Highly Efficient Electrocatalyst of 2D–2D gC₃N₄–MoS₂ Composites for Enhanced Overall Water Electrolysis

Abstract

For future clean and renewable energy technology, designing highly efficient and robust electrocatalysts is of great importance. Particularly, creating efficient bifunctional electrocatalysts capable of effectively catalyzing both hydrogen- and oxygen-evolution reactions (HERs and OERs) is vital for overall water electrolysis. In this study, we employ 2D molybdenum disulfide (MoS₂) nanosheets and pyrolytically fabricated 2D graphitic carbon nitride (gC₃N₄) nanosheets to create 2D gC₃N₄-decorated 2D MoS₂ (2D–2D gC₃N₄–MoS₂) nanocomposites using a facile sonochemical method. The 2D–2D gC₃N₄–MoS₂ nanocomposites show an interconnected and agglomerated structure of 2D gC₃N₄ nanosheets decorated on 2D MoS₂ nanosheets. For water electrolysis, the gC₃N₄–MoS₂ nanocomposites exhibit low overpotentials (OER: 225 mV, HER: 156 mV), small Tafel slope values (OER: 49 mV/dec, HER: 101 mV/dec), and excellent durability (up to 100 h for both OER and HER) at 10 mA/cm² in 1 M KOH. Furthermore, the gC₃N₄–MoS₂ nanocomposites show excellent overall water electrolysis performance with a low full-cell voltage (1.52 V at 10 mA/cm²) and outstanding long-term cell stability. The superb bifunctional activities of the gC₃N₄–MoS₂ nanocomposites are attributed to the synergistic effects of 2D gC₃N₄ (i.e., low charge-transfer resistance) and 2D MoS₂ (i.e., a large electrochemically active surface area). These findings suggest that the 2D–2D gC₃N₄–MoS₂ nanocomposites could serve as excellent bifunctional catalysts for overall water electrolysis.

1. Introduction

Fossil fuel scarcity, increasing energy demands, and environmental pollution concerns have stimulated significant interest in pursuing renewable, eco-friendly, and clean energy sources [1,2,3,4]. Recently, hydrogen has emerged as a leading alternative renewable energy resource because of its excellent mass–energy density, pollution-free characteristics, sustainability, renewability, and superior cleanliness [5,6,7,8]. Among all the available hydrogen production methods, electrocatalytic water electrolysis is considered a fascinating method for harvesting oxygen and hydrogen energies from water (via oxygen- and hydrogen-evolution reactions (OERs and HERs), respectively) because of its environmental friendliness, reusability, scalability, and feasibility [9,10,11,12]. Currently, Pt (HER catalyst) and IrO₂/RuO₂-based (OER catalyst) materials are widely believed to be highly efficient, stable, and reliable catalysts for water splitting, but their material scarcity and high cost hinder their practical applications [13,14]. Most of the reported electrocatalysts have not yet exhibited outstanding OER and HER activities simultaneously in a single electrolyte because of incompatible activity and stability [15,16,17,18]. Therefore, developing alternative catalysts with superior durability, excellent catalytic active sites, low cost, and earth-abundant resources is vital.

Recently, transition metal dichalcogenides (TMDs)-based materials have garnered considerable attention as prospective electrocatalysts for OERs and HERs because of their distinctive two-dimensional (2D) layered structure, low cost, and unique physicochemical characteristics [19,20,21,22]. Among the various TMDs, molybdenum disulfide (MoS₂) is considered a substantial catalyst for water splitting owing to its high intrinsic edge activity, layered structure, excellent stability, nontoxicity, natural abundance, low cost, unique electronic structures, and good electrochemical catalytic activity [23,24,25]. Generally, MoS₂ comprises two phases, i.e., 2H (semiconducting) and 1T (metallic) [26]. However, pure 2H phase MoS₂ demonstrates insufficient electrocatalytic performance due to its inferior electrical conductivity, insufficient active edge sites, sluggish charge-transfer kinetics, and high hydrogen adsorption–desorption energy [27,28]. Therefore, various methods have been used to hybridize 2H MoS₂ nanostructures with transition metal/metal oxide-based materials [29,30,31,32,33] and carbonaceous materials [34,35,36,37,38] in order to improve their electrochemical catalytic behavior. Recently, graphitic carbon nitride (gC₃N₄) has gained significant interest as a good co-catalyst for enhancing the catalytic activity of host catalysts owing to its impressive chemical stability, facile synthesis, affordability, high earth abundance, unique 2D layered structure, high nitrogen concentration, and easily adaptable structure [39,40]. The coupling of 2D MoS₂ and 2D gC₃N₄ can notably enhance the specific surface area, electrical conductivity, and density of active catalytic sites, all of which are beneficial for improving electrocatalytic performance [34,41]. For example, Fageria et al. [34] hydrothermally synthesized the highly efficient electrocatalyst of MoS₂-decorated gC₃N₄, demonstrating an HER overpotential of 240 mV at −10 mA/cm² in 0.5 M H₂SO₄. Additionally, Zhang et al. [35] synthesized MoS₂ nanosheets anchored on a gC₃N₄ substrate, exhibiting an HER overpotential of 158 mV at −10 mA/cm² in 0.5 M H₂SO₄. Recently, Liu et al. [42] used the hydrothermal method, followed by ultrasonic techniques, to fabricate a hybrid structure of the MoS₂/gC₃N₄ nanojunction, which exhibited effective HER activity with an overpotential of −200 mV to achieve a current density of 10 mA/cm² in 0.5 M H₂SO₄. He et al. [43] fabricated a CoO_x/mC@MoS₂@gC₃N₄ composite that showed the HER overpotential of 31 mV at −10 mA/cm² in 0.5 M H₂SO₄. Furthermore, Mehtab et al. [44] synthesized the MoS₂/gC₃N₄ heterostructure through a ball-milling technique and exhibited bifunctional OER and HER activities with overpotential values of 410 and 262 mV at 20 mA/cm² in 0.1 M KOH, respectively. Despite all the benefits of 2D MoS₂ and 2D gC₃N₄, hybrid 2D gC₃N₄-decorated 2D MoS₂ (2D–2D gC₃N₄–MoS₂) nanocomposites have rarely been investigated for bifunctional water electrolysis, particularly in regards to substantial OER and HER performance in alkaline medium.

Motivated by the abovementioned backgrounds, we synthesized 2D–2D gC₃N₄–MoS₂ nanocomposites using a facile sonochemical method and examined their bifunctional water electrolysis performances. The gC₃N₄–MoS₂ nanocomposites demonstrated excellent electrocatalytic water-splitting activities in 1 M KOH with small overpotential values (i.e., 225 mV for OER and 156 mV for HER at 10 mA/cm²). Furthermore, the assembled gC₃N₄–MoS₂||gC₃N₄–MoS₂ electrolyzer demonstrated a low full cell voltage of 1.52 V at a current density of 10 mA/cm², maintaining superior stability during prolonged operation for up to 100 h. Herein, the synthesis-to-electrocatalytic characteristics of the 2D–2D gC₃N₄–MoS₂ nanocomposites are assessed and deliberated in detail.

2. Experiment

2.1. MoS₂ Nanosheet Synthesis

Figure 1a depicts the facile sonochemical process for producing the hybrid 2D–2D gC₃N₄–MoS₂ nanocomposites. The 2D MoS₂ nanosheets were derived from commercial bulk MoS₂ (Sigma-Aldrich, St. Louis, MO, USA). Initially, 2 g of bulk MoS₂ was dispersed into 100 mL of deionized water and stirred continuously for 30 min. Next, the blended solution was sonicated (f_ultra = 35 kHz and P_ultra = 240 W) for 60 min. During the ultrasonication process, the ultrasonic waves induced alternating low and high pressures in the bulk MoS₂. This process facilitated cavitation-driven exfoliation of the MoS₂ sub-lattices, resulting in the successful formation of layered MoS₂ nanosheets. Afterward, the colloidal suspension was collected, cleaned, filtered, and parched in a 150 °C electric oven for 480 min. Finally, the MoS₂ nanosheets were collected in powder form.

2.2. gC₃N₄ Nanosheet Synthesis

The 2D gC₃N₄ nanosheets were synthesized from melamine using a facile pyrolysis method [39,45,46]. Initially, 5 g of melamine (Sigma-Aldrich, St. Louis, MO, USA) was transferred to an alumina crucible and encapsulated by the crucible cover. Thereafter, the crucible setup was loaded in a muffle furnace and thermally annealed at 550 °C for 240 min in an air atmosphere. Finally, yellow-colored powdered gC₃N₄ nanosheets were obtained.

2.3. gC₃N₄–MoS₂ Nanocomposite Fabrication

The 2D–2D gC₃N₄–MoS₂ nanocomposites were fabricated via an ultrasonication process using the pyrolytically fabricated 2D gC₃N₄ nanosheets and the sonochemically derived 2D MoS₂ nanosheets. Prior to discussing the experimental procedures, it is important to note that the gC₃N₄ to MoS₂ ratio of 1:0.3 was selected based on insights from previous literature [47,48,49,50,51,52], as this composition was found to provide optimal material properties and enhanced electrochemical performance. For this, first, the MoS₂ nanosheets (1 g) were interspersed in deionized water (100 mL) through vigorous stirring for 20 min. Then, the gC₃N₄ nanosheets (0.3 g) were subsequently added and mixed with the abovementioned MoS₂ solution through additional stirring for 20 min. Thereafter, the gC₃N₄–MoS₂ blended aqueous solution was ultrasonicated for 60 min (f_ultra = 35 kHz and P_ultra = 240 W). Finally, the ultrasonicated colloidal suspension was washed, accumulated, and parched at 150 °C for 6 h to obtain the nanopowder form of 2D–2D gC₃N₄–MoS₂.

2.4. Material Characterization

The morphological and elemental properties of MoS₂ and gC₃N₄–MoS₂ were monitored through field-emission scanning electron microscopy (FE-SEM, Clara LMH, Tescan Brno, Czech Republic) and in situ energy-dispersive X-ray spectroscopy (EDX) measurements, respectively. Moreover, the topographic features of the prepared catalysts were further assessed via transmission electron microscopy (TEM, JEM 2100F, JEOL, Tokyo, Japan). Crystallographic information was investigated through the X-ray diffraction (XRD; D8-Advance, Bruker, Billerica, MA, USA) analysis. The textural characteristics were assessed via the Brunauer–Emmett–Teller (BET, BELSORP-mini II system, MicrotracBEL, Osaka, Japan) and Barrett–Joyner–Halenda (BJH) techniques. The surface chemical states of the synthesized catalysts were examined through X-ray photoelectron spectroscopy (XPS, ESCALab250Xi system, Thermos Fisher Scientific, Waltham, MA, USA).

2.5. Electrocatalytic Measurements

The water splitting performance of the prepared MoS₂ nanosheets and gC₃N₄–MoS₂ nanocomposites was evaluated using the typical three-electrode method in an alkaline electrolyte (1 M KOH) solution using the VersaSTAT3 workstation (Ametek Scientific Company, Mahwah, NJ, USA). First, we assembled two different working electrodes using the prepared 2D MoS₂ nanosheets and 2D–2D gC₃N₄–MoS₂ nanocomposites. For this, 3 mg of each synthesized catalyst (i.e., either MoS₂ or gC₃N₄–MoS₂) was amalgamated with a 3 mL of N-methyl-2-pyrrolidinone solution and coated on nickel foam substrates (1 cm × 1 cm, 110 ppi) that were purchased from the MTI Korea Group, Seoul, Republic of Korea. Next, each substrate (i.e., catalyst-coated nickel foam) was dried at 180 °C for 480 min. Moreover, a saturated calomel electrode (SCE) (i.e., reference electrode) and a coiled Pt wire (i.e., counter electrode) were also prepared to set up the three-electrode system. After preparing all the electrode setups, the electrocatalytic characteristics were examined using the electrochemical workstation. First, the electrochemical cyclic voltammetry (CV) characteristics of the synthesized catalysts were examined in a 0–0.5 V potential range, where the scan rate was also varied from 10 to 100 mV/s. Next, the linear sweep voltammetry (LSV) characteristics were assessed at a fixed scan rate of 1 mV/s within certain potential windows (i.e., OER: −0.1–1.2 V, HER: −1–−1.8 V). Additionally, the electrochemical impedance spectroscopy (EIS) tests were performed using an AC signal amplitude of 10 mV in the 1 Hz to 10 kHz frequency range in 1 M KOH. Furthermore, the chronopotentiometry measurements were conducted at different current densities (i.e., OER: 10–100 mA/cm², HER: −10–−100 mA/cm²), for which each step was maintained for 10 min. Here, it should be noted that the reference was standardized to the reversible hydrogen electrode (RHE) scale. Furthermore, all the polarization values and their corresponding curves were i_R-corrected. The electrochemical double-layer capacitance (C_DL) and its corresponding electrochemically active surface area (ECSA) of the catalysts were determined from the non-Faradaic CV curves by using the following relationships [53,54,55]:

$$J_{DL}=C_{DL}\times v/A$$

(1)

$$ECSA=C_{DL}/C_e\text{,}$$

(2)

where J_DL, C_e, v, and A are the double-layer charging current, capacitance of the used electrolyte (0.04 mF/cm² for KOH), the potential scan rate, and the area of electrodes, respectively. The overpotential (η) and the Tafel slope (S_T) for the OER and HER were calculated using the following equations [14,56,57,58,59,60]:

$$E_{RHE}=E_{SCE}+E_{SCE}^0+0.059\mathrm{p}\mathrm{H}$$

(3)

$$\eta =E_{RHE}-1.23 \mathrm{V} \text{(for OER)}$$

(4)

$$\eta =E_{RHE} \text{(for OER)}$$

(5)

$$\eta =S_T\log \left(J\right)+c$$

(6)

where E⁰_SCE is the SCE’s standard potential, J is the current density, c is the fitting parameter, and E_RHE is the RHE’s standard potential.

3. Results and Discussion

The morphology and chemical composition of the bare 2D MoS₂ nanosheets and 2D–2D gC₃N₄–MoS₂ nanocomposites were characterized via FE-SEM and in situ EDX measurements. Figure 1b–e display the FE-SEM images of the pristine MoS₂ nanosheets and the hybrid gC₃N₄–MoS₂ nanocomposites. The MoS₂ sample exhibited stacked-layer nanosheet-like structures with an average length of 200–400 nm (Figure 1b,c). After ultrasonicating the bare MoS₂ nanosheets together with the gC₃N₄ nanosheets, the sample exhibited the interconnected and agglomerated structure of the 2D gC₃N₄-decorated 2D MoS₂ nanocomposites (Figure 1d,e). Next, the elemental composition of the bare MoS₂ nanosheets and gC₃N₄–MoS₂ nanocomposites was investigated. As shown in the EDX profiles (Figure 1f,g), both samples revealed their own intrinsic Mo, S, N, and C constituents, demonstrating that the prepared materials were highly pure and free of other impurities. Here, it should be noted that Pt detected in both samples arose from the conductive coating for FE-SEM measurements, which was applied to avoid the electron charging effect.

The topography of both the bare 2D MoS₂ nanosheets and 2D–2D gC₃N₄–MoS₂ nanocomposites was further analyzed by TEM measurements. Figure 2a,b show the bright-field TEM images of the MoS₂ nanosheets. The bare MoS₂ exhibited a stacked-layer nanosheet morphology. In the high-resolution image (Figure 2c), the bare MoS₂ shows an interlayer spacing of 0.62 nm, which corresponds to the (002) plane of hexagonal MoS₂ [61]. The well-defined diffraction lattice indicates that the bare MoS₂ nanosheets were single crystals corresponding to the hexagonal phase of MoS₂ (Figure 2d). Unlike the bare MoS₂ nanosheets, the hybrid gC₃N₄–MoS₂ nanocomposites exhibited an interconnected and agglomerated structure of gC₃N₄ nanosheets decorated on stacked-layer MoS₂ nanosheets (Figure 2e,f). From the high-resolution image of gC₃N₄–MoS₂ (Figure 2g), the lattice spacings of MoS₂ and gC₃N₄ were confirmed to be 0.62 and 0.32 nm, which are consistent with those of (002) MoS₂ and (002) gC₃N₄, respectively. Moreover, the SAED pattern of gC₃N₄–MoS₂ exhibited polycrystalline phases (Figure 2h) because of its microstructural hybridization.

To further understand the formation kinetics of the gC₃N₄–MoS₂ nanocomposites, we discuss the chemical mechanisms implicated during the ultrasonication process. The sonochemical technique possesses several advantages, including rapid synthesis, low energy consumption, and the ability to generate highly dispersed nanostructures with uniform morphology. Ultrasonic irradiation induces acoustic cavitation, producing localized high temperatures and pressures that promote fast nucleation and prevent agglomeration. These features result in materials with high surface area, better crystallinity, and enhanced active site exposure—critical for improving electrocatalytic performance. In an aqueous solution, ultrasonication generates two significant radicals from water (i.e., hydrogen (H^*) and hydroxyl (OH^*) radicals). During the sonication of bulk materials, these H^* and OH^* radicals serve as reductants [62,63,64]. Consequently, bulk MoS₂ (nMoS₂) could be diminished into stacked-layer MoS₂ nanosheets (MoS_2(n)) under the high ultrasonic power in water. This sonochemical exfoliation process (i.e., micro-cavitation and shock waves) can be described by the following reactions [63,65,66,67,68]:

$${\mathrm{H}}_2\mathrm{O}\underset{\mathrm{S}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}{\to }{\mathrm{O}\mathrm{H}}^{*}+{\mathrm{H}}^{*}$$

(7)

$$n{\mathrm{M}\mathrm{o}\mathrm{S}}_2+{\mathrm{O}\mathrm{H}}^{*}+{\mathrm{H}}^{*}\underset{\mathrm{S}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}{\to }{\mathrm{M}\mathrm{o}\mathrm{S}}_{2\left(\mathrm{n}\right)}$$

(8)

$$\mathrm{n}{\mathrm{M}\mathrm{o}\mathrm{S}}_2+{gC}_3{N}_4+{\mathrm{O}\mathrm{H}}^{*}+{\mathrm{H}}^{*}\underset{\mathrm{S}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}{\to }{\mathrm{M}\mathrm{o}\mathrm{S}}_{2\left(\mathrm{n}\right)}\text{–}{gC}_3{N}_4\text{–}{\mathrm{M}\mathrm{o}\mathrm{S}}_{2\left(\mathrm{n}\right)}\text{–}\cdots .$$

(9)

Figure 2. (a,b) Bright-field TEM images, (c) high-resolution TEM image, and (d) SAED pattern of MoS₂. (e,f) Bright-field TEM image, (g) high-resolution TEM image, and (h) SAED pattern of gC₃N₄–MoS₂.

Figure 2. (a,b) Bright-field TEM images, (c) high-resolution TEM image, and (d) SAED pattern of MoS₂. (e,f) Bright-field TEM image, (g) high-resolution TEM image, and (h) SAED pattern of gC₃N₄–MoS₂.

Next, the crystallographic phases of pristine MoS₂ nanosheets and gC₃N₄–MoS₂ nanocomposites were investigated using powder XRD measurements. Figure 3a shows the XRD patterns of MoS₂ and gC₃N₄–MoS₂. Both samples revealed diffraction angles at 14.38°, 29.14°, 32.72°, 33.58°, 35.94°, 39.67°, 44.25°, 49.94°, 56.07°, 58.36°, and 60.23°, which are attributed to the (002), (004), (100), (101), (102), (103), (006), (105), (106), (110), and (008) planes of 2H-phase hexagonal MoS₂, respectively (JCPDS No. 75-1539) [69,70,71]. The gC₃N₄–MoS₂ nanocomposites displayed two additional diffraction peaks at 12.88° and 25.77°, attributed to the (100) and (002) planes of hexagonal gC₃N₄, respectively (JCPDS No. 87-1526) [72,73]. The XRD pattern of gC₃N₄–MoS₂ displayed diffraction peaks from both MoS₂ and gC₃N₄, confirming the successful fabrication of the gC₃N₄–MoS₂ hybrid composite system. Moreover, no other secondary phases were observed in the fabricated materials, implying that the materials reflected high purity.

After confirming the successful hybridization of 2D gC₃N₄ and 2D MoS₂, the textural characteristics were assessed through BET and BJH measurements using N₂ adsorption–desorption isotherms (N₂-ADIs). Figure 3b shows the N₂-ADI curves of the bare MoS₂ nanosheets and gC₃N₄–MoS₂ nanocomposites. Both materials exhibited the Type-II isotherm characteristics, with a typical Type-H3 hysteresis loop (classified from IUPAC), demonstrating the distinctive features of mesoporous materials [74,75,76]. Based on the BET measurements, the specific surface areas of the bare MoS₂ nanosheets and gC₃N₄–MoS₂ nanocomposites were calculated to be 14 and 46 m²/g, respectively. Furthermore, BJH analysis revealed that the pore surface areas of the bare MoS₂ nanosheets and gC₃N₄–MoS₂ nanocomposites were 6 and 14 m²/g, respectively (Figure 3c). Additionally, the average pore diameter of MoS₂ and gC₃N₄–MoS₂ were 23.91 and 18.48 nm, respectively. Owing to the incorporation of layered gC₃N₄ nanosheets into the gC₃N₄–MoS₂ composite system, the total pore volume was greater for gC₃N₄–MoS₂ (0.09136 cm³/g) compared to that for MoS₂ (0.06124 cm³/g). The unique porosity and high surface area of the gC₃N₄–MoS₂ nanocomposites could be beneficial for enhancing their bifunctional water-splitting performance in alkaline media, which will be discussed later in detail.

The surface composition and ionic state interaction for both MoS₂ and gC₃N₄–MoS₂ were examined using XPS measurements. The XPS full survey spectrum of the bare MoS₂ nanosheets and gC₃N₄–MoS₂ nanocomposites clearly revealed their intrinsic elements, including Mo, S, C, N, and O (see Figure S1a,b). For the Mo 3d core-level spectra (Figure 4a), the bare MoS₂ displayed two predominant peaks at 229.46 and 232.58 eV, associating with the Mo 3d_5/2 and Mo 3d_3/2 spin-orbit splitting states of Mo⁴⁺, respectively [77]. In addition, the small peak observed at 235.91 eV is correlated with the orbital electron of Mo⁶⁺, indicating the existence of small amounts of MoO₃ due to the surface oxidation of MoS₂ [78]. The S 2s peak observed at 226.64 eV corresponds to the hexagonal phase of MoS₂ [79]. For S 2p (Figure 4b), the two peaks at 162.41 and 163.62 eV arise from the S 2p_3/2 and S 2p_1/2 orbit splitting states of the S²⁻ ion, respectively [80,81]. The hybrid gC₃N₄–MoS₂ nanocomposites clearly revealed similar features of the Mo 3d (Figure 4c) and S 2p (Figure 4d) core-level spectra. In the case of C 1s (Figure 4e), gC₃N₄–MoS₂ clearly exhibited two carbon peaks at 284.61 and 288.36 eV, attributed to the C–C and N–C=N bonds, respectively [41,82]. Moreover, as shown in Figure 4f, the N 1s core-level spectra contained two peaks at 398.24 and 399.93 eV, ascribed to the pyridinic C–N=C and triazine N–(C₃) bonds, respectively [34,83]. Additionally, the oxygen-related peaks were observed at 532.15 and 533.92 eV, attributed to the Mo–O bond and chemisorbed oxygen [34,41], respectively (see Figure S1c,d). These results clearly indicate that the gC₃N₄ nanosheets are well-decorated on the MoS₂ nanosheets in the composite system.

To investigate the impact of 2D–2D gC₃N₄–MoS₂ hybridization on its electrocatalytic performance, we assessed the electrochemical CV characteristics of the bare MoS₂ nanosheets and gC₃N₄–MoS₂ nanocomposites. As shown in Figure 5a,b, both the MoS₂ and gC₃N₄–MoS₂ catalysts display noticeable redox peaks in their CV curves. These peaks are mainly due to the electrochemical activity of the NF substrate, which undergoes a reversible Ni²⁺ to Ni³⁺ oxidation and Ni³⁺ to Ni²⁺ reduction in alkaline solution [84], while the catalysts may slightly influence the peak intensity or shape of the CV curves. These redox features are an indicative of the pseudocapacitive behavior that may affect both OER and HER activities. Here, it is noteworthy that the current density increased with an increasing scan rate. This means that the active catalyst material possesses a low diffusion resistance. Compared to the bare MoS₂ catalyst, the gC₃N₄–MoS₂ hybrid catalyst exhibited a wider CV window with a greater current response. This implies that the 2D–2D gC₃N₄–MoS₂ catalyst possessed a higher number of active sites than did the MoS₂ catalyst. We believe that the enhanced electrochemical activity of the 2D–2D gC₃N₄–MoS₂ hybrid catalyst is due to two possible reasons: the increased number of active sites [67,85,86] and the increased electrical conductivity [14,35,36,60,87]. The former will be explained below, while the latter is discussed later in the EIS section.

Figure 4. (a) Mo 3d and (b) S 2p core-level spectra of MoS₂ nanosheets. (c) Mo 3d, (d) S 2p, (e) C 1s, and (f) N 1s core-level spectra of gC₃N₄–MoS₂ nanocomposites.

Figure 4. (a) Mo 3d and (b) S 2p core-level spectra of MoS₂ nanosheets. (c) Mo 3d, (d) S 2p, (e) C 1s, and (f) N 1s core-level spectra of gC₃N₄–MoS₂ nanocomposites.

Figure 5. CV curves of (a) MoS₂ and (b) gC₃N₄–MoS₂. Non-faradaic J_DL at 1.10 V as a function of the potential scan rate for (c) MoS₂ and (d) gC₃N₄–MoS₂. Nyquist plots of (e) MoS₂ and (f) gC₃N₄–MoS₂. The insets of (e,f) illustrate the equivalent circuits of the fabricated working electrodes.

Figure 5. CV curves of (a) MoS₂ and (b) gC₃N₄–MoS₂. Non-faradaic J_DL at 1.10 V as a function of the potential scan rate for (c) MoS₂ and (d) gC₃N₄–MoS₂. Nyquist plots of (e) MoS₂ and (f) gC₃N₄–MoS₂. The insets of (e,f) illustrate the equivalent circuits of the fabricated working electrodes.

To elucidate the improved electrocatalytic performance of the gC₃N₄–MoS₂ hybrid catalyst, we first calculated the ECSA of the catalyst materials using Equations (1) and (2). The ECSA values were determined from the non-faradic CV regions (1.05 to 1.15 V) for both MoS₂ and gC₃N₄–MoS₂ (see Figure S2a,b). From the J_DL vs. v curves obtained from the non-faradaic CV region at approximately 1.10 V (Figure 5c,d), the C_DL values were calculated to be 1.33 and 6.29 mF/cm² for MoS₂ and gC₃N₄–MoS₂, respectively. Using Equations (1) and (2), the corresponding ECSA values of 33 and 158 cm² were estimated for MoS₂ and gC₃N₄–MoS₂, respectively. This indicates that the gC₃N₄–MoS₂ hybrid catalyst displayed a larger ECSA than did the bare MoS₂ catalyst (see Figure S3). Hence, it can be conjectured that the hybridization of gC₃N₄ and MoS₂ increased the number of electrochemically active sites in the entire composite medium of gC₃N₄–MoS₂. In short, hybridizing MoS₂ with gC₃N₄ enhanced the specific surface area compared to that of bare MoS₂. gC₃N₄ helps prevent the natural tendency of MoS₂ to restack by acting as a spacer, leading to a more open and porous structure [34,88]. This exposes more active edge sites, which are crucial for water splitting. The nitrogen-rich surface of gC₃N₄ also provides anchoring points for MoS₂, ensuring strong interaction and uniform dispersion [89,90]. In addition, electronic coupling between the two facilitates efficient charge transfer across the interface, while the rough, crumpled morphology created during assembly further increases surface roughness and accessibility [91,92,93].

Next, the electrochemical resistive behavior of the fabricated catalysts was examined through the EIS measurements. As shown in Figure 5e,f, the Nyquist plots of both MoS₂ and gC₃N₄–MoS₂ revealed straight lines at the low-frequency region, while they exhibited no parabolic curves at the high-frequency region. The former is relevant to the charge-transfer characteristics, which are directly associated with the series resistance (R_s) of the working electrode [34,94,95,96], and the latter is attributed to the electrolyte dispersion characteristics [36,67,97]. Through fitting the EIS data to the equivalent circuit model (Figure 5e,f, insets), the R_s values of MoS₂ and gC₃N₄–MoS₂ were estimated to be 0.71 and 0.48 Ω, respectively. Thus, it can be inferred that gC₃N₄–MoS₂ possess a better charge-transfer characteristic than does MoS₂. This can be interpreted as resulting from the large porosity and high electrical conductivity of the incorporated gC₃N₄ nanosheets.

The increased ECSA and the decreased R_s help to enhance rapid ion diffusion and swift electron transport, resulting in improved OER/HER performance. To verify this hypothesis, we performed LSV measurements for MoS₂, gC₃N₄, and gC₃N₄–MoS₂. Figure 6a shows the i_R-corrected OER LSV curves of the bare MoS₂ and gC₃N₄–MoS₂ hybrid catalysts at 1 mV/s (see also Figure S4a for bare gC₃N₄). Using Equations (3) and (4), the η₁₀ values of MoS₂, gC₃N₄, and gC₃N₄–MoS₂ were calculated to be 297, 325, and 225 mV, respectively, from the measured LSV curves at J = 10 mA/cm². The gC₃N₄–MoS₂ hybrid catalyst exhibited lower overpotential values (η₅₀ = 254 mV and η₁₀₀ = 271 mV), even at high current density (J = 50 and 100 mA/cm²), compared to those for MoS₂ (η₅₀ = 339 mV and η₁₀₀ = 378 mV) and gC₃N₄ (η₅₀ = 358 mV and η₁₀₀ = 381 mV) (Figure 6b). Notably, the η₁₀ values determined for MoS₂ and gC₃N₄–MoS₂ are in line with, and occasionally better than, established data from earlier studies (see Table S1). This means that the 2D–2D gC₃N₄–MoS₂ catalyst exhibits outstanding intrinsic reaction kinetics, resulting in significant OER performance in alkaline electrolytes [98,99,100]. Improved OER performance can also be confirmed by determining the S_T value. Using the Tafel equation in Equation (6), the small S_T values of MoS₂ (55 mV/dec), gC₃N₄ (58 mV/dec), and gC₃N₄–MoS₂ (49 mV/dec) were determined from their Tafel curves (Figure 6c,d and Figure S4b). The achieved S_T values suggest that the MoS₂, gC₃N₄, and gC₃N₄–MoS₂ catalysts follow the combined Volmer–Heyrovsky mechanism. In particular, gC₃N₄–MoS₂ exhibited a smaller S_T value than the values in the literature (Table S1). Namely, the 2D–2D gC₃N₄–MoS₂ catalyst exhibited outstanding intrinsic reaction kinetics owing to its decreased charge-transfer resistance, large porosity, and increased active surface area. Compared to MoS₂, the gC₃N₄–MoS₂ catalyst exhibited lower η and smaller S_T values, signifying the faster reaction rate of OH⁻ over the catalyst surface. The catalytic OER mechanism in an alkaline medium typically involves a four-electron process:

$$\mathrm{M}+{\mathrm{O}\mathrm{H}}^{-}\to \mathrm{M}{\mathrm{O}\mathrm{H}}_{\mathrm{a}\mathrm{d}\mathrm{s}}+{\mathrm{e}}^{-}$$

(10)

$$\mathrm{M}{\mathrm{O}\mathrm{H}}_{\mathrm{a}\mathrm{d}\mathrm{s}}+{\mathrm{O}\mathrm{H}}^{-}\to \mathrm{M}{\mathrm{O}}_{\mathrm{a}\mathrm{d}\mathrm{s}}+{\mathrm{e}}^{-}+{\mathrm{H}}_2\mathrm{O}$$

(11)

$$\mathrm{M}{\mathrm{O}}_{\mathrm{a}\mathrm{d}\mathrm{s}}+{\mathrm{O}\mathrm{H}}^{-}\to \mathrm{M}{\mathrm{O}\mathrm{O}\mathrm{H}}_{\mathrm{a}\mathrm{d}\mathrm{s}}+{\mathrm{e}}^{-}$$

(12)

$$\mathrm{M}{\mathrm{O}\mathrm{O}\mathrm{H}}_{\mathrm{a}\mathrm{d}\mathrm{s}}+{\mathrm{O}\mathrm{H}}^{-}\to \mathrm{M}+{\mathrm{e}}^{-}+{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}$$

(13)

where M specifies an active site of the prepared catalysts, and M-OOH, M-O, and M-OH are the reaction intermediates on the catalysts’ surface.

The improved intrinsic reaction kinetics of the catalyst can also impact its chronopotentiometric characteristics. As shown in Figure 6e, the gC₃N₄–MoS₂ catalyst displayed a smaller overpotential at each step (i.e., at different current densities) compared to that of the bare MoS₂ catalyst. This proves that the incorporation of gC₃N₄ could aid in enhancing ion storage performance and catalytic activity. The long-term OER stability of the MoS₂ and gC₃N₄–MoS₂ catalysts was systematically evaluated through CP measurements conducted at 10 and 100 mA/cm² for each 100 h, respectively. As shown in Figure 6f, both catalysts maintained stable performance and increased activity at diverse current densities. Furthermore, both samples exhibited almost indistinguishable LSV characteristic curves before and after the 100 h stability evaluation. (Figure S6). Compared to MoS₂, however, gC₃N₄–MoS₂ demonstrated superior long-term durability due to its trivial resistance and increased catalytic active areas. This indicates the excellent endurance of gC₃N₄–MoS₂ in an alkaline medium.

Figure 6. OER performance of fabricated electrocatalysts. (a) i_R-corrected LSV curves, (b) overpotential comparison, (c) Tafel plots, (d) Tafel slope comparison, (e) chronopotentiometric profiles at 10–100 mA/cm², and (f) long-term stability characteristics of MoS₂ and gC₃N₄–MoS₂.

Figure 6. OER performance of fabricated electrocatalysts. (a) i_R-corrected LSV curves, (b) overpotential comparison, (c) Tafel plots, (d) Tafel slope comparison, (e) chronopotentiometric profiles at 10–100 mA/cm², and (f) long-term stability characteristics of MoS₂ and gC₃N₄–MoS₂.

Next, we assessed the HER performance of the fabricated catalysts to examine their bifunctional water-splitting activity. Figure 7a shows the i_R-corrected LSV curves of gC₃N₄ and gC₃N₄–MoS₂ obtained during the HER process, which was conducted at a 1 mV/s scan rate in the KOH electrolyte (see also Figure S5a for bare gC₃N₄). Using Equations (3) and (5), the obtained η₁₀ values of MoS₂, gC₃N₄, and gC₃N₄–MoS₂ were 228, 236, and 156 mV, respectively, at a current density of −10 mA/cm². In addition, the gC₃N₄–MoS₂ catalyst revealed smaller η₅₀ (=244 mV) and η₁₀₀ (=275 mV) values compared to those for the MoS₂ catalyst (η₅₀ = 327 mV and η₁₀₀ = 374 mV) and gC₃N₄ catalyst (η₅₀ = 321 mV and η₁₀₀ = 378 mV) (Figure 7b). Furthermore, from the Tafel curves (Figure 7c,d and Figure S5b), the obtained S_T values for MoS₂, gC₃N₄, and gC₃N₄–MoS₂ were 145, 158, and 101 mV/dec, respectively. Notably, both η and S_T values obtained from gC₃N₄–MoS₂ compare favorably with and, in certain cases, surpass those reported in prior literature. (Table S2). In both the chronopotentiometric HER test (Figure 7e) and the long-term HER stability test (Figure 7f and Figure S7), gC₃N₄–MoS₂ also revealed better electrocatalytic activity than that of bare MoS₂.

The outstanding bifunctional OER and HER performances of the 2D–2D gC₃N₄–MoS₂ catalyst may allow for superior overall water-splitting performance. To verify this, we tested the overall water-splitting activities of gC₃N₄–MoS₂ using a two-electrode configuration in 1M KOH (Figure 8a). Figure 8b displays the LSV curve of the gC₃N₄–MoS₂|| gC₃N₄–MoS₂ electrolyzer. Remarkably, the bifunctional gC₃N₄–MoS₂||gC₃N₄–MoS₂ electrolyzer could drive specific current densities with low full-cell voltage values (e.g., 10 mA/cm² with 1.52 V and 100 mA/cm² with 1.85 V). These results depict an adequate catalytic activity of gC₃N₄–MoS₂ for efficient overall water electrolysis. The full-cell voltages achieved from the current gC₃N₄–MoS₂||gC₃N₄–MoS₂ electrolyzer are on par with or even lower than those previously reported for other metal oxide-based electrocatalysts (Table S3). Moreover, gC₃N₄–MoS₂||gC₃N₄–MoS₂ exhibited constant stability performance for 100 h at both 10 and 100 mA/cm² (Figure 8c). After overall water electrolysis stability tests, the morphology of both MoS₂ and gC₃N₄–MoS₂ maintained its initial structure (see Figure S8a,b). In EDX analysis, it was confirmed that the intrinsic elements Mo, S, C, and N clearly appear, consistent with the composition of the synthesized catalysts. However, the detection of additional Ni, K, and O signals suggests interfacial interactions between the catalyst surface, the electrolyte, and the underlying Ni foam substrate during electrochemical testing (see Figure S8c,d). In XRD testing, the gC₃N₄–MoS₂ still exhibited higher intensity compared to that of the bare MoS₂, although both samples showed decreased XRD intensities after the completion of all the water electrolysis tests (see Figure S9a,b). The above results suggest that the sonochemically hybridized 2D–2D gC₃N₄–MoS₂ nanocomposites exhibit excellent potential for use as a bifunctional electrocatalyst for superior overall water electrolysis.

Figure 7. HER performance of fabricated electrocatalysts. (a) i_R-corrected LSV curves, (b) overpotential comparison, (c) Tafel plots, (d) Tafel slope comparison, (e) chronopotentiometric profiles at −10–−100 mA/cm², and (f) long-term stability characteristics of MoS₂ and gC₃N₄–MoS₂.

Figure 7. HER performance of fabricated electrocatalysts. (a) i_R-corrected LSV curves, (b) overpotential comparison, (c) Tafel plots, (d) Tafel slope comparison, (e) chronopotentiometric profiles at −10–−100 mA/cm², and (f) long-term stability characteristics of MoS₂ and gC₃N₄–MoS₂.

Figure 8. Overall water-splitting performance of gC₃N₄–MoS₂||gC₃N₄–MoS₂: (a) illustration of the two-electrode cell setup for OWS measurements, (b) LSV curves before and after stability testing, and (c) long-term stability characteristics over 100 h under current densities of 10 and 100 mA/cm².

Figure 8. Overall water-splitting performance of gC₃N₄–MoS₂||gC₃N₄–MoS₂: (a) illustration of the two-electrode cell setup for OWS measurements, (b) LSV curves before and after stability testing, and (c) long-term stability characteristics over 100 h under current densities of 10 and 100 mA/cm².

4. Conclusions

High-performance 2D–2D gC₃N₄–MoS₂ nanocomposites were successfully fabricated via a facile ultrasonication process using sonochemically synthesized MoS₂ and pyrolytically derived gC₃N₄. The gC₃N₄–MoS₂ nanocomposites exhibited excellent bifunctional OER and HER performance in 1 M KOH. Namely, the gC₃N₄–MoS₂ catalyst demonstrated excellent OER performance (i.e., a low η of 225 mV and a small S_T of 49 mV/dec), as well as outstanding HER performance (i.e., a low η of 156 mV and a small S_T of 101 mV/dec). Moreover, the gC₃N₄–MoS₂ electrocatalyst revealed the superb overall water splitting with a low full-cell voltage of 1.52 V at 10 mA/cm². These results were attributed to both low R_s (0.48 Ω) and large ECSA (158 cm²), resulting from the hybridization of 2D MoS₂ and highly conductive gC₃N₄. Additionally, gC₃N₄–MoS₂ also demonstrated good long-term stability up to 100 h for overall water splitting. This work marks a significant improvement over previously reported MoS₂-based catalysts by introducing a scalable, energy-efficient synthesis strategy that simultaneously improves catalytic activity and long-term durability. The adoption of a sonochemical approach offers a green, cost-effective, and industrially viable pathway for the fabrication of high-performance bifunctional electrocatalysts. Future research could explore compositional optimization, integration into membrane-based electrolyzer systems, and rigorous evaluation under industrially relevant conditions to facilitate the practical deployment of this catalyst in sustainable hydrogen production technologies.