1. Introduction
Fossil fuels are the major contributor to the energy landscape, but their depletion and negative environmental impacts have prompted a search for alternative energy solutions [
1,
2,
3,
4]. Toward this perspective, various renewable energies have been studied and developed. Among them, hydrogen has been studied extensively and is considered a promising energy carrier due to its clean combustion and high energy density [
5,
6,
7,
8,
9,
10,
11]. To date, steam methane reforming is a widely employed route for the industrial-scale production of hydrogen. In addition to the high energy supply requirement for maintaining high temperature and pressure in the reforming process, this route also produces carbon mono- and di-oxides as byproducts. Therefore, this conventional method of hydrogen production needs to substitute urgently with a green technique that runs with zero carbon emission and does not require a high energy supply. Electrochemical water slitting is a green route to obtain molecular oxygen and hydrogen. Hence, to meet global energy demands and address environmental concerns, producing hydrogen through water splitting has recently gained increasing interest [
12,
13]. Platinum-based catalysts are currently the most effective for the hydrogen evolution reaction (HER), exhibiting nearly zero onset overpotential and a low Tafel slope [
14,
15]. However, their limited availability and high cost hindered widespread applicability. Consequently, considerable work has been carried out to develop HER electrocatalysts that are equally efficient and durable but more affordable [
16,
17,
18,
19,
20,
21,
22].
Transition metal dichalcogenides (TMDs) such as sulfide, selenide, and telluride of molybdenum and tungsten with layered structures have been paid considerable attention in various fields, including catalysts, energy storage, transistors, and photoelectrochemical devices [
23,
24,
25,
26,
27,
28,
29]. The unique two-dimensional (2D) architecture with excellent electrical properties and good catalytic activity offers TMDs as promising candidates for HER electrocatalysts [
30,
31,
32]. Among them, MoS
2 has been studied extensively and employed as an HER catalyst [
33,
34]. In the same group of TMDs, MoSe
2 bears similar layer-to-layer structures and exhibits versatile electrochemical properties with good stability. MoSe
2 possessed a metallic nature, offering superior electrical conductivity compared to MoS
2, which is beneficial for HER catalysts [
35,
36]. However, the high surface energy and
van der Waals attractions of MoSe
2 interlayers lead to the stacking of the layers, resulting in a loss of active catalytic sites [
37,
38]. Additionally, MoSe
2 suffered from low intrinsic conductivity and weak electrical contact with active sites, which negatively impacts its catalytic activity. To overcome these challenges, researchers are actively exploring various approaches to improve the catalytic performance of MoSe
2. These approaches include the manipulation of nanostructures to improve conductivity and electron transfer, optimizing the composition and surface modification, and exploring hybrid structures with conductive carbon materials [
39,
40,
41,
42,
43].
Herein, this work demonstrates the successful synthesis of the MoSe
2-Gr composite via a two-step process involving solvothermal synthesis of MoSe
2 using diethylene glycol solvent followed by thermal treatment to create MoSe
2-Gr. The resulting MoSe
2-Gr catalyst exhibited highly competitive electrocatalytic performance in acidic media for the HER. Specifically, this catalyst exhibited lower overpotentials of 161 and 250 mV at a current density of 10 and 50 mA.cm
−2, respectively, and a small Tafel slope of 67 mV dec
−1. It is worth noting that this performance lies on the top tire of MoSe
2 and their carbonous composites based HER electrocatalysts synthesized by various approaches (see,
Table S1 in the Supporting Information Section). Most importantly, it should be further noted that the MoSe
2-Gr catalyst designed in this work demonstrated a superior HER catalytic activity than that of the state-of-the-art Pt/C catalyst, especially when the electrolysis was performed at a high cathodic current density beyond ca. −55 mA cm
−2. Such catalysts exhibiting superior HER catalytic activity at high current density are particularly applicable for the industrial scale of hydrogen production. The observed superior HER performance of the MoSe
2-Gr catalyst can be owing to the amplified surface area and concentration of active edge sites present within the wider interlayer width of MoSe
2. Moreover, the resulting architecture of the composite facilitated the efficient movement of charges and ions between the MoSe
2 interlayers and electrolytes, thereby enhancing the charge and mass transfer capability. Additionally, the MoSe
2-Gr catalyst exhibited prolonged electrochemical stability in acidic electrolytes during the HER.
3. Results and Discussions
Figure 1a depicts the schematic diagram for the synthesis of pristine MoSe
2 and MoSe
2-Gr samples, which involved the solvothermal reaction of Mo and Se precursor in DEG solvent at 200 °C for 15 h to produce MoSe
2, followed by thermal treatment in an inert atmosphere at 600 °C for 5 h to form MoSe
2-Gr.
The SEM surface topography of the obtained MoSe
2 and MoSe
2-Gr samples is shown in
Figure 1b,c, respectively. Compared to the as-synthesized MoSe
2, the surface morphology of the MoSe
2-Gr after the calcination looks significantly different. The resulting MoSe
2-Gr nanosheets assembled to a porous architecture are clearly observable, indicating an enhanced surface area. This feature contributes to higher catalytic active sites with facile movements of ions and charges between the MoSe
2-Gr catalyst and electrolyte, thereby significantly improving the electrocatalyst performance. To gain further insight into the surface area and pore size distribution of MoSe
2 and MoSe
2-Gr, BET nitrogen adsorption/desorption isotherm measurements were conducted and are presented in
Figure S1.
Figure S1a,b show the nitrogen sorption isotherms of the pristine MoSe
2 and MoSe
2-Gr, respectively, both exhibiting type IV isotherms. The specific surface areas of MoSe
2 and MoSe
2-Gr were determined to be 13.15 m
2 g
−1 and 38.33 m
2 g
−1, respectively. The significantly higher specific surface area of MoSe
2-Gr compared to pristine MoSe
2 is attributed to the in-situ incorporation of graphene into the MoSe
2 interlayers. The pore size distribution of MoSe
2 and MoSe
2-Gr are displayed in
Figure S1c,d, respectively. The pore size distribution curves indicate the presence of mesoporous structures. These observations are in agreement with the SEM analysis.
Figure 1d illustrates the XRD patterns of MoSe
2 and MoSe
2-Gr, and the reference 2H-MoSe
2 (JCPDS #29-0914) has also been shown for comparison. The XRD patterns of the as-prepared MoSe
2 sample indicate that the sample was primarily composed of MoSe
2 with a small amount of MoO
3 present. Specifically, the diffraction peaks at 2θ angles of 27.61°, 31.42°, 47.50°, and 56.00° can be observed, which correspond to the lattice planes of (004), (100), (105), and (110) of the crystalline 2H-MoSe
2, respectively. However, the main (002) peak of the as-prepared MoSe
2 sample, which is typically observed at 13.70° in 2H-MoSe
2 is observed to be shifted significantly to a lower 2θ position of 8.00°. The observed shift indicates that the interlayer spacing of the MoSe
2 sample has been expanded, which may have been caused by the incorporation of DEG solvent molecules into the MoSe
2 layers. [
28,
47]. In addition, the XRD pattern of the MoSe
2 sample presents the presence of two peaks at 23.42° and 28.70°, which correspond to the diffraction from (110) and (130) planes of MoO
3 (JCPDS #05-0508). The presence of oxides can be due to various factors such as impurities present in the precursors, residual solvents, and moisture contamination from the surrounding during the synthesis process [
28,
29,
48]. After the thermal treatment, the diffraction of the MoSe
2-Gr sample exhibited intense peaks, indicating an improved crystallinity of the MoSe
2-Gr sample. The diffraction peaks observed at 12.60°, 26.95°, 31.68°, 37.76°, 47.88°, 55.99°, 65.82°, and 70.55° can be precisely assigned to the (002), (004), (100), (103), (105), (110), (200) and (203) crystal planes of 2H-MoSe
2 (JCPDS 29-0914). This finding confirms the presence of MoSe
2 in the MoSe
2-Gr sample. However, in this case, the (002) peak of the MoSe
2-Gr sample showed a slight shift compared to 2H-MoSe
2. This shift is attributed to the evaporation of trapped DEG and the carbonization of a small amount of DEG, leading to the formation of graphene within the MoSe
2 layers during the calcination process. Interestingly, after the thermal treatment, the peaks corresponding to the oxides (110) and (130) disappeared, indicating a reduction of the oxides present initially in the MoSe
2 sample. The Raman spectra of both MoSe
2 and MoSe
2-Gr samples were displayed in
Figure 1e. The vibration modes including E
1g at 188.82 cm
−1, out-of-plane A
1g at 237.25 cm
−1, and in-plane modes E
12g and B
12g at 280.35 cm
−1 and 337.40 cm
−1, respectively, can be observed. These modes correspond to the 2H phase of MoSe
2, confirming the presence of MoSe
2 in both samples [
49,
50,
51,
52]. The Raman shift results agree well with the XRD patterns, confirming the successful formation of 2H-MoSe
2. Additionally, in the Raman spectrum of the MoSe
2-Gr sample, two broad bands were observed. The band at approximately 1352.40 cm
−1 corresponds to the D-band, while the band at around 1599.29 cm
−1 represents the G-band, both characteristic of graphene. The G-band rose from the bond stretching between all pairs of sp
2 atoms in both rings and chains of graphene, indicating the presence of a well-ordered carbon lattice. On the other hand, the D-band was associated with disordered carbon defects or disorder-induced phonon scattering in graphene sheets. These defects can arise from the carbonization of trapped DEG during thermal treatment, resulting in the formation of graphene within the MoSe
2 layers. The observation of these bands further supported the presence of graphene in the MoSe
2-Gr sample.
The crystal lattices of the MoSe
2-Gr sample were further investigated using HR-TEM.
Figure 2a displays the HR-TEM image, which reveals that the interplanar spacing of the MoSe
2-Gr sample measures 0.73 nm, which can be attributed to the expanded interlayer distance of MoSe
2, specifically of the (002) facet. It is important to note that the standard interplanar distance of the MoSe
2 (002) plane is typically 0.65 nm. This implies the widening of the interlayer spacing of MoSe
2 in the MoSe
2-Gr sample. This broadened interplanar spacing observed in the MoSe
2-Gr sample suggests that the
in situ formed graphene during the calcination was incorporated within the interlayer of MoSe
2. Furthermore, the HR-TEM image (
Figure 2a) shows an additional d-spacing of 0.33 nm corresponding to the characteristic d-spacing of graphene. This observation further supports the presence of graphene within the MoSe
2-Gr sample. Moreover, the selected area electron diffraction (SAED) patterns in
Figure 2b reveal that the MoSe
2-Gr sample has a polycrystalline structure. The diffraction rings observed in the SAED patterns correspond well with the (100), (103), and (110) planes of the hexagonal 2H-MoSe
2 phase, which is consistent with the results obtained from the XRD analysis. Taken together, the XRD, Raman, and HR-TEM data provide conclusive evidence for the in-situ formation and incorporation of graphene within the interlayers of MoSe
2 in the MoSe
2-Gr sample.
Figure 3a–e displayed the TEM image, STEM-based high-angle annular dark-field (HAADF-STEM) image, and the corresponding TEM-based EDX elemental map** images of Mo, Se, and C in the MoSe
2-Gr sample. These images demonstrate a homogeneous distribution of each constituent in the MoSe
2-Gr sample. Additionally,
Figure 3f displays the EDX spectrum of the MoSe
2-Gr sample along with the atomic percentages of each element obtained from TEM-EDX-based analysis. The atomic percentages of Mo and Se were measured to be 11.15 and 23.50%, respectively, which closely align with the expected 1:2 ratio of MoSe
2. Since carbon-supported grids were employed for TEM characterization, the carbon content in the sample could not be measured precisely via TEM-based EDX analysis. As a result, for the analysis of carbon content, elemental analysis (N, C, and H) in the MoSe
2 and MoSe
2-Gr samples was conducted using an elemental analyzer (FLASH EA1112). The results are listed in
Table 1. The MoSe
2 sample exhibited a C content of 11.632 wt%. while the MoSe
2-Gr sample showed a decrease in C content to 7.210 wt%. This can be due to the evaporation of DEG and the formation of graphene by thermal treatment, leading to a decrease in the C content, while the N content, which could be detected from contamination from the background surrounding remained relatively unchanged. Overall, the TEM-EDX analysis, along with the elemental analysis provided information on the elemental composition and distribution in the MoSe
2-Gr sample, confirming the successful integration of graphene into the MoSe
2 structure.
The HER performance of the MoSe
2, MoSe
2-Gr, and the benchmark Pt/C electrocatalysts was evaluated in a typical three-electrode electrochemical cell. The working electrode was a drop-casted catalyst film on a GEC electrode. Prior to the measurements, the working electrodes were activated and stabilized via cycling at a scanning rate of 50 mVs
−1 in a potential window of +0.1 to −0.4 V vs. RHE until constant voltammograms were obtained. The electrocatalytic activity for the HER was accessed via linear sweep voltammograms (LSV) recorded without an iR correction.
Figure 4a displays the cathodic LSV curves of MoSe
2 and MoSe
2-Gr in 0.5 M H
2SO
4 solution. In addition, a commercial 20 wt.% Pt/C, which is often employed as the state-of-the-art HER electrocatalyst was also measured as the benchmark HER catalyst. The benchmark Pt/C electrocatalyst exhibited a nearly zero onset potential, reflecting superior catalytic activity toward HER. The LSV curve of the as-synthesized MoSe
2 sample demonstrated poorer catalytic performance, which is indicated by a high overpotential of 512 mV required to achieve an HER current density of −10 mA·cm
−2. However, after the thermal treatment, the obtained MoSe
2-Gr catalyst showed significantly improved HER activity compared to that of the MoSe
2 catalyst. This improvement is evident by the considerably reduced overpotential of MoSe
2-Gr, measuring 161 mV vs. RHE to drive an HER current density of −10 mA·cm
−2. Interestingly, the MoSe
2-Gr catalyst achieved an HER current density of −50 mA·cm
−2 at an overpotential of 250 mV vs. RHE, which is closer to that of the benchmark Pt/C (230 mV) shown in
Figure 4a.
Figure 4b provided a detailed comparison of the HER overpotentials of MoSe
2, MoSe
2-Gr, and the benchmark Pt/C. This comparison highlights the superior electrocatalytic performance of MoSe
2-Gr, exhibiting lower overpotentials compared to MoSe
2. Most importantly, the HER catalytic activity of the MoSe
2-Gr approached the performance of the benchmark Pt/C catalyst when the HER was performed at ca. −50 mA·cm
−2. Furthermore, the MoSe
2-Gr even exhibited a superior HER catalytic performance, particularly when the HER was performed at a current density higher than ca. −55 mA·cm
−2. Specifically, the MoSe
2-Gr achieved −150 mA·cm
−2 HER current density at an overpotential of 350 mV, while the benchmark Pt/C catalyst could achieve this current density only at a significantly higher HER overpotential of 591 mV (
Figure 4a,b).
The geometrical area-based current density discussed above in fact depends on the morphology of the deposited film in the given mass loading of the catalyst. Therefore, to examine the intrinsic catalytic activity of the catalysts toward HER, the mass activity-based LSV polarization curves of the MoSe
2, MoSe
2-Gr, and the benchmark Pt/C were measured and presented in
Figure S2a. In addition, the corresponding mass activity profile at a given HER overpotential was also displayed in
Figure S2b. In line with the geometrical area-based HER activity, the MoSe
2-Gr catalyst exhibited significantly higher mass activity toward HER compared to that of the as-prepared pristine MoSe
2 catalyst. Furthermore, the MoSe
2-Gr catalyst demonstrated a comparable mass activity to that of the Pt/C at an overpotential of 250 mV. Beyond this, the MoSe
2-Gr catalyst even exhibited superior mass activity toward the HER, indicating its excellent electrocatalytic performance. The Tafel slope, which is a fundamental parameter in evaluating the HER kinetics of electrocatalysts, was shown in
Figure 4c. The corresponding Tafel slopes of MoSe
2, MoSe
2-Gr, and Pt/C in a 0.5 M H
2SO
4 electrolyte were determined to be 95, 67, and 51 mV.dec
−1, respectively. A lower Tafel slope indicates that the HER took place at a faster rate on the surface of the MoSe
2-Gr catalyst-based cathode compared to that of the MoSe
2 catalyst. This finding suggests that the MoSe
2-Gr catalyst had an enhanced HER kinetic, further supporting its superior electrocatalytic performance. The electrochemical impedance spectroscopy (EIS) of MoSe
2 and MoSe
2-Gr was applied to gain insights into their electrochemical kinetics.
Figure 4d shows the Nyquist plots obtained from the EIS, which reveal that the charge transfer resistance for the HER process at the MoSe
2-Gr/electrolyte interface is significantly lower compared to that at the MoSe
2/electrolyte interface. This can be attributed to the expanded interlayer spacing of MoSe
2 and the presence of graphene in the MoSe
2-Gr catalyst. The hybrid structure of the MoSe
2-Gr catalyst has, thus, provided a higher surface area and more active sites, facilitating the movement of electrons and ions between the catalyst and electrolyte. This may be the key reason for an improved HER performance of the MoSe
2-Gr catalyst. Furthermore, the electrocatalytic HER performance of the MoSe
2-Gr catalyst in 0.5 M H
2SO
4 is highly competitive with other MoSe
2-based composites, as indicated in
Table S1. This demonstrates the favorable performance and potential applicability of the MoSe
2-Gr catalyst designed in this work for green hydron production via HER.
The MoSe
2-Gr electrode was subjected to a chronopotentiometry test at a cathodic bias of −10 mA.cm
−2 to evaluate its electrochemical stability. The resulting chronopotentiometry trace, as shown in
Figure 5a, indicates that the MoSe
2-Gr electrode exhibited excellent endurance over a period of 24 h. Furthermore, after the stability test for 24 h, a similar linear sweep voltammogram curve of the MoSe
2-Gr catalyst to that of the one before the stability test was recorded, as displayed in
Figure 5b. This finding confirms the long-term electrochemical stability of the MoSe
2-Gr catalyst during HER in acidic electrolytes. The electrode maintained its catalytic activity and performance even after the extended operation, highlighting its potential for practical applications requiring sustained electrochemical performance.