Abstract

The most important biological changes which have to be monitored is the mechanism of ageing in the human body where the mitochondria play a major role. Hydrogen peroxide (H2O2) is one of the important markers for the reactive oxygen species (ROS), which denatures the protein and DNA, that was the main contributory factor of ageing. So, it is very important to monitor H2O2 levels in the biological samples. Herein, we reported the preparation of 1D graphene nanoribbon/cobalt oxide nanorod (GNR/Co3O4) based nanocomposite-modified electrochemical sensor for H2O2. Firstly, GNR was synthesized by oxidative unzipping of multiwalled carbon nanotubes (MWCNTs). Secondly, cobalt oxide nanorods (Co3O4) were grown onto GNR by a chemical reduction process. As-prepared nanocomposite was characterized by UV-Visible spectroscopy (UV-Vis) and HR-TEM. Electrochemical properties of GNR/Co3O4-coated electrode were studied by cyclic voltammetry (CV) which showed two redox peaks at 0.93 and 0.88 V in phosphate buffer solution. Next, the electrocatalytic activity of GNR/Co3O4-coated electrode was studied against H2O2 oxidation. The electrochemical studies revealed that GNR/Co3O4-coated electrode exhibited high electrocatalytic activity for H2O2 oxidation at 0.925 V. This sensor showed a linear response for H2O2 oxidation from 10 to 200 μM. The limit of detection (LOD) was calculated to be 1.27 μM. The selectivity of the sensor was also studied with other biomolecules associated in the human body, and the results showed that interference effect is negligible. Thus, the proposed GNR/Co3O4-modified electrode can be used for H2O2 detection with excellent stability and selectivity.

1. Introduction

Hydrogen peroxide (H2O2) is a well-known oxidizing agent. The process of manufacturing of H2O2 and its applications in various sectors are well documented [1]. H2O2 has been used in different industrial applications such as environmental, textile, clinical, pharmaceutical, and food [2]. Also, the importance of H2O2 is reported in many biosynthetic reactions and immune cell activation process [3]. The majority of the biological reactions were catalyzed by different enzymes such as lactate oxidase, cholesterol oxidase, D-amino acid oxidase, glutamate oxidase, and lysine oxidase, which resulted in the formation of H2O2 as a by-product. Thereby, it acts as a reactive oxygen species (ROS) that can promote the denaturation of proteins and DNA by oxidative processes [4]. Though it has cytotoxic effects, it plays a vital role in controlling the physiological process of immune cell activation and vascular remodelling [5]. H2O2 involves in enzymatic reactions and acts as an intermediate for the level of glucose, lactose, and cholesterol [6]. Generally, H2O2 can enter into our body through consumption of instant coffee, green tea, or black tea that can raise the H2O2 concentration level above 100 μM which may enter into the gastrointestinal tract [7]. According to the US National Institute for Occupational Safety and Health, the permitted exposure limit of H2O2 is 1 ppm based on time-weighted average, but if it reaches above the level of 75 ppm, that is dangerous to health and may lead to death [8]. The anomalous level of H2O2 production in the human body may lead to carcinogenesis and severe complications such as formation of cancer and heart attack [9]. Thus, there is a vital need for monitoring H2O2 concentrations in the biological and environmental samples [10]. The analysis of H2O2 was generally carried out by traditional analytical methods such as spectrophotometry [11], colorimetry [12], fluorimetry [13], and electrochemical methods [1417]. Colorimetry and fluorimetry methods are considered to be expensive and may not be suitable for general laboratory use. Electrochemical sensing of H2O2 can be performed by enzymatic and nonenzymatic methods [18, 19]. Enzymatic electrochemical sensors are vulnerable to the environmental factors such as temperature, pressure, pH, and biological pathogens [20]. The above-stated problems can be neglected by using nonenzymatic electrochemical sensors [21]. Majumder et al. reported a nonenzymatic electrochemical sensor for H2O2 and hydrazine using 3D microsnowflake architectures of α-Fe2O3 [22]. In 2019, Liu et al. prepared an innovative electrochemical sensor for detection of H2O2 using the hollow CuO/polyaniline nanohybrid fibers which exhibited a good linear range, lowest LOD, high selectivity, and extended stability [23]. Baghayeri et al. have prepared magnetic graphene oxide functionalized with an amine-terminated material that showed excellent electrochemical activity against H2O2 in the linear range of 0.05 to 160 μM with LOD of 0.01 μM [24].

Recently, carbon nanomaterials such as carbon dots (0D), carbon nanotubes (1D) [25, 26], graphene (2D) [27, 28], and graphite (3D) [29, 30] have been exploited as electrode materials for various electrochemical applications [31, 32]. Graphene nanoribbons (GNR) is a narrow strip of graphene with a quasi one-dimensional structure, which holds complementary properties of graphene sheet [33]. From the theoretical analysis, GNR’s electrical nature can be tuned by the size and edge formation, and they can differ from metallic to semiconducting nature [34, 35]. The physical properties of the GNR are based on the length, width, and number of layers, which in the sequence depends on their preparation process. There are three methods available for the preparation of GNR: (i) cutting graphene by using different lithographic techniques [36, 37], (ii) bottom-up synthesis from polycyclic molecules [38, 39], and (iii) unzipping of carbon nanotubes [4042]. Although the bottom-up technique offers a method to definite edge control, the lithographic process can produce GNR with exact location. However, the unzipping process offers the benefit of bulk preparation of GNR on a large scale. GNR prepared by the unzipping process has been effectively used in many applications such as a catalyst support for lithium-ion batteries [43, 44]. Unzipping of multiwall carbon nanotubes (MWCNTs) are classified into four types: (a) the reductive intercalation-assisted method [45], (b) the oxidative unzipping [46], (c) the electrochemical unzipping [47], and (d) miscellaneous approaches [48]. The first method depends on the familiar capability of alkali compounds to interpolate by the expansion of graphite along the -axis direction. This process of unzipping of MWCNTs creates lattice expansion tempts high strain within the concentrical walls, succeeding in the breaking or opening of longitudinal directions of the tubes [33]. Common intercalants such as lithium and potassium metals have been used [33] which produced GNR with high conductivity, but they persisted multilayered flakes. GNR cannot be separated into single-layer ribbons because of the attraction between the surfaces. The oxidative method comprises treating of MWCNTs in a lower pH environment with a preparation as similar to synthesis of graphene oxide from the natural graphite [49]. Pristine graphene has an inert chemical surface without defects, and they exhibited poor water solubility. On the contrary, as-prepared GNRs are more reactive. The high reactivity of GNRs are linked to the adsorption of analytes via stacking interaction and by electrostatic or hydrogen bonding interactions with functionalities of the target molecules or the oxygen moieties located at the edges of the graphene material [50].

The purpose of using nanomaterials in the fabrication of sensors is to increase the sensitivity, enhance catalytic activity of the process, reduce the over potentials, and increase the electron transfer rate of the reaction [51, 52]. Transitional metals mainly boost the oxidation of several substrates because they can initiate the process of multielectron oxidation [53]. Also, nanoscale compounds can improve diffusion and offers a good active surface for electrocatalytic reactions [54]. Cobalt oxide-based nanomaterials (Co3O4) have been used in energy storage, electrochromic thin films, magneto-resistive devices, and heterogeneous catalysis [55]. Additionally, Co3O4 remains an attractive catalyst, mainly due to its outstanding electrocatalytic activity against ozone and oxygen.

Herein, a nanocomposite made of 1D GNR/Co3O4 was prepared by the top-down method and then used to fabricate an electrochemical H2O2 sensor. Firstly, Co3O4 nanorods were synthesized by a chemical reduction process. Secondly, GNR was synthesized by oxidative unzipping of multiwalled carbon nanotubes (MWCNTs). Followed by, GNR/Co3O4 nanocomposite was prepared and characterized by UV-visible (UV-Vis) and HR-TEM. Electrochemical properties of GNR/Co3O4 were studied by cyclic voltammetry (CV) which showed two redox peaks centered at 0.93 and 0.88 V in phosphate buffer solution. Next, electrochemical oxidation of H2O2 was performed on GNR/Co3O4-modified electrode. Further, the electrochemical studies were revealed that the H2O2 oxidation occurs at 0.925 V. By using CV, the linear response of the sensor was obtained from 10 to 200 μM H2O2, and LOD was calculated as 1.27 μM. The selectivity of the sensor was also studied for H2O2 oxidation in the presence of other biomolecules. It was concluded that the proposed GNR/Co3O4-modified electrode could be used for selective detection of H2O2.

2. Experimental

2.1. Materials and Methods

Cobalt nitrate hexahydrate, MWCNTs, sodium hydroxide, and ascorbic acid (AA) were obtained from Sigma-Aldrich, India. Sulphuric acid, H2O2, uric acid (UA), paracetamol (PA), glucose, lactose, L-isoleucine, L-tyrosine, and potassium permanganate (KMnO4) purchased from SRL, India. The phosphate buffer electrolyte solution was prepared using NaH2PO4 and Na2HPO4 () for the sensing of H2O2. All the chemicals were of analytical grade and used without any further purifications. All the solutions were prepared with the milli-Q-water (18.2 MΩ.cm @ ). Electrochemical studies were carried out using a three-electrode system with an electrochemical workstation (Model: CHI-760E, USA). UV-Vis spectra of GNR/Co3O4 nanocomposite was recorded using 2000c nanodrop spectrophotometer, Nanodrop Technologies, USA. All the experiments were done at room temperature (). The surface morphology of GNR/Co3O4 was examined by high-resolution transmission electron microscope (HR-TEM) (Model: 2100, JEOL, Japan). Samples for HR-TEM were prepared by coating of GNR/Co3O4 dispersion on the copper grid and dried in a vacuum oven at room temperature.

2.2. Synthesis of GNR by Unzipping of MWCNTs

0.025 g of MWCNTs was mixed with 7.5 mL of conc. sulfuric acid and then constantly stirred using a magnetic stirrer for 3 h at room temperature. After that, 0.125 g of potassium permanganate (KMnO4) was added, and then, solution temperature was maintained at below 10°C using an ice-bath for 2 h. After that, the temperature was increased up to 35°C for 30 min. and then further increased up to 75°C for the next 30 min. Finally, this mixture was added in to 300 mL of ice water containing 10 mL of 30% H2O2 and filtered. After that, the reaction mixture washed with distilled water until the solution pH reaches neutral. As-synthesized GNR was filtered and dried in the vacuum oven at 60°C for 10 h [56].

2.3. Synthesis of GNR/Co3O4 Nanorods Nanocomposite

Sodium hydroxide (2.0 M, 5.0 mL) solution was added into the cobalt nitrate hexahydrate solution (0.01 M, 50 mL). In this step, the color of the solution turned into pale pink and then dark pinkish brown. 5 mL of GNR (0.1 mg/mL) aqueous dispersion was added into the pinkish-brown solution which turned as a dark black dispersion. After stirring for 0.5 h, AA solution (0.6 M, 5.0 mL) was added drop-wise into the solution. The black color liquid was slowly formed which was stirred for 3 h at 55°C. The precipitate was collected by centrifugation and washed with distilled water to remove the impurities and finally dried in a vacuum oven at 60°C for 5 h. Additionally, the prepared GNR/Co3O4 nanocomposite was calcinated for 2 h at 300°C.

2.4. Preparation of GNR/Co3O4/GCE

Initially, GCE was polished on a polishing pad with the sequence of alumina powder (Al2O3, sizes with ~1 μm, ~0.3 μm, and ~0.05 μm). Next, the GCE was washed with ultrapure Milli-Q water and left to air-dry at room temperature. After that, 10 μL of homogeneous GNR/Co3O4 (1 mg/mL) nanocomposite dispersion was drop casted on the electrode surface and left to dry. To discard the unbounded particles, the GNR/Co3O4/GCE was rinsed with distilled water. In the same manner, GNR/GCE and Co3O4/GCE were prepared for the control experiments.

3. Results and Discussion

3.1. UV-Visible Spectroscopy

UV–Vis spectra of GNR dispersion showed a strong absorbance band at 250 nm, which denoted the transition of aromatic C=C bonds, that was assigned to the partially oxidized graphene (Figure 1(a), curve i) [57]. Figure 1(a) (curve ii) shows the absorbance bands for Co3O4 nanorods at 275 and 480 nm which denoted the charge transfer from (O2− → Co2+ and O2− → Co3+) [58]. The broad peak found at 770 nm can be assigned to the d-d transition band from 4A2 (F) →4T1 (p) transition of Co2+ [58]. Figure 1(a) (curve iii) shows the absorbance spectrum of GNR/Co3O4 nanocomposite, where cobalt charge transfer peaks were blue shifted and observed at 270 and 467 nm. The peak shift was also observed in the d-d transition peak at 760 nm, indicating that composite was formed successfully. Next, the XRD patterns of the GNR/Co3O4 nanocomposite was recorded as shown in Figure 1(b). GNRs exhibited a peak at of 11.9°, which confirmed the unzipping of MWCNTs, and confirmed the successful formation of GNR sheets [59]. The XRD peaks of MWCNTs were diminished and disappeared at 25.8° and 44° [60]. The other XRD bands were observed at (111), 30.7° (220), 38.6° (311), 44.8° (400), 58.7° (511), and 65.38° (440), which indicated the presence of Co3O4 [61].

3.2. HR-TEM

Figures 2(a) and 2(b) show the HR-TEM images of (a) GNR and (b) GNR/Co3O4 nanocomposite. GNR dispersion showed the unzipped MWCNTs clearly with the breadth of 97.2 nm. The GNR could be observed with few ribbons overlapped with each other as shown in Figure 2(a). Interestingly, for the nanocomposite film, Co3O4 nanorods were decorated on the surface of GNRs which can be observed from TEM image of GNR/Co3O4 nanocomposite.

3.3. Electrochemical Oxidation of H2O2

Electrochemical activities of GNR/Co3O4-modified GCE and bare GCE were performed by cyclic voltammetry (CV). GNR/Co3O4/GCE exhibited two redox peaks. In the first anodic scan, two oxidation peaks a1 and a2 were observed at 0.68 V and 0.93 V. In the reverse scan, two cathodic peaks (c1 and c2) were observed at 0.65 V and 0.88 V, respectively. The formal potential of the two redox peaks were found to be 0.665 V (peak I) and 0.905 V (peak II) (Figure 3 curve b). Cobalt oxide-modified electrode showed two pairs of redox peaks due to the Co(II)/Co(III) transition at the GCE surface as given in Equation (1) [62]. Electrocatalytic activity of the GNR/Co3O4-modified GCE (Figure 3, curve a) was recorded in 0.1 M phosphate buffer solution (PBS) containing 30 μM H2O2. Due to the high electrocatalytic activity of the GNR/Co3O4/GCE, oxidation peak of Co3O4 was enhanced due to H2O2 oxidation at 0.925 V (Figure 3 curve a) Equation (2). For comparison, CVs of bare GCE was recorded from 0.2 to 1.1 V at the scan rate of 50 mV/s (Figure 3, curve c). Bare GCE showed only the nonfaradic current, which clearly indicated that GNR/Co3O4-modified GCE had more electrocatalytic activity. The surface coverage of the GNR/Co3O4-modified electrode was calculated using Equation (3), where is the number of electrons (), is the faraday constant, is the gas constant, is the area of the electrode (0.0707 cm2), and is the temperature [6]. From the slope of the graph between H2O2 oxidation current and square roots of scan rate (Figure 4), the surface coverage (ᴦGNR/Co3O4) of the electrode was calculated as .

To ascertain about the effects of scan rate, CVs of GNR/Co3O4/GCE were recorded in 0.1 M PBS containing 30 μM H2O2 with different scan rates (Figure 4(a)). The anodic peak currents of GNR/Co3O4/GCE were increased linearly with the increase of the scanning rate from 10 to 200 mV/s. The linear equation was established between H2O2 oxidation peak currents and square root of scan rates with a correlation coefficient of () 0.9893 (Figure 4(b)). It was found that H2O2 oxidation on GNR/Co3O4/GCE was a diffusion-controlled process [18].

Next, electrocatalytic oxidation of H2O2 on the GNR/Co3O4/GCE-modified electrode was studied in 0.1 M PBS by varying the concentrations. The oxidation peak of H2O2 takes place at the oxidation potential of Co3+ (0.925 V) with each addition of 10 μM of H2O2. H2O2 oxidation peak currents were increased linearly from 10 to 200 μM (Figure 5(a)). Hence, a calibration plot was made between the concentration of H2O2 and the oxidation peak currents, which resulted in a linear equation of and () of 0.9966 (Figure 5(b)).

This data confirmed that GNR/Co3O4 can be used as an appropriate electrocatalyst for the analysis of H2O2. The limit of detection (LOD) was estimated as 1.27 μM. The sensitivity of the sensor was calculated as 5.10 μA μM-1 cm-2. We have also compared the analytical performance of this new sensor with other reported sensors in Table 1. It was clear that this new sensor is more promising than some of the reported H2O2 sensors.

3.4. Effect of pH and Interferent Studies on GNR/Co3O4/GCE

GNR/Co3O4/GCE was subjected to study the effect of pH. CVs were recorded in different pH electrolyte solutions (from pH 1 to 9). It was found that the redox peak II of GNR/Co3O4/GCE was pH dependent which showed that protons are involved in the electron-transfer process. Figure 6(a) shows the linear plot of vs. pH which gave a slope value of -80 mV/pH. This indicated that an unequal number of protons and electrons were involved in the electrochemical reactions [66]. For the electrochemical detection of H2O2, pH 7.4 was chosen as an appropriate electrolyte because of the physiological condition of the solutions.

The interferents analysis was carried out in the presence of H2O2 using the GNR/Co3O4/GCE. In this study, various common interfering biomolecules associated with the human body such as AA, PA, UA, glucose, lactose, L-isoleucine, and L-tyrosine were added (10 μM of each compound) with the analysis of 10 μM H2O2 (Figure 6(b)) [67]. These interfering molecules did not show any significant response on the GNR/Co3O4/GCE. Therefore, it was suggested that GNR/Co3O4/GCE may be used for selective analysis of H2O2 by electrochemical oxidation.

3.5. Electrochemical Impedance, Repeatability, and Stability Studies

Electrochemical impedance spectroscopy (EIS) could reveal about the solid–liquid interface process of the modified electrodes. Figure 7 shows Nyquist plots obtained for (i) bare GCE, (ii) GNR/GCE, (iii) Co3O4/GCE, and (iv) GNR/Co3O4/GCE in 0.1 M KCl containing 5 mM [Fe(CN)6]3−/4−. The Nyquist impedance spectra obtained for (i) bare GCE (580 Ω), (ii) GNR/GCE (500 Ω), (iii) Co3O4/GCE (490 Ω), and (iv) GNR/Co3O4/GCE (425 Ω) indicated the charge transfer resistance (Rct) of each electrode by various diameters of semicircles [16]. For bare GCE, a small semicircle was found with the resistance of 580 Ω due to good electron transfer process. However, after modification with GNR/GCE layer, the electron transfer resistance was slightly decreased to 500 Ω, and Co3O4/GCE showed Rct of about 490 Ω. Interestingly, as-prepared nanocomposite showed the lowest Rct of about 425 Ω [68, 69], which indicated the high conductivity of GNR in the nanocomposite. As shown in the inset of Figure 7, an equivalent electrical circuit model was prepared and fitted with the EIS data of GNR/Co3O4/GCE. In addition, this sensor retained 93% of the electrode response after 24 h of usage (Figure 8(a)) due to good stability of the materials. The repeatability of the GNR/Co3O4/GCE sensor was also tested with 30 μM H2O2. This new sensor response was only decreased about 5.4% after 50 potential cycles (Figure 8(b)).

3.6. Real Sample Analysis

The real-world sample analysis was carried out using a GNR/Co3O4/GCE sensor. In order to measure the concentration of H2O2 in a real-world sample, a commercial antiseptic sample was diluted for 10 times in phosphate buffer solution and spiked with 40, 50, and 60 μM of H2O2. These samples were analyzed by CV using a GNR/Co3O4/GCE sensor. The obtained results are shown in Table 2. The relative standard deviation (RSD) values for the detection of spiked H2O2 in three different samples were found to be 4.01%, 3.61%, and 3.26%. Also, the recoveries of spiked H2O2 were calculated using the same method, and the values were found in the range of 96.75 to 99.60%. Therefore, the above results indicated that H2O2 can be effectively determined in a real-world sample using GNR/Co3O4/GCE sensor. We believe that this new method can be adopted for the detection of H2O2 in various samples from the environmental and medical fields.

4. Conclusion

In this work, GNR was synthesized by unzipping of MWCNTs, and the Co3O4 nanorods were synthesized by AA as a reducing agent in alkaline condition. After that, GNR/Co3O4 nanocomposite was successfully prepared and used to modify the surface of GCE. UV-Vis and HR-TEM results showed the successful formation of Co3O4 nanorods on the GNR. The electrooxidation of H2O2 on the GNR/Co3O4-modified electrode taken place at the potential of 0.93 V. Furthermore, GNR/Co3O4 nanocomposite-coated GCE was successfully applied for the detection of H2O2 with high sensitivity and selectivity. The effect of scan rate on H2O2 oxidation at GNR/Co3O4 indicated the diffusion-controlled electrochemical process. Using CV, the linear range of H2O2 concentration was observed from 10 to 200 μM, and the LOD was estimated as 1.27 μM. The selectivity of the sensor was also studied in the presence of other biomolecules; it was confirmed that GNR/Co3O4-modified electrode may be used for selective detection of H2O2 with lower LOD.

Data Availability

The research data used to support the findings of this study are included in the article.

Disclosure

This research work was presented as an abstract at the International Conference on Technologies for Smart Green Connected Society, 2021 (SPAST Abstracts, 1(01); https://spast.org/techrep/article/view/1369).

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Authors’ Contributions

Preethika Murugan, Ashok K. Sundramoorthy, and Raji Atchudan contributed equally to this work.

Acknowledgments

AKS thanks the Science and Engineering Research Board (SERB) for funding through CRG/2021/001517. We thank the Department of Science and Technology (DST) (International Bilateral Cooperation Division), India, for financial support through “INDO-RUSSIA Project (File No. INT/RUS/RFBR/385).” This work was also funded by the Researchers Supporting Project number (RSP-2021/243), King Saud University, Riyadh, Saudi Arabia.