3.1 Electrochemical responses of different sulfide modified electrodes
In order to investigate the effect of different sulfide ores on electrocatalytic detection of H2O2, we selected four kinds of pure sulfide ores including CuFeS2, FeS2, MoS2 and Ag2S mixed with CB respectively to modify the GCE electrode. The experiments were conducted in N2 saturated 0.1 M phosphate buffer (pH 7.0) at applied potential of -0.55 V (vs. Ag/AgCl) with amperometric i-t method in Fig. S1. At this potential, oxygen competes with H2O2 and eventually lead to the overlapping of reduction currents [37]. The electrochemical responses of four modified electrodes toward 20 mM H2O2 were investigated. CuFeS2-CB/GCE and FeS2-CB/GCE respond to H2O2 with the cathodic current values of 50.7 µA and 24.7 µA, respectively. However, MoS2-CB/GCE and Ag2S-CB/GCE have almost no current responses to the detection of H2O2. These results indicate that CuFeS2-CB/GCE and FeS2-CB/GCE possess the electrocatalytic activity to the detection of H2O2. In addition, the response current of CuFeS2-CB/GCE was 2.1 times greater than that of FeS2-CB/GCE, illustrating that CuFeS2-CB/GCE exhibited the best electrocatalytic activity among the selected sulfide ores. Therefore, CuFeS2 was selected in the subsequent experiments.
3.2 Characterization of CuFeS2-CB/GCE sensor
The SEM images were taken to observe the morphologies of CuFeS2 (a) and CuFeS2-CB (b). In Fig. 2a, the CuFeS2 presents irregular cubic and blocky shape, which is similar with our previous research [30]. The dimensions of raw CuFeS2 are in the scale of micrometer. The SEM image of the CuFeS2-CB (shown in Fig. 2b) exhibit tiny particles, which is attributed to the presence of CB. In addition, the physical mixed chalcopyrite and carbon black are tightly adsorbed together. As shown in Fig. 2c, the main element compositions of the compounds were C, Cu, Fe and S, with these elements distributed uniformly.
The phase features of CuFeS2, CB and CuFeS2-CB were analyzed by XRD spectra and shown in Fig. S2. Three distinct strong diffraction peaks appear at 2θ = 29.4º, 48.1º, and 57.0º, corresponding to the diffraction peaks of CuFeS2 (PDF 37–0471) in the card library. Carbon black is an amorphous carbon with two amorphous peaks at about 24º and 44º [38, 39], respectively. Fig. S2c is the XRD pattern of the mixture of CuFeS2 and CB, including the characteristic peaks of both CuFeS2 and CB.
3.3 Electrochemical behaviors of the modified electrode surface
The analytical performances of the prepared electrodes were evaluated to investigate the synergistic effect of CuFeS2 and CB in Fig. 3. CV curves with and without H2O2 in N2-saturated 0.1 M PBS (pH 7.0) for four different modified electrodes were performed. As can be seen from Fig. 3a, bare GCE exhibits very small response current to the reduction of 20 mM H2O2, with a linear range from 2.5 to 150 mM, and a sensitivity of 1.63 µA mM-1 cm-2. The bare GCE exhibits a wide linear range, but a very low sensitivity compared with other electrodes. It is essentially an inert electrical support to evaluate the electrochemical differences for other three modified electrodes [40]. When CB was modified on the GCE surface, the background current is the largest among the four electrodes, and only exhibits small response current in more negative potential region. This phenomena is not suitable for low concentration H2O2 detection. The linear range of CB/GCE is 30–50 mM, with the sensitivity of 77.87 µA mM-1 cm-2 in high concentration range of H2O2 in Fig. 3b. It is not suitable for environmental monitoring since the detectable concentration range is too high. Figure 3c is the CV curves with and without H2O2 for CuFeS2/GCE. The reduction current for H2O2 starts from 0.2 V vs. Ag/AgCl, which demonstrating a good candidate as H2O2 sensor to avoid the competition reduction of dissolved oxygen. CuFeS2/GCE exhibited a linear range of 2.5 to 25 mM, with the sensitivity of 26.93 µA mM-1 cm-2. Compared with CuFeS2/GCE, CuFeS2-CB/GCE (Fig. 3d) exhibits an obvious enlargement of the catalytic currents for the reduction of 20 mM H2O2. The catalytic current value of CuFeS2-CB/GCE is about 33.9 times than that of CuFeS2/GCE at the potential of -0.6 V. The linear range for CuFeS2-CB/GCE is 2.5–60 mM, with the sensitivity of 74.98 µA mM-1 cm-2. Furthermore, the reduction toward H2O2 of CuFeS2-CB/GCE is still from 0.2 V. The combination of CuFeS2 and CB is useful for not only enhance the response current to H2O2, but also enlarge the detection linear range. It is safe to conclude that the synergistic effect of chalcopyrite and carbon black is essential to enhance the electrocatalytic activity to H2O2.
The surface morphology of the modified materials greatly affects the electrocatalytic activity and electron transfer performance for non-enzymatic electrochemical sensor. The electrochemical behaviors of CuFeS2-CB/GCE (blue), bare GCE (red) and CuFeS2/GCE (green) were performed in 5 mM [Fe(CN)6]3-/4- containing PBS (0.1 M, pH 7.0) at a scan rate of 100 mV/s in Fig. 4a. The peak potential differences (ΔE) between the oxidation peaks and reduction peaks of CuFeS2-CB/GCE, bare GCE, and CuFeS2/GCE were 98.1, 224.0, and 360.5 mV, respectively. The CuFeS2/GCE exhibits larger ΔE compared with bare GCE, which suggesting the semi-conductivity of sulfide ores. After doping with CB, the CuFeS2-CB/GCE exhibits the highest electron transfer rate due to the high conductivity of CB.
EIS is a powerful tool to evaluate the interface properties between the modified electrode surface and the electrolyte. The charge transfer resistance (Rct) can be confirmed based on the diameter of the semicircle in the Nyquist diagram [41]. The Rct is calculated by fitting the Nyquist diagram in terms of Randles equivalent circuit. It is a very useful tool to evaluate the electron transfer characteristics of the adsorbed layer on the electrode surface. Figure 4b shows the Nyquist diagrams of CuFeS2-CB/GCE (blue), bare GCE (red) and CuFeS2/GCE (green) obtained using [Fe(CN)6]3-/4- as an electrochemical redox probe, respectively. The Rct of CuFeS2-CB/GCE, bare GCE and CuFeS2/GCE were 125, 1195 and 4202 Ω, respectively, with the same tendency of CVs. The CuFeS2/GCE presented a larger Rct value compared with the bare GCE, which is attributed to the semi-conductivity of CuFeS2. CuFeS2-CB/GCE showed the smallest Rct value among the three modified electrodes, also indicating that CB can effectively improve the conductivity of the modified electrode. It is safe to conclude that the electron transfer rate on the CuFeS2-CB/GCE could be enhanced even for physical doping of CB in this study.
To investigate the reaction and solution phase mechanisms of the electron transfer process, the effect of scan rate was studied by CV method. Figure 4c shows the results of CuFeS2-CB/GCE at different scan rates in 5 mM [Fe(CN)6]3-/4- containing 0.1 M PBS (pH 7.0). The linear equation is Ipa = -0.08845υ − 4.082, R2 = 0.9801, Ipc = 0.08433υ + 4.219, R2 = 0.9806, respectively. The Ipa and Ipc gradually increased in a linear relationship with the square root of the scan rate in the range of 100 to 100 mV/s and 100 to 500 mV/s (Fig. S3). It demonstrated that this electrochemical process on CuFeS2-CB/GCE was a diffusion-controlled process [42].
3.4 Optimization of experimental conditions
Electrolyte pH is an important factor to affect the sensor performance. The optimization of electrolyte pH was performed by using CV method, with the current selected at -0.55 V vs. Ag/AgCl. As can be seen from Fig. 5a, under acidic conditions, CuFeS2-CB/GCE shows very small response currents to H2O2. The response current of H2O2 increases with the increasing of electrolyte pH. Alkaline is not suitable for H2O2 based electrochemical sensors because of the extra reactions between base and H2O2 which interferes the detection accuracy. In addition, strong acidic and alkaline solutions are not desirable from the view point of application [43]. Therefore, pH 7 was chosen in the following experiments. Fig. 5b shows the optimization of applied potential for the detection of 20 mM H2O2 in 0.1 M PBS (pH 7.0) by using amperometric i-t curve. The response currents towards the reduction of H2O2 increases in the potential range from − 0.4 V to -0.55 V, and reach a maximum value at -0.55 V. Thus, -0.55 V was chosen as the optimal potential for the following amperometric i-t experiments. The influence of the CuFeS2-CB immobilization time ranges from 0.5 h to 2.5 h on the reduction current response was also investigated in Fig. 5c. The reduction current toward H2O2 reaches to the highest value when the immobilization time is 1 h. To improve the performance of CuFeS2-based H2O2 sensors, the ratio of CuFeS2 and CB was also optimized as shown in Fig. 5d. A comparison of these composites showed that the reduction response current toward H2O2 exhibited the best result when the ratio of CuFeS2 and CB was 1:2. A certain amount of CB is essential to improve the electrical conductivity. Therefore, mass ratio of 1:2 for CuFeS2 and CB was used in the following experiments.
3.5 Analytical performance of CuFeS2-CB/GCE on the detection of H2O2
Table 1 Analytical performance of present work with other reported electrodes for H2O2 detection.
|
Electrode
|
Method
|
Linear range/mM
|
LOD/mM
|
Ref.
|
CuO/rGO/Cu2O/Cu
|
AP
|
0.005–8.266
|
0.001
|
23
|
NiO/α-Fe2O3
|
AP
|
0.5-3
|
0.05
|
44
|
Ni(II)-MOFs/CNTs/GCE
|
AP
|
0.01–51.6
|
0.002
|
45
|
FeS (F4)
|
AP
|
0.5–20.5
|
0.15
|
21
|
GO-Ag nanocomposite
|
AP
|
0.1–11
|
0.028
|
46
|
CuFeS2-CB/GCE
|
CV
|
2.5–60
|
1.0
|
This work
|
AP
|
0.1–100
|
0.04
|
rGO, reduced graphene oxide; Ni(II)-MOFs, Ni(II)-based metal-organic framework; CNTs, carbon nanotubes; FeS (F4), ferrous sulfide nanosheets; GO, graphene oxide.
|
The analytical performance of CuFeS2-CB/GCE for the detection of H2O2 was evaluated. Figure 6a exhibited the typical current-time curve of the prepared non-enzymatic sensor on the successive additions of various concentrations of H2O2 in N2-saturated PBS (pH 7.0) at an applied potential of -0.55 V vs. Ag/AgCl. The steady-state cathodic background current changed rapidly after the addition of H2O2 and reached another steady-state current within 6 s. It showed a linear range of H2O2 from 0.1 to 100 mM and the regression equation is I = 0.4539C − 0.1259, with a coefficient of 0.9999 (n = 3) in Fig. 6b. The detection limit (LOD) of H2O2 was found to be 0.04 mM. Table 1 summarizes the analytical performances of the reported electrodes for H2O2 detection compared with the present work. In consideration of using sulfide ores as electrochemical platform, the performance of the CuFeS2-CB/GCE is comparable.
3.6 Stability, anti-interference, reproducibility and lifetime of the CuFeS2-CB/GCE sensor
The steady-state amperometric method was used to evaluate the stability of the sensor by adding H2O2 for 16 times at the potential of -0.55 V vs. Ag/AgCl. It can be seen from Fig. 6c that the CuFeS2-CB/GCE sensor exhibited good stability in continuing detection of H2O2, and showed a good linear relationship between concentrations and response currents. The relative standard deviation (RSD) of the H2O2 was calculate to be 9.13%. For the enzyme-free sensor, selectivity is an important factor affecting the performance of the sensor. The selectivity of the CuFeS2-CB/GCE was evaluated by adding potentially interfering substances to 0.1 M PBS (pH 7.0) solution. Figure 6d exhibited the relative response currents of 5 mM H2O2, and 5 mM H2O2 in the presence of 5 mM glucose, fructose, urea, Catechol, Na+, K+, Mg+, ascorbic acid, respectively. These substances showed almost no interference of the prepared sensor. The results showed that the CuFeS2-CB/GCE sensor has good selectivity and reliable anti-interference performance. In addition, six electrodes were prepared under the same conditions to detect 20 mM H2O2 as shown in Fig. 6d. The RSD of the six electrodes was 0.96%, indicating that the CuFeS2-CB/GCE based sensor has excellent reproducibility. Lifetime of the same CuFeS2-CB/GCE was studied using the steady-state amperometric method towards the detection of 20 mM H2O2. The prepared sensor was stored at 20°C in a dry state when not in use. Figure 6f shows the peak current responses which were checked at the 1st day, the 5th day and the 10th day, respectively. It retained 94.3 and 63.1% of its original activity after storage for 5 and 10 days, respectively, indicating that the CuFeS2-CB/GCE possess acceptable long-term stability.
3.7 Real sample analysis
H2O2 is widely used in the food industry and is also an important substance in biological systems. Therefore, the detection of H2O2 in real samples is of great significance. To verify the utility of the CuFeS2-CB/GCE sensor, it was applied to detect H2O2 in real drinking water. As shown in Table 2, the presence of H2O2 was not detected in the three water samples. When a certain amount of H2O2 was added to the sample, the corresponding currents can be detected. The recovery range is 94.8% ~ 96% for the present sensors, with the RSD lower than 5% (n = 3), indicating that the prepared electrode could be used for the detection of H2O2 in real samples.
Table 2
CuFeS2-CB/GCE sensor applied to drinking water
Number | Detected (mM) | Added (mM) | Found (mM) | RSD(%) | Recovery |
1 | Not found | 10 | 9.6 | 1.93 | 96.0% |
2 | Not found | 30 | 28.7 | 4.41 | 95.6% |
3 | Not found | 50 | 47.4 | 2.03 | 94.8% |
RSD (%) calculated from three separate experiments |
3.8 Conclusions
In this work, a novel type of CuFeS2-CB/GCE based enzyme-free electrochemical sensor was successfully constructed by simple physical mixing method, which showed good electrocatalytic ability for H2O2. Under the optimum conditions, the CuFeS2-CB/GCE sensor exhibited a wide sensing linear range from 0.1 mM to 100 mM, with the sensitivity of 74.98 µA mM− 1 cm− 2 for H2O2. The CuFeS2-CB/GCE sensor possesses the advantages of cost-effective, easily available, simple preparation and good electrochemical performances including rapid response time, wide linear range, high sensitivity, good stability and practical usage.