Nafion/Multiwalled Carbon Nanotubes/Mesoporous Silica Composite Modified Glassy Carbon Electrode for the Simultaneous Electrochemical Determination of Indigo Carmine and Carbendazim by Differential Pulse Voltammetry

Abstract A simple and reliable method based on a composite of Nafion (Naf), multiwalled carbon nanotubes (MWCNTs), and mesoporous silica (SBA-15) modified glassy carbon electrode (GCE) is reported for individual and simultaneous determination of indigo carmine (IC, dye) and carbendazim (CBZ, pesticide). SBA-15 was synthesized and characterized and used to design a GCE/Naf/MWCNTs/SBA-15 sensor, which was characterized by electrochemical techniques and used for the determination of IC and CBZ. Using the optimum conditions for differential pulse voltammetry (DPV), the peak currents were linear from 1.0 µM to 11.0 µM. The limits of detection for individual measurements were 0.50 µM and 0.57 µM for IC and CBZ, respectively, while those for simultaneous detection were 0.28 µM and 0.30 µM. The interferences studies show that GCE/Naf/MWCNTs/SBA-15 exhibited excellent selectivity. The developed procedure was successfully applied to determine IC and CBZ in spring, tap, and lake water.


Introduction
A wide range of foods, textiles, cosmetics, drugs, and papers contain pollutants such as indigo carmine (IC) and carbendazim (CBZ) (Dong, Lijun, and Lei 2017).IC is a hydrophilic blue dye known to be toxic to humans, causes irritation to the skin and eyes, and contributes to the development of cancer (Tiago et al. 2016;Incebay and Kilic 2022), whereas CBZ is a benzimidazole fungicide known to cause severe health problems due to its high toxicity (Teadoum et al. 2016;Shweta et al. 2020).Therefore, the coexistence of these pollutants in the environment constitutes a serious problem, and hence, it is necessary to develop a determination platform for environmental security and human health.
The determination of IC and CBZ is carried out using fluorescence (Reyes, Barrales, and Dı az 2003;Yadi, Zhang, and Junhui 2012), capillary electrophoresis (Fusako et al. 2004;Javier et al. 2013), spectrophotometry (Berzas et al. 1999;Nahid, Rastegarzadeh, and Larki 2015), and liquid chromatography-ultravioletmass spectrometry (Ana et al. 2012;Tsai, Kuo, and Shih 2015;Temgoua et al. 2021).Compared to these methods, electrochemical methods are simple, need minimal cost, and rapid (Hazem et al. 2013;Majid and Ghodsi 2014).However, despite the fact that IC and CBZ are electroactive species, there is no report in the literature on their simultaneous electrochemical determination.Thus, here is reported an electrochemical method for the simultaneous determination of IC and CBZ.
Recently, many authors have proposed electrochemical methods for the individual determination of IC and CBZ (Teadoum et al. 2016;Majid et al. 2017;Chunhuan et al. 2019;Incebay and Kilic 2022), but the need of good sensitivity and selectivity is still a preoccupation.Therefore, for voluminous and complex compounds such as IC and CBZ, the preparation of composite materials from multiwalled carbon nanotubes (MWCNTs) and mesoporous silicate (SBA-15) is a good starting point for their simultaneous determination.The possibility of improving the sensitivity and selectivity of the analysis through the use of modified electrode Naf/MWCNTs/SBA-15 is expected (Eduardo, Santana, and Spinelli 2023).MWCNTs are important electrode materials, since they play a favorable role in the transfer of electron and have good electrocatalytic ability (Zen, Ting, and Shih 1998).
Nafion (Naf) is widely used to dissolve MWCNTs and it has a capacity to decrease the oxidation potential of some organic compounds (Sundari, Palaniappan, and Manisankar 2010).Nafion has a negative global charge, and therefore attracts positively charge species.Santa , also called SBA-15, is a porous material with a wide range of potential applications including shape-selective catalysis and sorption of large organic molecules (Piyali et al. 2017).Therefore, it is expected that, by combining Naf, MWCNTs, and SBA-15 into a unique electrode material, the synergistic effect of these three components may lead to a sensitive determination of IC and CBZ.
The present work describes the determination of IC and CBZ via differential pulse voltammetry (DPV) which is one of the most sensitive pulse methods (Deffo et al. 2023) using a glassy carbon electrode (GCE) modified by Naf/MWCNTs/SBA-15.
A l-Autolab potentiostat type III (Ecochemie, Holland) running with GPES software, was used for cyclic voltammetry (CV) and DPV.A standard single compartment threeelectrode cell was used with an Ag/AgCl/3M KCl reference electrode and a stainless-steel rod counter electrode.The working electrode was a bare glassy carbon electrode (GCE) modified with the composite Naf/MWCNTs/SBA-15 prepared as described below.
Electrochemical impedance spectroscopy (EIS) was performed using a PalmSens potentiostat (Bvt technology, Czech Republic).SBA-15 was characterized by Fourier transform infrared (FTIR) spectroscopy using a Thermo Scientific Nicolet 8700 apparatus (LabWrench, Netherlands) equipped with a specular reflectance accessory (Smart Collector).The crystallinity of the synthesized SBA-15 material was determined by Xray diffraction (XRD) using a Bruker D-8 Advance Swax instrument (Bruker, Germany) operated at 40.0 kV and 40.0 mA calibrated with a standard silicon sample using Ni-filtered Cu-Ka (0.15406 nm).The Brunauer-Emmett-Teller (BET) surface area of the sample was characterized by N 2 adsorption/desorption isotherms obtained using a Beckman Coulter SA 3100 (Akribis Scientific, United Kingdom) at 77.0 K.The surface area was determined using the linear part of the BET equation.
Prior to its modification by drop-coating, bare GCE (geometry surface 0.07 cm 2 ) was polished with alumina paste.It was then placed in a 1:1 ethanol-water and cleaned in a sonicator for 5 min to eliminate any remaining alumina particles.2.5 mg of MWCNTs were dispersed in 100 mL Naf (5%) and 400 mL of ethanol and the mixture was sonicated to give homogeneous Naf/MWCNTs.2.5 mg of SBA-15 in 500 mL of distilled water were added and Naf/MWCNTs/SBA-15 dispersion was obtained after sonication for 5 min.The clean GCE surface was coated with 8 lL of resulting dispersion and dried at 60 C for 10 min to obtain the working electrode (hereafter referred as GCE/Naf/MWCNTs/ SBA-15).For comparison purposes, bare GCE, GCE/Naf, GCE/Naf/MWCNTs/ and GCE/SBA-15 were also prepared in a similar way.

Electrochemical procedure
Cyclic voltammograms were recorded in H 2 SO 4 and phosphate buffer (PB) containing IC and CBZ from À 0.25 V to þ 0.90 V and þ 0.30 V to þ 1.40 V, respectively.For individual and simultaneous determination of IC and CBZ, DPV was performed using the following optimized parameters: pulse amplitude 50.0 mV, step potential 5.0 mV, equilibrium and pulse duration time 5.0 s, scan rate 50 mVs À1 , potential range from þ 0.50 V to þ 0.85 V, þ 0.60 V to þ 1.40 V, and þ 0.50 V to þ 1.40 V, respectively, for IC, CBZ, and IC þ CBZ.The bare and modified GCE were characterized using [Fe(CN) 6 ] 3-and [Ru(NH 3 ) 6 ] 3þ ions as redox probes in 0.10 M KCl to assess the ionexchange, permselectivity, and conductivity.EIS was carried out to evaluate the charge transfer resistance of each electrode at the frequency range from 10.0 kHz to 0.10 Hz and with an amplitude of 0.20 V.
For real analysis, three water samples of the city of Dschang (Cameroon, Central Africa) taken from polluted municipal lake water; Foto district spring water, and tap water were collected.Before analysis, all samples were passed through filter paper.

Results and discussion
Physicochemical characterization of SBA-15 FTIR spectroscopy was used to identify the functional groups present in SBA-15 as shown in Figure 1a.A stretching vibration at 3456 cm À1 is characteristic of hydrogen bonded Si-OH groups (Nongyue, Bao, and Xu 1998;Timofeeva et al. 2007).The adsorption bands at 1085 and 806 cm À1 are assigned to the stretching and bending vibrations of silica-oxygen tetrahedrons in SBA-15 (Timofeeva et al. 2007;Vandana, Sasidharan, and Bhaumik 2015).
X-ray diffraction of SBA-15 silica exhibited a single strong peak at 10.40 nm as shown in Figure 1b corresponding to the characteristic of the lattice d 100 plane.Two other reflections registered at small diffraction angles can be indexed as ( 110) and ( 200) planes of the 2D-hexagonal lattice (Zhao et al. 1998;Michal et al. 2000;Sudipta, Mondal, and Bhaumik 2013;Anna et al. 2016).The synthesized SBA-15 exhibited hexagonal ordering and characteristics of the mesoporous material.
The nitrogen adsorption isotherm for the SBA-15 (Figure 2a) is similar to reported in the literature (Zhao et al. 1998;Michal et al. 2000;Sudipta, Mondal, and Bhaumik 2013), characteristic of the high quality SBA-15.The sharpness of the type IV adsorption branch is indicative of a narrow mesoporous size distribution.The adsorption branch was located at relative pressures from 0.60 to 0.80 and the determined surface area was 734.0 m 2 g À1 .Further, a large desorption hysteresis of H1 type suggested the presence of large mesopores in the material.
The pore size distribution (PSD) for the SBA-15 sample is shown in Figure 2b.The pore size corresponding to the maximum is 8.26 nm.The total pore volume estimated from this N 2 sorption isotherm was 0.92 ccg À1 .The sharp maxima for pore size between 5.81 and 9.53 nm and desorption hysteresis are similar to the literature (Michal et al. 2000).

Permeability of the modified electrodes and EIS
Cyclic voltammetry was performed using the bare GCE, GCE/Naf, GCE/Naf/MWCNTs, GCE/SBA-15, and GCE/Naf/MWCNTs/SBA-15 in 0.10 M KCl and 10 À3 M [Fe(CN) 6 ] 3- or 10 À3 M [Ru(NH 3 ) 6 ] 3þ at pH 7.0.Both probes gave rise to CV responses as shown in Figure 3a and b.For the [Fe(CN) 6 ] 3-probe, when modified the surface of GCE with the successive materials, the peak current decreases.The GCE peak is 16.76 times higher than for the GCE/Naf/MWCNTs/SBA-15 due to electrostatic repulsion between the negative surface of Naf (link to its hydrophilic part SO 3 -) and the anionic probe [Fe(CN) 6 ] 3-.The CV signals were significantly enhanced using [Ru(NH 3 ) 6 ] 3þ , indicating the good affinity, porosity, and electrostatic attraction of the negative charge materials with the positive probe.After modifying the bare GCE with Nafion, the peak current increased due to the electrostatic attraction between the negative charge surface and the positive redox probe.When the bare GCE is modified with SBA-15, the peak current also increased due to the porosity and large specific area.The MWCNTs inside Nafion increased the sensitivity of the electrode because of the high conductivity of the MWCNTs with the high porosity.Therefore, the combination of the three materials at the surface of GCE demonstrates high synergy.The anodic peak current for the GCE/Naf/MWCNTs/SBA-15 is 29.54 times higher than for the GCE.
The electrochemical impedance spectroscopy measurements in 0.10 M KCl and 10 À3 M [Ru(NH 3 ) 6 ] 3þ are presented in Figure 3c.The Nyquist plots obtained for the electrodes have been used to determine the value of the charge-transfer resistance (Rct) by extrapolation of each curve.The lowest value of Rct is obtained with GCE/Naf/MWCNTs/SBA-15 showing the rapid transfer of electrons on that electrode, which is in accordance with the CV results in Figure 3b.
Figure 4b shows the electrochemical response of CBZ on GCE/Naf/MWCNTs/SBA-15 by MCV in 10 À3 M PB from 0.30 V to 1.40 V.A peak appeared at 1.10 V, indicating the oxidation of CBZ which is irreversible at the surface of GCE/Naf/MWCNTs/SBA-15.This oxidation of CBZ is attributed to the nitrogen in the imidazole ring and nitrogen in the amide, where each nitrogen atom is oxidized with four electrons as shown in Scheme 2 (Chunhuan et al. 2019;Xiaoning et al. 2019;Temgoua et al. 2021).
For both analytes, a decrease in the peak current was observed with the number of cycles due to the higher availability of the active side (pores) on the composite which was saturated.These results demonstrate the suitability of sensor to quantify IC and CBZ during the first cycle.

Influence of the modification of the GCE and kinetic study of IC and CBZ
Figure 5 shows that on the GCE/Naf/MWCNTs/SBA-15, the peak current is significantly higher than on the bare GCE, GCE/Naf, GCE/Naf/SBA-15, and on GCE/SBA-15.Also, the oxidation potential at the GCE/Naf/MWCNTs/SBA-15 shifted negatively from 0.77 V to 0.73 V for IC (Figure 5a) and from 1.06 V to 1.02 V for CBZ (Figure 5b).This phenomenon confirms the catalytic effect of SBA-15, Nafion, and MWCNTs (Rafael et al. 2013).
The influence of potential scan rate (v) and peak potential (Ep) on GCE/Naf/MWCNTs/SBA-15 was investigated by CV in 10 À3 M IC in pH 1.50 H 2 SO 4 (Figure 6a) and 10 À3 M CBZ in pH 2.0 PB (Figure 6b).The correlation coefficient 0.99 (IC) and 0.99 (CBZ) obtained by plotting Ip ¼ f (v 1/2 ) from Figure 6c reflects good linearity with the square root of the scan rate between 10.0 and 70.0 mV/s.These measurements demonstrate that the transfer to the electrode surface is controlled by diffusion.This result was confirmed by plotting Log Ip vs. f (Log v) as shown in Figure 6d where the value of the slope is 0.42 (IC) and 0.44 (CBZ), closed to the theoretical value of 0.50 indicating a diffusion-controlled process (Barbara and Dryhurst 1971;Tiago et al. 2016;Yu et al. 2019;Deffo et al. 2022).

Optimization of the sensor by DPV
The stability of the GCE/Naf/MWCNTs/SB-15 was characterized as shown in Supporting Information Figure S1.Five successive measurements showed a relative standard deviation of 4.82% for IC and 2.21% for CBZ, indicating that the proposed sensor has good reproducibility and stability.
The influence of the proportion of the film deposited at the surface of the electrode is represented in Supporting Information Figure S2.The peak current increased with the quantity of film up to 8 mL (40 mg) where it reaches a stable value due to saturation of the electrode.This enhanced peak current is caused by an increase in the number of active sites on the surface of the electrode and hence a larger analyte concentration on the electrode surface (Teadoum et al. 2016;Njanja et al. 2019;Deffo et al. 2021).Thus, a volume of 8.0 lL of film at the surface of GCE was employed in further investigations.
The IC and CBZ have different electrochemical properties as shown in the literature (Yujing et al. 2011;Tiago et al. 2016;Jamballi 2018;Chunhuan et al. 2019).Therefore, the optimization of this parameter was done in three buffer solutions for each analyte as shown in Supporting Information Figure S3.The largest peak currents, the lowest background currents, and the best shape of peaks were obtained with sulfuric acid for IC and in phosphate buffer for CBZ.
We have, therefore, varied the pH (1.0 to 4.0 for IC and 1.0 to 8.0 for CBZ) of these electrolytes to optimize the conditions and the results are shown in Figure 7a for IC and Figure 7c for CBZ.The anodic peak current of IC reaches a maximum from pH 1.0 to 1.50 and decreases with increasing pH (Figure 7b-1).The same results are obtained for CBZ where the maximum peak current is at pH 2.0 and decreases with pH (Figure 7d-1).Therefore, pH 1.50 was deemed to be optimum for IC and pH 2.0 for CBZ.The peak potentials of IC and CBZ at the surface of the modified electrode shifted to less positive values with increasing pH. Figure 7b-2,d-2 show a linear shift of the oxidation peak potential (Ep) toward low potential with pH that demonstrates that protons are directly involved in the oxidation of IC and CBZ.These responses are described by A slope of 51 mV and 69 mV per pH were obtained which show they follow the Nernst equation with equal numbers of electrons and protons (Jamballi 2018;Chunhuan et al. 2019;Deffo et al. 2022).

Quantitative measurements for individual and simultaneous detection
DPV was used for the determination of trace IC and CBZ using the optimum conditions.Figure 8a shows IC and Figure 8c CBZ voltammograms at the GCE/Naf/MWCNTs/SBA-15 in H 2 SO 4 (pH 1.50) and PB (pH 2.0), respectively.The analytical curves (Figure 8b inset for IC and Figure 8d inset for CBZ) were linear from 0.0 to 11.0 lM with high correlation coefficients.Limits of detection (LOD) of 0.50 mM for IC and 0.57 mM for CBZ were obtained using 3S b /m where S b is the standard deviation of the blank and m the slope of the calibration relationship (Deffo et al. 2021).
The performance of the GCE/Naf/MWCNTs/SBA-15 electrode was further investigated by simultaneously determining IC and CBZ from 1.0 to 11.0 mM.Two well-separated oxidation peaks were obtained at 0.68 V for IC and 1.15 V for CBZ as shown in Figure 8e.Upon increasing the concentration, the separation increased, and the peak currents were linearly related to the concentrations with good correlation coefficients (0.998 for IC and 0.998 for CBZ).Therefore, the linear relationships were obtained for IC Ip (lA) ¼ 0.76 þ 12.18 C IC (lM) for IC and for CBZ Ip (lA) ¼ 0.14 þ 17.84 C IC (mM) (Figure 8f).The limits of detection were 0.28 mM and 0.30 mM for IC and CBM.These results confirm that the GCE/Naf/MWCNTs/SBA-15 electrode allows the simultaneous determination of IC and CBZ, although the presence of one analyte slightly increases the response of the other.The analytical performance is comparable to the literature for both analytes as shown in Table 1.

Interferences study and water analysis
Several possible interfering inorganic ions, pesticides, dyes, and food additives were used to evaluate the selectivity of the sensor for the determination of IC and CBZ.5.0 Â 10 À6 M of interfering species and 5.0 Â 10 À6 M IC and CBZ were added.The results in Supporting Information Table S1 show no obvious interferences for the determination of IC and CBZ.The changes of peak current were below 5% which is evidence of good selectivity.
The developed sensor was also used for the simultaneous determination of IC and CBZ in environmental water samples.Tap, spring, and lake water were spiked with 1 mM IC and CBZ.Before spiking, no peak current was present for IC (0.72 V) and CBZ (1.10 V).After spiking with 1 mM IC and CBZ, electrochemical peaks were observed.The results in Table 2 show that the relative standard deviation was less than 1% and demonstrate a promising method for the simultaneous and sensitive determination of IC and CBZ in the environmental water.

Conclusion
The individual and simultaneous determination of indigo-carmine (IC) and carbendazim (CBZ) were demonstrated for the first time, showing the suitability of  GCE/Naf/MWCNTs/SBA-15 electrode.The results show that the oxidation peak currents of IC and CBZ on GCE/Naf/MWCNTs/SBA-15 are 3-fold and 6-fold higher compared to the bare GCE.The anodic peak currents of IC and CBZ were proportional to concentration from 1.0 to 11.0 mM with limits of detection of 0.50 mM and 0.57 mM for individual measurement.The values for simultaneous determination were 0.28 mM and 0.30 mM.The preparation of Naf/MWCNTs/SBA-15 modified GCE is simple, fast, and demonstrated to be selective and sensitive for the determination of IC and CBZ in water samples.Therefore the development of point-of-care testing for these analytes may be envisaged.

Figure 7 .
Figure 7. (a) Differential pulse voltammograms of indigo carmine (5.0 Â 10 À5 M) on the GCE/Naf/MWCNTs/SBA-15 in H 2 SO 4 from pH 1.0 to 4.0.(b) (1) Variation of the indigo carmine peak current with pH.(2) Variation of indigo carmine peak potential with pH.(c) DPV of carbendazim (5.0 Â 10 À5 M) on the GCE/Naf/MWCNTs/SBA-15 in phosphate buffer from pH 1 to 8. (d) (1) Variation of the carbendazim peak current with pH.(2) Variation of the carbendazim peak potential with pH.Potential scan rate: 50.0 mV s À1 .5.0 g/L of suspension, and 8 mL of film.Experiments were performed in triplicate.Ep is the peak potential.

Table 2 .
Recovery of carmine and carbendazim in spiked water samples.Water samples were collected from Dschang, West Cameroon.
a b Concentration determined using the calibration curve.