Improved Electrochemical Hydrogen Peroxide Detection Using a Nickel(II) Phthalimide-Substituted Porphyrazine Combined with Various Carbon Nanomaterials

A metal-free porphyrazine derivative with peripheral phthalimide substituents was metallated with a nickel(II) ion. The purity of the nickel macrocycle was confirmed using HPLC, and characterized by MS, UV–VIS, and 1D (1H, 13C) and 2D (1H–13C HSQC, 1H–13C HMBC, 1H–1H COSY) NMR techniques. The novel porphyrazine was combined with various carbon nanomaterials, such as carbon nanotubes—single walled (SWCNTs) and multi-walled (MWCNTs), and electrochemically reduced graphene oxide (rGO), to create hybrid electroactive electrode materials. The carbon nanomaterials’ effect on the electrocatalytic properties of nickel(II) cations was compared. As a result, an extensive electrochemical characterization of the synthesized metallated porphyrazine derivative on various carbon nanostructures was carried out using cyclic voltammetry (CV), chronoamperometry (CA), and electrochemical impedance spectroscopy (EIS). An electrode modified with carbon nanomaterials GC/MWCNTs, GC/SWCNTs, or GC/rGO, respectively, was shown to have a lower overpotential than a bare glassy carbon electrode (GC), allowing for the measurement of hydrogen peroxide in neutral conditions (pH 7.4). It was shown that among the tested carbon nanomaterials, the modified electrode GC/MWCNTs/Pz3 exhibited the best electrocatalytic properties in the direction of hydrogen peroxide oxidation/reduction. The prepared sensor was determined to enable a linear response to H2O2 in concentrations ranging between 20–1200 µM with the detection limit of 18.57 µM and sensitivity of 14.18 µA mM−1 cm−2. As a result of this research, the sensors produced here may find use in biomedical and environmental applications.


Fabrication of GC/MWCNTs, GC/MWCNTs/Pz3, GC/SWCNTs, GC/SWCNTs/Pz3, GC/rGO, and GC/rGO/Pz3 Modified Electrodes
All materials and reagents used for the electrochemical testing are described in detail in the Supplementary Materials, which also includes information about the apparatus and electrodes.
Prior to each electrochemical experiment, the GC electrode was polished on a polishing cloth with an aqueous suspension of Al 2 O 3 (Buehler, 50 nm average diameter), and any impurities were subsequently removed using an ultrasonic bath containing an acetone/water solution (1:1, v/v). Afterwards, the cleaned surface of the GC electrode was drop-cast with 2 µL of either MWCNTs or SWCNTs dispersion (1 mg mL −1 in DMF), followed by oven-drying of the electrode at 60 • C until the solvent evaporated. In the case of the graphene oxide-modified electrode, initially 2 µL of the graphene oxide aqueous dispersion were drop-cast onto the electrode surface and evaporated under the same conditions. Then graphene oxide (GO) was electrochemically reduced to reduced graphene oxide (rGO) in a KH 2 PO 4 /K 2 HPO 4 buffer (pH 7.4) in the potential ranging from 0.4 to −1.3 V (scanning rate 50 mV s −1 ). The electrochemical reduction of GO to rGO on the surface of the GC electrode is shown in Figure S4. The resulting GC/MWCNTs, GC/SWCNTs, and GC/rGO electrodes were dipped into the Pz3 solution in dichloromethane (1 mg mL −1 ). The porphyrazines were non-covalently immobilized by soaking the respective electrodes in the Pz3 solution. All electrodes were positioned in the desired electrolyte to conduct electrochemical testing. The glass cell holding the electrolyte was deoxygenated prior to the experiments with N 2 gas. All electrochemical tests were performed at room temperature (around 25 • C).

Synthesis and Physicochemical Characterization
Phthalimide-substituted magnesium(II) Pz (1) and its metal-free derivative (2) were prepared using a previously reported three-step synthetic pathway [27]. Next, by modifying a procedure from the literature [28], a Ni 2+ cation was introduced into the Pz2 macrocyclic core by heating Pz2 and nickel(II) acetate tetrahydrate in DMF, which led to the formation of nickel(II) symmetrical porphyrazine Pz3 (Scheme 1). The compounds obtained were isolated chromatographically and their properties evaluated by spectral methods-mass spectrometry and UV-VIS spectroscopy. NMR experiments were performed to confirm the structure of Pz3. The 1 H and 13 C NMR resonances were unambiguously assigned using a combination of one-dimensional ( 1 H, 13 C) and two-dimensional ( 1 H-13 C HSQC, 1 H-13 C HMBC and 1 H-1 H COSY) experiments. A detailed analysis of the NMR spectra can be found in Figures S1-S3 and Table S1. The signals at 4.34 ppm were assigned to the SCH 2 groups based on their correlation with the C2 and C3 pyrrole carbons at 140.5 ppm in the 1 H-13 C HMBC spectrum. The correlations between carbonyl carbon signals at 167.8 ppm and hydrogen atom signals at 4.19 ppm and 7.27 ppm allowed to assign the NCH 2 hydrogen atoms of the ethylsulfanyl linker and phthalimide protons, respectively. The identification of aromatic proton resonances of the phthalimide moieties, as well as protons of the ethylene groups, was supported by the two-dimensional experiments (Table S1). Furthermore, HPLC assessment of Pz3, performed in three different eluent systems, confirmed the purity of the new macrocycle as exceeding 98%, with detection simultaneously at 380 nm and 670 nm (see Supplementary Materials).
The UV-VIS spectra of nickel(II) porphyrazine Pz3 revealed the absorption maxima at 667 nm in dichloromethane and N,N-dimethylformamide, and 668 nm in dimethylsulfoxide ( Figure 1). The calculated values of logarithms of the molar absorption coefficients (log ε) for these bands were 4.58 for dichloromethane and N,N-dimethylformamide and 4.57 in the case of dimethylsulfoxide (Table S2). The comparison of the absorption spectra of Pz3 with that of the previously obtained magnesium(II) porphyrazine 1 and demetallated analog 2 is shown in Figure 1b [27]. Notably, the Q-band absorption of nickel(II) porphyrazine Pz3 was slightly hypsochromically shifted and much less intense than that of the magnesium(II) derivative 1 (the absorption maximum in dichloromethane reached 667 nm for Pz3 and 674 nm for Pz1, while the log ε values equaled 4.58 and 4.83 for Pz3 and Pz1, respectively). However, for both complexes, the Q-band was sharp and single in contrast to the broad and divided band of the demetallated derivative 2 (Figure 1b) [27]. The UV-VIS spectra of nickel(II) porphyrazine Pz3 revealed the absorption maxima at 667 nm in dichloromethane and N,N-dimethylformamide, and 668 nm in dimethylsulfoxide ( Figure 1). The calculated values of logarithms of the molar absorption coefficients (log ε) for these bands were 4.58 for dichloromethane and N,N-dimethylformamide and 4.57 in the case of dimethylsulfoxide (Table S2). The comparison of the absorption spectra of Pz3 with that of the previously obtained magnesium(II) porphyrazine 1 and demetallated analog 2 is shown in Figure 1b [27]. Notably, the Q-band absorption of nickel(II) porphyrazine Pz3 was slightly hypsochromically shifted and much less intense than that of the magnesium(II) derivative 1 (the absorption maximum in dichloromethane reached 667 nm for Pz3 and 674 nm for Pz1, while the log ε values equaled 4.58 and 4.83 for Pz3 and Pz1, respectively). However, for both complexes, the Q-band was sharp and single in contrast to the broad and divided band of the demetallated derivative 2 ( Figure  1b) [27].

Electrochemical Study of Pz3 Deposited on MWCNTs, SWCNTs, and rGO
The voltammetric responses of the assigned hybrid electrodes are shown in Figure 2 for the three distinct carbon nanomaterials (MWCNTs, SWCNTs, and rGO) on which the produced porphyrazine was individually immobilized. To determine the electrochemical

Electrochemical Study of Pz3 Deposited on MWCNTs, SWCNTs, and rGO
The voltammetric responses of the assigned hybrid electrodes are shown in Figure 2 for the three distinct carbon nanomaterials (MWCNTs, SWCNTs, and rGO) on which the produced porphyrazine was individually immobilized. To determine the electrochemical activity, voltammetric measurements were made in a buffered (pH 7.4) water-based electrolyte. For the bare GC electrode and MWCNTs-modified electrode, the cyclic voltamperograms acquired between −1.0 and 0.8 V vs. Ag/AgCl show conventional capacitive characteristics. The peak couples observed for GC/SWCNTs and GC/rGO at ca. 0.0 V can be ascribed to surface-confined processes involving oxygen groups attached to the surface, especially since rGO is not fully reduced and can contain various oxygen-based groups, such as quinones. After the immobilization of Pz3, only slight changes of CV response can be observed. However, no additional redox peaks are present. Unlike GC/SWCNTs/Pz3 and GC/rGO/Pz3, the MWCNTs/Pz3 presents significant redox features formed in the cathodic and anodic range. These redox pairs line up with the electrochemical transition of the phthalimides substituents. In our previous work, we observed the redox transition of the phtalimide groups at ca. −0.4 V in phosphate buffer electrolyte [27]. The minor peak at ca. 0.4 V should then be assigned to the oxidation of Ni 2+ to Ni 3+ . The obtained data suggest that Pz3 exhibits electrochemical activity only on the surface of MWCNTs. The loading of nickel can be calculated based on the charge corresponding to the voltammetric peak at ca. 0.4 V. After the integration of the anodic peak corresponding to the Ni 2+ to Ni 3+ transition, we estimated the metal loadings to be 11.1 ng. We can tentatively assume that the aromatic system of phthalimide offers additional π-π interactions between the porphyrazine and the MWCNT, thus attaching the Pz3 more firmly on the surface of the modified electrode, resulting in efficient electron transfer. Cyclic voltamperometry measurements were performed at scanning rates ranging from 10 to 100 mV s −1 to evaluate the kinetics of electron transport on the surface of the Cyclic voltamperometry measurements were performed at scanning rates ranging from 10 to 100 mV s −1 to evaluate the kinetics of electron transport on the surface of the electrodes modified with GC/MWCNTs/Pz3 (Figure 3). Peak currents increased linearly with scanning rate, pointing to redox activities occurring at the surface. Surface-limited redox characterization is made possible by the π-π stacking of the conjugated porphyrazine macroring with the highly delocalized π-bonding network of carbon nanomaterials. Cyclic voltamperometry measurements were performed at scanning rates rangin from 10 to 100 mV s −1 to evaluate the kinetics of electron transport on the surface of t electrodes modified with GC/MWCNTs/Pz3 (Figure 3). Peak currents increased linear with scanning rate, pointing to redox activities occurring at the surface. Surface-limit redox characterization is made possible by the π-π stacking of the conjugated porphyr zine macroring with the highly delocalized π-bonding network of carbon nanomaterial Comparative CVs for the hybrid materials functionalized with Pz3 (MWCNTs/Pz SWCNTs/Pz3, and rGO/Pz3) employed in this study are shown in Figure 4 in the presen of the [Fe(CN)6] 3−/4− redox couple. As seen in Figure 4A, for the GC/Pz3 electrode, a d Comparative CVs for the hybrid materials functionalized with Pz3 (MWCNTs/Pz3, SWCNTs/Pz3, and rGO/Pz3) employed in this study are shown in Figure 4 in the presence of the [Fe(CN) 6 ] 3−/4− redox couple. As seen in Figure 4A, for the GC/Pz3 electrode, a decrease in peak current was observed in relation to the bare GC electrode. In addition, a significant increase of peak-to-peak separation from 84 to 155 mV was observed at GC/Pz3 when compared to bare GC. This suggests hampered electron transfer kinetics at GC/Pz3, probably due to the insulating nature of Pz3. The modification of the MWCNTs resulted in an increase in the peak currents (compared to the bare GC electrode). It is well-known that MWCNTs have a highly porous structure and good electron transfer characteristics [29]. After the addition of porphyrazine to the GC/MWCNTs system, the peak-to-peak separation increased slightly. Additionally, the peak currents of the [Fe(CN) 6 ] 3−/4− redox couple remained almost unchanged for MWCNTs/Pz3. Such behavior suggests that Pz3 can strongly interact with the surface of MWCNTs via π-π stacking. As a result, electron transfer kinetics at MWCNTs/Pz3 are fast, which is crucial for electrocatalytic sensing applications. In the case of the GC/SWCNTs/Pz3 modified electrode, there was a decrease in both the peak current and capacitance in relation to the GC/SWCNTs electrode ( Figure 4B). The same dependence can be observed for the GC/rGO and GC/rGO/Pz3 modified electrodes ( Figure 4C). Deterioration of the electrochemical properties at these two electrodes was thus observed. Table 1 summarizes the separation peak values for all tested electrodes. GC/MWCNTs/Pz3 electrodes indicates fast electron transfer, which in turn improves the electrocatalytic properties. The results are in good agreement with the cyclic voltammograms displayed in Figure 4 A, B, C. In addition, for the electrodes GC/SWCNTs and GC/SWCNTs/Pz3, as well as GC/rGO and GC/rGO/Pz3, smaller semicircles were observed in comparison to bare GC and GC/Pz3.

The Influence of Hydrogen Peroxide on the GC, GC/Pz3, GC/MWCNTs, GC/MWCNTs/Pz3, GC/SWCNTs, GC/SWCNTs/Pz3, GC/rGO, and GC/rGO/Pz3 Electrodes
It is acknowledged that porphyrazines with transition metal ion centers are interesting candidates for the electrocatalysis of hydrogen peroxide oxidation/reduction [23]. Thus, the electrocatalytic performance of nickel(II)-containing porphyrazine (Pz3) was investigated in the presence of H2O2 ( Figure 5). The efficiency of the constructed hybrid platforms (GC/MWCNTs/Pz3, GC/SWCNTs/Pz3, and GC/rGO/Pz3) was compared with those of bare GC, GC/Pz3, GC/MWCNTs, GC/SWCNTs, and GC/rGO electrodes. The reduction of H2O2 may occur on unmodified GC at a very negative overpotential ( Figure  5A), while the measured reductive current was rather small; an even smaller reduction current was observed in the case of the GC/Pz3 electrode ( Figure 5B). When a GC/MWCNTs electrode was employed, a slight improvement in the H2O2 redox behavior was observed ( Figure 5C). In this case, both cathodic and anodic current waves were seen. Significant electrocatalytic activity was observed for the hybrid electrode of GC/MWCNTs/Pz3. Figure 5D (curve b) illustrates how the redox peaks were greatly enhanced when 2 mM H2O2 was added. At GC/MWCNTs/Pz3, hydrogen peroxide was shown to be reduced and oxidized using a small overpotential and high current. This result indicates that GC/MWCNTs/Pz3 is a suitable electrode for hydrogen peroxide  Table 2 presents the results of the calculated electroactive surface areas for the modified electrodes. We can conclude that by modifying the surface of the GC electrode with the studied nanomaterials, its surface area increases. However, covering nanomaterials with Pz3 causes a decrease in porosity, and a decrease in the electroactive surface of the electrodes is also observed. The EIS measurements (Nyquist plots) for all the tested electrodes are shown in Figure 4D-F. The high-frequency area (semicircle in Nyquist plots) is responsible for the charge transfer resistance on the electrode surface. The relatively large semicircle observed for the GC and GC/Pz3 electrodes (black and red lines, respectively) indicates slow electron transfer. The absence of a semicircle in Figure 4D for the GC/MWCNTs and GC/MWCNTs/Pz3 electrodes indicates fast electron transfer, which in turn improves the electrocatalytic properties. The results are in good agreement with the cyclic voltammograms displayed in Figure 4A-C. In addition, for the electrodes GC/SWCNTs and GC/SWCNTs/Pz3, as well as GC/rGO and GC/rGO/Pz3, smaller semicircles were observed in comparison to bare GC and GC/Pz3. It is acknowledged that porphyrazines with transition metal ion centers are interesting candidates for the electrocatalysis of hydrogen peroxide oxidation/reduction [23]. Thus, the electrocatalytic performance of nickel(II)-containing porphyrazine (Pz3) was investigated in the presence of H 2 O 2 ( Figure 5). The efficiency of the constructed hybrid platforms (GC/MWCNTs/Pz3, GC/SWCNTs/Pz3, and GC/rGO/Pz3) was compared with those of bare GC, GC/Pz3, GC/MWCNTs, GC/SWCNTs, and GC/rGO electrodes. The reduction of H 2 O 2 may occur on unmodified GC at a very negative overpotential ( Figure 5A), while the measured reductive current was rather small; an even smaller reduction current was observed in the case of the GC/Pz3 electrode ( Figure 5B). When a GC/MWCNTs electrode was employed, a slight improvement in the H 2 O 2 redox behavior was observed ( Figure 5C). In this case, both cathodic and anodic current waves were seen. Significant electrocatalytic activity was observed for the hybrid electrode of GC/MWCNTs/Pz3. Figure 5D (curve b) illustrates how the redox peaks were greatly enhanced when 2 mM H 2 O 2 was added. At GC/MWCNTs/Pz3, hydrogen peroxide was shown to be reduced and oxidized using a small overpotential and high current. This result indicates that GC/MWCNTs/Pz3 is a suitable electrode for hydrogen peroxide electrocatalysis. A well-defined H 2 O 2 anode peak current was also observed, suggesting that GC/MWCNTs/Pz3 has good electrocatalytic performance. Regarding electrode modifications, a comparison of SWCNTs and rGO ( Figure 5E,G) with SWCNTs/Pz3 and rGO/Pz3 ( Figure 5F,H) showed only a slight amplification of the peak cathode and anode currents, which indicates that SWCNTs and rGO are not promising platforms for Pz3 immobilization in the context of H 2 O 2 electrocatalysis. Nanomaterials 2023, 13, x FOR PEER REVIEW 9 of 15 electrocatalysis. A well-defined H2O2 anode peak current was also observed, suggesting that GC/MWCNTs/Pz3 has good electrocatalytic performance. Regarding electrode modifications, a comparison of SWCNTs and rGO ( Figure 5E,G) with SWCNTs/Pz3 and rGO/Pz3 ( Figure 5F,H) showed only a slight amplification of the peak cathode and anode currents, which indicates that SWCNTs and rGO are not promising platforms for Pz3 immobilization in the context of H2O2 electrocatalysis.   Furthermore, under stirring conditions, chronoamperometric measurements were made for GC/MWCNTs/Pz3, GC/SWCNTs/Pz3, and GC/rGO/Pz3. The oxidation of H 2 O 2 at the electrodes was driven by the applied anodic potential of +0.6 V. Figure 6A shows the increase of the amperometric signal after the addition of small amounts of H 2 O 2 . The GC/MWCNTs/Pz3 electrode expressed linearity within 20-1200 µM of H 2 O 2 . For GC/SWCNTs/Pz3, linearity was seen in the H 2 O 2 concentration range of 10 to 980 µM. For the GC/rGO/Pz3 electrode, linearity was noted in the range of analyte concentrations from 20 to 750 µM. Figure 6B,E,H also show the amperometry results recorded for the electrode modified with only carbon-based nanomaterials-MWCNTs, SWCNTs, and rGO, respectively. In connection with the above, it was shown that the linearity in the widest range of hydrogen peroxide concentrations was recorded for the GC/MWCNTs/Pz3 electrode. In addition, in the case of carbon nanomaterials (SWCNTs and rGO), a decrease in the current signal was observed after Pz3 adsorption ( Figure 6F,I). The estimated limits of detection (LODs) were 18.57, 9.42, and 9.15 µM for GC/MWCNTs/Pz3, GC/SWCNTs/Pz3 and GC/rGO/Pz3, respectively, when the signal-to-noise ratio of 3 was taken into account. The limits of quantification (LOQs), which were 56.27, 28.56, and 58.04 µM, respectively, were also compared.
In Table 3, the performance of the investigated electrodes for hydrogen peroxide electroanalysis is compared with other electrodes found in the literature. and rGO), a decrease in the current signal was observed after Pz3 adsorption (Figure 6 F, I). The estimated limits of detection (LODs) were 18.57, 9.42, and 9.15 µ M for GC/MWCNTs/Pz3, GC/SWCNTs/Pz3 and GC/rGO/Pz3, respectively, when the signal-to-noise ratio of 3 was taken into account. The limits of quantification (LOQs), which were 56.27, 28.56, and 58.04 µ M, respectively, were also compared.
In Table 3, the performance of the investigated electrodes for hydrogen peroxide electroanalysis is compared with other electrodes found in the literature.  To assess the selectivity of the GC/MWCNTs/Pz3 electrode, the amperometric response was monitored in the presence of different interferents that can commonly occur in a variety of bodily fluids (in this work: glucose, fructose, lactose, maltose, saccharose, caffeine, and sodium chloride) at +0.6 V with stirring ( Figure 7). While H 2 O 2 addition gave a fast current response, none of the inserted interferents resulted in a current reply. Therefore, the obtained sensor has a satisfactory selectivity. In addition, analogous tests were carried out for the modified electrodes: GC/SWCNTs/Pz3 and GC/rGO/Pz3. The results of these analyses are shown in Figure S5. As in the case of the GC/MWCNTs/Pz3 electrode, no interference effect was observed. A decrease in the current signal was observed over time.

Conclusions
In this study, a new phthalimide-substituted sulfanyl porphyrazine deriv sessing an Ni(II) ion in the core was obtained via chemical synthesis. The for this molecule was confirmed using spectral techniques, including NMR spec mass spectrometry, and UV-VIS spectrophotometry. Additionally, detailed spectral studies were performed. The Pz3 was subsequently used to etch glas electrodes, modified earlier with either multi-walled carbon nanotubes, sing carbon nanotubes, or reduced graphene oxide. Among the studied carbon nano it was found that multi-walled carbon nanotubes constitute a suitable matr immobilization of Pz3 porphyrazine on the surface of the GC electrode. It shown that the GC/MWCNTs/Pz3 modified electrode has excellent electr properties in the detection of hydrogen peroxide oxidation/reduction (LOD: 1 linear range: 20-1200 μM) and may be regarded as a potential sensor of this i molecule. The results obtained in this study show that a determination of the ap conductive material for electrode modification is crucial to achieving the syne fect and satisfactory electrocatalytic properties.  Figure S1: NMR data of Pz3. Table S1: 1 H and 13 C NMR data obtained for Pz3. Figure S2: 1 H NMR spectrum of Pz3 (400 MHz, chloroform-d 1 , 298 K). Figure S3: 13 C NMR spectrum recorded for Pz3 (100 MHz, chloroform-d 1 , 298 K). Table S2: UV-VIS absorption maxima (λ Abs ) and logarithms of molar absorption coefficients (logε) of Pz3 in selected organic solvents. Figure S4: Electrochemical reduction of GC/GO to GC/rGO in phosphate buffer (PB, pH 7.4) in the potential range from 0.4 to −1.3 V (scanning rate 50 mV s −1 ). Figure S5: Chronoamperometric responses recorded at (A) GC/SWCNTs/Pz3 and (B) GC/rGO/Pz3 sensor upon addition of 100 µM H 2 O 2 and selected interferents. The applied potential was +0.6 V. The supporting electrolyte was PB (pH 7.4).