Introduction

There is a substantial need for altermative methods for the determination of hydrogen peroxide (H2O2) [1]. Many researchers pay much attention on electrochemical method [2,3,4,5]. However, the direct detection of H2O2 on bare electrode is not easy. Because the reduction or oxidation of H2O2 on bare electrode requires a high overpotential with a low response, such high overpotential usually results in interference and the low response makes it difficult to detect H2O2 with a low concentration. Therefore, in order to ensure a lower overpotential and a higher response, studying modified electrodes is vital. For example, Jin [6] reported a glassy carbon electrode modified with FeS nanosheets as a highly sensitive for H2O2 assay in 0.1 M NaOH solution. Ni [7] described a nonenzymatic amperometric method for H2O2 assay which used a nanocomposite consisting of Co3O4 nanoparticles and mesoporous carbon nanofibers. The synergetic combination of the electrocatalytic activity of the Co3O4 nanoparticles and the electrical conductivity of mesoporous carbon nanofibers made the method exhibit good electrocatalytic performance in 0.1 M NaOH solution. In addition, many other metal nanoparticles [8, 9] were also employed for H2O2 assay. A non-enzymatic electrochemical method was obtained by modifying a glassy carbon electrode with nanocomposites containing nanoporous copper and carbon black. The result indicated that the method possessed enhanced electrocatalytic activities towards H2O2 assay at 0.75 V [8]. Wu [9] synthesized a sandwich structured nanocomposite consisting of mildly reduced graphene oxide modified with silver nanoparticles supported on Co3O4. Then, they employed such nanocomposite to fabricate a nonenzymatic electrochemical method for H2O2 detection. The integration of mildly reduced graphene oxide, Co3O4 and silver nanoparticles into a nanocomposite brought out remarkable properties to catalyze H2O2 with high sensitivity and wide linear range.

Some transition metal oxides usually worked under high pH conditions and noble metal was expensive, which limited their wide applications. As an important semiconductor, cuprous oxide (Cu2O) has been employed to detect H2O2 because of its catalytic properties and low cost [10,11,12,13]. However, the low conductivity and severe aggregation of Cu2O limit its applications [14]. On the other hand, an appropriate supporting material possessing large surface area, high conductivity and large numbers of functional groups can prevent electrocatalytic material from aggregating and improve its conductivity [15]. Reduced graphene oxide (rGO) is a two-dimensional carbon material with large surface area and high conductivity [16]. Polyaniline (PANI) is a conductive polymer possessing excellent conductivity and large numbers of nitrogen-containing groups [17]. Therefore, the modification of rGO with PANI will enable PANI/rGO a promising support material. Some rGO/PANI-based nanocomposites were reported [18,19,20,21,22], in which electrochemical or chemical oxidative polymerization was employed to prepare PANI/rGO. Then, rGO/PANI-based nanocomposites were obtained by hydrothermal or other treatment. Yan [21] prepared RGO/PANI/Cu2O via one-step in situ redox method using ethanol as the solvent. RGO/PANI/Cu2O nanocomposites presented a flower-like structure with an average size of 2.0 μm in diameter and the product exhibits excellent microwave absorption property. The method was very simple and the product exhibits excellent microwave absorption property. However, the size of the product was big. The size of nanocomposite possessed influence on its properties and smaller size possessed larger surface area. Miao [22] synthesized Cu2O/PANI/rGO via a one-pot method in the presence of cubic Cu2O nanoparticles, where Cu2O nanoparticles were synthesized first. Yan [23] reported that granular nanowires with a diameter of about 60 nm were fabricated from Cu2O by an electrochemical method using anodic aluminium oxide as the template. This work was interesting. However, the synthesis of rGO/PANI-based nanocomposites and granular Cu2O nanowires usually required complicate steps. Therefore, developing approaches to synthesize Cu2O/PANI/rGO for applications is still necessary.

Here, a one-step hydrothermal method was employed to prepare Cu2O/PANI/rGO nanocomposites. Aniline was employed as both the precursor of PANI and the reductant for Cu2+ and GO. The polymerization of aniline and the formation of Cu2O and rGO occur simultaneously. After that, Cu2O/PANI/rGO nanocomposites were employed for H2O2 assay. The composition, morphology and electrochemical properties of Cu2O/PANI/rGO nanocomposite were investigated.

Experimental

Reagents and materials

Shanghai Yuanju Biotechnology (Shanghai, China, http://www.yjbiotech.cn/) supplied graphite powder, aniline (C6H7N, 99.9%) and Cu(NO3)2·3H2O. Sigma Company (http://www.sigma.com) supplied uric acid (UA), ascorbic acid (AA) and glucose (Glu). Na2HPO4 and NaH2PO4 were mixed to obtain the phosphate buffer (PB).

Apparatus

TEM images were recorded with Tecnai G2 F20 S-TWIN (FEI, USA). D/MAX-3C (Rigaku, Japan) and TENSIR 27 (Bruker, German) were employed to record XRD patterns and FTIR spectra, respectively. CHI 660 electrochemical workstation (China) was used to measure electrochemical properties, in which glassy carbon electrode (GCE), saturated calomel electrode and platinum wire were employed as working electrode, reference electrode and counter electrode, respectively. All potentials given in thiswork were referred to the saturated calomel electrode.

Synthesis of Cu2O/PANI/rGO

Graphite powder was employed to prepare GO through modified Hummers method [24]. 0.5 g of graphite and 0.5 g of NaNO3 in 23 mL of 12.1 M H2SO4 were stirred in an ice bath for 15 min. Then 4.0 g of KMnO4 was slowly added in an ice bath to yield a purple-green mixture. This suspension was transferred to a 40 °C water bath and magnetically stirred for 90 min. The dark brown colored paste was diluted with the slow addition of 50 mL of deionized water and allowed to stir for a further 10 min. A 6 mL portion of H2O2 was slowly added to quench the solution to produce a golden-brown sol. 50 mL of water was added, and the resultant product centrifuged at 5000 rpm and washed with deionized water repeatedly. Finally the product was dried at 80 °C for 24 h. Then, Cu2O/PANI/rGO was synthesized as follows: 0.193 g Cu(NO3)2·3H2O and 160 μL of aniline were added into the suspension of GO in ethanol (6.8 mg GO, 13.6 mL) under stirring. After 0.5 h, the solution was transferred into Teflon-lined autoclave (40 mL). The temperature kept at 160 °C for 360 min. When the Teflon-lined autoclave cooled, Cu2O/PANI/rGO was centrifuged at 8000 rpm for 5 min. After the supernatant solution was removed, the precipitate was washed with deionized water. Finally the product was dried at 50 °Cfor 24 h.

Electrode modification

The casting method was employed to fabricate modified electrode. Before modifying, GCE was polished by alumina powder. Water was used for dissolution Cu2O/PANI/rGO. Nafion was diluted with ethanol to 0.05 wt%. 6 μL of Cu2O/PANI/rGO solution (2 mg·mL−1) was dropped on electrode surface. Then 6 μL of nafion solution (0.05 wt%) was dropped on electrode surface when the solvent (water) was allowed to evaporate at ambient temperature. The modified electrode was expressed as Cu2O/PANI/rGO/GCE. GO/GCE was fabricated in the similar way.

Results and discussion

Choice of materials

rGO attracts much attention for various applications due to its specific high surface area, exceptional electrical, mechanical, and thermal properties. PANI exhibits excellent conductivity and possesses large numbers of nitrogen-containing groups. Therefore, it is desired to employ PANI to modify rGO an advanced support material. Cu2O is an attractive material because of its catalytic properties and low cost. Thus, decorating Cu2O on PANI/rGO would display an excellent electrochemical activity toward H2O2 reduction.

Characterizations of Cu2O/PANI/rGO

TEM and XRD are employed to characterize morphology and structure of GO and Cu2O/PANI/rGO, respectively. Some wrinkles and folds are observed on GO nanosheet (Fig. 1a). In the case of Cu2O/PANI/rGO (Fig. 1b, c), abundant Cu2O nanorods are deposited on PANI/rGO. This is benefited from the large numbers of anchor sites provided by PANI/rGO. During the synthesis of Cu2O/PANI/rGO, Cu2+ and GO were reduced by aniline [25, 26] and aniline was polymerized on rGO. The modification of rGO with PANI made PANI/rGO possess many functional groups. Such functional groups offer large numbers of anchor sites for the growth of dispersed Cu2O. Moreover, the XRD pattern of Cu2O/PANI/rGO (Fig. 1d) reveal several obvious diffraction peaks at 36.3°, 42.1°, 61.4° and 73.7°, which are indexed to (111), (200), (220) and (311) planes of Cu2O [27]. This confirms that the nanorods decorating on PANI/rGO are Cu2O.

Fig. 1
figure 1

TEM images of nanocomposites: a GO, b, c Cu2O/PANI/rGO and XRD of (d) Cu2O/PANI/rGO

Full size image

The chemical composition of Cu2O/PANI/rGO is recorded with EDS (Fig. 2a), which displays C, O, N, Cu element. FTIR spectrum characterizes chemical structures of GO and Cu2O/PANI/rGO. Curve a in Fig. 3b displays some characteristic peaks of GO, including the stretching of −OH, −COOH, C=C, C − O at 3427, 1720, 1628 and 1082 cm−1, respectively [28]. Peaks of Cu2O/PANI/rGO at 1720 cm−1 and 1082 cm−1 almost disappears, indicating the reduction of GO [29]. Some new peaks at 627, 1292 and 1485 cm−1 appears. These peaks relates to Cu-O vibration [30], C=C stretching of benzenoid ring and C-N stretching of secondary aromatic amine [31], respectively. Fig. S1 in the supporting information shows the raman spectroscopy of GO and rGO. rGO shows a relatively higher intensity of D to G bands than that of GO. This indicates the decrease in the size of the in-plane sp2 domains, the removal of the oxygen functional groups in the graphene oxide nanosheets, and the reduction of GO. Two new peaks at around 212 and 634 cm−1 appears, which are associated with Cu2O. From above points, Cu2O/PANI/rGO was successfully prepared.

Fig. 2
figure 2

a EDS and (b) FTIR spectra of Cu2O/PANI/rGO nanocomposites

Full size image
Fig. 3
figure 3

Cyclic voltammograms obtained at bare GCE (a, a’) GO/GCE (b, b’) and Cu2O/PANI/rGO (c, c’) in N2-saturated phosphate buffer (pH 7.4) in the absence (a, b, c) and presence (a’, b’, c’) of 3.0 mM H2O2 at a scan rate of 100 mV s−1

Full size image

Electrochemical properties of Cu2O/PANI/rGO nanocomposites

Electrochemical performance of diffirent modified electrodes was recorded with cyclic voltammograms (CVs). As shown in Fig. 3a, no electrochemical response is found on GCE (a), GO/GCE (b). However, Cu2O/PANI/rGO/GCE (c) reveals a strong redox couple, where anodic peak is ascribed to the oxidation of Cu2O to CuO and cathodic peak resulted from the reduction of CuO to Cu2O [23]. In presence of H2O2, GCE (a’) and GO/GCE (b’) exhibit poor response, indicating that the direct reduction of H2O2 on bare electrode or GO/GCE is difficult. In the case of Cu2O/PANI/rGO/GCE (c’), cathodic peak enhances accompanying with anodic peak current decreasing. This indicates that Cu2O/PANI/rGO possesses excellent electrocatalytic activity toward H2O2 reduction. The excellent electrocatalytic activity is due to that PANI/rGO increase conductivity of Cu2O and also prevents Cu2O from aggregating, which provides more electroative sites for H2O2 reaction. The mechanism of H2O2 electroreduction on Cu2O/PANI/rGO/GCE is as follows [23]:

$$ {\displaystyle \begin{array}{c}{\mathrm{Cu}}_2\mathrm{O}+{\mathrm{H}}_2{\mathrm{O}}_2\to 2\mathrm{CuO}+{\mathrm{H}}_2\mathrm{O}\\ {}2\mathrm{CuO}+2{\mathrm{e}}^{-}+2{\mathrm{H}}^{+}\to {\mathrm{Cu}}_2\mathrm{O}+{\mathrm{H}}_2\mathrm{O}\end{array}} $$

Fig. S2 in the supporting information shows the response of Cu2O/PANI/rGO/GCE in the presence of 1.0 mM H2O2 in phosphate buffer with various pH. The cathodic peak current enhanced firstly and then decreased with the cathodic peak shifting positively. With increasing the pH solutions, cathodic peak changes to positive values, which indicates that the redox process is pH dependence. Considering that the cathodic peak current was larger at pH 7.4. Therefore pH 7.4 phosphate buffer was chosen as the supporting electrolyte in this work.

Figure 4 displays CVs of teh modified GCE at different scan rates, which shows that the cathodic current increases with the increasing of scan rates (Fig. 4a) and a linear relationship between cathodic peak current and square root of scan rate is obtained (Fig. 4b). Therefore, electroreduction of H2O2 on Cu2O/PANI/rGO/GCE is a diffusion-controlled process.

Fig. 4
figure 4

a Cyclic voltammograms obtained at Cu2O/PANI/rGO/GCE in presence of 3.0 mM H2O2 in N2-saturated phosphate buffer (pH 7.4) at different scan rates (from a to j: 20, 40, 60, 80, 100, 120, 140, 160, 180 and 200 mV s−1). b Linear fitting program of current versus the square root of scan rate

Full size image

Electrocatalytic reduction of H2O2

Amperometric detection is an important detection method in electrochemical analysis. The resulting current is proportional to the concentration of the species generating the current, and the quantification of H2O2 can be achieved via the electrochemical detection. Electrocatalytic properties of the modified GCE were studied. The reduction current of H2O2 obtained at working potential of of −0.2 V is higher than that of 0 V and −0.1 V. Compared with the potential of −0.3 V and −0.4 V, the background noise is low at such lower working potential. Therefore, − 0.2 V was chosen as operating potential. Because such potential was beneficial to ensure less interference and low background noise. Amperometric response current increases when H2O2 is added (Fig. 5a). Figure 5b reveals calibration curve. A linear regression equation of Ip (μA) = −1.376 + (−2.76)·C (mM) is obtained in the range of 0.8 μM to 12.78 mM. The sensitivity and detection limit are 39.4 μA mM−1 cm−2 and 0.5 μM (S/N = 3), respectively. The sensitivity is estimated from the slope of the calibration curve and electrode surface area. As shown in Table 1, many methods exhibits an excellent electrocatalytic activity toward H2O2 reduction. These methods have their own advantages and disadvantages. For example, the method for H2O2 determination based on Ti3C2Tx is the most sensitive device described so far with a detection limit of 0.7 nM. However, Ti3C2Tx modified GCE for H2O2 assay usually operated an applied potential of - 0.5 V and such operating potential may lead to more interference of other electroactive species in the solution. Ti3C2Tx was prepared by HF treatment protocol, which made the synthesis of Ti3C2Tx complicated. Ag nanoparticles and MnO2 can exhibit an excellent performance to H2O2. However, transition metal oxides usually worked under high pH conditions and noble metal was expensive, which limited their wide applications. Comparing with other modified electrodes (Table 1), Cu2O/PANI/rGO/GCE exhibits an excellent electrocatalytic activity toward H2O2 reduction with a wide linear range and a low detection limit. These benefited from the combination of Cu2O and PANI/rGO. PANI/rGO acted as the support material and the support material provided many anchor sites for the growth of dispersed Cu2O. Therefore, Cu2O/PANI/rGO offered more electroative sites for H2O2 molecules to react. Cu2O/PANI enhanced the conductivity of Cu2O and thus was beneficial to improve electrocatalytic activity. However, Cu2O/PANI/rGO/GCE possessed some limits. For example, the sensitivity was low. Therefore, further works for improving the sensitivity are on our schedule.

Fig. 5
figure 5

a Amperometric curve of Cu2O/PANI/rGO/GCE for successive additions of H2O2 in N2-saturated phosphate buffer (pH 7.4) at −0.2 V. b Calibration curve of H2O2 versus its concentration

Full size image
Table 1 Comparision of present work with several modified electrode for H2O2 detection
Full size table

Repeatability, stability, selectivity and real sample analysis

The repeatability and stability of the Cu2O/PANI/rGO/GCE were studied in the linear range of H2O2. The relative standard deviation (RSD) was 1.8% for eight successive measurements of 5 mM H2O2 in pH 7.4 phosphate buffer. This indicates that the modified electrode possesses good repeatability. The current responses to 5 mM H2O2 showed no obvious change after 30 cycles, and then decreased slowly with the increase of the number of cycles, indicating that the Cu2O/PANI/rGO/GCE was stable. The storage stability of the method was further investigated. The amperometric measurements were measured using the same electrode and it retained above 96% of its initial response after being stored at 4 °C for 1 month. These results displayed that the modified electrode had good stability. The fabrication reproducibility of six electrodes, made independently, showed an acceptable reproducibility with the relative standard deviations of 3.5% and 4.0% for the current determination of 5 mM H2O2.

Influence of AA, Glu, NaCl and UA on the detection of H2O2 was studied (Fig. 6). The GCE exhibits low response when AA, Glu, NaCl and UA were added, but the addition of 0.04 mM H2O2 shows an obvious current. The interference effects were also investigated by testing the amperometric responses of acetaminophen and dopamine (Fig. S3A). The successive addition of each interfering species brings out hardly discernible current response. These results suggest that this GCE have selectivity towards H2O2. It is seen obviously (Fig. S3B) that the amplitude of increase in peak current is similar in the absence and presence of air after the addition of H2O2. This suggests the feasibility of H2O2 detection in the presence of air. However, it is found that the background noise increased under air. Therefore, we used the N2-saturated solution in order to maintain the stability of the system in the detection process.

Fig. 6
figure 6

Amperometric response of 0.04 mM H2O2, 0.01 mM AA, Glu, NaCl and UA at Cu2O/PANI/rGO/GCE in N2-saturated phosphate buffer (pH 7.4) at −0.2 V

Full size image

Analysis of real samples has also been investigated. As shown in Table 2, the H2O2 concentration in the real sample of disinfectant containing 3.5% H2O2 is detected. Before experiments, the real samples were diluted with double-distilled water. It indicats that the method can be used for the H2O2 sample analysis.

Table 2 Determination of H2O2 in disinfector sample by the present method and the titration method
Full size table

Conclusions

Cuprous oxide/polyaniline/reduced graphene oxide nanocomposites were successfully synthesized by a hydrothermal approach. The experimental results revealed that Cu2+ and GO were reduced by aniline. Aniline was polymerized on rGO. The preparation method did not need complicated steps. PANI/rGO prevented Cu2O from aggregating and improved its conductivity. Therefore, the method exhibited excellent electrochemical response toward H2O2. Cu2O/PANI/rGO nanocomposites can be employed for H2O2 assay. The one-step hydrothermal approach may be extended to the fabrication of other metal oxide/conductive polymer/reduced graphene oxide nanocomposites for applications. Although Cu2O/PANI/rGO/GCE exhibited an excellent electrocatalytic activity toward H2O2 reduction with a wide linear range and a low detection limit, the sensitivity was not very high, further works for improving sensitivity are on our schedule.