Hybridization of Co3S4 and Graphitic Carbon Nitride Nanosheets for High-performance Nonenzymatic Sensing of H2O2

The development of efficient H2O2 sensors is crucial because of their multiple functions inside and outside the biological system and the adverse effects that a higher concentration can cause. This work reports a highly sensitive and selective non-enzymatic electrochemical H2O2 sensor achieved through the hybridization of Co3S4 and graphitic carbon nitride nanosheets (GCNNS). The Co3S4 is synthesized via a hydrothermal method, and the bulk g-C3N4 (b-GCN) is prepared by the thermal polycondensation of melamine. The as-prepared b-GCN is exfoliated into nanosheets using solvent exfoliation, and the composite with Co3S4 is formed during nanosheet formation. Compared to the performances of pure components, the hybrid structure demonstrates excellent electroreduction towards H2O2. We investigate the H2O2-sensing performance of the composite by cyclic voltammetry, differential pulse voltammetry, and amperometry. As an amperometric sensor, the Co3S4/GCNNS exhibits high sensitivity over a broad linear range from 10 nM to 1.5 mM H2O2 with a high detection limit of 70 nM and fast response of 3 s. The excellent electrocatalytic properties of the composite strengthen its potential application as a sensor to monitor H2O2 in real samples. The remarkable enhancement of the electrocatalytic activity of the composite for H2O2 reduction is attributed to the synergistic effect between Co3S4 and GCNNS.


Introduction
H 2 O 2 is an important chemical in many fields, including the biomedical, pharmaceutical, food, and textile industries [1][2][3]. It is also a reactive oxygen species abundant in living organisms and is essential for maintaining regular biological functions. At normal concentrations, it functions as a signaling molecule for neural development and cell proliferation and is a byproduct of cellular metabolism [4]. However, abnormal levels of H 2 O 2 in a cell cause oxidative stress, leading to aging and diseases such as Parkinson's, Alzheimer's, cardiovascular diseases, cancers, or inflammation, and is, therefore, a biomarker [5,6]. As a result, the accurate and efficient determination of H 2 O 2 is imperative, and the drive to develop inexpensive and highly sensitive H 2 O 2 sensors has increased significantly.
There are different methods to detect H 2 O 2 , such as fluorimetry, titrimetry, chemiluminescence, and spectrophotometry [7][8][9][10]. However, electrochemical methods offer a better platform for fast, sensitive, inexpensive, and portable sensing [6,11]. For the electrochemical detection of H 2 O 2 , enzymatic or non-enzymatic sensors can be used. With enzymatic the increased electron transfer kinetics at the composite electrode due to the collective effect between Co 3 S 4 and GCNNS. The practical application of the developed sensor was evaluated by estimating H 2 O 2 in the human serum. Co 3 S 4 micro flowers were synthesized by hydrothermal route [33]. First, 6 mmol of cobalt (II) nitrate hexahydrate was dissolved in 60 mL of distilled water. After stirring for 10 min, 8 mmol of thiourea was added and vigorously stirred for another 15 min. Then, 4 mL of ethylenediamine was added to the above solution, and the color of the solution changed to brown. The solution was transferred to a 100 mL Teflon-lined autoclave and held at 200 • C for 12 h. After that, the reaction mixture was cooled to room temperature. A black-colored product was obtained, which was washed three times with distilled water and ethanol. The product was dried at 60 • C overnight.

GCNNS
b-GCN was synthesized via the thermal polycondensation of melamine. 4 g of melamine was heated at 600 • C in a crucible for 2 h at 25 • C min −1 to obtain a yellowcolored b-GCN. The nanosheets were prepared by the solvent exfoliation of b-GCN. 10 mg of b-GCN was dispersed in 10 mL of distilled water and sonicated at 40 kHz for 2 h to obtain the nanosheets [30].

Co 3 S 4 /GCNNS
First, 10 mg b-GCN was dispersed in 10 mL of distilled water and sonicated for 2 h at 40 kHz to obtain the nanosheets. Then, 2 mg of Co 3 S 4 was added to the above dispersion and was further sonicated for 45 min to obtain the Co 3 S 4 /GCNNS [34,35].

Fabrication of the Electrodes
The glassy carbon electrode (GCE) was polished with 0.3 and 0.05 µm alumina slurry using a polishing cloth, sonicated for 3 min in distilled water and dried at room temperature. Drop-casting is a simple, efficient, and fast method to modify electrode surfaces [36]. 10 µL of the prepared composite was drop-cast onto the mirror-polished GCE and dried overnight at room temperature. The fabricated Co 3 S 4 /GCNNS-modified electrode was washed with distilled water and used for electrochemical studies. GCNNS-modified GCE and Co 3 S 4modified GCE were prepared by a similar procedure. A scheme for synthesizing the materials and the fabrication of Co 3 S 4/ GCNNS is shown in Figure 1.

Crystallographic Studies
Powder X-ray diffraction (XRD) studies of the as-prepared Co3S4, b-GCN, GCNNS, and Co3S4/GCNNS ( Figure 2) help reveal the crystal and phase structure of the prepared materials. The XRD spectra of b-GCN and GCNNS are identical, and the characteristic peaks exist at 2θ values of 12.87° and 27.58° due to diffraction from (100) and (002) crystallographic plane and agree well with the standard JCPDS card no# 87-1526. The peak centered at 12.87° is a feature of the repeating tris triazine structural motifs present in b-GCN and GCNNS, and the peak at 27.58° ascribes to the periodic stacking of layers along the c-axis. The XRD spectrum of Co3S4 shows a cubic phase with the diffraction peaks at 2θ values of 31.1°, 36.04°, 47.13°, and 55.0° corresponding to the crystal planes of (311), (400), (422), and (440) aligning well with standard JCPDS card no# . As depicted in Figure 2, the XRD pattern of the Co3S4/GCNNS nanocomposite shows the presence of both GCNNS and Co3S4, ensuring the successful formation of the composite. XRD analysis validated the synthesis of the materials, crystalline behavior, and phase purity.

Crystallographic Studies
Powder X-ray diffraction (XRD) studies of the as-prepared Co 3 S 4 , b-GCN, GCNNS, and Co 3 S 4 /GCNNS ( Figure 2) help reveal the crystal and phase structure of the prepared materials. The XRD spectra of b-GCN and GCNNS are identical, and the characteristic peaks exist at 2θ values of 12.87 • and 27.58 • due to diffraction from (100) and (002) crystallographic plane and agree well with the standard JCPDS card no# 87-1526. The peak centered at 12.87 • is a feature of the repeating tris triazine structural motifs present in b-GCN and GCNNS, and the peak at 27.58 • ascribes to the periodic stacking of layers along the c-axis. The XRD spectrum of Co 3 S 4 shows a cubic phase with the diffraction peaks at 2θ values of 31.1 • , 36.04 • , 47.13 • , and 55.0 • corresponding to the crystal planes of (311), (400), (422), and (440) aligning well with standard JCPDS card no# 42-1448. As depicted in Figure 2, the XRD pattern of the Co 3 S 4 /GCNNS nanocomposite shows the presence of both GCNNS and Co 3 S 4 , ensuring the successful formation of the composite. XRD analysis validated the synthesis of the materials, crystalline behavior, and phase purity. XPS studies were performed to acquire informatio elements in Co3S4, b-GCN, GCNNS, and Co3S4/GCNN GCN was determined to be 0.77, with the atomic perce and 55.4, respectively. This confirms the graphitic natur tains C and N in a nearly 3:4 ratio [37,38]. The spectra o ure 3, and those of GCNNS and Co3S4 are in Figure S full-scan spectrum of Co3S4/GCNNS in Figure 3a revea in the composite. The O detected is due to the oxidatio sample in the air [39]. As shown in Figure 3b, the high deconvoluted into two peaks at 284.8 and 288.0 eV. Th graphitic or amorphous carbon present in GCNNS or ad atoms in the N = C (N) group in GCNNS yield a p resolution spectrum of N 1s (Figure 3c) combines fou 404.6 eV. The peak at 398.4 eV is the N sp 2 bond in C

XPS Analysis
XPS studies were performed to acquire information about the chemical states of the elements in Co 3 S 4 , b-GCN, GCNNS, and Co 3 S 4 /GCNNS. The C/N atomic ratio in the b-GCN was determined to be 0.77, with the atomic percentages of C 1s and N 1s being 44.6 and 55.4, respectively. This confirms the graphitic nature of the prepared b-GCN as it contains C and N in a nearly 3:4 ratio [37,38]. The spectra of Co 3 S 4 /GCNNS are shown in Figure 3, and those of GCNNS and Co 3 S 4 are in Figure S1 (Supplementary Materials). The full-scan spectrum of Co 3 S 4 /GCNNS in Figure 3a reveals the presence of C, N, Co, and S in the composite. The O detected is due to the oxidation or absorption of oxygen by the sample in the air [39]. As shown in Figure 3b, the high-resolution spectra of C 1s can be deconvoluted into two peaks at 284.8 and 288.0 eV. The peak at 284.8 eV is attributed to graphitic or amorphous carbon present in GCNNS or adsorbed on the surface. The carbon atoms in the N = C − (N) 2 group in GCNNS yield a peak at 288.0 eV [35,40]. The high-resolution spectrum of N 1s (Figure 3c) combines four peaks at 398.4, 399.3, 400.5, and 404.6 eV. The peak at 398.4 eV is the N sp 2 bond in C − N = C in the triazine ring, and the peaks at 399.3 and 400.5 eV correspond to the tertiary nitrogen group (N − (C) 3 ) and the quaternary N three-carbon atom amino functional group (N − H) in the aromatic ring, respectively. The peak at 404.6 eV corresponds to the π excitation of C = N in GCNNS [41].
in the hybrid structure. Meanwhile, the electronegativity of Co (1.88) is lower than that of C and N, further proving that the electron of Co3S4 tends to donate to GCNNS. These results prove that the integration between Co3S4 and GCNNS in the composite is not just a physical mixture but that there is heterojunction formation with strong electronic interactions [44].

UV-Visible
The UV-visible spectra of Co3S4, GCNNS, and Co3S4/GCNNS are shown in Figure 4. GCNNS shows a characteristic absorption band at 318 nm due to the π-π* transition in GCNNS, as previously reported [30]. The corresponding absorption in Co3S4/GCNNS appears at 321 nm, although the intensity of the absorption has diminished. The slight wavelength shift and decrease in adsorption intensity are attributed to composite formation. Co3S4 displays a broad absorption in the 400-800 nm region. Due to the integration of Co3S4 into the GCNNS matrix, the Co3S4/GCNNS exhibits greater absorption in the visible light region than that of GCNNS. The high-resolution spectra of Co 2p in the composite (Figure 3d) can be deconvoluted into two spin-orbit doublets of Co 2+ and Co 3+ . The peaks 14.96 eV apart at 783.3 and 798.26 eV correspond to the 2p 3/2 and 2p 1/2 orbitals of Co 2+ in Co 3 S 4 . Whereas the 2p 3/2 and 2p 1/2 doublet of Co 3+ appears at 780.7 and 796.1 eV with a spacing of 15.4 eV, suggesting the existence of Co in the +2 and +3 oxidation states in Co 3 S 4 /GCNNS and a satellite peak appears at 804.2 eV [42,43]. The S 2p spectra (Figure 3e) of Co 3 S 4 /GCNNS are fitted into four peaks, the peaks at 161.9 and 164.9 eV are indexed to S 2p 3/2 and S 2p 1/2 of S in Co 3 S 4 and the two peaks at 166.8 and 171.48 eV belong to S in SO 2− 3 and SO 2− 4 , respectively [41]. The results demonstrate the successful formation of the Co 3 S 4 /GCNNS composite. As shown in Figure S1a, the survey scan of GCNNS showed the presence of C 1s and N 1s, while Co 3 S 4 showed the presence of Co 2p and S 2p, and the deconvoluted spectra are shown in Figure S1b,c,e,f. The spectral peaks of the materials are presented in Table S1 for comparison. Interestingly, the binding energy of Co 2p has increased in the composite compared to pure Co 3 S 4 , and the binding energy of C 1s has decreased compared to its pure counterpart, strongly suggesting the electron transfer between Co 3 S 4 and GCNNS in the composite. The decrease in electron density of Co 3 S 4 in the composite causes an increase in its binding energy, attributed to the transfer of electrons from Co 3 S 4 to GCNNS in the hybrid structure. Meanwhile, the electronegativity of Co (1.88) is lower than that of C and N, further proving that the electron of Co 3 S 4 tends to donate to GCNNS. These results prove that the integration between Co 3 S 4 and GCNNS in the composite is not just a physical mixture but that there is heterojunction formation with strong electronic interactions [44].

UV-Visible
The UV-visible spectra of Co 3 S 4 , GCNNS, and Co 3 S 4 /GCNNS are shown in Figure 4. GCNNS shows a characteristic absorption band at 318 nm due to the π-π* transition in GCNNS, as previously reported [30]. The corresponding absorption in Co 3 S 4 /GCNNS appears at 321 nm, although the intensity of the absorption has diminished. The slight wavelength shift and decrease in adsorption intensity are attributed to composite formation. Co 3 S 4 displays a broad absorption in the 400-800 nm region. Due to the integration of Co 3 S 4 into the GCNNS matrix, the Co 3 S 4 /GCNNS exhibits greater absorption in the visible light region than that of GCNNS.

. SEM and EDS
The morphology of the synthesized materials was in Figure 5. The SEM micrograph of the b-GCN shown i structure. Further GCNNS was drop-cast and dried on ord the SEM images ( Figure 5b). Interestingly, wrink thicknesses are observed, confirming the formation of t micrographs of Co3S4 ( Figure 5c) show a micron-sized f with the literature [33]. Co3S4 micro flowers sonicated in GC plate, and examined by SEM to understand whether sonication. No morphological changes were observed modified on the GC plate was analyzed, and as shown in both GCNNS and Co3S4 micro flowers, with Co3S4 embe Co3S4 and GCNNS retain their structure in the composi

SEM and EDS
The morphology of the synthesized materials was examined by SEM and is shown in Figure 5. The SEM micrograph of the b-GCN shown in Figure 5a is a highly aggregated structure. Further GCNNS was drop-cast and dried on a glassy carbon (GC) plate to record the SEM images ( Figure 5b). Interestingly, wrinkled layers with a few-nanometer thicknesses are observed, confirming the formation of the nanosheets. The recorded SEM micrographs of Co 3 S 4 ( Figure 5c) show a micron-sized flower-like morphology consistent with the literature [33]. Co 3 S 4 micro flowers sonicated in water were drop-cast, dried on a GC plate, and examined by SEM to understand whether structural distortion occurs under sonication. No morphological changes were observed. Furthermore, the Co 3 S 4 /GCNNS modified on the GC plate was analyzed, and as shown in Figure 5d, the composite showed both GCNNS and Co 3 S 4 micro flowers, with Co 3 S 4 embedded on the GCNNS surface. The Co 3 S 4 and GCNNS retain their structure in the composite; however, slight agglomeration was observed in the composite. The elemental composition of the materials was analyzed using EDS. As shown in Figure 5e, the EDS spectrum of Co 3 S 4 /GCNNS showed the presence of C, N, Co, and S, confirming the formation of the composite. The respective atomic and weight percentages of the elements are presented in Figure 5f. The uniform distribution of the elements can be understood from the elemental mapping shown in Figure 5g.
both GCNNS and Co3S4 micro flowers, with Co3S4 embedded on the GCNNS surface. The Co3S4 and GCNNS retain their structure in the composite; however, slight agglomeration was observed in the composite. The elemental composition of the materials was analyzed using EDS. As shown in Figure 5e, the EDS spectrum of Co3S4/GCNNS showed the presence of C, N, Co, and S, confirming the formation of the composite. The respective atomic and weight percentages of the elements are presented in Figure 5f. The uniform distribution of the elements can be understood from the elemental mapping shown in Figure 5g. It can be observed that the Co 3 S 4 fringes are closely surrounded by GCNNS fringes in the composite, indicating the strong interfacial contact between the materials, which is favorable for the faster electron transfer to promote the electrocatalytic reduction of H 2 O 2 . As shown in Figure 6d, the SAED pattern of Co 3 S 4 /GCNNS has a concentric ring structure, revealing the polycrystalline nature of the composite. TEM analysis confirms the successful coupling of GCNNS and Co 3 S 4 in the composite.

Response of Co 3 S 4 and Co 3 S 4 /GCNNS Modified GCE Electrode in NaOH
The formation of the Co 3 S 4 and Co 3 S 4 /GCNNS layer on the GCE surface was examined by recording cyclic voltammograms of the modified electrodes in 0.1 M NaOH at different scan rates of 10-100 mV/s and is shown in Figures 7a,b, respectively. The cyclic voltammograms of Co 3 S 4 show four characteristic peaks, including two anodic peaks at 0.22, and 0.50 V and two cathodic peaks at 0.20 and 0.51 V, as shown in Figure 6a, and are consistent with the literature [15,45]. As the scan rate increases, the peak currents increase. These peaks arise from reversible electrochemical redox reactions of cobalt in different oxidation states in the Co 3 S 4 . Although the mechanisms of these reactions in Co 3 S 4 are not well understood, they are expected to be similar to the well-reported redox reactions of Co(OH) 2 since the CV profile is similar. There is only a slight deviation in the anodic and cathodic peak potentials of Co 3 S 4 from that of Co(OH) 2 . Oxygen and sulfur also belong to the same group [46]. Furthermore, GCE modified with Co 3 S 4 /GCNNS (0.2 mg/mL) was characterized in the same way, as illustrated in Figure 6b. The corresponding anodic peak observed in Co 3 S 4 can also be seen in the composite in the positive potential scan, and there is an increase in the peak currents, and the peak at 0.22 V becomes prominent. During the negative potential scan, two cathodic peaks corresponding to that in the pure Co 3 S 4 are visible, and the reduction peak currents have increased. The results confirm the presence of cobalt in both Co 3 S 4 and Co 3 S 4 /GCNNS electrodes and suggest an improvement in the redox activity of the Co 3 S 4 by the composite formation with GCNNS. Co3S4 are not well understood, they are expected to be similar to the well-reported redox reactions of Co(OH)2 since the CV profile is similar. There is only a slight deviation in the anodic and cathodic peak potentials of Co3S4 from that of Co(OH)2. Oxygen and sulfur also belong to the same group [46]. Furthermore, GCE modified with Co3S4/GCNNS (0.2 mg/mL) was characterized in the same way, as illustrated in Figure 6b. The corresponding anodic peak observed in Co3S4 can also be seen in the composite in the positive potential scan, and there is an increase in the peak currents, and the peak at 0.22 V becomes prominent. During the negative potential scan, two cathodic peaks corresponding to that in the pure Co3S4 are visible, and the reduction peak currents have increased. The results confirm the presence of cobalt in both Co3S4 and Co3S4/GCNNS electrodes and suggest an improvement in the redox activity of the Co3S4 by the composite formation with GCNNS.   6 ] each, from a potential range of 0.6 to −0.2 V vs. Ag/AgCl at a scan rate of 50mV/s and are shown in Figure 7c. A well-defined reversible redox peak for the K 4 [Fe(CN) 6 ]/K 3 [Fe(CN) 6 ] system is observed. Bare GCE showed an oxidation peak current of 10.2 µA, and the redox peaks are separated by 67 mV (∆E p ). The oxidation peak current was increased when GCE was modified with Co 3 S 4 to a value of 14.88 µA, although the ∆Ep value has increased to 140 mV. Meanwhile, the GCNNS-modified electrode yielded an oxidation current of 18.57 µA, 1.25 times higher than Co 3 S 4 , and displayed a peak separation of 90 mV lower than for Co 3 S 4 . The increase in the response of GCNNS to Fe 2+ /Fe 3+ is because the nanosheet morphology of GCNNS favors electron transfer between the analyte and the electrode. These results illustrate the superior electrocatalytic activity of GCNNS towards the Fe 2+ /Fe 3+ couple compared to Co 3 S 4 . The Co 3 S 4 /GCNNS-modified GCE showed a maximum peak current of 28.20 µA, 1.89 and 1.51 times higher than Co 3 S 4 and GCNNS with a ∆Ep of 82 mV. Interestingly, the hybridization of Co 3 S 4 with GCNNS resulted in a significant increase in the oxidation peak current and a reduced ∆Ep compared to its counterparts, strongly indicating the enhanced electrocatalytic aspects and the importance of the synergistic effects of the composite. The electroactive surface area (EAS) of the fabricated electrodes was calculated using the Randles-Sevcik Equation (1), where I p is the peak current at the respective electrode, n is the number of electrons involved (1 for [Fe(CN) 6 ] 3−/4− ), A is the EAS of the electrode in cm 2 , D is the diffusion coefficient (6.7 × 10 6 cm 2 /s), C is the concentration of the redox couple (1 mM), and ν is the scan rate (50 mV/s). The EAS of GCE, Co 3 S 4 , GCNNS, and Co 3 S 4 /GCNNS-modified GCE was estimated to be 0.067, 0.097, 0.12, and 0.18 cm 2 , respectively. The Co 3 S 4 /GCNNSmodified GCE has the highest EAS, 1.9 times higher than Co 3 S 4 and 1.5 times higher than GCNNS which gives the composite electrode high electrocatalytic activity compared to their pure counterparts.

Electrochemical Impedance Studies
The impedance changes occurring at the electrode surface were further investigated using electrochemical impedance spectroscopy. Nyquist plots of bare GCE, Co 3 S 4 , GCNNS, and Co 3 S 4 /GCNNS-modified GCE electrodes were recorded in 0.1 M KCl containing 1 mM K 4 [Fe(CN) 6 ]/K 3 [Fe(CN) 6 ], and the obtained plots are shown in Figure 7d. The impedance data were simulated using the Randles equivalent circuit model and are shown in Figure 7d (inset), where R S is the ohmic resistance of the electrolyte, R CT is the charge transfer resistance, C dl is the double layer capacitance, and Z W is the Warburg impedance. The semicircular region in the Nyquist plots corresponds to the electron transfer limited process, and the diameter is equal to the R CT . The R CT values of the electrodes were determined to be 15.80, 6.28, 2.88, and 1.40 kΩ for bare, Co 3 S 4 , GCNNS, and Co 3 S 4 /GCNNS modified GCEs, respectively. When modified with Co 3 S 4 or GCNNS, the R CT values become lower than bare GCE values. Meanwhile, Co 3 S 4 /GCNNS has the lowest R CT value, demonstrating fast electron transfer mediated by the composite, compared to the pure counterparts. The heterogeneous electron transfer rate constant (k et ) of K 4 [Fe(CN) 6 ]/K 3 [Fe(CN) 6 ] at the modified electrode is calculated using Equation (2), where R is the gas constant (8.314 J.mol The results confirm that an efficient electrical network through Co 3 S 4 anchored on the surface of GCNNS facilitates electron transfer in the composite-modified GCE. To further investigate the charge transfer, the photoluminescence (PL) of the materials was measured and is shown in Figure S2. At an excitation wavelength of 330 nm, both GCNNS and Co 3 S 4 /GCNNS show PL peaks around 450 nm. The GCNNS showed the highest PL intensity, indicating the faster recombination of photogenerated eand h + [38]. The decrease in PL intensity in the composite is due to the quenching of carrier recombination by Co 3 S 4 , and these electrons can be transferred at the interface. The results indicate that the heterogeneous electron transfer rate in Co 3 S 4 /GCNNS is larger than in GCNNS and is beneficial for H 2 O 2 sensing.   Figure 8 (curve a). Figure 8 (curve b) shows that the GCE modified with GCNNS reduces H 2 O 2 at a potential of −0.65 V to yield a peak current of −12.21 µA. The amino or cyano groups caused the activity on the surface of the GCNNS. The abundance of N atoms with lone pairs in their sp 2 orbitals in the GCNNS also aids in the adsorption of tiny molecules, such as H 2 O 2 [27,32]. In this case, this favorable interaction between the analyte and the material facilitates electrocatalytic reduction. On the other hand, the electrochemical reduction of H 2 O 2 at the Co 3 S 4 -modified GCE has the advantage of a lower reduction potential, as shown by Figure 8 (curve c). H 2 O 2 is reduced at 200 mV less negative potential than the GCNNS electrode, but the peak current is less (−8.14 µA).

Electrochemical Reduction of H2O2 at Co3S4/GCNNS Modified GCE
The electrocatalytic properties of the fabricated electrodes towards H2O2 reduction were initially investigated using CV in a 0.2 M phosphate-buffered solution (PBS) of pH 7.2. Figure 8 shows the cyclic voltammograms obtained for bare and modified electrodes in the absence or presence of 1 mM H2O2, recorded between a potential range of 0.4 to −0.95 V at a scan rate of 50 mV/s. A bare GCE shows no redox features in the presence of H2O2 in Figure 8 (curve a). Figure 8 (curve b) shows that the GCE modified with GCNNS reduces H2O2 at a potential of −0.65 V to yield a peak current of −12.21 µA. The amino or cyano groups caused the activity on the surface of the GCNNS. The abundance of N atoms with lone pairs in their sp 2 orbitals in the GCNNS also aids in the adsorption of tiny molecules, such as H2O2 [27,32]. In this case, this favorable interaction between the analyte and the material facilitates electrocatalytic reduction. On the other hand, the electrochemical reduction of H2O2 at the Co3S4-modified GCE has the advantage of a lower reduction potential, as shown by Figure 8 (curve c). H2O2 is reduced at 200 mV less negative potential than the GCNNS electrode, but the peak current is less (−8.14 µA).
At the Co3S4/GCNNS electrode, a sharp and enhanced reduction peak can be observed at a potential of −0.6 V for the electroreduction of H2O2 in Figure 8 (curve d), while the electrode shows no response in the absence of H2O2, as seen in Figure 8 (curve e). The composite electrode delivers a large current of −58.2 µA, 4.7 times higher and 7.1 times higher than GCNNS and Co3S4-modified GCEs, respectively. Interestingly, the H2O2 reduction potential at Co3S4/GCNNS is reduced to 50 mV less negative than that at the GCNNS electrode. The cyclic voltammetric results show that the GCE modified with Co3S4/GCNNS shows the best performance towards H2O2 reduction. The remarkable enhancement in the reduction peak currents can be attributed to the synergistic effect between Co3S4 and GCNNS that might have fastened the electron transfer with the H2O2 at the electrode surface. Co3S4 has good catalytic activity towards the reduction of H2O2 as it reduces the analyte at a lower potential. Still, the reduction peak current is lower, possibly due to the lower conductivity and small surface area of the Co3S4 micro flowers (0.097 cm 2 ). When combined with GCNNS, the effective surface area and  Figure 8 (curve e). The composite electrode delivers a large current of −58.2 µA, 4.7 times higher and 7.1 times higher than GCNNS and Co 3 S 4 -modified GCEs, respectively. Interestingly, the H 2 O 2 reduction potential at Co 3 S 4 /GCNNS is reduced to 50 mV less negative than that at the GCNNS electrode. The cyclic voltammetric results show that the GCE modified with Co 3 S 4 /GCNNS shows the best performance towards H 2 O 2 reduction.
The remarkable enhancement in the reduction peak currents can be attributed to the synergistic effect between Co 3 S 4 and GCNNS that might have fastened the electron transfer with the H 2 O 2 at the electrode surface. Co 3 S 4 has good catalytic activity towards the reduction of H 2 O 2 as it reduces the analyte at a lower potential. Still, the reduction peak current is lower, possibly due to the lower conductivity and small surface area of the Co 3 S 4 micro flowers (0.097 cm 2 ). When combined with GCNNS, the effective surface area and conductivity are improved. As discussed earlier, the Co 3 S 4 /GCNNS electrode has a high electroactive surface area (0.18 cm 2 ), the lowest R CT value (1.40 kΩ), indicating improved conductivity, and the highest charge transfer constant to facilitate faster electron transfer between the electrode and the analyte. These parameters are decisive in the superior electrocatalytic activity of the Co 3 S 4 /GCNNS electrode towards H 2 O 2 reduction. All of these observations emphasize the excellent electrocatalytic activity of the Co 3 S 4 /GCNNS electrode.
The possible mechanism for the electroreduction of H 2 O 2 at Co 3 S 4 /GCNNS can be expressed as in Equations (3)-(8). H 2 O 2 reduction at the composite electrode occurs via the direct two-electron transfer pathway H 2 O 2 + 2e − → 2OH − and is converted to H 2 O. In the composite, both GCNNS and Co 3 S 4 involve the reduction mechanism. The reaction at the GCNNS surface can be expressed as in Equations (3)-(5), where the active sites are the nitrogen-containing functional groups. While in Co 3 S 4 , H 2 O 2 is reduced by the Co(II) ions and is converted to Co 3 S 4 (OH). Co(III) in Co 3 S 4 (OH) is again electro-reduced to Co(II), and Co 3 S 4 is regenerated at the electrode surface. The corresponding mechanism is shown by Equations (6)- (8).
OH ads + e − → OH − (4) As indicated by XPS, there is electron transfer between GCNNS and Co 3 S 4 , which could increase the active sites available for redox reactions. Additionally, the coordination bonds formed between the cobalt ion and the lone nitrogen pairs stabilize Co 3 S 4 on the GCNNS surface, and since GCNNS readily adsorbs H 2 O 2 on its surface, this also increases the interaction of H 2 O 2 with Co(II) active sites and thereby favors a stronger reduction.
The response of Co 3 S 4 /GCNNS to 1 mM H 2 O 2 was further studied by varying the scan rate from 10 to 60 mV/s, as shown in Figure 9a. Upon increasing the scan rates, the cathodic peak currents were increased. The peak current vs. scan rate was plotted, as shown in Figure 9b, and it showed a good linear relationship. The regression expression is I = −0.66 × ν -22.58 with an R 2 value of 0.99. This indicates that the reduction of H 2 O 2 at the Co 3 S 4 /GCNNS electrode is a surface adsorption-controlled process. The electrochemical reduction behavior of the composite electrode towards 1 mM H 2 O 2 in different pH ranges from 3 to 11 was studied using CV in 0.2 M PBS at a scan rate of 50 mV/s. With an increase in the pH, the reduction peak potential of H 2 O 2 on the composite electrode shifts to more negative values, as seen in Figure 9c. The shift in the peak potentials suggests that protons are involved in the electrochemical reduction. The reduction peak currents increase upon increasing the pH from 3 to 7 and decrease thereafter. The peak current achieves the maximum at a pH of 7; hence it is chosen for the electrochemical studies to determine H 2 O 2 . The relationship of reduction peak potential vs. pH is plotted in Figure 9d, showing a linear relationship. The regression equation is E = −0.05 × pH − 0.28 with an R 2 value of 0.99.
The slope of the curve is around 59 mV, which matches the theoretical value for the two protons and two electrons reaction.
Biosensors 2023, 13, x FOR PEER REVIEW 13 of 20 current and onset potential increase. This is possible because above 0.2 mg, Co3S4 agglomerates on the GCNNS surface, reducing the number of catalytically active sites available for H2O2 reduction. Furthermore, the weight of GCNNS was varied to 0.5, 1, and 1.5 mg for a weight of Co3S4 of 0.2 mg/mL water, and the recorded CV curves are shown in Figure S3b. At a loading of 0.5 mg/mL, the peak reduction current is lower; with an increase in the GCNNS level, there is an increase in peak current, but above 1 mg/mL, the activity decreases. This is due to the agglomeration of the nanosheets and the loss of active sites on the electrode surface due to the increased loading. Therefore, the optimal combination to produce the best-performing Co3S4/GCNNS composite electrode was determined to be 0.2 mg Co3S4 and 1 mg GCNNS in 1 mL distilled water, which was then used for subsequent electrochemical studies.

Sensitive Determination of H2O2
As illustrated in Figure 10a, differential pulse voltammetry (DPV) was achieved for the successive injection of 1 µM of H2O2 in 0.2 M PBS (pH 7.2) at the Co3S4/GCNNS electrode. The reduction current rose linearly with each addition of H2O2. Still, the reduction potential remained constant at −0.52 V. This suggests that it is possible to sensitively measure H2O2 at Co3S4/GCNNS without impacting its reduction potential. The regression relation I = −0.05 × c − 1.24 represents the calibration plot between the reduction current and concentration of H2O2 in Figure 10b, and it displays a strong linear relationship with an R 2 value of 0.99.  Figure S3a, in which the peak current increases up to a load of 0.2 mg; beyond that, the current decreases, and the shape of the CV curve changes as the capacitance current and onset potential increase. This is possible because above 0.2 mg, Co 3 S 4 agglomerates on the GCNNS surface, reducing the number of catalytically active sites available for H 2 O 2 reduction.
Furthermore, the weight of GCNNS was varied to 0.5, 1, and 1.5 mg for a weight of Co 3 S 4 of 0.2 mg/mL water, and the recorded CV curves are shown in Figure S3b. At a loading of 0.5 mg/mL, the peak reduction current is lower; with an increase in the GCNNS level, there is an increase in peak current, but above 1 mg/mL, the activity decreases. This is due to the agglomeration of the nanosheets and the loss of active sites on the electrode surface due to the increased loading. Therefore, the optimal combination to produce the best-performing Co 3 S 4 /GCNNS composite electrode was determined to be 0.2 mg Co 3 S 4 and 1 mg GCNNS in 1 mL distilled water, which was then used for subsequent electrochemical studies.

Sensitive Determination of H 2 O 2
As illustrated in Figure 10a

Amperometric Sensing of H2O2 at Co3S4/GCNNS Modified GCE
Wide-range amperometry was employed to evaluate the application of the Co3S4/GCNNS electrode in the detection of H2O2, as depicted in Figure 11a. The amperometric i-t curve was obtained at a constant potential of −0.7 V, and the concentration of H2O2 varied from 10 nM to 1.5 mM in 0.2 M PBS of pH 7.2. During each addition of H2O2, a well-defined increase in the current was observed. The sharp increase in the current is due to the electroreduction of H2O2 at the Co3S4/GCNNS electrode. A steady state of the final reduction current was reached in 3 s, indicating the rapid response of the electrode to H2O2. The amperometric response also demonstrates the excellent electrocatalytic performance of the Co3S4/GCNNS composite electrode. The linear increase in the reduction current with the increasing H2O2 concentration validates its practical application as an H2O2 sensor. As shown in Figure 11b, the calibration curve for the sensor was plotted between the H2O2 concentrations and the reduction currents obtained. The regression equation for the calibration curve is obtained as I = −0.08 × c − 2.02 showing good linearity with an R 2 value of 0.99. The limit of detection of the sensor was calculated to be 70 nM (S/N = 3), and it shows a sensitivity of 1.16 µAµM −1 cm −2 (slope/area). The analytical performance of Co3S4/GCNNS is compared to various non-enzymatic sensors in Table S2. Shu et al. developed a TiO2/SiO2 composite as a phosphorescence sensor for H2O2 [47]. Although it detects H2O2 over a wide range of analyte concentrations, the LOD is as low as 0.16 µM. The graphene-CdS electroluminescence sensor detects H2O2 in a range from 5 µM to 1 mM but has a low LOD of 1.7 µM [48]. Gan et al. explored MoS2 quantum dots as a fluorescence sensor and determined H2O2 in a narrow range of 2-20 µΜ [49]. Calorimetric detection of H2O2 has been demonstrated by porphyrin iron-grafted mesoporous silica composites, but the detection limit is low as 67 µΜ [50]. Ding et al. presented an optical sensor for H2O2 in the 1 µΜ to 10 mM range by growing Pt nanoparticles inside the pores of fibrous silica particles and with a low LOD of 15 µΜ [51]. Co3S4/GCNNS has a very high LOD and sensitivity than these sensors. Further, the performance of Co3S4/GCNNS is compared with related electrochemical sensors (Table S2). Co3O4 hollow-sphere-based H2O2 sensor has low LOD and sensitivity compared to Co3S4/GCNNS [52]. Au/Cu bimetallic nanoparticles reported by Gowthaman et al. are expensive and have low LOD and sensitivity [53]. Chen et al. presented Cu2S@Co3S4 for the excellent electrocatalytic reduction of H2O2 using CV but did not explore the sensor aspects in detail [24]. GCN hollow spheres operate in a short range and have low sensitivity and LOD [27]. Liu et al. synthesized ZnO/GCNNS for H2O2 oxidation, but the sensor has a low LOD [31]. Atacan et al. prepared CuO/GCN nanoflakes. They demonstrated the sensing of H2O2 using DPV with high sensitivity, but the operation is limited to a short range, and the LOD is in the  [51]. Co 3 S 4 /GCNNS has a very high LOD and sensitivity than these sensors. Further, the performance of Co 3 S 4 /GCNNS is compared with related electrochemical sensors (Table S2). Co 3 O 4 hollow-sphere-based H 2 O 2 sensor has low LOD and sensitivity compared to Co 3 S 4 /GCNNS [52]. Au/Cu bimetallic nanoparticles reported by Gowthaman et al. are expensive and have low LOD and sensitivity [53]. Chen et al. presented Cu 2 S@Co 3 S 4 for the excellent electrocatalytic reduction of H 2 O 2 using CV but did not explore the sensor aspects in detail [24]. GCN hollow spheres operate in a short range and have low sensitivity and LOD [27]. Liu

Stability of Co3S4/GCNNS Modified GCE Sensor
Cyclic voltammetric responses of Co3S4/GCNNS-modified GCE towards 1 mM of H2O2 in 0.2 M PBS (pH = 7.2) were obtained for many 1, 2, 3, 4, 5, 6, and 10 days to confirm the long-term stability and repeatability of the proposed sensor, and the recorded cyclic voltammograms are shown in Figure 12a. The H2O2 reduction current showed no noticeable change, and the reduction potential remained unaltered, indicating the good stability of the fabricated sensor toward H2O2 reduction. Furthermore, the amperometric i-t curve of the Co3S4/GCNNS electrode was recorded for 4000 s at an applied potential of −0.7 V in 0.2 M PBS (pH 7.2) and is shown in Figure 12b. The current response was constant throughout the experiment, showing good amperometric stability of the Co3S4/GCNNS electrode in PBS. One of the most common problems when using sensors to determine H 2 O 2 in real samples is their response to the interfering species. To investigate the selectivity of Co 3 S 4 /GCNNS towards H 2 O 2 detection, the influence of other common components in blood serum was examined by amperometry at −0.7 V in 0.2 M PBS of pH 7.2. It is reported that the H 2 O 2 concentration present in human blood serum is generally less than 10 µM. In contrast, creatinine and xanthine are present in blood serum at concentrations below 119.3 µM and 2 mM, respectively. For non-diabetics, the normal glucose level is between 3.9 and 7.1 mM. The uric acid concentration ranges from 208 to 428 µM, and the urea concentration ranges between 2.5 and 7.5 mM [54][55][56][57]. On average, these species are present at less than or around 500 times the serum H 2 O 2 concentration. Therefore, 500-fold concentrations of these potential interferences are chosen to study the selectivity of Co 3 S 4 /GCNNS towards the electroreduction of 1 µM H 2 O 2 . As shown in Figure 11c, after attaining a stable response in PBS, 1 µM H 2 O 2 was added to the PBS, and the corresponding increase in the reduction current indicated the proper functioning of the Co 3 S 4 /GCNNS electrode. Then, 1 µM H 2 O 2 was added in another step to obtain the same increase in reduction current as in the previous step. Further, every 60 s, 500 µM of urea, glucose, creatinine, uric acid, and xanthine were added, respectively, to the region marked as a-e ( Figure 11c). As shown, no appreciable increase in the current was observed upon their addition, confirming that they show no interference at the Co 3 S 4 /GCNNS electrode. The amperometric current at the modified electrode increases for two consecutive additions of 1 µM of H 2 O 2 , indicating that the previously added species has no effect on the reduction of H 2 O 2 at Co 3 S 4 /GCNNS. Furthermore, the influence of the common cations present in the blood serum was studied by adding 500 µM of salts of each Na + , K + , Ca 2+ , Mg 2+, and Fe 2+ , respectively (f-j); nevertheless, no change in current was observed, still a spike of 1 µM of H 2 O 2 increases the current response indicating the selectivity of Co 3 S 4 /GCNNS electrode towards H 2 O 2 than to these cations. Further 500 µM concentrations of anions, Cl − , CO 3 2− , CH 3 COO − , and C 2 O 4 2− were each added at 60 s intervals, as shown in the region; k-n, and only a negligible change in current was observed, whereas two successive additions of H 2 O 2 resulted in the increase of current due to the reduction of H 2 O 2 . The results confirm that 500-fold additions of these entities do not alter the current response due to H 2 O 2 reduction, proving the high selectivity of the Co 3 S 4 /GCNNS electrode as an H 2 O 2 sensor.

Stability of Co 3 S 4 /GCNNS Modified GCE Sensor
Cyclic voltammetric responses of Co 3 S 4 /GCNNS-modified GCE towards 1 mM of H 2 O 2 in 0.2 M PBS (pH = 7.2) were obtained for many 1, 2, 3, 4, 5, 6, and 10 days to confirm the long-term stability and repeatability of the proposed sensor, and the recorded cyclic voltammograms are shown in Figure 12a. The H 2 O 2 reduction current showed no noticeable change, and the reduction potential remained unaltered, indicating the good stability of the fabricated sensor toward H 2 O 2 reduction. Furthermore, the amperometric i-t curve of the Co 3 S 4 /GCNNS electrode was recorded for 4000 s at an applied potential of −0.7 V in 0.2 M PBS (pH 7.2) and is shown in Figure 12b. The current response was constant throughout the experiment, showing good amperometric stability of the Co 3 S 4 /GCNNS electrode in PBS. 2) were obtained for many 1, 2, 3, 4, 5, 6, and 10 days to confirm the long-term stability and repeatability of the proposed sensor, and the recorded cyclic voltammograms are shown in Figure 12a. The H2O2 reduction current showed no noticeable change, and the reduction potential remained unaltered, indicating the good stability of the fabricated sensor toward H2O2 reduction. Furthermore, the amperometric i-t curve of the Co3S4/GCNNS electrode was recorded for 4000 s at an applied potential of −0.7 V in 0.2 M PBS (pH 7.2) and is shown in Figure 12b. The current response was constant throughout the experiment, showing good amperometric stability of the Co3S4/GCNNS electrode in PBS.

Real Sample Analysis
The determination of H 2 O 2 in human blood serum was performed at the Co 3 S 4 /GCNNS electrode using DPV in 0.2 M PBS of pH 7.2 to check the practical applicability of the proposed sensor, which is shown in Figure 13 (curve a). The reduction peak for H 2 O 2 in the serum sample appears around −0.52 V. The H 2 O 2 concentration in the serum was found to be 3.2 µM, which is within the normal range. An additional 50 µM of H 2 O 2 was added to the serum, and the reduction peak current increased without affecting the reduction potential, as shown in Figure 13 (curve b). A recovery of 98.5 % was achieved. This indicates that the proposed sensor can be employed for monitoring H 2 O 2 in real samples.
for H2O2 in the serum sample appears around −0.52 V. The H2O2 concentration in the serum was found to be 3.2 µM, which is within the normal range. An additional 50 µM of H2O2 was added to the serum, and the reduction peak current increased without affecting the reduction potential, as shown in Figure 13 (curve b). A recovery of 98.5 % was achieved. This indicates that the proposed sensor can be employed for monitoring H2O2 in real samples.

Conclusions
Co3S4/GCNNS were synthesized as a promising material for the electrocatalytic reduction of H2O2 by a simple and low-cost method. Physical and chemical characterizations helped us to confirm the successful formation of the materials. Nanosheets of GCNNS and micro flowers of Co3S4 were observed by SEM. From CV, the Co3S4/GCNNS composite was observed to reduce H2O2 and provide 4.7 and 7.1 times higher reduction current compared to pure Co3S4 and GCNNS, respectively, showing excellent electrocatalytic activity. The hybridization between Co3S4 and GCNNS enhanced the electroactive surface area and conductivity of the proposed sensor, which are crucial for superior electrocatalytic activity. At the same time, as evident from XPS studies, the electronic interactions between GCNNS and Co3S4 enhance the catalytically active redox centers on the Co3S4/GCNNS surface for H2O2 reduction. Co3S4/GCNNS was further explored for the amperometric sensing of H2O2 to show high performance over a wide range from 10 nM to 1.5 mM with a high detection limit of 70 nM. The sensor showed a fast response and excellent selectivity against potential interferences, and the practicality of the sensor was evaluated by the determination of H2O2 in the human serum. The current work demonstrates the potential of Co3S4/GCNNS as an ideal material for constructing high-performance H2O2 sensors.

Conclusions
Co 3 S 4 /GCNNS were synthesized as a promising material for the electrocatalytic reduction of H 2 O 2 by a simple and low-cost method. Physical and chemical characterizations helped us to confirm the successful formation of the materials. Nanosheets of GCNNS and micro flowers of Co 3 S 4 were observed by SEM. From CV, the Co 3 S 4 /GCNNS composite was observed to reduce H 2 O 2 and provide 4.7 and 7.1 times higher reduction current compared to pure Co 3 S 4 and GCNNS, respectively, showing excellent electrocatalytic activity. The hybridization between Co 3 S 4 and GCNNS enhanced the electroactive surface area and conductivity of the proposed sensor, which are crucial for superior electrocatalytic activity. At the same time, as evident from XPS studies, the electronic interactions between GCNNS and Co 3 S 4 enhance the catalytically active redox centers on the Co 3 S 4 /GCNNS surface for H 2 O 2 reduction. Co 3 S 4 /GCNNS was further explored for the amperometric sensing of H 2 O 2 to show high performance over a wide range from 10 nM to 1.5 mM with a high detection limit of 70 nM. The sensor showed a fast response and excellent selectivity against potential interferences, and the practicality of the sensor was evaluated by the determination of H 2 O 2 in the human serum. The current work demonstrates the potential of Co 3 S 4 /GCNNS as an ideal material for constructing high-performance H 2 O 2 sensors.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/bios13010108/s1, Details on materials, characterization, and electrochemical studies are provided. Figure S1: XPS spectra of GCNNS, (a) survey scan, high-resolution spectra of (b) C 1s and (c) N1s. XPS spectra of Co 3 S 4 , (d) survey scan, high-resolution spectra of (e) Co 2p and (f) S 2p., Figure S2: Comparison of PL spectra of Co 3 S 4 , GCNNS, and Co 3 S 4 /GCNNS measured at an excitation wavelength of 330 nm., Figure S3: (a) CVs obtained for the reduction of 1 mM H 2 O 2 at 1 mg/mL GCNNS containing various amounts of Co 3 S 4 ; 0.1, 0.15, 0.2, and 0.25 mg/mL in 0.2 M PBS pH 7.2, (b) CVs obtained for the reduction of 1 mM H 2 O 2 at 0.2 mg/mL Co 3 S 4 containing various amount of GCNNS of 0.5, 1 and 1.5 mg/mL. Table S1: Comparison of the binding energy of the core levels in Co 3 S 4 , GCN, and Co 3 S 4 /GCNNS.  Institutional Review Board Statement: Normal human serum used in this study was purchased from Lakshmi clinic Chinnalappeti, Tamilnadu, India, and is a biological product. Thus, it is not applicable to an ethics statement.

Informed Consent Statement: Not applicable.
Data Availability Statement: Data can be made available upon request.