolet–visible absorption spectra (UV–vis), X-ray diffraction (XRD), energy-dispersive X-ray spectroscopy (EDS), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). Furthermore, electrochemical sensors were applied for the electrochemical determination of Cd2+ and Pb2+ using the differential pulse voltammetry (DPV) method. The results indicated that 10 wt % poly(BPE)/g-C3N4 composite-modified electrode exhibited linear detection ranging from 0.12 to 7.2 μM and 0.08 to 7.2 μM for Cd2+ and Pb2+, with detection limits (S/N = 3) of 0.018 μM and 0.00324 μM. Interference analysis suggested that the 10 wt % poly(BPE)/g-C3N4-modified electrode can be applied for the detection of the Cd2+ and Pb2+ in real samples.


Introduction
Industrial waste water usually contains heavy metal ion-such as Zn 2+ , Cd 2+ , Pb 2+ , Cu 2+ and Hg 2+ , the accumulation of these heavy metal ions in the human body can cause many chronic diseases [1][2][3]. For example, an imbalance of Zn 2+ can reduce the amount of vitamin C and iron in the body, and cause iron deficiency anemia. Cd 2+ accumulation to a certain value in the human body will lead to renal failure. Both lead and mercury affect the brain and nervous system, and result in mental retardation and brain damage in children [4]. Therefore, developing a simple and sensitive analytical method to detect these heavy metal ions is essential [5].
Up to now, various techniques have been used for detection of heavy metals, including inductively coupled plasma mass spectrometry [6], inductively coupled plasma atomic emission spectrometry [7], atomic fluorescence spectrometry [8], atomic absorption spectrometry [9], surface enhanced Raman spectrometry, and electrochemical analysis technology [5,10]. However, most of these methods require expensive equipment, complicated procedures, and specialized training. Due to its high sensitivity, low operating cost, and fast response, electrochemical analysis has been considered as an effective technique for the determination of heavy metal ions. Among the current developed electrochemical approaches, anodic stripping voltammetry (ASV) has been widely accepted as one of

Materials
EDOT (3,4-ethylenedioxythiophene), n-butyl lithium, 3,6-dibromopyridazide, and ferric chloride were bought from Shanghai Chemical Reagents Company (Shanghai, China). All the other chemicals and solvents, including urea, zinc chloride, sodium acetate (NaAc), acetic acid (HAc), and chloroform were used as received without further purification. The 0.1 M acetate buffer solution (ABS) was obtain through mixing different volume ratio stock solutions of 0.1 M NaAc and HAc. Standard solutions of 1 × 10 −3 M of Cd 2+ and Pb 2+ were prepared by dissolving lead acetate and cadmium acetate in ultrapure water, respectively.

Instruments and Characterizations
The structure and properties of the composites were investigated by Fourier transform infrared spectroscopy (FT-IR), ultraviolet-visible absorption spectra (UV-vis), X-ray diffraction (XRD), energy-dispersive X-ray spectroscopy (EDS), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). The FT-IR spectra of the composites were recorded with a BRUKER-QEUINOX-55 FTIR spectrometer (Bruker, Billerica, MA, USA) using KBr pellets. The UV-vis spectra of the samples were recorded with a UV-vis spectrophotometer (UV4802, Unico, Dayton, NJ, USA). XRD patterns were obtained by using a Bruker AXS D8 diffractometer (Bruker, Billerica, MA, USA), scan range (2θ), which was 10-80 • with monochromatic CuKα radiation source (λ = 0.15418 nm). The SEM images were carried out on a scanning electron microscope (SEM, Hitachi, Chiyoda-ku, Japan, S-4800, operating voltage, 5 kV), with powdered samples scattered on the conducting resin. Before SEM imaging, the samples were sputtered with thin layers of aurum under vacuum. Transmission electron microscopy (TEM, Akishima, Tokyo, Japan, model 2100) was performed with an accelerating voltage of 100 kV. The elemental percentages of samples were measured using energy-dispersive X-ray spectroscopy, which was taken on a Leo1430VP microscope (Carl Zeiss Inc., Oberkochen, Germany) with operating voltage 5 kV.

Preparation of g-C 3 N 4
The graphitic carbon nitride samples were prepared by a pyrolysis method using urea as precursor [32]. Typically, 10 g of urea was loaded into a 40 mL crucible with a cover. The crucible was placed in a muffle furnace under air condition and heated to 500 • C with a heating rate of 10 • C/min. Then, it was heated at 500 • C for 2 h and at 550 • C for another 2 h. Finally, the crucible was naturally cooled to room temperature, and a pale-yellow powder was obtained. The sample was washed with deionized water and absolute ethanol three times and dried at 60 • C for 24 h.

Preparation of Monomer BPE
The monomer of BPE was synthesized based on the previous report and reaction process as shown in Scheme 1 [33]. EDOT (5.0 g, 35.2 mmol) was added to 100 mL of anhydrous THF. Upon cooling to −78 • C, n-BuLi was added to the solution drop-wise (14.7 mL, 2.4 M) and stirred for 60 min, then ZnCl 2 (4.8 g, 16.7 mmol) was added. The mixture solution was stirred at 0 • C for 60 min before Pd(PPh 3 ) 4 (1.0 g, 0.865 mmol) and 2,5-dibromopyridine (2.08 g, 8.78 mmol) in 25 mL THF was added. The reaction mixture was stirred at 80 • C for another 72 h. Finally, the mixture was cooled to room temperature, and the THF was removed under reduced pressure. The product was extracted using chloroform, then further purified by recrystallization by methanol to obtain canary yellow solid. The structure of the monomer was characterized by 1 H-NMR (in Figure 1). 1

Preparation of Poly(BPE)/g-C3N4 Composites
To prepare the poly(BPE)/g-C3N4 composites, a certain weight of g-C3N4 was dispersed in 20 mL chloroform with ultrasonication for 30 min. Monomer BPE dissolved in 10 mL chloroform was added and ultrasonic dispersion applied for another 30 min. After that, FeCl3 was dispersed in 10 mL chloroform and added to the above solution drop by drop as oxidant. The reaction was stirred under magnetic stirring for 24 h. Finally, the sample was washed several times with chloroform, methanol, and distilled water, and then dried in an oven at 60 °C for 12 h.
Different weight percentages of poly(BPE)/g-C3N4 composite were obtained with a similar method. The pure poly(BPE) was prepared by a similar method without the addition of g-C3N4.

Preparation of Modified Electrodes
The bare glassy carbon electrode (GCE) was polished with 0.3 and 0.05 μm alumina slurries in sequence. The GCE was modified by a simple casting method. Five microliters of poly(BPE)/g-C3N4 dispersion (1 mg/mL) was spread onto the surface of the cleansed electrode and left to dry at 40 °C. The different weight percentages of poly(BPE)/g-C3N4, pure poly(BPE), and g-C3N4-modified GCE were prepared using the same procedure.

Electrochemical Measurements
All electrochemical measurements were carried out on CHI660C electrochemical workstation (ChenHua Instruments Co., Shanghai, China). In the three-electrode system, composite-modified electrode, platinum electrode, and saturated calomel electrode were used, respectively, as the working electrode, counter electrode, and reference electrode.

Preparation of Poly(BPE)/g-C3N4 Composites
To prepare the poly(BPE)/g-C3N4 composites, a certain weight of g-C3N4 was dispersed in 20 mL chloroform with ultrasonication for 30 min. Monomer BPE dissolved in 10 mL chloroform was added and ultrasonic dispersion applied for another 30 min. After that, FeCl3 was dispersed in 10 mL chloroform and added to the above solution drop by drop as oxidant. The reaction was stirred under magnetic stirring for 24 h. Finally, the sample was washed several times with chloroform, methanol, and distilled water, and then dried in an oven at 60 °C for 12 h.
Different weight percentages of poly(BPE)/g-C3N4 composite were obtained with a similar method. The pure poly(BPE) was prepared by a similar method without the addition of g-C3N4.

Preparation of Modified Electrodes
The bare glassy carbon electrode (GCE) was polished with 0.3 and 0.05 μm alumina slurries in sequence. The GCE was modified by a simple casting method. Five microliters of poly(BPE)/g-C3N4 dispersion (1 mg/mL) was spread onto the surface of the cleansed electrode and left to dry at 40 °C. The different weight percentages of poly(BPE)/g-C3N4, pure poly(BPE), and g-C3N4-modified GCE were prepared using the same procedure.

Electrochemical Measurements
All electrochemical measurements were carried out on CHI660C electrochemical workstation (ChenHua Instruments Co., Shanghai, China). In the three-electrode system, composite-modified electrode, platinum electrode, and saturated calomel electrode were used, respectively, as the working electrode, counter electrode, and reference electrode.

Preparation of Poly(BPE)/g-C 3 N 4 Composites
To prepare the poly(BPE)/g-C 3 N 4 composites, a certain weight of g-C 3 N 4 was dispersed in 20 mL chloroform with ultrasonication for 30 min. Monomer BPE dissolved in 10 mL chloroform was added and ultrasonic dispersion applied for another 30 min. After that, FeCl 3 was dispersed in 10 mL chloroform and added to the above solution drop by drop as oxidant. The reaction was stirred under magnetic stirring for 24 h. Finally, the sample was washed several times with chloroform, methanol, and distilled water, and then dried in an oven at 60 • C for 12 h.
Different weight percentages of poly(BPE)/g-C 3 N 4 composite were obtained with a similar method. The pure poly(BPE) was prepared by a similar method without the addition of g-C 3 N 4 .

Preparation of Modified Electrodes
The bare glassy carbon electrode (GCE) was polished with 0.3 and 0.05 µm alumina slurries in sequence. The GCE was modified by a simple casting method. Five microliters of poly(BPE)/g-C 3 N 4 dispersion (1 mg/mL) was spread onto the surface of the cleansed electrode and left to dry at 40 • C. The different weight percentages of poly(BPE)/g-C 3 N 4 , pure poly(BPE), and g-C 3 N 4 -modified GCE were prepared using the same procedure.

Electrochemical Measurements
All electrochemical measurements were carried out on CHI660C electrochemical workstation (ChenHua Instruments Co., Shanghai, China). In the three-electrode system, composite-modified electrode, platinum electrode, and saturated calomel electrode were used, respectively, as the working electrode, counter electrode, and reference electrode.  Figure 2A represents the FT-IR spectra of g-C 3 N 4 , poly(BPE), and poly(BPE)/g-C 3 N 4 composites. In the spectra of g-C 3 N 4 , the band at 812 cm −1 is attributed to the bending vibration of the triazine ring modes out of plane. The weak peak at 890 cm −1 is assigned to the cross-linked heptazine deformation mode [34]. The fingerprint regions at the range of 1230-1650 cm −1 correspond to the stretching vibrations of C-N and C=N in heterocycles [35]. A broad vibration band appears in the region of 3000-3500 cm −1 , which can be assigned to the stretching vibrations of aromatic N-H bonds of the uncondensed amino group (-NH 2 ) [36]. For the pure poly(BPE), the band at 2800-3100 cm −1 corresponds to the aromatic C-H stretching vibrations. The peaks at 1439, 1358, and 1300 cm −1 are due to the C=C asymmetric stretching vibration and C-C stretching vibration in the poly(BPE) ring. The bands at 1663 cm −1 and 1658 cm −1 are C=N stretching in the pyridine. There are only a few discrepancies between the spectra of g-C 3 N 4 and poly(BPE)/g-C 3 N 4 composite, and it is probably because the peaks of g-C 3 N 4 and poly(BPE) at 1230-1650 cm −1 have been superimposed. The new weak band at 1081 cm −1 was observed after incorporating the polymer into the composite. This band is assigned to the presence of ν(C-O-C) in the ethylenedioxy group of EDOT, due to the strong interaction between poly(BPE) and g-C 3 N 4 [37][38][39].  Figure 2A represents the FT-IR spectra of g-C3N4, poly(BPE), and poly(BPE)/g-C3N4 composites. In the spectra of g-C3N4, the band at 812 cm −1 is attributed to the bending vibration of the triazine ring modes out of plane. The weak peak at 890 cm −1 is assigned to the cross-linked heptazine deformation mode [34]. The fingerprint regions at the range of 1230-1650 cm −1 correspond to the stretching vibrations of C-N and C=N in heterocycles [35]. A broad vibration band appears in the region of 3000-3500 cm −1 , which can be assigned to the stretching vibrations of aromatic N-H bonds of the uncondensed amino group (-NH2) [36]. For the pure poly(BPE), the band at 2800-3100 cm −1 corresponds to the aromatic C-H stretching vibrations. The peaks at 1439, 1358, and 1300 cm −1 are due to the C=C asymmetric stretching vibration and C-C stretching vibration in the poly(BPE) ring. The bands at 1663 cm −1 and 1658 cm −1 are C=N stretching in the pyridine. There are only a few discrepancies between the spectra of g-C3N4 and poly(BPE)/g-C3N4 composite, and it is probably because the peaks of g-C3N4 and poly(BPE) at 1230-1650 cm −1 have been superimposed. The new weak band at 1081 cm −1 was observed after incorporating the polymer into the composite. This band is assigned to the presence of ν(C-O-C) in the ethylenedioxy group of EDOT, due to the strong interaction between poly(BPE) and g-C3N4 [37][38][39].   Figure 2B represents the UV-vis spectra of g-C 3 N 4 , poly(BPE), and poly(BPE)/g-C 3 N 4 composites. The UV-vis spectra of g-C 3 N 4 shows characteristic absorption peaks from 270 to 430 nm, which are characteristic peaks of the carbon nitride. Furthermore, the absorption peaks of poly(BPE) appear at 348, 490, and 586 nm, which are assigned to the π-π* transition of the thiophene ring [39]. In the case of poly(BPE)/g-C 3 N 4 composites, aside from the characteristic peak of g-C 3 N 4 , the absorption peak appears at 490 nm for 10 wt % poly(BPE)/g-C 3 N 4 and 15 wt % poly(BPE)/g-C 3 N 4 . The results suggested that poly(BPE) was successfully incorporated into the g-C 3 N 4 matrix. Figure 2C exhibits the XRD patterns of poly(BPE), g-C 3 N 4 , and poly(BPE)/g-C 3 N 4 composites. The poly(BPE) shows broad characteristic peaks at about 2θ = 22.5 • , which is assigned to the π-π* stacking within the molecule. Moreover, the sharp diffraction peaks at 2θ ≈ 33 • , 35 • , 41 • , 49 • , and 54 • are associated with the FeCl 4 doping agent [38,39]. Diffraction peaks of g-C 3 N 4 are located at 27.1 • and 13.2 • , which can be indexed as (002) and (100) diffraction planes for graphitic materials (JCPDS 87-1526). The strongest diffraction peak of graphite appears at 2θ = 27.1 • , which is attributed to the planes of graphitic structures. The minor peak at 13 • corresponds to the hole-to-hole arrays of tri-s-triazine units. For the composites, no other peaks appeared, indicating the crystal structure was not changed. However, it can be seen that as the intensity of (100) planes of g-C 3 N 4 decreases, the amount of poly (BPE) increases, suggesting that the original ordered intralayer structures of g-C 3 N 4 probably suffered from damage in the presence of poly(BPE) [40,41]. Figure 2D depicts the EDS of g-C 3 N 4 and poly(BPE)/g-C 3 N 4 composites and poly(BPE). As shown in Figure 2D, the EDS spectrum of the pure g-C 3 N 4 samples reveal the existence of C and N elements [42]. In the spectrum of pure poly(BPE), the C, N, O, and S elements are the major chemical elements, and a small amount of Fe element was detected due to the addition of ferric chloride [38]. For the poly(BPE)/g-C 3 N 4 , except the C and N elements, the O and S elements were also detected, which provided powerful evidence for the existence of poly(BPE). In addition, as the percentage of poly(BPE) increases, the weight percentage of S element in the composite increases from 0.21% to 1.14%. Figure 3 presents the SEM and TEM images of g-C 3 N 4 , poly(BPE), and 10 wt % poly(BEP)/g-C 3 N 4 composite. As shown in Figure 3A,D, the sheet-like g-C 3 N 4 was like the fold of the sheet structure and appears as ultrathin and well-spread sheets. As shown in Figure 3B,E, the pure poly(BPE) exhibits uneven thickness of the irregular lump material. As shown in Figure 3C,F, slightly less poly(BPE)/g-C 3 N 4 (at 10 wt %) grows on the surface of g-C 3 N 4 compared to pure g-C 3 N 4 and poly(BPE), forming a net-like structure. Due to the richness of amino, g-C 3 N 4 can be easily dispersion by the method of ultrasonication. The BPE monomer can be adsorbed on the surface of sheet-like g-C 3 N 4 by the π-π aromatic interaction and electrostatic attraction. During the in situ chemical polymerization from the effect of ferric chloride, the monomer adsorbed on the surface of g-C 3 N 4 grew on g-C 3 N 4 to form poly(BPE) [43].

Structure Characterization of Poly(BPE)/g-C3N4 Composites
Materials 2018, 11, x FOR PEER REVIEW 6 of 16 Figure 2B represents the UV-vis spectra of g-C3N4, poly(BPE), and poly(BPE)/g-C3N4 composites. The UV-vis spectra of g-C3N4 shows characteristic absorption peaks from 270 to 430 nm, which are characteristic peaks of the carbon nitride. Furthermore, the absorption peaks of poly(BPE) appear at 348, 490, and 586 nm, which are assigned to the π-π* transition of the thiophene ring [39]. In the case of poly(BPE)/g-C3N4 composites, aside from the characteristic peak of g-C3N4, the absorption peak appears at 490 nm for 10 wt % poly(BPE)/g-C3N4 and 15 wt % poly(BPE)/g-C3N4. The results suggested that poly(BPE) was successfully incorporated into the g-C3N4 matrix. Figure 2C exhibits the XRD patterns of poly(BPE), g-C3N4, and poly(BPE)/g-C3N4 composites. The poly(BPE) shows broad characteristic peaks at about 2θ = 22.5°, which is assigned to the π-π* stacking within the molecule. Moreover, the sharp diffraction peaks at 2θ ≈ 33°, 35°, 41°, 49°, and 54° are associated with the FeCl4 -doping agent [38,39]. Diffraction peaks of g-C3N4 are located at 27.1° and 13.2°, which can be indexed as (002) and (100) diffraction planes for graphitic materials (JCPDS 87-1526). The strongest diffraction peak of graphite appears at 2θ = 27.1°, which is attributed to the planes of graphitic structures. The minor peak at 13° corresponds to the hole-to-hole arrays of tri-striazine units. For the composites, no other peaks appeared, indicating the crystal structure was not changed. However, it can be seen that as the intensity of (100) planes of g-C3N4 decreases, the amount of poly (BPE) increases, suggesting that the original ordered intralayer structures of g-C3N4 probably suffered from damage in the presence of poly(BPE) [40,41]. Figure 2D depicts the EDS of g-C3N4 and poly(BPE)/g-C3N4 composites and poly(BPE). As shown in Figure 2D, the EDS spectrum of the pure g-C3N4 samples reveal the existence of C and N elements [42]. In the spectrum of pure poly(BPE), the C, N, O, and S elements are the major chemical elements, and a small amount of Fe element was detected due to the addition of ferric chloride [38]. For the poly(BPE)/g-C3N4, except the C and N elements, the O and S elements were also detected, which provided powerful evidence for the existence of poly(BPE). In addition, as the percentage of poly(BPE) increases, the weight percentage of S element in the composite increases from 0.21% to 1.14%. Figure 3 presents the SEM and TEM images of g-C3N4, poly(BPE), and 10 wt % poly(BEP)/g-C3N4 composite. As shown in Figure 3A,D, the sheet-like g-C3N4 was like the fold of the sheet structure and appears as ultrathin and well-spread sheets. As shown in Figure 3B,E, the pure poly(BPE) exhibits uneven thickness of the irregular lump material. As shown in Figure 3C,F, slightly less poly(BPE)/g-C3N4 (at 10 wt %) grows on the surface of g-C3N4 compared to pure g-C3N4 and poly(BPE), forming a net-like structure. Due to the richness of amino, g-C3N4 can be easily dispersion by the method of ultrasonication. The BPE monomer can be adsorbed on the surface of sheet-like g-C3N4 by the π-π aromatic interaction and electrostatic attraction. During the in situ chemical polymerization from the effect of ferric chloride, the monomer adsorbed on the surface of g-C3N4 grew on g-C3N4 to form poly(BPE) [43].

Electrochemical Characterization of Poly(BPE)/g-C 3 N 4 Composites
The electrochemical activity of differently modified GCEs was evaluated using CV in a redox probe solution containing 5 mM [Fe(CN) 6 ] 3−/4− in 0.1 M KCl. As shown in Figure 4A, all the modified electrodes show well-defined redox peaks, which are related to the [Fe(CN) 6 ] 3−/4− redox processes. Compared to poly(BEP) and g-C 3 N 4 -modified GCE, the redox peak of the poly(BEP)/g-C 3 N 4 composite-modified GCE is obviously enhanced. This indicates that more electrochemically active sites were present on the surface of poly(BEP)/g-C 3 N 4 . Also, it is probably due to the excellent electron transfer property of poly(BEP) [44]. The peak-to-peak potential (∆E p = E anodicpeak − E cathodicpeak ) at 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE is about 110 mV, while those at the g-C 3 N 4 , 5 wt % poly(BPE)/g-C 3 N 4 , and 15 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE are 280, 247, and 155 mV, respectively (detailed data in Table 1). Meanwhile, the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE presents the largest background current and peak current compared to other modified GCEs, indicating that more electrochemical active sites present on the surface of 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE [45]. Under the Randles-Sevcik equation, I p = 2.69 × 10 5 n 3/2 ACD 0 1/2 ν 1/2 , where I p is the anodic peak current, n is the number of electrons transferred, A is electroactive surface area, C is concentration, ν is the potential scanning rate, and D 0 is the diffusion coefficient [44,46,47]. The effective surface areas of the g-C 3 N 4 , 5 wt % poly(BPE)/g-C 3 N 4 , 10 wt % poly(BPE)/g-C 3 N 4 and 15 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE are estimated to be 0.0781 cm 2 , 0.0851 cm 2 , 0.1234 cm 2 , and 0.0989 cm 2 , respectively. The 10 wt % poly(BPE)/g-C 3 N 4 has the largest effective surface area. Therefore, the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE could be used as an excellent sensor for electroactive species.

Electrochemical Characterization of Poly(BPE)/g-C3N4 Composites
The electrochemical activity of differently modified GCEs was evaluated using CV in a redox probe solution containing 5 mM [Fe(CN)6] 3−/4− in 0.1 M KCl. As shown in Figure 4A, all the modified electrodes show well-defined redox peaks, which are related to the [Fe(CN)6] 3−/4− redox processes. Compared to poly(BEP) and g-C3N4-modified GCE, the redox peak of the poly(BEP)/g-C3N4 composite-modified GCE is obviously enhanced. This indicates that more electrochemically active sites were present on the surface of poly(BEP)/g-C3N4. Also, it is probably due to the excellent electron transfer property of poly(BEP) [44]. The peak-to-peak potential (∆Ep = Eanodicpeak − Ecathodicpeak) at 10 wt % poly(BPE)/g-C3N4 composite-modified GCE is about 110 mV, while those at the g-C3N4, 5 wt % poly(BPE)/g-C3N4, and 15 wt % poly(BPE)/g-C3N4 composite-modified GCE are 280, 247, and 155 mV, respectively (detailed data in Table 1). Meanwhile, the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE presents the largest background current and peak current compared to other modified GCEs, indicating that more electrochemical active sites present on the surface of 10 wt % poly(BPE)/g-C3N4 composite-modified GCE [45]. Under the Randles-Sevcik equation, Ip = 2.69 × 10 5 n 3/2 ACD0 1/2 ν 1/2 , where Ip is the anodic peak current, n is the number of electrons transferred, A is electroactive surface area, C is concentration, ν is the potential scanning rate, and D0 is the diffusion coefficient [44,46,47]. The effective surface areas of the g-C3N4, 5 wt % poly(BPE)/g-C3N4, 10 wt % poly(BPE)/g-C3N4 and 15 wt % poly(BPE)/g-C3N4 composite-modified GCE are estimated to be 0.0781 cm 2 , 0.0851 cm 2 , 0.1234 cm 2 , and 0.0989 cm 2 , respectively. The 10 wt % poly(BPE)/g-C3N4 has the largest effective surface area. Therefore, the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE could be used as an excellent sensor for electroactive species.   Figure 4B shows the DPV of the differently modified GCEs in 0.1 M ABS (pH = 4.5) containing 2.0 μM Cd 2+ and Pb 2+ . As shown, the distance between the individual peaks is large enough, with individual peaks at approximately −0.834 V and −0.586 V for Cd 2+ and Pb 2+ , respectively. The peak current of GCE modified by 10 wt % poly(BPE)/g-C3N4 composite increased significantly, the main reason being that poly(BPE) not only effectively improves the electron transfer rate of the electrode   Figure 4B shows the DPV of the differently modified GCEs in 0.1 M ABS (pH = 4.5) containing 2.0 µM Cd 2+ and Pb 2+ . As shown, the distance between the individual peaks is large enough, with individual peaks at approximately −0.834 V and −0.586 V for Cd 2+ and Pb 2+ , respectively. The peak current of GCE modified by 10 wt % poly(BPE)/g-C 3 N 4 composite increased significantly, the main reason being that poly(BPE) not only effectively improves the electron transfer rate of the electrode surface, but also strongly interacts with the conjugated structure of g-C 3 N 4 by π-π stacking, which results from electrode materials having strong adsorption capacity [48].

Optimization of Experimental Conditions
To optimize the experimental conditions, simultaneous determination of 2.0 µM Cd 2+ and Pb 2+ at the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE under different pH values were evaluated (deposition potential: −1.4 V, deposition time: 210 s, pulse width: 50 ms; pulse period: 100 ms, increment potential: 2 mV). As shown in Figure 5A, maximum current responses appeared at pH = 4.5. The lower pH values (3.5 and 4.0) could results in a reduction peak current, which is possibly due to the protonation of the hydrophilic groups reducing the absorption of metal ions. The peak current at the higher pH values (5.0 and 5.5) decreased, which is possibly due to the hydrolysis of Cd 2+ and Pb 2+ . Thus, pH = 4.5 was chosen as the best condition for the electrochemical measurements. surface, but also strongly interacts with the conjugated structure of g-C3N4 by π-π stacking, which results from electrode materials having strong adsorption capacity [48].

Optimization of Experimental Conditions
To optimize the experimental conditions, simultaneous determination of 2.0 μM Cd 2+ and Pb 2+ at the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE under different pH values were evaluated (deposition potential: −1.4 V, deposition time: 210 s, pulse width: 50 ms; pulse period: 100 ms, increment potential: 2 mV). As shown in Figure 5A, maximum current responses appeared at pH = 4.5. The lower pH values (3.5 and 4.0) could results in a reduction peak current, which is possibly due to the protonation of the hydrophilic groups reducing the absorption of metal ions. The peak current at the higher pH values (5.0 and 5.5) decreased, which is possibly due to the hydrolysis of Cd 2+ and Pb 2+ . Thus, pH = 4.5 was chosen as the best condition for the electrochemical measurements. The effect of the deposition potential on the performance of the modified electrode was investigated in the range from −1.0 to −1.6 V, and the results are shown in Figure 5B. The maximum current peak could be observed at −1.4 V. However, the peak currents decreased gradually with the potential moving to the negative direction. Thus, the deposition potential of −1.4 V was chosen as the optimum value for detection of two heavy metals. Figure 5C exhibits the DPV current response of 2.0 μM Cd 2+ and Pb 2+ over the accumulation time of 60-390 s. As shown in Figure 5C, for the time of 60-210 s, the peak currents are almost linearly proportional to accumulation time, and this may be attributed to the fact that the amount of metal ions at the modified electrode surface greatly increases due to electrochemical deposition. When deposition time was above 210 s, the increase rate of the peak current of Pb 2+ changed, and this is probably due to the working electrode surface saturation. Under the consideration of sensitivity, a determination time of 210 s was selected for the deposition of the ions.

Individual Determination of Cd 2+ and Pb 2+
Under the optimized conditions, DPV was used as an analytical method for the electrochemical detection of Cd 2+ and Pb 2+ using various modified GCEs in 0.1 M ABS (pH = 4.5). Figure 6 exhibits the DPV responses of 10 wt % poly(BEP)/g-C3N4 composite-modified GCE toward Cd 2+ and Pb 2+ . The figure also shows the linear relationship between peak currents and concentrations of the two ions, and the inset shows as well as their linear equations and correlation coefficient (Figure 6 inset). The linear range of Cd 2+ is 0.1-6.8 μM with a detection limit of 0.0097 μM. The linear range of Pb 2+ is 0.1-6.4 μM with a detection limit of 0.00327 μM. Detailed results are shown in Table 2. The effect of the deposition potential on the performance of the modified electrode was investigated in the range from −1.0 to −1.6 V, and the results are shown in Figure 5B. The maximum current peak could be observed at −1.4 V. However, the peak currents decreased gradually with the potential moving to the negative direction. Thus, the deposition potential of −1.4 V was chosen as the optimum value for detection of two heavy metals. Figure 5C exhibits the DPV current response of 2.0 µM Cd 2+ and Pb 2+ over the accumulation time of 60-390 s. As shown in Figure 5C, for the time of 60-210 s, the peak currents are almost linearly proportional to accumulation time, and this may be attributed to the fact that the amount of metal ions at the modified electrode surface greatly increases due to electrochemical deposition. When deposition time was above 210 s, the increase rate of the peak current of Pb 2+ changed, and this is probably due to the working electrode surface saturation. Under the consideration of sensitivity, a determination time of 210 s was selected for the deposition of the ions.

Individual Determination of Cd 2+ and Pb 2+
Under the optimized conditions, DPV was used as an analytical method for the electrochemical detection of Cd 2+ and Pb 2+ using various modified GCEs in 0.1 M ABS (pH = 4.5). Figure 6 exhibits the DPV responses of 10 wt % poly(BEP)/g-C 3 N 4 composite-modified GCE toward Cd 2+ and Pb 2+ . The figure also shows the linear relationship between peak currents and concentrations of the two ions, and the inset shows as well as their linear equations and correlation coefficient (Figure 6 inset).
The linear range of Cd 2+ is 0.1-6.8 µM with a detection limit of 0.0097 µM. The linear range of Pb 2+ is 0.1-6.4 µM with a detection limit of 0.00327 µM. Detailed results are shown in Table 2.  From the above results, it is clear that the 10 wt % poly(BPE)/g-C3N4 composite-modified electrode showed a wide detection range and low detection limit, and the detection limits were lower than the those of WHO standards. It should be noticed that the electrochemical analysis of 10 wt % poly(BPE)/g-C3N4 composite for trace metal ions may be attributed to the lone-pair electrons of nitrogen in the g-C3N4. The report shows that the highly ordered tri-s-triazine units contain many ideal coordination sites and thus metal ions can intercalate into g-C3N4 through the lone-pair electrons of nitrogen [26][27][28][29]. The combination of poly(BPE) and g-C3N4 not only improved the conduction pathway on the electrode surface, but also produced a strong conjugate effect on them, enhancing the adsorption of metal ions.

Simultaneous Determination of Cd 2+ and Pb 2+
The analytical performance of the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE was investigated by simultaneous determination of Cd 2+ and Pb 2+ in 0.1 M ABS (pH = 4.5). As shown in Figure 7A, the current response of Cd 2+ and Pb 2+ appeared at −0.82 V and −0.58 V, respectively. The distance between each individual peak is large enough to simultaneous detect these heavy metal ions using the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE. Figure 7B,C shows the linear relationship between peak current and concentration of the two heavy metal ions, as well as their linear equations and the correlation coefficient. The linear ranges of Cd 2+ and Pb 2+ are 0.12-7.2 μM and 0.08-7.2 μM, respectively. The detection limits of Cd 2+ and Pb 2+ are 0.018 μM and 0.00324 μM, respectively. Other detailed results are shown in Table 1.  From the above results, it is clear that the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified electrode showed a wide detection range and low detection limit, and the detection limits were lower than the those of WHO standards. It should be noticed that the electrochemical analysis of 10 wt % poly(BPE)/g-C 3 N 4 composite for trace metal ions may be attributed to the lone-pair electrons of nitrogen in the g-C 3 N 4 . The report shows that the highly ordered tri-s-triazine units contain many ideal coordination sites and thus metal ions can intercalate into g-C 3 N 4 through the lone-pair electrons of nitrogen [26][27][28][29]. The combination of poly(BPE) and g-C 3 N 4 not only improved the conduction pathway on the electrode surface, but also produced a strong conjugate effect on them, enhancing the adsorption of metal ions.

Simultaneous Determination of Cd 2+ and Pb 2+
The analytical performance of the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE was investigated by simultaneous determination of Cd 2+ and Pb 2+ in 0.1 M ABS (pH = 4.5). As shown in Figure 7A, the current response of Cd 2+ and Pb 2+ appeared at −0.82 V and −0.58 V, respectively. The distance between each individual peak is large enough to simultaneous detect these heavy metal ions using the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE. Figure 7B,C shows the linear relationship between peak current and concentration of the two heavy metal ions, as well as their linear equations and the correlation coefficient. The linear ranges of Cd 2+ and Pb 2+ are 0.12-7.2 µM and 0.08-7.2 µM, respectively. The detection limits of Cd 2+ and Pb 2+ are 0.018 µM and 0.00324 µM, respectively. Other detailed results are shown in Table 1. In the DPV technique, interfering molecules in the sample solution may be co-deposited on the active sites of the electrode surface, which result in changes in the stripping peak current. The interference may be the result of two main factors: (i) intermetallic compound formation and (ii) the competition between analytes and interferent ions for active sites on the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE surface. In order to understand whether there is interference between Cd 2+ and Pb 2+ in simultaneous detection, we performed the following experiment. The effect of a single species on the multispecies was performed by changing one species' concentration while the other species was unchanged. As shown in Figure 8, the DPV response of Cd 2+ and Pb 2+ increased linearly with the increase of the target ion's concentration, while the other ion was kept at a constant concentration of 2 μM. From Figure 8A, it can be seen that the peak current of Pb 2+ is practically unaltered with increasing of Cd 2+ concentration, and the peak current of Cd 2+ is practically unaltered with the increasing of Pb 2+ concentration ( Figure 8B). These results indicate that the other coexisting ion did not interfere with the determination of Cd 2+ or Pb 2+ . In addition, comparisons of the differently modified electrodes toward the simultaneous detection of Cd 2+ and Pb 2+ are shown in Table 3. From Table 3, it can be deduced that the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE could be an ideal sensor for simultaneous detection of Cd 2+ and Pb 2+ . In the DPV technique, interfering molecules in the sample solution may be co-deposited on the active sites of the electrode surface, which result in changes in the stripping peak current. The interference may be the result of two main factors: (i) intermetallic compound formation and (ii) the competition between analytes and interferent ions for active sites on the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE surface. In order to understand whether there is interference between Cd 2+ and Pb 2+ in simultaneous detection, we performed the following experiment. The effect of a single species on the multispecies was performed by changing one species' concentration while the other species was unchanged. As shown in Figure 8, the DPV response of Cd 2+ and Pb 2+ increased linearly with the increase of the target ion's concentration, while the other ion was kept at a constant concentration of 2 µM. From Figure 8A, it can be seen that the peak current of Pb 2+ is practically unaltered with increasing of Cd 2+ concentration, and the peak current of Cd 2+ is practically unaltered with the increasing of Pb 2+ concentration ( Figure 8B). These results indicate that the other coexisting ion did not interfere with the determination of Cd 2+ or Pb 2+ . In addition, comparisons of the differently modified electrodes toward the simultaneous detection of Cd 2+ and Pb 2+ are shown in Table 3. From Table 3, it can be deduced that the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE could be an ideal sensor for simultaneous detection of Cd 2+ and Pb 2+ .

Interference Study
It is known that, in practical applications, interference ions might co-deposit on an electrode with Cd 2+ and Pb 2+ . The interference study was performed by adding various potentially interfering metal cations including Na + , K + , Ca 2+ , Mg 2+ , Al 3+ , Fe 3+ , Co 2+ , Ni 2+ , Cu 2+ , and Zn 2+ in 50-fold excess with target metal ions into a standard solution containing 2 μM Cd 2+ and Pb 2+ under the optimized working conditions. As listed in Table 4, the change in the peak current of Cd 2+ and Pb 2+ was less than 10% after adding interfering ions. Thus, the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE displayed high selectivity for Cd 2+ and Pb 2+ in the heavy metal analysis.

Interference Study
It is known that, in practical applications, interference ions might co-deposit on an electrode with Cd 2+ and Pb 2+ . The interference study was performed by adding various potentially interfering metal cations including Na + , K + , Ca 2+ , Mg 2+ , Al 3+ , Fe 3+ , Co 2+ , Ni 2+ , Cu 2+ , and Zn 2+ in 50-fold excess with target metal ions into a standard solution containing 2 µM Cd 2+ and Pb 2+ under the optimized working conditions. As listed in Table 4, the change in the peak current of Cd 2+ and Pb 2+ was less than 10% after adding interfering ions. Thus, the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE displayed high selectivity for Cd 2+ and Pb 2+ in the heavy metal analysis.

Reproducibility of Modified Electrode
To further evaluate the sensing performance, the repeatability of the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE was tested with 2 µM Cd 2+ and Pb 2+ under the optimized conditions. Electrolyte: 0.1 M ABS (pH = 4.5), deposition potential: −1.4 V, deposition time: 210 s. As shown in Figure 9, the reproducibility was estimated with five different 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCEs that were prepared independently by the same procedure. The values of relative standard deviation (RSD) were 5.61% for Cd 2+ and 2.86% for Pb 2+ in the presence of 2 µM of metal ions, which demonstrated the reliability of the fabrication procedure. Repeatability of the developed method was evaluated by detecting 2 µM Cd 2+ and Pb 2+ at the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE for 10 measurements. The values of RSD were 1.58% for Cd 2+ and 1.71% for Pb 2+ . Hence, the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE shows ideal reproducibility and repeatability.

Reproducibility of Modified Electrode
To further evaluate the sensing performance, the repeatability of the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE was tested with 2 μM Cd 2+ and Pb 2+ under the optimized conditions. Electrolyte: 0.1 M ABS (pH = 4.5), deposition potential: −1.4 V, deposition time: 210 s. As shown in Figure 9, the reproducibility was estimated with five different 10 wt % poly(BPE)/g-C3N4 compositemodified GCEs that were prepared independently by the same procedure. The values of relative standard deviation (RSD) were 5.61% for Cd 2+ and 2.86% for Pb 2+ in the presence of 2 μM of metal ions, which demonstrated the reliability of the fabrication procedure. Repeatability of the developed method was evaluated by detecting 2 μM Cd 2+ and Pb 2+ at the 10 wt % poly(BPE)/g-C3N4 compositemodified GCE for 10 measurements. The values of RSD were 1.58% for Cd 2+ and 1.71% for Pb 2+ . Hence, the 10 wt % poly(BPE)/g-C3N4 composite-modified GCE shows ideal reproducibility and repeatability.

Real Sample Analysis
The 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE for simultaneous determination of Cd 2+ and Pb 2+ showed high sensitivity and better reproducibility. Tap water samples were taken to carry out a further study. Firstly, certain amounts of tap water in 0.1 M ABS (pH = 4.5) were prepared. Subsequently, standard solutions of Cd 2+ and Pb 2+ with different concentration were added to the tap water samples. The results are illustrated in Table 5. The recoveries of the Cd 2+ and Pb 2+ are 98.64-106.74% and 99.81-113.15%, respectively. Results indicate that the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified GCE could be applied in the detection of Cd 2+ and Pb 2+ in tap water samples.

Conclusions
In summary, a novel poly(BPE)/g-C 3 N 4 composite has been successfully synthesized via chemical oxidative polymerization and used for determination of Cd 2+ and Pb 2+ . The highly ordered tri-s-triazine units contain many ideal coordination sites, and thus metal ions can intercalate into g-C 3 N 4 through the lone-pair electrons of nitrogen. Besides that, the composite-modified electrode possesses a wide detection range and excellent sensitivity towards the simultaneous detection of Cd 2+ and Pb 2+ .The results demonstrate that the 10 wt % poly(BPE)/g-C 3 N 4 composite-modified electrode possesses high sensitively, wide linear range, and low detection limit for the determination of Cd 2+ and Pb 2+ .
Author Contributions: Shuai Ding, Ahmat Ali and Ruxangul Jamal conceived and designed the experiments; Shuai Ding performed the experiments; Ling Xiang and Ziping Zhong analyzed the data; Tursun Abdiryim contributed reagents/materials/analysis tools; Shuai Ding and Tursun Abdiryim wrote the paper.