Impedance and voltammetry detection of bromate in food samples using NiPcMWCNTs modified glassy carbon electrode

Abstract

A sensitive bromate sensor was developed using nickel phthalocyanine multi-walled carbon nanotubes nanocomposite modified on a glassy carbon electrode. The NiPcMWCNTs nanocomposite was prepared from nickel nanoparticles, phthalocyanine, and functionalized MWCNTs via ultrasonication. UV–visible spectroscopy, SEM, XRD, TEM, and EDX techniques were used to verify the successful fabrication of the nanomaterials. The results of the EIS and CV experiments conducted in 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS (pH 7) revealed that the NiPcMWCNTs/GCE exhibited higher current response, faster electron transfer, and high specific capacitance compared to other electrodes. The electrochemical reduction of bromate was actualized in 0.1 M H₂SO₄ (pH 1) using EIS and SWV techniques. Using the EIS technique, an LoD of 6.72 μM was obtained with a sensitivity of 483.7 μA μM⁻¹ over a linear dynamic range (LDR) of 24–100 μM. Whereas, with the SWV technique, a lower LoD (1.47 μM) was obtained with a higher sensitivity (1293 μA μM⁻¹) over an LDR of 12–56 μM. The developed sensor was characterized by good selectivity, high stability (95.5%), and good reproducibility (% RSD; 3.5%). The fabricated sensor was effectively used to detect bromate in bread samples with a good recovery rate, demonstrating the practical application of the sensor to detect bromate in real samples (bread).

Introduction

Despite bromate's usefulness in water treatment and food production, there are several instances of its negative impact on human health. Peripheral neuropathy, renal problems, and anemia have all been specifically linked to bromate intake (Crofton 2006; Fawell and Walker 2006) if ingested beyond the World Health Organization's authorized threshold of 25 µg L⁻¹ (WHO 1996). It has also been linked to malignant growth in laboratory animals. Additionally, increased incidences of kidney tumors, thyroid cells, and peritoneal mesotheliomas have been associated with bromate in mice’s and rats' drinking water. Human and animal auditory functions are also affected due to high bromate intake, according to scientific evidence (Balamurugan and Chen 2007). The World Health Organization, together with the US Environmental Protection Agency (USEPA), has set 0.078 µM (10 µg L⁻¹) as the maximum acceptable level (MAL) owing to its carcinogenicity (Organization and WHO 2004). Bromate in food and water has been linked to cancer risks for lifetime exposures, and the toxicological investigations by WHO (1996) have categorized bromate as a group B2 carcinogen in foods and water (Shanmugavel et al. 2020). Also, the present techniques for its detection have significant limitations that necessitate the development of new and more efficient techniques for detecting BrO₃⁻ at trace levels, entailing fast, economic, reliable, highly selective, and sensitive analytical techniques.

Ion chromatography, gas chromatography, and spectrophotometry liquid chromatography (Kim and Shin 2012; Rahali et al. 2011; Snyder et al. 2005; Zakaria et al. 2011) are previously employed techniques to detect bromate. However, tedious and time-demanding, special sample preparation, multiple extractions and hydrolysis, low sensitivity, high temperature requirements, costly and sophisticated instrumentation limit the usage of these techniques for bromate detection (Menendez-Miranda et al. 2013). Most of these setbacks can be addressed to a greater extent by employing electrochemical methods, which are recently being utilized because of their quick analysis period, no sample pre-treatment, accessibility, and environmental friendliness. Besides these, they offer high selectivity, high sensitivity, and a lower detection limit (Balogun and Fayemi 2022).

In recent times, metal nanoparticles have been extensively utilized in magnetic storage, sensors, ferrofluids, and imaging applications owing to their electronic, thermal, magnetic, catalytic, and optical properties compared to bulk metals (Goyal 2017). Remarkably, the magnetic nanoparticles of nickel, iron, and cobalt have received greater attention in the past decade as a result of their remarkable magnetic properties, such as high Curie temperatures, colossal magnetoresistance, low saturation magnetization, and a very large coercivity as well as their potential uses in numerous fields, such as sensors, catalysis, and memory storage devices (Balogun and Fayemi 2022; Eluri and Paul 2012). Nickel nanoparticles (NiNPs) stand out among these nanoparticles due to their potential applications in various fields and extraordinary properties (Ali et al. 2019). Nickel nanoparticles have demonstrated good electrocatalytic activity towards dopamine, ascorbic acid, uric acid (He et al. 2016), nitrite, glucose (Arikan et al. 2022; Kumar et al. 2020), ethanol, methanol, oxygen (Luo et al. 2022; Roodbari et al. 2022), catechol, and hydroquinone (Hassine et al. 2021).

Carbon nanotubes (CNTs) belong to the family of fullerene, which is composed of sp² hybridized carbon and forms a tube-like structure of graphene units. The use of CNTs as a sensor fabricating material has been strongly influenced by their electronic conductivity, specifically the rolled-up concentric graphene units of the multi-walled carbon nanotubes (MWCNTs). Scientists have come to appreciate MWCNTs as electrode modifiers in electrochemical sensor fabrication owing to their electronic conductivity, huge surface area, and chemical inertness (Oliveira and Morais 2018). Furthermore, it has been found that MWCNTs contribute to the enhancement of the stability and dispersion of metal oxide nanoparticles in a composite, including both metal oxide nanoparticles and MWCNTs (Puangjan et al. 2017). The improved chemical properties and the intrinsic electronic conductivity of MWCNTs favour their usage as a modifier for BrO₃⁻ sensor (Roodbari et al. 2022; Vilian et al. 2016; Zhou et al. 2012). This might be attributed to electrostatic interaction as well as potential π–π interaction between BrO₃⁻ and MWCNTs. Multi-walled carbon nanotubes are now used as electrochemical sensing materials to facilitate electron transfer reactions. This is because they have a large surface area, are strong mechanically, have great electrical conductivity, and are chemically stable (Balogun and Fayemi 2022).

A phthalocyanine (Pc) is a macrocyclic compound with a central planar molecule and a delocalized 18-π electron system on carbon–nitrogen double bonds (Balogun and Fayemi 2022). Phthalocyanine is one of the important chemicals with numerous applications, including electrochemical sensors, photovoltaic cells, electrocatalysis, liquid crystal materials, photodynamic therapy, catalysts, and photosensitizers, because of its good spectral performance and high stability (Das et al. 2012; Kaya et al. 2014; Wang et al. 2021). The presence of hydrogen bonds, coordination interactions, and van der Waal forces allows Pc to bind a variety of analytes in an indifferent way (Reddy 2020). In addition, the ability to accommodate about 70 different metal atoms within the rings and to vary the side chain substituents results in exceptional and efficient nanomaterials with different selectivity, stability, and sensitivity (Balogun and Fayemi 2022). Recently, the metal phthalocyanines (MPc) have garnered much attention due to their excellent catalytic properties. In metal phthalocyanine, the central metal ions play a significant role in its high catalytic activity (Qiu et al. 2022; Yadav et al. 2016). Besides, the exceptional electrocatalytic activity of nickel phthalocyanines (NiPc) for a variety of compounds, its quick electron transfer capability, and its rich redox chemistry have made it more widely used in sensor development (Shao et al. 2022). Additionally, the core metal of the NiPc can easily alter its coordination environment, thus significantly boosting its selectivity and catalytic activity. The ease with which organic groups can be substituted in the complex's axial and equatorial regions, resulting in a variety of functions for anchoring complexes in solid substrates, is one of NiPc's outstanding properties that makes it a superior material in fabricating electrochemical sensors (Han et al. 2017; Jiang et al. 2019; Li et al. 2022; Mei et al. 2022).

Phthalocyanines have been extensively researched for functionalizing CNTs since they exhibit diverse electrical and photoelectronic properties that increase the efficiency of CNTs-based devices. Recent studies revealed that MPc-CNTs composites show improved current responses than individual MPc or CNTs (Mphuthi et al. 2017). Studies have shown that impregnating MPc composites with other highly conductive materials on working electrodes reduces redox overpotentials and improves electrode faradaic currents (Kantize et al. 2019; Winiarski et al. 2020). Notably, the conductivity of the metal phthalocyanine-based films is enhanced with the addition of metallic nanoparticles as well as carbon-based materials (MWCNTs, quantum dots, graphene, and SWCNTs) (Guaraldo et al. 2019). An electron transport improvement is achieved between the electrode’s surface and the analytes adsorbed on the thin film by adding these conductivity-enhancing materials (Kantize et al. 2019).

Nickel phthalocyanine multi-walled carbon nanotubes (NiPcMWCNTs) nanocomposites are now used to make electrochemical sensors that can detect a wide range of analytes with high selectivity, a low limit of detection, a large current response, high sensitivity, great reproducibility and repeatability, a large capacitance, and excellent stability over a wide concentration range. Thus, this is the primary motive behind the widespread usage of these nanocomposites in constructing sensors with other conducting nanomaterials. The study, therefore, explored the advantages of the phthalocyanine and MWCNTs sizable surface area, as well as the MWCNTs and nickel nanoparticles' high electrical conductivity to fabricate a fast, low-cost, sensitive, and selective electrochemical sensor for bromate detection. As far as we know, no research has ever been done on using the glassy carbon electrode (GCE) modified with NiPcMWCNTs for bromate detection.

Materials, reagents, and instruments

Potassium bromate (KBrO₃), multi-walled carbon nanotubes (98%), ethylene glycol, 29H, 31H⁻ phthalocyanine (Pc), potassium hexacyanoferrate IV (K₄(Fe(CN)₆), N,N-dimethyl formamide (DMF), potassium hexacyanoferrate III (K₃(Fe(CN)₆), sodium dihydrogen phosphate (NaH₂PO₄), and disodium hydrogen phosphate (Na₂HPO₄) were supplied by Sigma-Aldrich (Darmstadt, Germany). Sulfuric acid, hydrochloric acid (HCl), nickel chloride, hydrazine hydrate, sodium hydroxide (NaOH), and methanol were products of Glassworld Chemicals (Johannesburg, South Africa). All reagents were of analytical quality.

The nanoparticles and nanocomposites were characterized morphologically and structurally using various techniques. Transmission electron microscopy (TEM) was performed using a JEOL2100 instrument equipped with a LaB 6 electron gun from JEOL Ltd (Tokyo, Japan). X-ray diffraction spectrophotometry (XRD) was conducted using equipment from Bruker Company (Karlsruhe, Germany). Scanning electron microscopy (SEM) was carried out using a JEOL JSM-6610 LV instrument from Dearborn, Peabody, MA, USA. UV–visible spectrophotometry was performed using an Agilent Technology, Cary 300 series UV–Vis spectrometer from Darmstadt, Germany. Energy diffraction X-ray (EDX) was conducted using a JEOL JSM-6610 LV instrument from Dearborn, Peabody, MA, USA. Electrochemical and impedance spectroscopy (EIS) analyses were performed utilizing an Autolab Potentiostat PGSTAT 302 (Eco Chemie, Utrecht, The Netherlands), managed by GPES (version 4.9) software (Utrecht, The Netherlands). Simultaneously, the electrochemical impedance spectroscopy (EIS) measurements were acquired utilizing the Metrohm Autolab (FRA 32) NOVA 2.1.3 software (Utrecht, The Netherlands) with a frequency span of 100 kHz–0.1 Hz. The experiments were conducted using distilled water.

Nickel nanoparticles synthesis

Nickel nanoparticles (NiNPs) synthesis was achieved by adding 8 mL of 1 M NaOH to a solution of 800 mL of ethylene glycol containing a mixture of 2 g of nickel chloride and 10 mL of hydrazine hydrate. The reaction mixture was continuously stirred at 70 °C for 45 min in a closed bottle. The resultant black particles were washed thoroughly with absolute ethyl alcohol and kept under vacuum at 27 °C for 24 h (Wu et al. 2012).

Functionalization of raw MWCNTs

The raw MWCNTs underwent a treatment process to introduce –COOH functional groups onto their surface. The functionalization process entails introducing 100 mg of pristine multi-walled carbon nanotubes (MWCNTs) into a solution comprising 150 mL of concentrated sulfuric acid (H₂SO₄) and 50 mL of concentrated nitric acid (HNO₃). The concoction underwent thermal treatment for 5 h at 65 °C and an additional 8 h at 80 °C. The resulting mixture was subjected to multiple washes with distilled water until its pH was neutralized. The obtained black particles were kept in an oven at a temperature of 70 °C for 12 h (Lee et al. 2017).

NiPcMWCNTs nanocomposites preparation

The preparation of the nanocomposites entails solubilizing 3 mg of phthalocyanine Pc and 3 mg of NiNPs in 1 mL of DMF followed by sonication for 2 h. After adding 3 mg of fMWCNTs, the resultant mixture underwent sonication for an additional 2 h. By means of ultrasonication-assisted spontaneous adsorption, the fMWCNTs' surface was coated with nickel phthalocyanine nanoparticles (NiPc) (Balogun and Fayemi 2022).

Modification of GCE with nanocatalysts

Before the bare GCE was modified with the synthesized nanoparticles and the nanocomposite, preliminary electrode treatment was carried out. The bare GCE was initially cleaned on a Sic Emery paper containing aluminum oxide slurry. The bare GCE in this state was immediately washed with a copious amount of water to remove traces of aluminum oxide particles. To ensure complete cleaning, the bare GCE was subjected to ultrasonication inside a beaker containing distilled water for about 5 min, and afterward in methanol for another 5 min.

Using the drop-dry approach, the cleaned, unmodified GCE was treated with each of the nanocatalyst pastes. About 5 µL of the prepared paste of NiNPs, fMWCNTs, Pc, and NiPcMWCNTs were dropped on the cleaned GCE and dried in an oven for about 5 min at 50 °C. The resultant modified electrodes were labelled NiPcMWCNTs/GCE, fMWCNTs/GCE, Pc/GCE, and Ni/GCE.

Results and discussion

Energy dispersive X-ray (EDX)

The EDX spectra of nickel nanoparticles and NiPcMWCNTs nanocomposites, as shown in Fig. 1, reflect their elemental composition. The presence of oxygen and carbon in Fig. 1a (NiNPs spectrum) indicates intruded impurities from the precursors. Similarly, the chlorine and sulphur in Fig. 1b were from the reagents (concentrated HCl and H₂SO₄) used for functionalizing the rMWCNTs. The existence of nitrogen and carbon (signifying phthalocyanine), carbon and oxygen (signifying fMWCNTs), and nickel in Fig. 1b suggests that the NiPcMWCNTs nanocomposite was successfully fabricated.

Fig. 1

EDX spectra of a NiNPs, b NiPcMWCNTs

X-ray diffraction (XRD)

The crystalline structures of the synthesized NiNPs, raw MWCNTs, fMWCNTs, and NiPcMWCNTs nanocomposite were determined using XRD. Figure 2a depicts the XRD diffractogram of NiNPs with three major diffraction peaks at 2θ = 44.5° (1 1 1), 51.9° (2 0 0), 76.5° (2 2 0) corresponding to face-centered cubic (FCC) of pure nickel nanoparticles. These diffraction peaks appear similar to the ones reported for NiNPs in literature (Eluri and Paul 2012; Wu et al. 2012; Wu et al. 2010), affirming the successful synthesis of FCC pure nickel nanoparticles. Using Scherrer’s equation (Klug and Alexander 1974), 12 nm was obtained as the average crystallite size (D) for the NiNPs, as shown in Eq. 1.

$$D = \frac{{0.89\uplambda }}{{{\text{B cos }}\uptheta }}$$

(1)

where B denotes full-width half-maximum, λ stands for wavelength (0.15418 nm), and θ represents Bragg angle of XRD peak.

Fig. 2

XRD patterns of a NiNPs, b fMWCNTs and rMWCNTs, and c NiPcMWCNTs nanocomposite

The raw MWCNTs in Fig. 2b revealed low-intensity diffraction peak at angles 2θ = 26.35°. In comparison, the functionalized MWCNTs exhibited a very intense diffraction peak at the angles 2θ = 26.3° and 25.97° (002) and 42.31° indexed to the (002) and (101) lattice plane reflections of graphite-like materials characteristic of CNTs. The peak intensity indicated highly improved crystallinity, confirming the attachment of the oxygen-containing covalent functional groups on the MWCNTs walls (Cao et al. 2003; Li et al. 2011). The XRD diffractogram of NiPcMWCNTs nanocomposite, as depicted in Fig. 2c, shows eight peaks at 2θ = 6.92°, 10.97°, 15.08°, 17.06°, 26.26°, 44.64°, 52.02° and 76.60°. The eclectic and strong peak at 26.26° (002) is assigned to the fMWCNTs, while 44.64°, 52.02°, and 76.60° are attributed to NiNPs. The remaining peaks (6.92°, 10.97°, 15.08°, and 17.07°) are distinctive peaks of phthalocyanine (Cao et al. 2003; Li et al. 2011).

Scanning electron microscopy (SEM)

The SEM images of the nanomaterials and the nanocomposite are depicted in Fig. 3a–d. A smooth spherical particle shape but irregular particle morphology with a grain size of 18 nm was observed for NiNPs’s TEM image in Fig. 3a. The value is more in line with the result obtained by applying Scherrer’s equation. The micrograph (Fig. 3d) displays NiNPs and Pc attaching to the fMWCNTs surface, thereby suggesting a successful NiPcMWCNTs nanocomposite fabrication.

Fig. 3

SEM images of a NiNPs, b Pc, c fMWCNTs, and d NiPcMWCNTs nanocomposite

Ultraviolet–visible characterization

The UV–visible spectra of (a) nickel nanoparticles (b) Phthalocyanine (c) raw MWCNTs & functionalized MWCNTs, and (d) NiPcMWCNTs nanocomposite are illustrated in Fig. 4. Figure 4a represents the NiNPs spectrum with a prominent peak at 296 nm. The UV–visible spectrum of rMWCNTs and fMWCNTs, as depicted in Fig. 4b shows that the absorption peaks at 289 nm (rMWCNTs) and 294 nm (fMWCNTs) differ noticeably, which confirms that the rMWCNTs were successfully functionalized.

Fig. 4

UV–visible spectra for a NiNPs, b fMWCNTs and rMWCNTs, c Pc, and d NiNPs, fMWCNTs, Pc, and NiPcMWCNTs nanocomposites

Two predominant peaks are peculiar with Phthalocyanine: the Q-band and the B-band. The Q-band takes place at 668 nm as a result of the electronic transitions from the lowest unoccupied molecular orbital to the highest occupied molecular orbital (π → π* transition), while the B-band takes place in the ultraviolet region between 310–400 nm due to the transition from π* to π. At a wavelength of 668 nm, there was Q-band splitting in the phthalocyanine spectrum due to the compound's D_2h symmetry. A similar Q-band splitting is observed in the NiPcMWCNTs nanocomposite spectrum (Fig. 4d), which includes all the identified peaks for Pc, NiNPs, and fMWCNTs.

Electrochemical characterization

To investigate the electron transport behaviors of nanomaterials and the nanocomposite, the electrochemical characterization of all the electrodes was carried out using CV. All the electrodes produced a pair of redox peaks in 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS (pH 7) at 25 mVs⁻¹, as displayed in the cyclic voltammograms (Fig. 5). According to Table 1, the values of the peak separation (Ep) obtained for the electrodes were GCE (0.39 V), Pc/GCE (0.25 V), Ni/GCE (0.18 V), fMWCNTs/GCE (0.13 V), and NiPcMWCNTs/GCE (0.18 V). The Ep values obtained are greater than 59 mV predicted for rapid one-electron transport, indicating a quasi-reversible process at the electrode surfaces (Haque et al. 2013; Zangeneh Kamali et al. 2014). Table 1 further demonstrates that all electrodes' Ipa/Ipc values are about equal to one. This implies reversibility on all modified electrode surfaces, including the bare electrode. Figure 5 shows the relative current responsiveness of each electrode, with NiPcMWCNTs/GCE (457.0 μA) coming in first place, followed by fMWCNTs/GCE (58.0 μA), Ni/GCE (35.6 μA), Pc/GCE (21.5 μA), and GCE (25.0 μA). An anodic peak current (Ipa) of 457.0 μA, roughly 18 times more than GCE’s peak current and about 8 times more than other modified electrodes, was recorded for the NiPcMWCNTs/GCE modified electrode. This suggests that the NiPcMWCNTs/GCE electrode offers the greatest current response of all the electrodes. The combined cooperative effect of fMWCNTs, NiNPs, and Pc is responsible for the improved current response displayed at the NiPcMWCNTs/GCE electrode. This effect is brought on by the phthalocyanine and fMWCNTs sizable surface area, as well as the fMWCNTs and nickel nanoparticles high electrical conductivity (Mphuthi et al. 2017).

Fig. 5

a Cyclic voltammograms of the electrodes in 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS at 25 mV s⁻¹, and b Histogram of the current responses for GCE, Pc/GCE, Ni/GCE, fMWCNTs/GCE, and NiPcMWCNTs/GCE electrodes

Table 1 Cyclic voltammetry data of all the electrodes in 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS (pH 7)

Equally, the electrodes' specific capacitance (Sc) was calculated from the individual voltammograms of the modified electrodes using Eq. 2. Noteworthy, a larger capacitive current was exhibited by the NiPcMWCNTs/GCE in comparison to the other electrodes, indicating its superiority in charge and discharge capacity over them.

$${\text{Sc}} = \frac{{\text{A}}}{{2{\text{mk}}\left( {{\text{V}}_{2} { }{-}{\text{V}}_{1} } \right)}}$$

(2)

where A denotes the area of CV curve (AV), m denotes the mass of active material (g), k stands for scan rate (mV s⁻¹⁾, and (V₂ − V₁) symbolizes the potential window (V s⁻¹).

In reference to Table 1, the Sc values of the electrodes were 0.17 F g⁻¹ (GCE), 0.26 F g⁻¹ (Pc/GCE), 0.31 F g⁻¹ (Ni/GCE), 0.63 F g⁻¹ (fMWCNTs/GCE), and 6.80 F g⁻¹ (NiPcMWCNTs/GCE). This data makes it evident that NiPcMWCNTs/GCE offered a substantially greater Sc compared to the other electrodes. The enhanced capacitive performance demonstrated by the NiPcMWCNTs/GCE has also been accredited to the Phthalocyanine and fMWCNTs sizable surface area, as well as the fMWCNTs and nickel nanoparticles high electrical conductivity. These electrodes' capacitive properties were further investigated using the EIS method.

Scan rate study

The impact of scan rate variation (25–300 mV s⁻¹) on the electrochemical processes occurring on the surface of the NiPcMWCNTs/GCE electrode was studied in a solution of 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS. An increase in the scan rate is precisely proportional to the peak currents, as seen in Fig. 6a. The plot of the scan rate's square root vs peak current in Fig. 6b, shows a linearity, resulting in the regression equation below.

$$\begin{gathered} I_{pa} = \, 0.0054 \, v^{1/2} - \, 0.0004\quad \quad (R^{2} = \, 0.9987) \hfill \\ I_{pc} = \, 0.0048 \, v^{1/2} + \, 0.0003\quad \quad (R^{2} = \, 0.9990) \hfill \\ \end{gathered}$$

The linear connection between the scan rate's square root and the peak currents suggests a common diffusion-controlled mechanism for the NiPcMWCNTs/GCE.

Fig. 6

Cyclic voltammogram of a NiPcMWCNTs/GCE from 25 to 300 mV s⁻¹ in 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS, and b linear plot of scan rate’s square root vs peak current

Electrochemical impedance spectroscopy (EIS) study

To fully examine the capacitive properties and the electron transport behaviours of these electrodes, an EIS study was performed in 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS (pH 7) via a Metrohm Autolab Potentiostat. Having 0.3 V as its constant potential against Ag/AgCl, saturated with 3 M KCl, with an operating frequency of 100 kHz–0.1 Hz. The experiment was conducted at 0.3 V since it had the highest peak current and capacitance at this potential. Figure 7 displays the Nyquist plots for the fitted data of the electrodes in 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS, While X denotes the circuit employed to fit the obtained data. The circuit's components contain the following parameters: Capacitance (C), constant phase element (CPE), solution resistance (R_s), charge transfer resistance (R_ct), and warburg impedance (W). Table 2 provides the obtained EIS fitted data along with individual chi-square (x²) values. The EIS data were well fitted, as evidenced by the small % errors in the parentheses and the negative x² values obtained.

Fig. 7

Nyquist plots of the electrodes in a 5 mM K₃(Fe(CN)₆/K₄(Fe(CN)₆ made in 0.1 M of PBS and b Circuit employed to fit the obtained EIS data

Table 2 Overview of the fitted EIS parameters of the electrodes

The R_ct values recorded for the electrodes as presented in Table 2, were NiPcMWCNTs/GCE (0.18), fMWCNTs/GCE (0.44), Pc/GCE (0.89), Ni/GCE (1.39), and GCE (8.73 KΩ) with the NiPcMWCNTs/GCE having the lowest R_ct, indicating its rapid electrons transport capability over other electrodes. The NiPcMWCNTs/GCE fast electron transfer ability is due to the functionalized MWCNTs' high conductivity, which serves as an excellent electron conductor between Pc and NiNPs and the NiPcMWCNTs/GCE’s surface. The reduction in R_ct values of the modified electrodes, suggests that the bare electrode was properly modified with the nanocatalysts. The result obtained here is similar to the current response result in the CV. This also led to the huge Sc values obtained by the modified electrodes, particularly the NiPcMWCNTs/GCE. The n values of all the electrodes, which range from 0.72 to 0.97, may be responsible for the easy diffusion of ions across the solution/electrode interface.

Choice of electrolyte

Prior to the detection of bromate, an impedance spectroscopy experiment was carried out in four dissimilar supporting electrolytes with NiPcMWCNTs/GCE. This study was done to ascertain the most suitable supporting electrolyte for bromate detection. The four electrolytes investigated were 0.1 M sodium acetate buffer (SAB) containing 0.1 M KBrO₃ (pH 7), 0.1 M Na₂SO₄ containing 0.1 M KBrO₃ (pH 2), 0.1 M H₂SO₄ containing 0.1 M KBrO₃ (pH 1), and 0.1 M PBS containing 0.1 M KBrO₃ (pH 7). Figure 8 depicts the Nyquist plot of NiPcMWCNTs/GCE in H₂SO₄, SAB, PBS, and Na₂SO₄ supporting electrolytes. The R_ct values obtained for the NiPcMWCNTs/GCE in the four supporting electrolytes were 0.06, 0.36, 1.98, and 9.04 kΩ for the H₂SO₄, SAB, PBS, and Na₂SO₄, respectively (as shown in Table 3). The lowest R_ct value obtained for NiPcMWCNTs/GCE was in H₂SO₄ electrolyte, while the highest was in Na₂SO₄. This indicates that the H₂SO₄ electrolyte is a better medium for the rapid transfer of electrons than the other electrolytes investigated.

Fig. 8

Nyquist plots of a NiPcMWCNTs/GCE in H₂SO₄, PBS, Na₂SO₄, and SAB electrolyte, and b circuit employed to fit the obtained EIS data

Table 3 EIS results of NiPcMWCNTs/GCE in four dissimilar electrolytes

The results of the EIS of the NiPcMWCNTs/GCE in the four dissimilar electrolytes show that the nanocomposite electrode is a suitable bromate sensor, particularly in the H₂SO₄ electrolyte. A lot of studies on the usage of H₂SO₄ for bromate detection have been reported in the literature (Ding et al. 2014; Hassan et al. 2013; Majidi et al. 2015; Zhang et al. 2020). Hence, the fast electron transport capability of the nanocomposite electrode in the H₂SO₄ electrolyte accounted for its choice as the suitable electrolyte for bromate detection.

Electroanalysis of bromate at NiPcMWCNTs/GCE

Utilizing cyclic voltammetry, the electrocatalytic property of bromate in 0.1 M H₂SO₄ (pH 1) containing 1 mM KBrO₃ at 25 mV/s was investigated. The cyclic voltammograms (CV) of the electrodes are depicted in Fig. 9a. Expectedly, the NiPcMWCNTs/GCE has the highest reduction peak current. This supports the finding that NiPcMWCNTs/GCE is superior to all other electrodes. In comparison with other electrodes, the NiPcMWCNTs/GCE displayed a better potential and a well-defined bromate cathodic (reduction) peak current. From Table 4, the bromate reduction potential (E_pc) and current (I_pc) recorded for the NiPcMWCNTs/GCE, fMWCNTs/GCE, Ni/GCE, Pc/GCE, and GCE are − 1299.10 μA (− 0.0430 V), − 306.59 μA (− 0.2208 V), − 11.78 μA (− 0.2978 V), − 36.70 μA (− 0.3124 V), and − 23.19 μA (− 0.2332 V), respectively. In addition to the reduction (cathodic) peak demonstrated by all of the electrodes, the modified NiPcMWCNTs/GCE also displayed an anodic peak, as seen in Fig. 9; however, this peak vanishes at higher scan rates. The NiPcMWCNTs-GCE outperformed other electrodes in terms of electrocatalytic response for bromate reduction.

Fig. 9

Cyclic voltammogram of the electrodes in a 1 mM KBrO₃ prepared in 0.1 M H₂SO₄ (pH 1) and b histogram of the electrodes’ current response

Table 4 Summary of CV results of the electrodes in 1 mM KBrO₃ prepared in 0.1 M H₂SO₄ (pH 1) at 25 mV/s

The good electronic conductivity of nanocomposite components (Pc, NiNPs, and fMWCNTs), their biocompatibility with the bromate, and their large surface area are responsible for this improved current response. Remarkably, a visible bromate reduction peak given by NiPcMWCNTs/GCE resulted from the electron transfer that happened quickly during the bromate reduction process.

Impact of scan rate variation

The impact of varied scan rate at NiPcMWCNTs/GCE in 0.1 M H₂SO₄ solution containing 1 mM bromate was examined with CV. An increment from 25 to 300 mV/s, gradually increased the cathodic peak current with a positive shift of the peak potentials, as shown in Fig. 10a. A linear connection exists between the scan rate’s square root and the cathodic peak current, giving a linear regression (Eq. 3). Thereby, suggesting that the NiPcMWCNTs/GCE reduction process is a diffusion-controlled process (Wierzbicka and Sulka 2016).

$$I_{pc} = \, - 1.283E - 04 \, v^{1/2} - \, 6.7592E - 4\quad \quad \left( {R^{2} \, = \, 0.9899} \right)$$

(3)

$$E_{pc} = \, - 0.27133 \, + \, 0.32003\; \, log \, v\quad \quad \left( {R^{2} \, = \, 0.99849} \right)$$

(4)

Fig. 10

Cyclic voltammogram of a NiPcMWCNTs/GCE (25–300 mV s⁻¹) in 1 mM KBrO₃ prepared in 0.1 M H₂SO₄ (pH 1), b linear graph of scan rate against cathodic peak current I_pc, c linear graph of scan rate’s square root against cathodic potential I_pc, and d linear graph of logarithm of scan rate against cathodic potential E_pc

A linear connection also exists between the logarithm of the scan rate (50–175 mV s⁻¹) and the peak potential, as depicted in Fig. 10d, with a linear equation represented in Eq. 4. Using the Laviron equation, given that the slope of the graph of log v against E_cp is b1 (Eq. 5), the electron transfer coefficient (α), as well as the number of electrons transferred (n), can then be calculated.

$$b1 = \frac{{{\text{RT}}}}{{ \propto {\text{nF}}}}$$

(5)

$$E_{p} = \frac{{\text{b}}}{2}\log v + K$$

(6)

where n, F, R, K, α, v, and T denote the number of electrons in cathodic reaction, Faraday constant (96,485 C mol⁻¹), universal gas constant (8.314 J K⁻¹ mol⁻¹), reaction rate constant, electron transfer coefficient, potential scan rate, and temperature Kelvin.

The Tafel slope b is calculated as 225 mV dec⁻¹ from Eq. 6. A Tafel slope obtain within the range of 60–120 mV dec⁻¹ suggests a one-electron process at the rate-determining step, while a Tafel slope greater than 120 mV dec⁻¹ suggests an adsorptive process at the electrode surface (Ndebele et al. 2021). The strong binding force between the electrode modifier and the analyte has been attributed as the cause of this kind of adsorption (Fang and Liu 2014). The Tafel slope for bromate reduction (225 mV dec⁻¹) is larger than 120 mV dec⁻¹ implying the occurrence of an adsorptive process at NiPcMWCNTs/GCE electrode (Ndebele et al. 2021). The proposed mechanism of bromate reduction at NiPcMWCNTs/GCE, as illustrated in Scheme 1, involves two phases. The first phase is the oxidation of Ni^IIPc to Ni^IIIPc, while the second phase is the reduction of bromate to bromide and Ni^IIPc regeneration.

Scheme 1

Mechanism of bromate reduction

Effects of pH

Using CV in 0.1 M H₂SO₄ solution at varied pH values (1, 2, 3, & 5), the effect of the varied electrolyte pH on the electrochemical performance of NiPcMWCNTs/GCE towards bromate reduction was investigated. The primary goal of this investigation was to attain the best reduction peak resolution for bromate detection. With an increasing pH of the electrolyte, a negative shift in the cathodic peak potentials was noticed, suggesting that the electrocatalytic reduction of bromate involves proton transfer at the modified nanocomposite electrode. The highest cathodic peak current of bromate was attained at the pH of 1 and subsequently reduced with increasing pH. For this reason, a pH of 1 was used throughout the experiment. Interestingly, the electroanalysis of bromate at a pH of 1 has been widely reported (Li et al. 2006, Zhou et al. 2012).

Concentration studies

EIS detection of bromate at NiPcMWCNTs/GCE

The use of NiPcMWCNTs/GCE for bromate detection was favoured among other electrodes due to its lower R_ct value and higher current response. EIS technique was adopted for BrO₃⁻ detection at the developed sensor. Figure 11a shows the Nyquist plot of NiPcMWCNTs/GCE in 24–100 µM BrO₃⁻. Expectedly, as the BrO3 concentration increases, the charge-transfer resistance R_ct reduces. The charge-transfer resistance and BrO₃⁻ concentration were found to be linearly related over a concentration range of 2.4 × 10^–5− 1.0 × 10^–4 μM for NiPcMWCNTs/GCE (Fig. 11b). The LoD and LoQ for BrO₃⁻ for NiPcMWCNTs/GCE were calculated using Eq. 7 and 8, respectively.

$$LoD = \frac{{3.3\updelta }}{{\text{m}}}$$

(7)

$$LoQ = \frac{{10\updelta }}{{\text{m}}}$$

(8)

where δ denotes the standard deviation of the intercept, m represents the slope of the intercept.

Fig. 11

Nyquist plot of a NiPcMWCNTs/GCE over BrO₃⁻ concentration range of 24–100 µM at pH 1. b plot of R_ct versus concentration

The LoD and LoQ obtained for the NiPcMWCNTs/GCE were 6.72 μM and 20.38 µM, respectively, with a sensitivity of 483.7 μM Ω⁻¹. The LoD obtained for the fabricated sensor is compared in Table 5 with other reported bromate sensors in the literature.

Table 5 Comparison of the proposed sensor with the previous bromate sensors

SWV detection of bromate at NiPcMWCNTs/GCE

At the NiPcMWCNTs/GCE electrode, the electrocatalytic reduction of BrO₃⁻ was further examined via the square wave voltammetry (SWV) technique. The voltammogram of NiPcMWCNTs/GCE in the bromate concentration range of 12–56 μM is depicted in Fig. 12a. Here, the peak current also increased with the BrO₃⁻ concentration. The current response and BrO₃⁻ concentration were found to be linearly related, as depicted in Fig. 12b. At the developed sensor, there is a linear relationship between the BrO₃⁻ concentration and the current over an LDR of 12 to 56 μM.

Fig. 12

Square wave voltammograms of a NiPcMWCNTs-GCE over BrO₃⁻ concentration range of 12–56 μM at pH 1, and b plot of current versus concentration obtained from (a)

The developed sensor's LoD and LoQ were estimated as 1.47 and 4.46 μM, respectively. Compared to the EIS results, the LoD and LoQ achieved using SWV are superior. Additionally, the sensor has an extremely high sensitivity (1293 μA μM⁻¹).

Interference studies

One of the qualities of a good sensor is its ability to selectively detect the target analyte amidst the likely interfering ions in a real sample. As a result, the performance of the developed sensor is examined in relation to several potential interfering species. The selectivity of the NiPcMWCNTs/GCE towards bromate detection was investigated through EIS technique. The interference study was accomplished by comparing the EIS response (R_ct value) of 0.1 mM bromate in 0.1 M H₂SO₄ solution before and after adding certain concentrations of the interfering ions to a known bromate concentration at a voltage of − 0.3. The Nyquist plot (not shown) demonstrates no change in the R_ct value of bromate concentration with the addition of the interfering ions concentration except for a few ions. The difference between the R_ct obtained in 0.1 mM bromate alone, and the R_ct of the mixture of interfering ion concentration plus 0.1 mM bromate serves as a basis for the sensor selectivity. As depicted in Table 6, the addition of tenfold excesses of these species (K⁺, Na⁺, SO₄²⁻, CO₃²⁻, ClO₃⁻, and Cl⁻) does not affect the R^ct value of the bromate, except for IO₃⁻ and Mg²⁺ that show a slight degree of interference. There was about 2.2% and 3.2% drop in R_ct value of the proposed sensor upon the addition of IO₃⁻ and Mg²⁺. The IO₃²⁻ reacts and undergoes oxidation in the H₂SO₄ solution, which is mainly responsible for its interference. The interference caused by Mg²⁺ is possibly due to the precipitates formed by the Mg²⁺ in the electrolyte solution.

Table 6 Impact of interfering ions on bromate detection at NiPcMWCNTs/GCE

Real sample analysis

Prior to the analysis of the real samples, a preliminary qualitative test was carried out on a small portion of the bread samples by adding a few drops of a mixture of 1 mL of 0.5 M hydrochloric acid and 5 mL of 0.05 M promethazine solution. When the bread sample solution changes to pink, it confirms that bromate is present. This preliminary qualitative test is important to determine the presence of bromate in each bread sample. Three different bread samples were used for this analysis.

The real sample analysis of bromate in the bread samples was investigated with the constructed sensor (NIPcMWCNTs/GCE) utilizing the standard addition procedure. About 5 g of dried bread samples were ground to powder using a mortar and pestle and dissolved in distilled water to form a homogeneous paste by continuously shaking the mixture at room temperature for 60 min. The resultant paste was sufficiently diluted and left for 24 h. The solution was filtered and centrifuged at 3000 rpm to obtain a clear supernatant. The supernatant was spiked with a known bromate concentration (20, 30, and 50 μM), and the bromate content of each spiked bread sample was analyzed using the square wave voltammetry technique. The preliminary test result shows that none of the three bread samples test positive.

Table 7 summarizes the mean percentage recovery of bromate from the bread samples after a triplicate determination. The mean percentage recovery obtained for the NIPcMWCNTs/GCE sensor is 103.3%, 100.9%, and 102.8%, with the percentage relative standard deviation (% RSD) of 1.58%, 3.60%, and 2.24% for sample X, Y, and Z, respectively.

Table 7 Bromate analysis in bread samples at NIPcMWCNTs/GCE electrode

The results in Table 7 show that NIPcMWCNTs/GCE exhibited a good percentage recovery. The bromate % recovery offered by the this sensor competes favorably with similar fabricated bromate sensors in various real samples (Kirowa‐Eisner et al. 1994; Zhang et al. 2022a, b). This result also established the reliability of the NIPcMWCNTs/GCE as suitable sensor for bromate detection in food products.

Stability and reproducibility

The stability of NiPcMWCNTs/GCE was examined via CV with 20 repeated scans (75 mV s⁻¹) in 0.1 M H₂SO₄ (pH 1) with 1 mM KBrO₃ in it. There was only a 4.5% current drop (Fig. 13), indicating that the NiPcMWCNTs/GCE is very stable. This electrode offered high stability (95.5%) compared with other fabricated bromate sensors (Mao et al. 2015; Zhong et al. 2016).

Fig. 13

Cyclic voltammograms of NiPcMWCNTs/GCE (20 scans) in 0.1 M H₂SO₄ containing 1 mM KBrO₃ solution (pH 1)

The reproducibility of NiPcMWCNTs/GCE was investigated using cyclic voltammetry in 0.1 M H₂SO₄ (pH 1) with 1 mM KBrO₃ in it. The bare electrode was modified with NiPcMWCNTs nanocomposite at three different times before the electroanalysis of the bromate, and the produced cyclic voltammogram was recorded. Figure 14 depicts the NiPcMWCNTs/GCE cyclic voltammograms. After the three successive modifications, the cathodic current responses recorded for the NiPcMWCNTs/GCE gave the percentage relative standard deviation (% RSD) of 3.5%. The NiPcMWCNTs/GCE with a low % RSD, suggesting that the NiPcMWCNTs/GCE displayed a more reproducible result. The fabricated sensor offers a high reproducibility compared to various bromate sensors in literature (Li et al. 2005; Mao et al. 2014).

Fig. 14

Cyclic voltammograms of NiPcMWCNTs/GCE in 0.1 M H₂SO₄ containing 1 mM KBrO₃ solution (pH 1) at 25 mV s⁻¹ (scan rate) for three different electrode modification trials

Conclusion

The study explored the use of the NiPcMWCNTs/GCE to detect bromate in food samples using EIS and SWV techniques. Energy dispersive X-Ray, X-ray diffraction, UV–visible spectroscopy, TEM, and SEM established the successful fabrication of the nanomaterials. The EIS and CV studies revealed that the nanocomposite modified electrode demonstrated a higher current response and a quicker electron transport capability over other modified electrodes. It also showed that the nanocomposite electrode displayed quick electron transfer capability, higher power density, and super-capacitive behaviour in the H₂SO₄ electrolyte than the other investigated electrolytes (Na₂SO₄, PBS, and SAB).

The electrochemical detection of bromate at NiPcMWCNTs/GCE via SWV gave a lower LoD and a higher sensitivity than the EIS technique. In contrast, the EIS offered a wider LDR than the SWV. The mechanism of bromate at the interface of the modified electrode surfaces of NiPcMWCNTs/GCE was found to be diffusion controlled. The developed sensor was characterized by good selectivity, high stability (95.5%), and good reproducibility (% RSD; 3.5%). The developed sensor was effectively adopted to determine bromate in bread samples with a good recovery. Thus, it is a suitable and reliable sensor for bromate detection in real samples (bread).