Modification of NiSe₂ Nanoparticles by ZIF-8-Derived NC for Boosting H₂O₂ Production from Electrochemical Oxygen Reduction in Acidic Media

Abstract

The two-electron oxygen reduction reaction (2e⁻ ORR) has emerged as an attractive alternative for H₂O₂ production. Developing efficient earth-abundant transition metal electrocatalysts and reaction mechanism exploration for H₂O₂ production are important but remain challenging. Herein, a nitrogen-doped carbon-coated NiSe₂ (NiSe₂@NC) electrocatalyst was prepared by successive annealing treatment. Benefiting from the synergistic effect between the NiSe₂ nanoparticles and NC, the 2e⁻ ORR activity, selectivity, and stability of NiSe₂@NC in 0.1 M HClO₄ was greatly enhanced, with the yield of H₂O₂ being 4.4 times that of the bare NiSe₂ nanoparticles. The in situ Raman spectra and density functional theory (DFT) calculation revealed that the presence of NC was beneficial for regulating the electronic state of NiSe₂ and optimizing the adsorption free energy of *OOH, which could enhance the adsorption of O₂, stabilize the O-O bond, and boost the production of H₂O₂. This work provides an effective strategy to improve the performance of the transition metal chalcogenide for 2e⁻ ORR to H₂O₂.

1. Introduction

Hydrogen peroxide (H₂O₂) has been a high-value-added chemical with wide applications [1,2,3]. The industrial production of H₂O₂ uses an anthraquinone process, which has many disadvantages such as high energy consumption and pollution [4,5,6]. In contrast, the preparation of H₂O₂ using electrocatalytic oxygen reduction can be carried out at room temperature and pressure. The environment-friendly electrosynthesis of H₂O₂ through 2e⁻ ORR is regarded as a promising alternative [7,8]. As is generally known, ORR is an electrocatalytic reaction proceeding with a four-electron (4e⁻) pathway or two-electron (2e⁻) pathway when applying different electrocatalysts. The 4e⁻ ORR reaction has been intensively studied and applied in battery conversion systems such as fuel cells or metal–air batteries [9,10,11], while the development of efficient electrocatalysts for the selective 2e⁻ ORR pathway toward H₂O₂ remains challenging at present [1,9].

To date, various catalysts have been developed for electrocatalytic oxygen reduction to produce H₂O₂, but most of them showed a significantly improved 2e⁻ ORR performance only under alkaline media [12,13,14,15]. However, from the perspective of practical application, H₂O₂ is prone to decomposition in the alkaline environment for a long time, which greatly limits its application in actual industry and daily life [16,17,18]. On the contrary, acidic media can not only effectively stabilize H₂O₂, but it can also be more suitable for application scenarios such as electro-Fenton, showing more attractive advantages [7,19,20,21]. Therefore, researchers tend to develop high-performance electrocatalysts under acidic media. Precious metal-based catalysts have been approved to show the best 2e⁻ ORR performance under acidic media, but are limited by their high cost and scarcity [22,23,24]. Therefore, transition metal chalcogenides have attracted more attention because of their advantages such as their low cost, abundant reserves, high stability, and easy adjustment of electronic structure [25,26,27].

In particular, transition metal selenides (TMSs) have been regarded as a promising 2e⁻ ORR catalyst under acidic media [28,29,30]. NiSe₂ is a typical TMS catalyst, but most reports focus on the field of water splitting or energy storage in previous studies [31,32]. Recently, NiSe₂ has also been confirmed to be effective for 2e⁻ ORR reaction [33]. Due to the large atomic radius, high electronegativity of the Se element, and metal-like property, NiSe₂ can effectively weaken the interactions of the outer metal electrons and then optimize the activation of oxygen-containing intermediates to promote the generation of H₂O₂. However, NiSe₂ often exhibits poor electrical conductivity and tends to aggregate into larger particles, which could affect the electron transfer rate [34,35]. It is well-known that the carbon materials have a high electrical conductivity and stability; in the meantime, they could stabilize the nanoparticles as the supports or coating agents [36,37]. Accordingly, compositing NiSe₂ with highly conductive carbon materials could make use of both advantages to facilitate the formation of H₂O₂ from 2e⁻ ORR in acidic media. In addition, metal–organic frameworks (MOFs) have the advantages of large porosity, large surface area, and tunable morphology and structure, and they are often used as an effective precursor for preparing efficient electrocatalysts [38]. Therefore, using MOFs precursors could provide a good platform for the effective combination of NiSe₂ with a carbon material.

Based on the above considerations, in order to enhance the 2e⁻ ORR performance of NiSe₂, a nitrogen-doped carbon-coated NiSe₂ (NiSe₂@NC) electrocatalyst was fabricated using ZIF-8 as the N-doped carbon precursor. The results showed that the introduction of NC could promote NiSe₂ to form small-particle-size nanoparticles. In comparing with the bare NiSe₂, NiSe₂@NC could achieve a significantly improved H₂O₂ Faraday efficiency of 73% to 81%, H₂O₂ yield of 12 ppm to 53 ppm in 2 h if electrolysis, and long-term stability of 10 h in an H-type cell. In situ Raman spectroscopy and DFT calculation revealed that NiSe₂@NC was beneficial for adjusting the electronic state of NiSe₂ and optimizing the adsorption free energy of *OOH, which was conducive to promoting 2e⁻ ORR to produce H₂O₂. This work hopefully provides a valuable reference for the development of more efficient transition metal electrocatalysts toward 2e⁻ ORR.

2. Results and Discussion

2.1. Material Characterizations

The synthetic procedure of NiSe₂@NC is displayed in Figure 1, which is prepared by the impregnation of Ni²⁺ with ZIF-8 and the successive annealing treatment, respectively. Firstly, Ni²⁺ ions were introduced into the self-assembled ZIF-8 to form Ni-ZIF-8. Then, the annealing treatment at 1000 °C under an Ar atmosphere was applied to form Ni-N-C. The subsequent pyrolysis process with selenium powder at 500 °C was performed to obtain the NiSe₂@NC catalyst.

The morphologies of the catalysts were characterized by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Figure S1a, the ZIF-8 samples exhibit a rhombic dodecahedral morphology with the diameter of 200~250 nm. Figure 2a displays the N-doped carbon (NC) samples with a pristine morphology by the annealing treatment of ZIF-8 at 1000 °C under an Ar atmosphere, which have a smaller size (100~200 nm) than ZIF-8. And there is no formation of metal nanoparticles inside (Figure 2d), implying that the Zn NPs formed by annealing are completely removed under high-temperature calcination. The as-prepared NiSe₂@NC samples (Figure 2b,e) show a disordered structure and a rough surface containing small nanoparticles (10~20 nm), indicating the successful composite of metal and NC. In contrast, pure NiSe₂ particles (Figure 2c,f) show much larger aggregated particles in the absence of carbon carriers. Furthermore, the high-resolution TEM image (HRTEM) in Figure 2g shows that the lattice fringe of the nanoparticles is 0.266 nm, which corresponds to the (210) plane of NiSe₂ [33], again validating the successful combination of NiSe₂ and NC. Moreover, the elemental mapping images of NiSe₂@NC shown in Figure 2h–k further indicate the presence of C, Ni, Se, and N elements and uniform distribution. These results imply that N-doped carbon-coated NiSe₂ catalysts were prepared successfully. Meanwhile, the NiSe₂ nanoparticles with a small particle size in NiSe₂@NC could be beneficial to expose more reactive active sites and thus obtain a higher ORR activity.

The composition and the crystal structure of the samples were characterized by X-ray diffraction (XRD). The characteristic diffraction peaks of ZIF-8 (Figure S1b) are consistent with previous reports [39]. As shown in Figure 3a, NC has wide diffraction peaks at 24.8° and 43.6°, which can be indexed to the (002) and (100) planes of carbon, respectively. Both NiSe₂@NC and NiSe₂ show peaks at 29.8°, 33.5°, 36.8°, 42.8°, 50.6°, 55.4°, 57.8°, 62.2°, and 72.6°, corresponding to the (200), (210), (211), (220), (311), (230), (321), (400), and (421) planes of NiSe₂ (PDF#65-1843) [40], respectively, which further confirms the successful synthesis of the NiSe₂@NC catalyst. Combined with the Scherrer equation, the average crystal size of NiSe₂ nanoparticles in NiSe₂@NC could be calculated to be 20.2 nm, which is consistent with the TEM characterization result. In addition, the carbon properties were characterized by Raman spectroscopy as shown in Figure 3b. Both NC and NiSe₂@NC have two characteristic peaks at 1350 cm⁻¹ and 1590 cm⁻¹, corresponding to the D and G bands of carbon, respectively. The D band represents disordered or defective carbon and the G band represents graphitized carbon [41]. The I_D/I_G values of NC and NiSe₂@NC are 0.86 and 0.90, respectively, indicating the presence of defects in both catalysts, which have been reported to favor ORR activity [42].

For comparison, the other two NiSe₂@NC samples (NiSe₂@NC-L and NiSe₂@NC-H) with lower and higher Ni addition were prepared under the same condition, respectively. Their morphologies and compositions were also observed. As shown in Figure S2a,b, the morphology of NiSe₂@NC-H is similar to that of NiSe₂@NC, while a large number of tubular structures exist in NiSe₂@NC-L. The XRD pattern of NiSe₂@NC-L (Figure S2c) shows a series of characteristic diffraction peaks of NiSe₂ (PDF#65-1843) and a sharp diffraction peak at 26°, which is attributed to the (002) plane of graphitized carbon, corresponding to the formed carbon nanotubes [43]. In the case of increasing the Ni²⁺ addition, the diffraction peaks of NiSe₂ disappear in NiSe₂@NC-H, and new characteristic peaks appear at 33.0°, 44.6°, 50.5°, 60.2°, 61.6°, and 69.6°, which can be well indexed to Ni_0.85Se (PDF#18-0888) [44]. These results indicate that the amount of Ni precursor addition could affect the formation of nickel selenide, and the NiSe₂ phase could gradually transform into the Ni_0.85Se phase with the increase in Ni addition.

The composition and valence state of NiSe₂@NC were further determined by X-ray photoelectron spectroscopy (XPS). The signal peaks of C 1s, N 1s, Ni 2p, and Se 3d can be observed in Figure S3, confirming the existence of the four elements. The C 1s spectrum (Figure 4a) can be divided into two peaks ascribed to C-C (284.8 eV) and C-N (285.7 eV) [45]. The N 1s spectrum can be divided into three peaks at 398.2 eV, 400.0 eV, and 401.8 eV, corresponding to pyridinic N, pyrrolic N, and graphitic N, respectively (Figure 4b) [46]. The Ni 2p spectrum in Figure 4c shows the peak signals of Ni 2p_3/2 and Ni 2p_1/2. The fitting peaks at the binding energy of 853.4 eV and 870.9 eV are related to Ni²⁺, and the peaks at 855.6 eV and 873.8 eV are ascribed to Ni³⁺ (oxidation state). And the other two peaks at 860.1 eV and 879.1 eV can be assigned to the satellite peaks [47]. The Se 3d spectrum in Figure 4d has two peaks at 54.8 eV and 55.6 eV, which correspond to Se 3d_5/2 and Se 3d_3/2, indicating the existence form of Se₂²⁻ [48]. And the peak at 59.0 eV can be assigned to oxidized Se (SeO_x) [37,48].

2.2. Electrocatalytic Performance

The 2e⁻ ORR performance of NC, NiSe₂@NC, and NiSe₂ were evaluated by the RRDE system in O₂-saturated 0.1 M HClO₄. According to the polarization curves shown in Figure 5a, NC has the most positive onset potential (0.57 V vs. RHE), followed by NiSe₂@NC (0.55 V vs. RHE) and NiSe₂ (0.54 V vs. RHE). The disk current densities also present the same changing trend, which means NC has the highest ORR reaction kinetics among the three catalysts. Obviously, NiSe₂@NC also shows a better ORR activity than bare NiSe₂ due to the combination with N-doped carbon. Figure 5b shows the ring current densities of the three catalysts at different potentials. Compared with the other two catalysts, NiSe₂@NC has the highest ring current density, which is about 1.2 and 1.8 times that of NC and NiSe₂ at 0 V vs. RHE, respectively. This trend means that more O₂ is converted to H₂O₂, which can be confirmed by the H₂O₂ selectivity and electron transfer number in Figure 5c,d. NiSe₂@NC has the highest H₂O₂ selectivity of 77% and the lowest electron transfer number at 0.3 V vs. RHE. The selectivity of H₂O₂ of NiSe₂ is 68% at the same potential. It can also be seen that the selectivity of NC is the lowest, only 38~59%. These results indicate that the combination of NiSe₂ and NC is more beneficial to the 2e⁻ pathway than that of the pure component. Among the catalysts with different degrees of selenization, NiSe₂@NC-H shows the lowest ORR activity (Figure S4), which may be related to the existence form of Ni_0.85Se that is not conducive to the reaction under acidic media. Moreover, NiSe₂@NC also exhibits the best 2e⁻ ORR activity and selectivity.

In addition, the ORR performance of NiSe₂@NC under alkaline media (0.1 M KOH) was also evaluated, as shown in Figure S5. NiSe₂@NC exhibits an excellent ORR activity with an initial potential of 0.79 V vs. RHE and a disk current density of 2.8 mA cm⁻² at 0 V vs. RHE. However, the ring current density is drastically reduced with only ~35% H₂O₂ selectivity, which is much lower than the performance tested under acidic media. The results suggest that NiSe₂@NC is more favorable for electrocatalytic oxygen reduction for H₂O₂ production under acidic media.

To reveal the effect of the combination of NiSe₂ and NC on the improved catalytic performance, other electrochemical properties were further tested. Figure 6a shows the Nyquist plots and fitted charge transfer resistance of the three catalysts to reveal the electronic conductivity [49,50]. The metal-free NC has the smallest semicircular diameter, corresponding to the highest disk current in the LSV curves. In addition, compared with the bare NiSe₂, NiSe₂@NC also exhibits a smaller semicircular diameter and lower charge transfer resistance, implying a higher electronic conductivity than NiSe₂. The cyclic voltammetry curves of the double-layer region (Figure S6) were also performed at different scan rates, which could be used to measure the electrochemical active surface area (ECSA). As seen in Figure 6b, the double-layer capacitance (C_dl) of NC is 2.9 mF cm⁻². It is followed by NiSe₂@NC with the C_dl value of 0.86 mF cm⁻², which is 2.7 times higher than that of NiSe₂ (0.32 mF cm⁻²), revealing a higher active surface area [15]. These results indicate that the combination with NC can not only improve the conductivity of NiSe₂, but also expose more active sites.

The Faraday efficiency (FE) and yield of H₂O₂ are also important indicators to evaluate the practical application of catalysts [51]. Therefore, the actual H₂O₂ production capacity of NiSe₂@NC was evaluated by electrolysis of 7200 s in an H-type cell, combined with UV–Vis spectra of Ce⁴⁺ at 319 nm. The electrolytic tests were performed at different potentials to determine the optimal potential of NiSe₂@NC. According to Figure S7a,b, the chronoamperometry curves show significantly increased current densities with the negative shift of potential. And the absorbance of Ce⁴⁺ also decreases correspondingly, indicating that more H₂O₂ is produced. This could be confirmed from the calculated Faraday efficiency (FE) and H₂O₂ concentration of NiSe₂@NC in Figure 7a. The highest Faraday efficiency of 81% can be achieved at 0.3 V vs. RHE. Meanwhile, the H₂O₂ concentration increases continuously as the potential becomes negative, with the maximum of 68 ppm at 0.2 V vs. RHE. In general, NiSe₂@NC has the best catalytic performance at 0.3 V vs. RHE. Subsequently, the electrochemical performance of H₂O₂ production from NC, NiSe₂@NC, and NiSe₂ was compared at 0.3 V vs. RHE, as shown in Figure 7b. Note that the NiSe₂@NC has a higher Faraday efficiency (81%) and H₂O₂ concentration (53 ppm) than that of NC (59%, 38 ppm) and NiSe₂ (73%, 12 ppm), which is consistent with the RRDE trend. It can be seen that NiSe₂@NC has the best H₂O₂ production capacity, which also proves that the combination of NiSe₂ and NC plays a great role in enhancing the performance of 2e⁻ ORR.

The stability is another important factor to evaluate the performance of catalysts in practical applications. Therefore, the 10 h electrolysis experiments were performed at 0.4 V vs. RHE to evaluate the stability of NiSe₂@NC and NiSe₂ (Figure 7c). Obviously, the chronoamperometry curve of bare NiSe₂ exhibits a much higher loss in current density than NiSe₂@NC. According to the XRD patterns in Figure 7d and the wide range of patterns in Figure S8, the crystal structure of NiSe₂@NC could not be destroyed after the long-term electrolysis. Combined with the above analysis results, the introduction of NC can significantly improve the stability of the catalytic system in 0.1 M HClO₄, which is more conducive to the production of H₂O₂. Table S1 compares the 2e⁻ ORR performance of other reported transition metal chalcogenides. It can be seen that NiSe₂@NC is a promising 2e⁻ ORR electrocatalyst with a high selectivity and H₂O₂ production capacity under acidic media.

2.3. Electrocatalytic Mechanism on NiSe₂@NC

To further investigate the catalytic mechanism of NiSe₂@NC during the ORR process, in situ Raman measurement was used to detect the intermediates produced during the reaction in O₂-saturated 0.1 M HClO₄. The Raman spectra of different potentials were collected at a 638 nm excitation wavelength. As shown in Figure 8a, the signal peak at 933 cm⁻¹ can be associated to the symmetric stretching mode of ClO₄⁻ [52]. Another peak at 463 cm⁻¹ also presents at all potentials, which may be related to the adsorption of oxygen-containing species on the surface of the catalyst, resulting in the formation of a Ni-O bond [53]. Furthermore, only the peaks of the ClO₄⁻ and Ni-O bond could be identified at the open circuit potential (OCP). The new peak signal located at 733 cm⁻¹ appears when the potential is shifted negatively to 0.4 V vs. RHE, which is assigned to the vibration of the key intermediate *OOH [54]. And this peak intensity increases with the negative shift of potential, implying the accumulation of *OOH. These results indicate that NiSe₂@NC is an effective 2e⁻ ORR catalyst that can stabilize *OOH on the surface without further O-O bond breaking.

Based on the above in situ Raman results, density functional theoretical (DFT) calculations were carried out to further illustrate how the introduction of NC enhanced the 2e⁻ ORR activity and selectivity. The optimized simulation models of *OOH absorbed on NiSe₂ and NiSe₂@NC are exhibited in Figures S9 and S10. Undoubtedly, the pathway of the oxygen reduction reaction is mainly determined by the *OOH adsorption strength on the surface of the catalyst, which is also closely related to the electronic state of the active site [33]. Therefore, Bader charge analysis was performed to reveal the different electronic states of NiSe₂ and NiSe₂@NC. As shown in Figure 8b, the active sites of NiSe₂ and NiSe₂@NC obtain 0.30 e and 0.35 e, respectively. Therefore, the active site of NiSe₂@NC has more electrons and carries more negative charge than NiSe₂, which is conducive to the adsorption of O₂ and H⁺ during the 2e⁻ ORR reaction pathway. These results verify that the introduction of NC could effectively promote the adsorption and activation of oxygen on the active site to further produce *OOH intermediates.

The ideal 2e⁻ ORR reaction requires an optimum balance between the adsorption and desorption of *OOH, with neither too strong nor too weak an adsorption strength. Then, the *OOH adsorption free energies (ΔG_*OOH) of the two catalysts were calculated at the equilibrium potential (U = 0.7 V) and standard condition (U = 0 V) to evaluate the adsorption strength of *OOH. As shown in Figure 8c,d, the *OOH adsorption energies of NiSe₂@NC at U = 0.7 V and 0 V are 2.70 eV and 4.10 eV, respectively, both of which are closer to the ideal values (3.52 eV and 4.22 eV) than NiSe₂ (2.46 eV and 3.86 eV) [55,56]. From the above analysis, the conversion from *OOH to H₂O₂ production on the surface of the NiSe₂@NC catalyst is more favorable than for NiSe₂. In addition, the differential charge density distributions between adsorbed *OOH and substrates were also simulated and are displayed in Figure S11. In comparing the two catalysts, the adsorption site of NiSe₂ has more positive charges than that of NiSe₂@NC, indicating that NiSe₂ exhibits stronger charge interactions with *OOH, which may contribute to the breaking of the O-O bond. On the contrary, the *OOH adsorption on NiSe₂@NC is moderate, which is conducive to preserving the O-O bond and promoting the next hydrogenation to H₂O₂.

In conclusion, the above DFT results indicate that the introduction of NC could not only change the electronic state of NiSe₂ to promote the adsorption and activation of oxygen, but also optimize the adsorption free energy of *OOH to preserve the O-O bond and convert it to H₂O₂ during the oxygen reduction reaction. Therefore, NiSe₂@NC shows a significantly increased 2e⁻ ORR activity and selectivity.

3. Materials and Methods

3.1. Reagent and Chemicals

All reagents were of analytical grade and used without further purification. Zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O), nickel chloride hexahydrate (NiCl₂·6H₂O), and potassium hydroxide (KOH) were purchased from Aladdin Bio-technology Co., Ltd., Shanghai, China. Selenium powder (Se), methanol (CH₃OH), and anhydrous ethanol (C₂H₅OH) were obtained from Yuanli Chemical Co., Ltd., Tianjin, China. The compound 2-methylimidazole (2-melm) was purchased from Bailingwei Technology Co., Ltd., Beijing, China. The Nafion solution was purchased from Yingke United Co., Ltd., Tianjin, China.

3.2. Material Synthesis

3.2.1. Synthesis of ZIF-8

The compound ZIF-8 was synthesized based on a previous reported method with a slight adjustment [39]. Typically, 1.487 g of Zn(NO₃)₂·6H₂O and 1.644 g of 2-methylimidazole were dissolved in 50 mL of methanol solution, respectively. After mixing the two solutions, the mixture was stirred rapidly for 30 min and left at room temperature for 24 h. Finally, the white product was centrifuged, washed with ethanol, and fully dried in a vacuum drying oven at 60 °C.

3.2.2. Synthesis of NiSe₂@NC

A total of 200 mg of ZIF-8 was ultrasonically dispersed in 60 mL of ethanol. Then, 100 mg of NiCl₂·6H₂O was added to the above solution and stirred at room temperature for 12 h. The precipitate was centrifuged and dried to obtain Ni-ZIF-8. Subsequently, the Ni-ZIF-8 was transferred to a tube furnace and annealed at 1000 °C for 3 h with a heating rate of 5 °C min⁻¹ under an Ar atmosphere to form Ni-N-C. Finally, a 2:1 mass ratio of Se powder and Ni-N-C were loaded into two porcelain boats. One of the porcelain boats containing Se powder was placed upstream of the tube furnace and annealed at 500 °C for 2 h with a heating rate of 2 °C min⁻¹ under an Ar atmosphere. The samples after pyrolysis were recorded as NiSe₂@NC. In addition, the content of NiCl₂·6H₂O was varied during catalyst preparation by adding 50 mg and 150 mg of NiCl₂·6H₂O, respectively. Other steps were carried out under the same conditions. The final samples were respectively labeled as NiSe₂@NC-L and NiSe₂@NC-H, respectively.

3.2.3. Synthesis of NC and NiSe₂

The NC catalyst was obtained by the direct calcination of ZIF-8 under the same conditions. The NiSe₂ catalyst was synthesized by a one-step hydrothermal reaction. In total, 0.24 g of NiCl₂·6H₂O, 0.18 g of Se powder, and 3 g of KOH were added to 20 mL of deionized water and stirred for 30 min at room temperature. Subsequently, the mixture was transferred to an oven for 12 h at 150 °C. After the solution was cooled to room temperature, it was washed with deionized water and finally dried in a vacuum drying oven at 60 °C. The resulting samples were denoted as NiSe₂.

3.3. Materials Characterization

The morphology and structure of the catalyst were measured by scanning electron microscopy (SEM, S-4800, Hitachi, Tokyo, Japan) and transmission electron microscopy (TEM, JEM-2100F, Japan Electronics, Islamabad, Pakistan). The distribution of elements was determined by energy dispersive X-ray spectroscopy (EDX) using scanning electron microscopy. The crystal structure was measured using X-ray diffraction (XRD, D8 Focus, Bruck, Billerica, MA, USA) using a Cu-Kα radiation source at a scan rate of 5 °C min⁻¹. X-ray photoelectron spectroscopy (XPS, K-Alpha+, Thermo, Waltham, MA, USA) was performed to determine the composition and chemical state. Raman spectra (Horiba, Kyoto, Japan) were measured at 638 nm to determine the phase composition. The optical properties were detected by an ultraviolet–visible spectrophotometer (UV–Vis, T2600, Yoke, Shanghai, China). The crystallites’ size from the XRD pattern could be calculated by the Scherrer equation.

3.4. Electrochemical Measurements

All electrochemical tests in this paper were performed using a CHI 760E electrochemical workstation (Chenhua Instrument Co., Ltd., Shanghai, China) and 0.1 M HClO₄. The 2e⁻ ORR performance of catalysts was evaluated using a rotating ring disk electrode (RRDE, disk area: 0.247 cm², platinum ring area: 0.186 cm²) as the working electrode. Carbon rods were used as the counter electrodes and saturated calomel electrodes (SCE) as the reference electrodes to form the test system. The ink was prepared by mixing 2.5 mg of catalyst, 500 μL of ethanol, 500 μL of H₂O, and 20 μL of Nafion solution. After ultrasonic dispersion, 10 μL of ink was added to the RRDE surface with the catalyst loading of 0.1 mg cm⁻². Before the catalytic performance measurement, cyclic voltammetry (CV) was performed on catalyst-loaded RRDE under a N₂ atmosphere, scanning in the potential range from 0 to 0.8 V vs. a reversible hydrogen electrode (RHE) at 50 mV s⁻¹ for around 20 cycles. After the CV curves remained stable, the linear voltametric sweep (LSV) test was performed in O₂-saturated 0.1 M HClO₄. In the same potential range, LSV curves were measured at a scan rate of 5 mV s⁻¹ and a rotating speed of 1600 rpm. Moreover, the Pt ring voltage was kept at 1.3 V vs. RHE during the test. The selectivity of H₂O₂ and electron transfer number (n) were calculated according to the following formula:

$${\mathrm{H}}_2{\mathrm{O}}_2\left(\%\right)=200 \times {\mathrm{I}}_{\mathrm{r}}/\left[\right(\mathrm{N} \times {\mathrm{I}}_{\mathrm{d}})+{\mathrm{I}}_{\mathrm{r}}]$$

(1)

$$\mathrm{n}=4 \times {\mathrm{I}}_{\mathrm{d}}/({\mathrm{I}}_{\mathrm{d}}+{\mathrm{I}}_{\mathrm{r}}/\mathrm{N})$$

(2)

where I_d is the disk current, I_r is the ring current, and N is the collection efficiency of the Pt ring, which could be identified by the LSV curves at different speeds. As shown in Figure S12, the collection efficiency is 37%.

The Faraday efficiency (FE) and yield of H₂O₂ are determined by the chromogenic reaction of Ce(SO₄)₂ (2Ce⁴⁺ + H₂O₂ → 2Ce³⁺ + 2H⁺ + O₂). In the presence of H₂O₂, Ce(SO₄)₂ (yellow) can be converted to Ce³⁺ (colorless), which could reduce the absorbance of UV–Vis at 319 nm. Therefore, the H₂O₂ production capacity of the catalyst can be measured by the change in absorbance. The standard solution of Ce(SO₄)₂ with different concentrations was configured by dissolving Ce(SO₄)₂ in a 0.5 M H₂SO₄ solution. According to the UV–Vis peak absorbance and Ce⁴⁺ concentration, the standard curve is drawn in Figure S13 (y = 4.084x − 0.0312). The electrolytic process is carried out in an H-type cell, as shown in Figure S14. A 1 × 1 cm² carbon paper supported with the catalyst was used as the working electrode, saturated calomel electrodes as the reference electrodes, and the Pt sheet as the counter electrode. After two hours of electrolysis, 50 μL of electrolytic solution was added to 3 mL of 0.5 mM Ce(SO₄)₂ solution, and then the change in the peak absorbance of Ce⁴⁺ was recorded. The FE and yield of H₂O₂ can be calculated according to the following formula:

$$\mathrm{FE}\left(\%\right)=200 \times \mathrm{C} \times \mathrm{V} \times \mathrm{F}/\mathrm{Q}$$

(3)

$$\left[{\mathrm{H}}_2{\mathrm{O}}_2\right] \left(\mathrm{ppm}\right)=\frac{1}{2} \times \frac{{\mathrm{V}}_1 \times {\mathrm{C}}_{\mathrm{before}} - ({\mathrm{V}}_1+{\mathrm{V}}_2) \times {\mathrm{C}}_{\mathrm{after}}}{{\mathrm{V}}_2} \times 34.01$$

(4)

where C is the concentration of the H₂O₂ product (mol L⁻¹), V is the volume of the solution in the electrolytic cell (L), F is the Faraday constant (96,485 C mol⁻¹), V₁ is the volume of Ce⁴⁺ before the reaction (mL), C_before is the concentration of Ce⁴⁺ before the reaction (mmol L⁻¹), V₂ is the volume of the added electrolyte solution (mL), and C_after is the concentration of Ce⁴⁺ after the reaction (mmol L⁻¹).

And in this work, all of the measured SCE potentials were converted to the reversible hydrogen electrode (RHE) potentials according to the following formula:

$${\mathrm{E}}_{\mathrm{RHE}}={\mathrm{E}}_{\mathrm{SCE}}+0.059 \mathrm{pH}+0.233$$

(5)

3.5. Computational Method

All calculations in this work were performed using the VASP software package (5.4.4). The GGA-PBE function is used to describe the exchange correlation interaction, the PAW method is used to describe the direct interaction between nuclei and valence electrons, and the DFT-D3 method is used to modify the van der Waals interaction. The convergence criteria for energy and force are 10⁻⁴ eV and 0.05 eV/Å. The plane wave truncation energy is set to 400 eV, and a 3 × 3 × 1 K-point grid is used to optimize the geometric structure and calculate the frequency and electronic characteristics of the catalyst. The optimal NiSe₂ exposed crystal surface is (210), the catalyst is set to a thickness of 3 layers, and the NiSe₂ is p (2 × 1) period units. The graphene carrier is a 4 × 3 supercell containing 48 carbon atoms, and a 15 Å vacuum layer is added to reduce the interaction between adjacent layers. The free energy is calculated according to the following formula:

$$\mathrm{G}={\mathrm{E}}_{\mathrm{elect}}+{\mathrm{E}}_{\mathrm{ZPE}}-\mathrm{TS}+{\mathrm{G}}_{\mathrm{U}}$$

(6)

where E_elect is the electronic energy calculated by DFT, E_ZPE and S are the zero-point energy and entropy, T is 298.15 K, and G_U = −n_eU.

4. Conclusions

In summary, we present a nitrogen-doped carbon-coated NiSe₂ (NiSe₂@NC) electrocatalyst for boosting the H₂O₂ production from 2e⁻ ORR in acidic media. NiSe₂@NC exhibited the better 2e⁻ ORR activity and selectivity, with the yield of H₂O₂ being 4.4 times that of bare NiSe₂. In addition, the long-term stability of 10 h could be achieved in the electrolysis process. This significantly improved 2e⁻ ORR performance was mainly attributed to the synergistic effect of NiSe₂ and NC. The introduction of NC could reduce the charge transfer resistance and stabilize the NiSe₂ nanoparticles to expose a greater electrochemically active surface area. Moreover, in situ Raman tests and DFT calculations were further performed to reveal the role of the composition of NiSe₂ with NC. The introduction of NC could not only adjust the electronic state of NiSe₂ to carry more negative charges on the active site, but it could also optimize the adsorption free energy of *OOH, thus further promoting 2e⁻ ORR for the production of H₂O₂. This work provides an effective strategy to modify the transition metal selenides for selectively producing H₂O₂ by the 2e⁻ ORR reaction under acidic media.