Introduction

Several practical applications including industrial and clinical fields use hydrogen peroxide (H2O2), which is a by-product of many biological reactions catalyzed by oxidase. Even though H2O2 is utilized as bleaching agent, disinfectant, and oxidizing agent, excess of H2O2 can damage the proteins and DNA in humans [1,2,3,4]. The tolerable exposure limit of H2O2 for human beings given by the Occupational Safety and Health Administration (OSHA) is 1 ppm or 1.4 mg/m3 time-weighted average (TWA) [5]. Thus, it is essential to monitor the quantity of H2O2 at the present environment. Many analytical methods such as titrimetry [6], high-performance liquid chromatography [7], spectrophotometry [8], fluorimetry [9], and chemiluminescence [10] have been employed for the determination of H2O2 in recent years. However, the main obstacles such as high cost, complicated procedure, time consumption, and low selectivity of the abovementioned methods are considered to be replaced. In order to overcome these obstacles, the researchers have switched over to electrochemical methods as they offer more flexibility, quick response, high sensitivity, and selectivity [11,12,13]. The chemically modified electrodes with highly active catalysts have been tremendously explored and used for the fabrication of electrochemical sensors and biosensors to increase the sensitivity, selectivity, and stability of the sensing interferences. The model enzymes such as horseradish peroxidase [14], myoglobin [15], hemoglobin [16], microperoxidase-11 [17], and catalase [18] immobilized active catalyst biosensors have been fabricated for the selective and sensitive detection of H2O2. However, the enzymatic biosensors used for the determination of H2O2 were not capable to overcome the issues including poor stability, forbearance, high cost, poor repeatability, and reproducibility. Thus, researches focusing on enzyme-free sensors for H2O2 determination are more desirable in recent years [19].

For the enzyme-free detection of H2O2, many materials have been utilized to chemically modify the electrodes. Remarkably, the transition metal oxides [ZnO, NiO, CuO, Cu2O, Co3O4, SnO2, SmO, and etc.] have attracted the present-day researchers towards the electrochemical detection of H2O2 in terms of high stability, good catalytic activity, environmental compatibility, and unusually useful electronic and magnetic properties [20,21,22,23,24,25]. Unlike other transition metal oxides, manganese oxides (MnO2, Mn2O3) have been chosen for the efficient and sensitive detection of H2O2 [26,27,28,29,30]. The important features of MnO2 as a transition metal oxide, including higher abundance, ecofriendly nature, and unique physical and chemical properties, have made it as a potential metal oxide for the catalysis and energy storage device applications [31]. Besides, the conductance of MnO2 has been increased by the co-synthesis of carbon materials such as activated carbon [32], graphene [33], and MWCNT [34].

Carbon nanotubes (CNTs) are cylindrical allotropes of carbon made up of hexagonally arranged graphene sheets [35]. Among the single-walled carbon nanotubes (SWCNTs) and multi-walled carbon nanotubes (MWCNTs), the latter is more desirable for the current researches owing to their inimitable electrical, mechanical, structural, and optical properties in the field of sensing and bio-sensing research [36, 37]. Conversely, MWCNTs suffer from poor dispersion ability in aqueous medium. However, the surface alteration through oxidation can result in good adsorption and dispersion ability of CNTs was evident from the previous literatures. Thus, MWCNTs carrying the functional groups (f-MWCNTs) such as OH, COOH, and CO have received more attention from the researchers [38, 39].

In our present study, f-MWCNTs with manganese oxide nanoflakes (MnO2 NFs) have been prepared via a simple chemical approach. The formation of f-MWCNTs/MnO2 NFs was confirmed by field emission scanning electron microscopy (FESEM), energy-dispersive X-ray spectroscopy (EDX), X-ray diffraction (XRD) measurements, and X-ray photoelectron spectroscopy (XPS). The prepared nanocomposite was coated on the surface of a glassy carbon electrode (GCE) for the electro catalysis applications. Our fabricated modified electrode was used to detect H2O2 by electrochemical methods. f-MWCNTs/MnO2 NFs/GCE showed an excellent electrocatalytic activity towards the electro-reduction of hydrogen peroxide in terms of its high sensitivity, selectivity, low limits of detection, and feasible practicality. Moreover, the obtained values of limits of detection (LOD), sensitivity, and linear range are comparable with the formerly reported MnO2-based electrochemical hydrogen peroxide sensor [40,41,42,43,44,45,46,47,48,49].

Experimental

Reagents and apparatus

MWCNTs (powder, <20 μm), KMnO4, and H2O2 were purchased from Sigma-Aldrich. All other chemicals used were of ACS-certified reagent grade and used without further purification. The supporting electrolyte used for all the electrochemical studies was 0.05 M phosphate–buffer (pH 7) solution (PBS). Prior to each experiment, all the solutions were deoxygenated with pre-purified N2 gas for 15 min unless otherwise specified. Double-distilled water with a conductivity of ≥18 MΩ cm−1 was used for all the experiments.

Preparation and fabrication of f-MWCNTs/MnO2 NFs/GCE

f-MWCNTs were prepared as given in the previous literatures [40]. Firstly, 0.5 g of crude MWCNTs was mixed with 60 ml aqueous solution of 0.4 M HCl in a beaker. It was sonicated in a water bath for 4 h, and 3:1 of concentrated H2SO4/HNO3 (60 ml) was added to the above mixture under constant stirring. After refluxing for 4 h, the mixture was allowed to attain room temperature and it was diluted with double-distilled water (400 ml). The diluted mixture was filtered several times till its pH value reaches 7. The resulting f-MWCNTs solid was dried for 12 h in 70 °C oven. Later, this f-MWCNTs was further used in the preparation of f-MWCNT/MnO2 NF composite.

Secondly, f-MWCNTs/MnO2 NFs were synthesized through a simple chemical method as follows: 0.225 g KMnO4 and f-MWCNTs (1 mg/ml) were dissolved in 40 ml double-distilled water. The solution was then transferred into a round bottom flask. One milliliter of concentrated HCl solution was injected in to the abovementioned solution and stirred for 30 min. The mixture was heated at 110 °C for 24 h. A brown-colored MnO2 NF product was obtained after the completion of reaction. The centrifugal filtration method was used to collect the powdered MnO2 NFs. Later, it was dried at 100 °C in order to remove the excess oxygen in MnO2 NFs. For the comparison, the same procedure was followed to prepare only MnO2 NFs in the absence of f-MWCNTs. Buehler polishing kit was used to polish the surface of a GCE with 0.05 μm alumina slurry. The polished electrode was cleaned and dried. Later, 10 μl of f-MWCNTs/MnO2 NFs (1 mg|ml) was drop-cast on the surface of a pre-cleaned GCE and dried at ambient conditions for performing the further electrochemical experiments. Similarly, MnO2 NF and f-MWCNT-modified GCEs were also prepared individually for the comparative studies. A schematic depiction for the preparation and fabrication of f-MWCNT/MnO2 NF-modified GCE for the detection of H2O2 is given in Scheme 1.

Scheme 1
scheme 1

A schematic depiction for the preparation and fabrication of f-MWCNT/MnO2 NF-modified GCE towards the electro-reduction of H2O2

Full size image

Result and discussion

Characterization of prepared f-MWCNT/MnO2 NF composite

FESEM technique was used to characterize the morphology of our prepared nanocomposites. Figure 1 shows the FESEM images of MnO2 NFs (a) and f-MWCNTs/MnO2 NFs (b). It can be clearly seen that the higher magnification FESEM image of MnO2 nanocomposite shows the highly dense nanoflakes assembly and the lower magnification FESEM image of MnO2 nanocomposite shows the nanoflake-interconnected flower-like structrue with a diameter of 3 μm (inset to Fig. 1a). In addition, f-MWCNTs interconnected with the surface of MnO2 NFs were observed in the FESEM image of f-MWCNT/MnO2 NF composite. In order to study the crystal plane of our prepared composite, XRD technique was employed. Figure 1c shows the XRD pattern observed for MnO2 NFs and f-MWCNTs/MnO2 NFs. Three different diffraction peaks were observed for MnO2 NFs at the 2θ values of 12.08°, 37.78°, and 60.71° with their corresponding crystal planes at (001), (201), and (311), respectively. The XRD pattern of MnO2 NFs was in concordance with the previously reported monoclinic potassium birnessite-type MnO2 composite [41]. In addition, a new crystal plane was observed at 25.34°, confirming the graphitic network of f-MWCNTs [42]. Moreover, the observed crystal plane of f-MWCNTs with low intensity reveals the higher incorporation of MnO2 NFs with f-MWCNTs. Energy-dispersive X-ray (EDX) spectrum of f-MWCNTs/MnO2 NFs was presented in Fig. 1d. From the figure, the EDX signals of carbon, oxygen, and manganese with their corresponding weight percentages of 8.36, 42.88, and 51.38 were clearly visible. Yet again, the lower weight percentage of carbon in the elemental analysis validates the higher incorporation of MnO2 NFs with f-MWCNTs. X-ray photo electron spectroscopy results given in Fig. 2 also strongly substantiate the oxidation state of Mn and elemental composition of the prepared composite. As shown in the figure, the corresponding peaks of O 1s, C 1s, Mn 2p3/2, and Mn 2p1/2 were seen at the binding energies of 528.6, 283.2, 641, and 653 eV, respectively. The XPS results observed for f-MWCNTs/MnO2 NFs were in close agreement with the formerly reported XPS results of the literatures [43].

Fig. 1
figure 1

FESEM images of MnO2 NFs (a) and f-MWCNTs/MnO2 NFs (b). XRD (c) and EDX (d) results of f-MWCNTs/MnO2 NFs

Full size image
Fig. 2
figure 2

XPS spectra of f-MWCNTs/MnO2 NFs

Full size image

Electro-reduction of H2O2 at various modified electrodes

Figure 3a displays the CVs at MnO2 NF (a), f-MWCNT (b), and f-MWCNT/MnO2 NF (c) film-modified electrodes in 0.05 M PBS (pH 7) in the absence and presence of 1 mM H2O2. MnO2 NF-modified GCE showed a sharp cathodic peak at the potential of −0.65 V with a peak current of −74.4 μA for 1 mM H2O2. It can be due to the good catalytic activity of MnO2 NFs towards the reduction of H2O2, whereas f-MWCNTs/GCE exhibited a feeble cathodic peak at the potential of −0.432 V and the value of peak current was found to be −12.2 μA. An enhanced reduction peak current (−526 μA) appeared at MWCNTs/MnO2 NFs/GCE. The obtained peak current is fivefold higher than MnO2 NFs/GCE and several folds higher than that of f-MWCNTs/GCE. From Fig. 3a–c, the overall cathodic peak current (E pc) towards the electrochemical reduction of H2O2 is in the following order: f-MWCNTs > MnO2 NFs > f-MWCNTs/MnO2 NFs. Moreover, after optimizing the concentration of MWCNTs/MnO2 NFs (1 mg/ ml) at GCE, we concluded that 10 μl of 1 mg/ml MWCNTs/MnO2 NFs showed an enhanced catalytic activity (inset in Fig. 3c). The better electro-reduction of H2O2 at f-MWCNTs/MnO2 NFs/GCE than that of other modified electrodes can be due to the presence of hydrophilic COOH-functionalized MWCNTs. These functional groups increase the incorporation of a large quantity of MnO2 NFs with f-MWCNTs. As a result, these f-MWCNTs/MnO2 NFs provide higher surface area and catalytic sites for the effective electrochemical reduction of H2O2. The probable mechanism for the electro-reduction of H2O2 at f-MWCNTs/MnO2 NFs/GCE is given in Eqs. 1 and 2 [44].

$$ \mathrm{Mn}{\mathrm{O}}_2+{\mathrm{H}}_2{\mathrm{O}}_2\hbox{--} >\mathrm{M}{\mathrm{n}}_2{\mathrm{O}}_3+{\mathrm{H}}_2\mathrm{O}+{\mathrm{O}}_2 $$
(1)
$$ \mathrm{M}{\mathrm{n}}_2{\mathrm{O}}_3+2\mathrm{O}{\mathrm{H}}^{-}\hbox{--} >2\mathrm{Mn}{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}+{2\mathrm{e}}^{-} $$
(2)
Fig. 3
figure 3

Cyclic voltammograms obtained at MnO2 NFs (a), f-MWCNTs (b) and f-MWCNTs/MnO2 NFs (c) film-modified electrodes in 0.05 M PBS (pH 7) for the absence (violet curve) and presence (red curve) of 1 mM H2O2. (d). Cyclic voltammograms obtained at f-MWCNTs/MnO2 NFs/GCE in 0.05 M PBS (pH 7) at different scan rates from 0.02 to 0.16 V s−1 (a–h) in the presence of 1 mM H2O2. Inset: calibration plot between I p vs ν1/2

Full size image

The influence of scan rate at f-MWCNTs/MnO2 NFs/GCE towards the detection of H2O2 in 0.05 M PBS (pH 7) was studied using a cyclic voltammetry and represented in Fig. 3d. The scan rate was varied from 0.02 to 0.16 V/s (a–h). An increase in cathodic peak was observed for the increasing scan rate from 0.02 to 0.16 V/s which in turn leads to the potential shift in the negative direction. The linear dependency of the cathodic peak current with the square root of the scan rate was evident from the inset calibration plot. Therefore, the electro-reduction of H2O2 at f-MWCNT/MnO2 NF film-coated GCE follows a diffusion controlled process.

Amperometric i–t determination of H2O2 at f-MWCNTs/MnO2 NFs/GCE

In a three-electrode cell, the commonly used hydrodynamic working electrode is a rotating disc electrode. A flux of analyte is induced to the electrode during its rotation in the amperometric i–t measurement. Thus, it shows high sensitivity towards the detection of a desired analyte [44]. In this present work, f-MWCNT/MnO2 NF film modified at rotating glassy carbon electrode (fixed rpm, 1500 rpm) was utilized for the amperometric determination of H2O2. The displayed amperograms of our modified electrode for the consecutive additions of H2O2 in 0.05 M PBS (pH 7) with increasing concentration (5, 50, and 500 μM) was shown in Fig. 4. −0.4 V was fixed to be the applied potential for the amperometric determination of H2O2. A steady and quick cathodic peak current response towards each addition of H2O2 reveals that f-MWCNTs/MnO2 NFs/GCE attained 95% of steady-state current response within 4 s during H2O2 detection in this study. Each successive addition of H2O2 resulted in an increasing cathodic peak current with the increase of H2O2 concentration. In addition, the discrimination of peaks for the varying concentration of H2O2 can be clearly seen from the presented figure. The linear relationship of cathodic peak current response with H2O2 concentration is evident from the equivalent calibration plot (inset in Fig. 4). The relevant electroanalytical parameters namely linear range, LOD, and sensitivity for our modified electrode were calculated to be 5 to 4530 μM, 0.952 μM, and 219.05 μA mM−1. The equation, LOD = 3s b/S (where s b = standard deviation of blank signal and S = sensitivity) was used to calculate the LOD value of our modified electrode [18]. The evaluated values of linear range, LOD, and sensitivity of f-MWCNTs/MnO2 NFs/GCE were in close agreement with the electroanalytical parameters for the related sensors as previously reported in the literatures (Table 1).

Fig. 4
figure 4

Amperograms obtained at f-MWCNT/MnO2 NF composite-modified rotating disk electrode upon the successive additions of H2O2 into 0.05 M PBS (pH 7). Working potential = −0.4 V. Inset: calibration plot of I pa vs [H2O2]

Full size image
Table 1 Comparison of electroanalytical parameters for the H2O2 determination at f-MWCNT/MnO2 NF-modified GCE with other film-modified electrodes
Full size table

Repeatability, reproducibility, and stability

The results of repeatability and reproducibility tests were also acquired from the amperometric i–t technique. In order to examine the repeatability behavior of f-MWCNTs/MnO2 NFs/GCE, six repetitive measurements were performed by using a single modified electrode in 0.05 M PBS (pH 7) containing 1 mM H2O2 at a scan rate of 50 mV/s, and as a result, an appreciable repeatability with the relative standard deviation (RSD) of 2.8% was obtained. Six individual f-MWCNT/MnO2 NF-modified GCEs were fabricated to study the reproducing capability of our sensor under the abovementioned experimental conditions. A significant reproducibility of f-MWCNTs/MnO2 NFs/GCE was achieved, and the relative standard deviation (RSD) was found to be 3.4%. From the tests for repeatability and reproducibility, it is evident that our fabricated modified electrode was not affected by the oxidation or reduction by-products and it can be used frequently with an excellent reproducibility. After the completion of every day experiment towards H2O2 determination for a time period of 30 days, the modified electrode was stored in 0.05 M PBS (pH 7) at 4 °C. Even after a month, the amperometric response at f-MWCNTs/MnO2 NFs/GCE for H2O2 detection resulted in 94.5% of the initial response current which reveals the good operational and storage stability of f-MWCNTs/MnO2 NF-modified GCE.

Selectivity studies

Similarly, amperometry was again used to study the selective nature of our sensor towards H2O2 (Fig. 5). The interfering biomolecules including 0.5 mM glucose (b), nitrate (c), nitrite (d), ascorbic acid (e), dopamine (f), and uric acid (g) were chosen for the selectivity study of f-MWCNTs/MnO2 NFs/GCE. However, our fabricated electrode exposed a well-defined and superior peak current response for 100 μM concentration of H2O2. Since low potential detection is possible for H2O2, there were no considerable peaks for the other biologically interfering species. This study reveals the high selectivity of our f-MWCNT/MnO2 NF-modified electrode towards H2O2 determination.

Fig. 5
figure 5

Amperometric i–t response at f-MWCNT/MnO2 NF composite-modified rotating disc electrode for the addition of 0.1 mM H2O2 (a) and 0.5 mM addition of glucose (b), nitrate (c), nitrite (d), ascorbic acid (e), dopamine (f), and uric acid (g) in 0.05 M PBS (pH 7)

Full size image

Real sample analysis

The practical feasibility of f-MWCNTs/MnO2 NFs/GCE was demonstrated in commercial cleaning solution for contact lens (purchased from local Watson medical store, Taipei, Taiwan) which contains 3% H2O2 of the total compounds comprising pluronic 17R4, phosphate, sodium chloride, and phosphonic acid. The obtained three different real clinical samples were diluted with 0.05 M PBS (pH 7) and used for the determination of H2O2. The response for the determination of H2O2 was recorded in 0.05 M PBS (pH 7) at 25 °C (room temperature) with the simultaneous addition of lab and real H2O2 samples (Fig. 6). The addition of commercial lens cleaning solution does not influence the steady-state response of H2O2 at f-MWCNT/MnO2 NF-modified GCE. Standard addition method was used for further three continuous measurements taken in this study, and the obtained adequate recovery results were given in Table 2. Therefore, our modified fabricated electrode holds good for the practical applications.

Fig. 6
figure 6

Amperometric i–t response for lab (a) and real samples of H2O2 (b–d) in 0.05 M PBS (pH 7) at f-MWCNTs/MnO2 NFs/GCE. Applied potential, E app = −0.4 V, rotation speed = 1500 rpm

Full size image
Table 2 Determination of H2O2 in clinical lens solutions using f-MWCNT/MnO2 NF film-modified GCE (n = 3)
Full size table

Conclusions

Finally, we have successfully constructed f-MWCNT/MnO2 NF film-modified GCE for H2O2 detection with high sensitivity and selectivity by following a simple electrochemical methodology. The incorporation of MnO2 NFs with f-MWCNTs was confirmed from FESEM (surface morphology), XRD (crystal structure), EDX, and XPS (elemental composition) techniques. An excellent electro-reduction of H2O2 at f-MWCNTs/MnO2 NFs/GCE was evident from the steady and well-defined cathodic peak current responses that appeared in the studies including cyclic voltammetry and amperometry. In this work, the evaluated electroanalytical parameters such as a wide linear range, low LOD, and high sensitivity exhibit the good electrocatalytic activity of our fabricated electrode towards determination of H2O2. The accurate detection of H2O2 in commercially used clinical lens cleaning solutions validates the practical feasibility of our sensor. Thus, near-future applications (biosensors, electronics, and optics) can make use of f-MWCNTs/MnO2 NFs/GCE as a better stand for the potential developments.