Hydrogen Evolution Reaction Property of Molybdenum Disulﬁde/Nickel Phosphide Hybrids in Alkaline Solution

: The hydrogen evolution reaction (HER) property of molybdenum disulﬁde (MoS 2 ) is undesirable because of the insufﬁcient active edge sites and the poor conductivity. To enhance HER performance of MoS 2 , nickel phosphide (Ni 2 P) was combined with this catalyst and three MoS 2 /Ni 2 P hybrids (38 wt % Ni 2 P addition for MoS 2 /Ni 2 P-38, 50 wt % Ni 2 P addition for MoS 2 /Ni 2 P-50, and 58 wt % Ni 2 P addition for MoS 2 /Ni 2 P-58) were fabricated via a hydrothermal synthesis process. Morphologies, crystallinities, chemical components, speciﬁc surface areas, and HER properties of the fabricated MoS 2 /Ni 2 P samples in an alkaline electrolyte were characterized and tested. In addition, the insight into the HER properties of as-prepared catalysts were revealed by the density functional theory (DFT) calculation. Additionally, the stabilities of pure MoS 2 , Ni 2 P, and MoS 2 /Ni 2 P-50 samples were evaluated. The results show that the addition of Ni 2 P can enhance the HER property of the MoS 2 catalyst. Although HER properties of the above-mentioned three MoS 2 /Ni 2 P hybrids are inferior to that of pure Ni 2 P, they are much higher than that of MoS 2 . Among as-prepared three hybrids, MoS 2 /Ni 2 P-50 exhibits the best HER performance, which may be due to its uniform morphology, large speciﬁc surface area, and excellent stability. The MoS 2 /Ni 2 P-50 hybrid shows a high cathodic current density (70 mA/cm 2 at − 0.48 V), small Tafel slope (~58 mV/decade), and a low charge transfer resistance (0.83 k Ω · cm 2 ). of


Introduction
With growing concerns about environmental pollution and energy crises resulting from overconsumption of coal and fossil fuels, the exploitation of renewable clean energies, such as solar energy, wind energy, hydraulic power, biological energy, fuel cell, and hydrogen energy come to the forefront [1][2][3]. Among the clean energies mentioned above, hydrogen energy is attracting ever-growing attention due to the convenient production and effective cost [4,5]. Apart from the traditional method through steam from fossil fuels, the process of electrochemical water splitting is considered as an alternative to produce hydrogen through the HER [6][7][8][9]. Until the present, the well-known platinum and platinum-based alloys show the best electrocatalytic performance for HER [10][11][12][13]. However, their high-cost and scarcity impede their wide applications in practice [14][15][16].

Synthesis of the Ni 2 P Nanosheet
Under vigorous stirring condition, 1.63 g red phosphorus and 1.25 g NiCl 2 ·6H 2 O were dissolved in 15 mL deionized water for 30 min and a slurry-like mixture was obtained. Then this mixture solution was transferred into a 25 mL Teflon-lined stainless steel autoclave and followed by placed into a muffle furnace (WRN-010, Eurasian, Tianjin, China), which was preheated to 180 • C. The hydrothermal synthesis process was carried out at 180 • C for 24 h. When the temperature of the muffle furnace was cooled to room temperature naturally, the formed particles were separated by centrifugation and washed with ethanol twice and deionized water three times. Lastly, the fabricated gray-black Ni 2 P powders were dried in a vacuum oven (DZF-6050, Boxun Industrial Co. Ltd., Shanghai, China) and retained for use.

Synthesis of the MoS 2 /Ni 2 P Hybrids
A total of 1.41 g (NH 4 ) 6 Mo 7 O 24 ·4H 2 O, 0.26 g CH 4 N 2 S and three different amounts of Ni 2 P (38 wt %, 50 wt %, and 58 wt %) were dispersed in 20 mL of distilled water. The mixture solution was vigorously stirred for 1 h at room temperature. After that, the solution was transferred into a 25 mL Teflon-lined stainless steel autoclave placed in a muffle furnace, and then reacted at 200 • C for 24 h. As the solution temperature was cooled to room temperature naturally, the resultant three black samples (MoS 2 /Ni 2 P-38, MoS 2 /Ni 2 P-50, and MoS 2 /Ni 2 P-58) were centrifuged and adequately washed with ethanol and deionized water. Finally, the as-synthesized samples were dried under a vacuum atmosphere at 60 • C for 12 h.

Characterization
Crystal structures of the synthesized MoS 2 /Ni 2 P samples were determined by an X-ray diffractometer (XRD, SmartLab, Tokyo, Japan) using Cu Kα radiation (λ = 1.5418 Å) from 5 to 100 angles at a scanning rate of 5 • /min. The morphologies of the samples were obtained using a scanning electron microscope (SEM, S-4800, Hitachi, Tokyo, Japan) at an accelerating voltage of 5 kV. The elemental compositions and chemical states of these three MoS 2 /Ni 2 P samples were characterized by X-ray photoelectron spectra (XPS, ESCALAB MK II, Thermo Fisher Scientific, Waltham, MA, USA) using Mg Kα as the excitation source. The Brunauer-Emmett-Teller (BET) specific surface areas of the obtained samples were examined by N 2 adsorption/desorption measurement on a Nova 4000e analyzer (Quantachrome Instrument, Boynton Beach, FL, USA) at 77 K.

Electrochemical Measurements
Before preparing the working electrode, a glassy carbon electrode (GCE) of 3 mm in diameter was firstly polished with 1000# water sandpaper and cleaned with ethanol and deionized water. Typically, 5 mg of catalyst (Ni 2 P, MoS 2 , MoS 2 /Ni 2 P hybrids, and the commercial 20 wt % Pt/C) and 30 µL Nafion solution (5 wt %) were dispersed in 1 mL solution consisting of 250 µL absolute ethanol and 750 µL deionized water, and followed by sonication for 1 h to form a homogeneous ink. Then, 5 µL of the dispersion solution was loaded onto the surface of the polished GCE and the electrode was dried at room temperature; herein, the excessive ethanol present in the slurry was removed as much as possible by extending the drying time. Prior to each electrochemical test, the electrolyte solution was degassed by bubbling pure nitrogen gas for 30 min to remove the dissolved oxygen.
All electrochemical tests were conducted in a typical three-electrode system attached to an electrochemical workstation (CHI 650C, Chenhua Co. Ltd., Shanghai, China) in 1.0 mol/L KOH electrolyte. The GCE modified by the MoS 2 /Ni 2 P hybrids (MoS 2 , Ni 2 P or Pt/C) acted as the working electrode, while a Ag/AgCl electrode and a platinum foil were employed as the reference and counter electrodes, respectively. The acquired potential values relevant to the Ag/AgCl electrode were converted to the reversible hydrogen electrode (RHE) scale: E RHE = E Ag/AgCl + 0.059 pH + 0.209 V, i.e., the value of potential throughout this manuscript is relative to RHE. Linear sweep voltammetry (LSV) was analyzed in the potential of −0.8 to 0.2 V at a scan rate of 2 mV/s. Tafel polarization curves were measured in a potential window of 0.3-0.65 V with a scan rate of 2 mV/s. Cyclic voltammetry (CV) tests were conducted in a potential window from −0.6 to 0.8 V at a scan rate of 10 mV/s. The electrochemical impedance spectroscopy (EIS) measurements were carried out at a cathodic overpotential of 0.7 V by employing the sinusoidal signal amplitude of 5 mV; the frequency ranged from 10 5 to 0.01 Hz.

Computational Details
The calculations based on DFT were implemented in the Materials Studio DMol 3 (version 7.0, Accelrys Inc., San Diego, CA, USA) [50,51]. Pulay's direct inversion in the iterative subspace (DIIS) technique, as well as double numerical plus polarization functions (DNP) and the revised Perdew-Burke-Ernzerhof (RPBE) functional were employed. The effective core potential was applied to treat the core electrons of nickel atoms. With respect to the aqueous HER process, a continuum solvation model (COSMO) was used and water with the dielectric constant of 78.54 as the solvent [52]. In addition, for the purpose of taking into account of the weak interactions (hydrogen bond and van der Waals force), the TS (Tkatchenko-Scheffler) method for DFT-D correction was employed [52].
As is well known, the basal plane of MoS 2 has been validated as chemically inert, thus, only the Mo edge-type structure of this catalyst was considered [30]. Three kinds of geometries for MoS 2 , Ni 2 P, and MoS 2 /Ni 2 P catalysts were employed: a (4 × 5 × 1) supercell with two S-Mo-S trilayer for MoS 2 , a (2 × 1 × 1) supercell derived from Ni 2 P (111) crystal surface with four layers, and a single (4 × 5 × 1)-sized trilayer MoS 2 combined by three (2 × 1 × 1)-sized Ni 2 P layers for MoS 2 /Ni 2 P. A vacuum with the thickness of 15 Å in the z-direction was used to separate neighboring slabs and to minimize the interactions between them. During the calculations, only the first layers of MoS 2 and Ni 2 P were relaxed in order to save time and computational cost.

X-ray Diffraction (XRD) Analysis
The XRD patterns of MoS 2 , Ni 2 P, and three MoS 2 /Ni 2 P hybrids (MoS 2 /Ni 2 P-38, MoS 2 /Ni 2 P-50, and MoS 2 /Ni 2 P-58) are displayed in Figure 1. The characteristics peaks at 2θ = 33.8 • and 57.1 • (curve 1) can be identified, corresponding to (100) and (110) crystal planes of MoS 2 , respectively. Herein, it should be noted that the diffraction peak of (002) plane of MoS 2 is undetected. The absence of the (002) diffraction peak of MoS 2 indicates a low stacking height along this direction [5,53] [54], and another weak peak at 2θ = 35.4 • can be assigned to the presence of Ni 12 P 5 phase [55]. XRD patterns of the MoS 2 /Ni 2 P hybrids are presented in curves 3-5. The diffraction patterns of MoS 2 /Ni 2 P hybrids consist of very weak diffraction peak of MoS 2 , which indicates that the MoS 2 is amorphous [5]. In addition, the diffraction peaks related to MoS 2 and Ni 2 P are observed, thus suggesting the successful combination between MoS 2 and Ni 2 P. It may be indexed to a proof of the fabrication of MoS 2 /Ni 2 P hybrid. Compared with pure Ni 2 P, the diffraction intensity of Ni 2 P crystal for three MoS 2 /Ni 2 P hybrids becomes weak. However, with the increase of Ni 2 P addition, intensities of diffraction peaks attributed to Ni 2 P become strong and those of MoS 2 show a descending trend, which indicates that Ni 2 P plays a major role in the crystallinity of the hybrids.

Scanning Electron Microscope (SEM)
The morphologies of MoS2, Ni2P, and three MoS2/Ni2P hybrids are given in Figure 2. As shown in Figure 2a, the nanoflower-like structure of MoS2 can be ascertained; of course, there are some nanosheets randomly stacking, which may lead to the decrease in active sites. The fabricated Ni2P particle exhibits a nanosheet-type morphology (Figure 2b), which would be advantageous to electrochemical conduction among the active sites. Morphologies of the three hybrids are observed in Figure 2c (MoS2/Ni2P-38), 2d (MoS2/Ni2P-50) and 2e (MoS2/Ni2P-58). All of them exhibit a mixed morphology of flower-like and lamellar structures. As for the SEM pictures of hybrids, with the increasing addition of Ni2P, the flower-shaped structure can be more easily observed, the presence of the Ni2P nanosheet will be helpful for the fabrication of MoS2 nanoflowers. For the sample of MoS2/Ni2P-38, the small amount of Ni2P cannot provide an abundant depositing area, thereby resulting in the serious agglomeration of MoS2 ( Figure  2c). For MoS2/Ni2P-50 sample (the mass ratio of MoS2 and Ni2P is 1:1), it can be observed that the petal structure and the plate-type structure uniformly coexist (Figure 2d), and more well-proportioned nanoflower-like MoS2 particles distribute on the surface of Ni2P. Whereas, for MoS2/Ni2P-58 ( Figure   Figure 1. XRD patterns of MoS 2 , Ni 2 P, and three MoS 2 /Ni 2 P hybrids.

Scanning Electron Microscope (SEM)
The morphologies of MoS 2 , Ni 2 P, and three MoS 2 /Ni 2 P hybrids are given in Figure 2. As shown in Figure 2a, the nanoflower-like structure of MoS 2 can be ascertained; of course, there are some nanosheets randomly stacking, which may lead to the decrease in active sites. The fabricated Ni 2 P particle exhibits a nanosheet-type morphology (Figure 2b), which would be advantageous to electrochemical conduction among the active sites. Morphologies of the three hybrids are observed in Figure 2c (MoS 2 /Ni 2 P-38), 2d (MoS 2 /Ni 2 P-50) and 2e (MoS 2 /Ni 2 P-58). All of them exhibit a mixed morphology of flower-like and lamellar structures.

Scanning Electron Microscope (SEM)
The morphologies of MoS2, Ni2P, and three MoS2/Ni2P hybrids are given in Figure 2. As shown in Figure 2a, the nanoflower-like structure of MoS2 can be ascertained; of course, there are some nanosheets randomly stacking, which may lead to the decrease in active sites. The fabricated Ni2P particle exhibits a nanosheet-type morphology (Figure 2b), which would be advantageous to electrochemical conduction among the active sites. Morphologies of the three hybrids are observed in Figure 2c (MoS2/Ni2P-38), 2d (MoS2/Ni2P-50) and 2e (MoS2/Ni2P-58). All of them exhibit a mixed morphology of flower-like and lamellar structures. As for the SEM pictures of hybrids, with the increasing addition of Ni2P, the flower-shaped structure can be more easily observed, the presence of the Ni2P nanosheet will be helpful for the fabrication of MoS2 nanoflowers. For the sample of MoS2/Ni2P-38, the small amount of Ni2P cannot provide an abundant depositing area, thereby resulting in the serious agglomeration of MoS2 ( Figure  2c). For MoS2/Ni2P-50 sample (the mass ratio of MoS2 and Ni2P is 1:1), it can be observed that the petal structure and the plate-type structure uniformly coexist (Figure 2d), and more well-proportioned As for the SEM pictures of hybrids, with the increasing addition of Ni 2 P, the flower-shaped structure can be more easily observed, the presence of the Ni 2 P nanosheet will be helpful for the fabrication of MoS 2 nanoflowers. For the sample of MoS 2 /Ni 2 P-38, the small amount of Ni 2 P cannot provide an abundant depositing area, thereby resulting in the serious agglomeration of MoS 2 (Figure 2c). For MoS 2 /Ni 2 P-50 sample (the mass ratio of MoS 2 and Ni 2 P is 1:1), it can be observed that the petal structure and the plate-type structure uniformly coexist (Figure 2d), and more well-proportioned nanoflower-like MoS 2 particles distribute on the surface of Ni 2 P. Whereas, for MoS 2 /Ni 2 P-58 (Figure 2e), with the large addition of Ni 2 P, it is explicit that the dense MoS 2 flower structure can be obtained, showing a resemblance to that of MoS 2 (Figure 2a). By a comparison of morphologies of the three hybrids, for the MoS 2 /Ni 2 P-50 sample, the MoS 2 nanoparticles evenly disperses on the surface of the Ni 2 P nanosheet.

X-ray Photoelectron Spectroscopy (XPS)
The elemental compositions and their valences of three MoS 2 /Ni 2 P hybrids were measured by XPS. The Mo 3d, S 2p, Ni 2p, P 2p and O 1s high-resolution spectra of three MoS 2 /Ni 2 P hybrids are respectively presented in Figure 3, and the corresponding bond energies and assignments are given in Table 1. As shown in Figure 3a, the characteristic peaks located at 229.0 eV and 231.9 eV (take MoS 2 /Ni 2 P-38 as an example, the following is the same) can be assigned to Mo 3d 5/2 and Mo 3d 3/2 respectively, suggesting the existence of Mo 4+ . The two peaks at 232.9 eV and 236.0 eV are ascribed to the compounds MoO 3 or MoO 4 2− (Mo 6+ ) due to the oxidation of the samples. In addition, the weak peak appearing at 226.4 eV corresponds to S 2s of MoS 2 [56]. For S 2p spectra, which is shown in Figure 3b, the two peaks at 161.8 eV and 163.3 eV are attributed to the 2p 3/2 and 2p 1/2 orbitals of divalent sulfide ion (S 2− ), while the other two peaks at 168.9 eV and 170.0 eV represent the existence of tetravalent sulfur in the form of SO 3 2− (S 4+ ). Generally, the sulphur atom with a +4 state locates at the edge of MoS 2 layered structure due to the oxidation of MoS 2 particle [57].

X-ray Photoelectron Spectroscopy (XPS)
The elemental compositions and their valences of three MoS2/Ni2P hybrids were measured by XPS. The Mo 3d, S 2p, Ni 2p, P 2p and O 1s high-resolution spectra of three MoS2/Ni2P hybrids are respectively presented in Figure 3, and the corresponding bond energies and assignments are given in Table 1. As shown in Figure 3a, the characteristic peaks located at 229.0 eV and 231.9 eV (take MoS2/Ni2P-38 as an example, the following is the same) can be assigned to Mo 3d5/2 and Mo 3d3/2 respectively, suggesting the existence of Mo 4+ . The two peaks at 232.9 eV and 236.0 eV are ascribed to the compounds MoO3 or MoO4 2− (Mo 6+ ) due to the oxidation of the samples. In addition, the weak peak appearing at 226.4 eV corresponds to S 2s of MoS2 [56]. For S 2p spectra, which is shown in Figure 3b, the two peaks at 161.8 eV and 163.3 eV are attributed to the 2p3/2 and 2p1/2 orbitals of divalent sulfide ion (S 2− ), while the other two peaks at 168.9 eV and 170.0 eV represent the existence of tetravalent sulfur in the form of SO3 2− (S 4+ ). Generally, the sulphur atom with a +4 state locates at the edge of MoS2 layered structure due to the oxidation of MoS2 particle [57]. As shown in Figure 3c, the three peaks observed at 857.2 eV, 858.9 eV and 862.8 eV may be assigned to Ni δ+ (0 < δ < 2) in Ni2P, oxidized Ni species (Ni 2+ ) and the satellite of Ni 2p3/2, respectively, As shown in Figure 3c, the three peaks observed at 857.2 eV, 858.9 eV and 862.8 eV may be assigned to Ni δ+ (0 < δ < 2) in Ni 2 P, oxidized Ni species (Ni 2+ ) and the satellite of Ni 2p 3/2 , respectively, while the other three peaks at 875.0 eV, 876.8 eV, and 881.2 eV can be indexed to Ni δ+ in Ni 2 P, oxidized Ni species, and the satellite of Ni 2p 1/2 [58]. As for the P 2p spectra (Figure 3d), the peak at 130.2 eV is mark of metal-P bonds in metal phosphides, the peak at 133.7 eV can be assumed to the oxidized P species because the samples are exposed to the air [59].
Additionally, the O 1s peak (Figure 3e), with the binding energies of 532.0 eV, indicates the existence of Mo(IV)-O bond [60]. Thus, to some extent, it is also confirmed that the three hybrids were oxidized during the preparation process. The specific surface areas of the as-synthesized pure MoS 2 , Ni 2 P, and three MoS 2 /Ni 2 P hybrids were characterized by N 2 adsorption/desorption isotherm measurement ( Figure 4). All curves reveal a type IV isotherm with a distinct hysteresis loop, indicating the mesoporous structures of the above samples. The specific surface areas of aforementioned three hybrids are in a trend: MoS 2 /Ni 2 P-50 (2.42 m 2 /g) > MoS 2 /Ni 2 P-58 (1.95 m 2 /g) > MoS 2 /Ni 2 P-38 (1.86 m 2 /g). The specific surface areas of these three hybrids are smaller than that of pure Ni 2 P (4.62 m 2 /g). For pure MoS 2 , although its specific surface area (7.04 m 2 /g) is larger than the other four samples, its basal surface is HER inert, so it cannot provide abundant active sites for hydrogen adsorption, which means an inferior HER property. Among the three hybrids, MoS 2 /Ni 2 P-50 possess the largest specific surface area, the HER performance of this hybrid will be more outstanding than the other two hybrids. However, its HER characteristic cannot be comparable to that of Ni 2 P, because of its smaller determined specific surface area (ca. 52.4% of that of Ni 2 P).

Linear Sweep Voltammetry (LSV)
The LSV curves of all samples in 1.0 mol/L KOH at a scan rate of 2 mV/s are depicted in Figure  5. The obtained onset overpotentials (ηonset) [61] are tabulated in Table 2. All samples were measured within the potential window of −0.8~0.2 V. The pure MoS2 shows the lower HER activity which is attributed to its inert basal surface and inferior conductivity [28]. By contrast, HER performances of the three hybrids are higher than pure MoS2 because of the excellent conductivity and affordable active sites. Although the catalytic performances of three MoS2/Ni2P hybrids are lower than those of Ni2P and commercial Pt/C, they can still raise attention in hydrogen evolution reaction. When the potential was kept at −0.1 V, the cathodic current density of HER for tested samples is in the following sequence: MoS2 (1.56 mA/cm 2 ) < MoS2/Ni2P-38 (1.86 mA/cm 2 ) < MoS2/Ni2P-58 (2.52 mA/cm 2 ) < MoS2/Ni2P-50 (3.53 mA/cm 2 ) < Ni2P (5.71 mA/cm 2 ) < Pt/C (8.57 mA/cm 2 ). Herein, it should be mentioned that although Ni2P shows a better HER feature than the produced three hybrids, its stability will be worse in an alkaline solution [14]. Among the three hybrids, MoS2/Ni2P-50 shows a high cathodic current density (70 mA/cm 2 at −0.48 V), thus exhibiting the best HER performance. The HER performance of this hybrid might be due to the excellent conductivity, as well as the coexistence of nanosheet-type Ni2P crystal on the bottom and nanoflower-like MoS2 on the surface. Without doubt, the incorporation of Ni2P brings a great enhancement in HER performance of MoS2.

Linear Sweep Voltammetry (LSV)
The LSV curves of all samples in 1.0 mol/L KOH at a scan rate of 2 mV/s are depicted in Figure 5. The obtained onset overpotentials (η onset ) [61] are tabulated in Table 2. All samples were measured within the potential window of −0.8~0.2 V. The pure MoS 2 shows the lower HER activity which is attributed to its inert basal surface and inferior conductivity [28]. By contrast, HER performances of the three hybrids are higher than pure MoS 2 because of the excellent conductivity and affordable active sites. Although the catalytic performances of three MoS 2 /Ni 2 P hybrids are lower than those of Ni 2 P and commercial Pt/C, they can still raise attention in hydrogen evolution reaction. When the potential was kept at −0.1 V, the cathodic current density of HER for tested samples is in the following sequence: MoS 2 (1.56 mA/cm 2 ) < MoS 2 /Ni 2 P-38 (1.86 mA/cm 2 ) < MoS 2 /Ni 2 P-58 (2.52 mA/cm 2 ) < MoS 2 /Ni 2 P-50 (3.53 mA/cm 2 ) < Ni 2 P (5.71 mA/cm 2 ) < Pt/C (8.57 mA/cm 2 ). Herein, it should be mentioned that although Ni 2 P shows a better HER feature than the produced three hybrids, its stability will be worse in an alkaline solution [14]. Among the three hybrids, MoS 2 /Ni 2 P-50 shows a high cathodic current density (70 mA/cm 2 at −0.48 V), thus exhibiting the best HER performance. The HER performance of this hybrid might be due to the excellent conductivity, as well as the coexistence of nanosheet-type Ni 2 P crystal on the bottom and nanoflower-like MoS 2 on the surface. Without doubt, the incorporation of Ni 2 P brings a great enhancement in HER performance of MoS 2 .

Linear Sweep Voltammetry (LSV)
The LSV curves of all samples in 1.0 mol/L KOH at a scan rate of 2 mV/s are depicted in Figure  5. The obtained onset overpotentials (ηonset) [61] are tabulated in Table 2. All samples were measured within the potential window of −0.8~0.2 V. The pure MoS2 shows the lower HER activity which is attributed to its inert basal surface and inferior conductivity [28]. By contrast, HER performances of the three hybrids are higher than pure MoS2 because of the excellent conductivity and affordable active sites. Although the catalytic performances of three MoS2/Ni2P hybrids are lower than those of Ni2P and commercial Pt/C, they can still raise attention in hydrogen evolution reaction. When the potential was kept at −0.1 V, the cathodic current density of HER for tested samples is in the following sequence: MoS2 (1.56 mA/cm 2 ) < MoS2/Ni2P-38 (1.86 mA/cm 2 ) < MoS2/Ni2P-58 (2.52 mA/cm 2 ) < MoS2/Ni2P-50 (3.53 mA/cm 2 ) < Ni2P (5.71 mA/cm 2 ) < Pt/C (8.57 mA/cm 2 ). Herein, it should be mentioned that although Ni2P shows a better HER feature than the produced three hybrids, its stability will be worse in an alkaline solution [14]. Among the three hybrids, MoS2/Ni2P-50 shows a high cathodic current density (70 mA/cm 2 at −0.48 V), thus exhibiting the best HER performance. The HER performance of this hybrid might be due to the excellent conductivity, as well as the coexistence of nanosheet-type Ni2P crystal on the bottom and nanoflower-like MoS2 on the surface. Without doubt, the incorporation of Ni2P brings a great enhancement in HER performance of MoS2.

Tafel Polarization
Equation (1) was employed to describe the relationship of overpotential (η, V) and current density (j, mA/cm 2 ); the charge transfer coefficient (α) can be obtained by Equation (2) [62]: where b and R are the Tafel slope (mV/decade) and gas constant (8.314 J/(K·mol)), respectively. A is an analyzed constant, T is the absolute temperature, m e is number of electrons transferred, and F is the Faraday constant (96,485 C/mol). According to published studies [12,13], the HER in alkaline solutions is typically considered as the Volmer-Heyrovsky process or Volmer-Tafel pathways, they are depicted as follows: The kinetic mechanism in the HER process is determined by the rate-determining step (rds) of a multi-step reaction. The Tafel equation can play an important role in estimating the kinetic mechanism. The parameter of α corresponds to the rds for multi-step reactions, this is depicted as follows: the value of α is~0.5 when the rds is Volmer reaction step, the value of α is~1.5 when the rds is Heyrovsky reaction step, and the value of α will be close to 2 when the rds is step Tafel reaction [63,64]. Additionally, the Tafel slope is an important factor to describe the HER rate by examining the change of current density with overpotential. Generally, a small Tafel slope is desirable to drive a large catalytic current density at low overpotentials [65]. The Tafel plots of these samples derived from the polarization curves are shown in Figure 6. The HER performance of MoS 2 is inferior due to its large Tafel slope (~128 mV/decade) and small α (0.46). Within three MoS 2 /Ni 2 P hybrids, MoS 2 /Ni 2 P-50 has a minimum Tafel slope (~58 mV/decade), indicative of the excellent HER performance. It might be a substitute to noble metals in HER application.

Cyclic Voltammetry (CV)
Cyclic voltammetry (CV) measurement (in a potential window of −0.6~0.8 V) can be helpful to reveal the reversibility of the electrochemical reaction. CV curves of MoS2/Ni2P-38, MoS2/Ni2P-50 and MoS2/Ni2P-58 samples are exhibited in Figure 7; the inset shows those of MoS2/Ni2P-50, pure MoS2, pure Ni2P and Pt/C. It can be seen from the determined CV curves that each of them is composed of an anodic oxidation and a cathodic reduction peak. The quasi-reversible redox peaks may be related to the processes of electrochemical hydrogen adsorption and electrochemical hydrogen desorption [28]. For the three hybrids, cathodic reduction potentials of them are as follows: −0.21 V for MoS2/Ni2P-38, −0.11 V for MoS2/Ni2P-50, and −0.15 V for MoS2/Ni2P-58. In addition, the current density follows the same tendency. Thus, a positive potential and the large current density for MoS2/Ni2P-50 composite can be validated, inferring its higher HER property than those of the other two hybrids.

Cyclic Voltammetry (CV)
Cyclic voltammetry (CV) measurement (in a potential window of −0.6~0.8 V) can be helpful to reveal the reversibility of the electrochemical reaction. CV curves of MoS 2 /Ni 2 P-38, MoS 2 /Ni 2 P-50 and MoS 2 /Ni 2 P-58 samples are exhibited in Figure 7; the inset shows those of MoS 2 /Ni 2 P-50, pure MoS 2 , pure Ni 2 P and Pt/C. It can be seen from the determined CV curves that each of them is composed of an anodic oxidation and a cathodic reduction peak. The quasi-reversible redox peaks may be related to the processes of electrochemical hydrogen adsorption and electrochemical hydrogen desorption [28]. For the three hybrids, cathodic reduction potentials of them are as follows: −0.21 V for MoS 2 /Ni 2 P-38, −0.11 V for MoS 2 /Ni 2 P-50, and −0.15 V for MoS 2 /Ni 2 P-58. In addition, the current density follows the same tendency. Thus, a positive potential and the large current density for MoS 2 /Ni 2 P-50 composite can be validated, inferring its higher HER property than those of the other two hybrids.

Cyclic Voltammetry (CV)
Cyclic voltammetry (CV) measurement (in a potential window of −0.6~0.8 V) can be helpful to reveal the reversibility of the electrochemical reaction. CV curves of MoS2/Ni2P-38, MoS2/Ni2P-50 and MoS2/Ni2P-58 samples are exhibited in Figure 7; the inset shows those of MoS2/Ni2P-50, pure MoS2, pure Ni2P and Pt/C. It can be seen from the determined CV curves that each of them is composed of an anodic oxidation and a cathodic reduction peak. The quasi-reversible redox peaks may be related to the processes of electrochemical hydrogen adsorption and electrochemical hydrogen desorption [28]. For the three hybrids, cathodic reduction potentials of them are as follows: −0.21 V for MoS2/Ni2P-38, −0.11 V for MoS2/Ni2P-50, and −0.15 V for MoS2/Ni2P-58. In addition, the current density follows the same tendency. Thus, a positive potential and the large current density for MoS2/Ni2P-50 composite can be validated, inferring its higher HER property than those of the other two hybrids.

Electrochemical Impedance Spectroscopy (EIS)
The obtained Nyquist plots for the pure MoS 2 , Ni 2 P and three MoS 2 /Ni 2 P samples are shown in Figure 8. The R s (CPER ct ) equivalent circuit was adopted to analyze the obtained EIS data by use of Zsimpwin software. Where R s (Ω·cm 2 ) and R ct (kΩ·cm 2 ) are the solution resistance and charge transfer resistance. The double layer capacitance (C dl ) and the exchange current density (j 0 ) of the electrode are calculated by Equations (6) and (7) [6,66,67]: Herein, Q is the capacitance coefficient (Ω −1 ·s n ·cm −2 ) of constant phase element (CPE), and n is the phase angle of constant phase element. R is the gas constant (J/(K·mol)), T is the absolute temperature (K), m e is the number of switched electrons (the parameter of m e is the same as that in Equation (2)) and F is the Faraday constant. The calculated values of them are also reported in Table 3.
Of all the tested samples, it is no doubt that Pt/C, with the feature of the largest C dl and j 0 , exhibits the best HER performance. Apart from the commercial Pt/C specimen, among other five lab-made samples, pure Ni 2 P has the largest C dl and j 0 , followed by the three hybrids and, lastly, by the pure MoS 2 . From the above results, it seems that Ni 2 P is an ideal HER catalyst, however, the demerit of its instability in alkaline solution will retard its application. Although C dl and j 0 of the three hybrids are smaller than Ni 2 P, they are much larger than that of MoS 2 . Among these catalysts, the HER performance of MoS 2 /Ni 2 P-50 is remarkable. The obtained Nyquist plots for the pure MoS2, Ni2P and three MoS2/Ni2P samples are shown in Figure 8. The Rs (CPERct) equivalent circuit was adopted to analyze the obtained EIS data by use of Zsimpwin software. Where Rs (Ω·cm 2 ) and Rct (kΩ·cm 2 ) are the solution resistance and charge transfer resistance. The double layer capacitance (Cdl) and the exchange current density (j0) of the electrode are calculated by Equations (6) and (7) Herein, Q is the capacitance coefficient (Ω −1 ·s n ·cm −2 ) of constant phase element (CPE), and n is the phase angle of constant phase element. R is the gas constant (J/(K·mol)), T is the absolute temperature (K), me is the number of switched electrons (the parameter of me is the same as that in Equation (2)) and F is the Faraday constant. The calculated values of them are also reported in Table 3.
Of all the tested samples, it is no doubt that Pt/C, with the feature of the largest Cdl and j0, exhibits the best HER performance. Apart from the commercial Pt/C specimen, among other five lab-made samples, pure Ni2P has the largest Cdl and j0, followed by the three hybrids and, lastly, by the pure MoS2. From the above results, it seems that Ni2P is an ideal HER catalyst, however, the demerit of its instability in alkaline solution will retard its application. Although Cdl and j0 of the three hybrids are smaller than Ni2P, they are much larger than that of MoS2. Among these catalysts, the HER performance of MoS2/Ni2P-50 is remarkable.

DFT Calculations
The optimized structures of the three catalysts of MoS 2 , Ni 2 P, and MoS 2 /Ni 2 P without hydrogen adsorption, as well as those with one and two adsorbed hydrogen atoms are shown in Figure 9. The adsorption energy (∆E ads ) and Gibbs free energy of adsorption (∆G ads ) for the uptake of hydrogen by the catalyst at 298 K were calculated using Equations (8) and (9) [30].
where E(catalyst + xH, x = 1 or 2) is the COSMO-corrected total energy for the system containing the catalyst support and x adsorbed hydrogen atoms, E(catalyst + (x − 1)H) is that of for (x − 1) adsorbed hydrogen atoms-including system, and E(H 2 ) is the energy of an isolated hydrogen molecule. G(X) is the computed temperature-corrected free energy of the aforesaid species (X) at 298 K. ∆E ads shows the absorption capability catalyst toward the hydrogen atom; the more negative value of ∆E ads suggests the stronger interaction between them, and the geometry with the most negative ∆E ads will be more stable. The negative value of ∆G ads convinces the spontaneous characteristic of the adsorption interaction process. As for pristine MoS 2 , there are six edge sites (identified by numbers in Figure 9a) that can absorb the hydrogen atom. After one hydrogen atom capture, ∆E ads (MoS 2 + H) values are as follows: −16.12 kcal/mol for site 1, −16.07 for site 2, −16.08 for site 3, −12.91 for site 4, −12.98 for site 5, and −12.97 kcal/mol for site 6. Therefore, among these six sites, site 1 is more suitable than others for the anchor of hydrogen atom (Figure 9b). When one hydrogen atom is absorbed, the negative ∆G ads of all six hydrogen-containing configurations (−13.47, −13.34, −13.35, −9.55, −9.67, and −9.61 kcal/mol for hydrogen atom at sites 1-6) indicates the spontaneous characteristic of the hydrogen uptake. After one hydrogen atom adsorption, there are only five edge sites (2-6) for trapping another hydrogen atom. Among geometries for the two hydrogen atoms, the structure with ∆E ads of −14.78 kcal/mol for MoS 2 with sites 1 and 3 for the capture of hydrogen atoms (Figure 9c) is most stable; ∆E ads for others are: −13.52 kcal/mol for sites 1 and 2, −2.45 kcal/mol for sites 1 and 4, −12.37 kcal/mol for sites 1 and 5, and −13.19 kcal/mol for sites 1 and 6. The values of ∆G ads of these five compounds are negative, also inferring the easy uptake of the second hydrogen atom. Of course, by comparison of ∆E ads , the capture of the first hydrogen atom is easier than that of the second one.
For the case of hydrogen adsorption by Ni 2 P, all four sites (labeled by numbers in Figure 9d) are employed to trap hydrogen atoms. Site 3 is desirable for the capture of the first hydrogen atom (Figure 9e), and the value ∆E ads for the hydrogen-capturing compound is −6.095 kcal/mol; those for the other three compounds are −1.658 kcal/mol at site 1, −4.889 kcal/mol at site 2, and −4.898 kcal/mol at site 4. Values of ∆G ads for hydrogen trapper at sites 1-4 are −0.235 kcal/mol, −4.037 kcal/mol, −4.045 kcal/mol and −4.036 kcal/mol, respectively. As for the three systems containing two hydrogens, ∆E ads and ∆G ads of them are −0.31 and 1.74 kcal/mol for sites 3 and 1, 1.78 and 0.33 kcal/mol for sites 3 and 2, and −4.19 and −2.23 kcal/mol for sites 3 and 4. Therefore, the Ni 2 P catalyst tends to use sites 3 and 4 to trap hydrogen atoms (Figure 9f).
Lastly, for the hydrogen capture by the MoS 2 /Ni 2 P hybrid, among the six edge sites of top MoS 2 layer (Figure 9g), which is dissimilar to that of the pure MoS 2 , site 5 is more competent for trapping the first hydrogen atom (Figure 9h), in view of ∆E ads and ∆G ads as follows: −4.79 and −3.57 kcal/mol for site 1, 9.15 and 11.25 kcal/mol for site 2, −4.89 and −4.83 kcal/mol for site 3, −4.99 and −4.64 kcal/mol for site 4, −6.78 and −8.36 kcal/mol for site 5, and −5.2 and −6.87 kcal/mol for site 6. Additionally, for the adsorption of two hydrogen atoms, sites 5 and 3 (Figure 9i) are the most acceptable, because of the following values of ∆E ads and ∆G ads for the second hydrogen uptake: −3.81 and −1.88 kcal/mol for site 1, −3.56 and −1.77 kcal/mol for site 2, −6.30 and −2.73 kcal/mol for site 3, −6.29 and −2.58 kcal/mol for site 4, 7.32 and 11.90 kcal/mol for site 6. Furthermore, it should be mentioned that for S atoms at the sites 4 and 6, the distances between which and the second hydrogen atom are extended, and this hydrogen atom tends to close to the S atoms at sites 3 and 5.
In terms of the DFT calculations results, for the first and second hydrogen uptakes, ∆E ads and ∆G ads of MoS 2 are the most negative, followed by MoS 2 /Ni 2 P and Ni 2 P. Compared with that of MoS 2 , affinities of MoS 2 /Ni 2 P and Ni 2 P to hydrogen neither too strong nor too weak, indicating its excellent HER performance [51,68]. In addition, the ∆G ads of Ni 2 P that capturing hydrogen atoms is more close to zero than those of the other two systems, it is sure that the HER property of Ni 2 P is higher than those of MoS 2 and MoS 2 /Ni 2 P hybrids [51,69]. Thus, it can be concluded that the HER performances for these catalysts are in an order of Ni 2 P > MoS 2 /Ni 2 P > MoS 2 . This deduction is consistent with the experimental tests. Herein, the combination between MoS 2 and Ni 2 P, and the role of Ni 2 P in the enhancement of MoS 2 HER property will be investigated in a further study. the sites 4 and 6, the distances between which and the second hydrogen atom are extended, and this hydrogen atom tends to close to the S atoms at sites 3 and 5.
In terms of the DFT calculations results, for the first and second hydrogen uptakes, ΔEads and ΔGads of MoS2 are the most negative, followed by MoS2/Ni2P and Ni2P. Compared with that of MoS2, affinities of MoS2/Ni2P and Ni2P to hydrogen neither too strong nor too weak, indicating its excellent HER performance [51,68]. In addition, the ΔGads of Ni2P that capturing hydrogen atoms is more close to zero than those of the other two systems, it is sure that the HER property of Ni2P is higher than those of MoS2 and MoS2/Ni2P hybrids [51,69]. Thus, it can be concluded that the HER performances for these catalysts are in an order of Ni2P > MoS2/Ni2P > MoS2. This deduction is consistent with the experimental tests. Herein, the combination between MoS2 and Ni2P, and the role of Ni2P in the enhancement of MoS2 HER property will be investigated in a further study.

HER Stability of the MoS2/Ni2P Hybrids
For practical applications, in addition to the HER activity mentioned above, the stability of electrocatalysts is another important criterion to evaluate the catalytic activity. To assess the durability of pure MoS2, Ni2P, and the MoS2/Ni2P-50 hybrid, continuous CV tests for 1000 cycles with the potential in range of −1.0 to 0.4 V were conducted at a scan rate of 50 mV/s in 1.0 mol/L KOH ( Figure 10). By comparison of polarization curves of these three samples before and after 1000 cycles, the cathodic current densities of them decrease somewhat after tests. Compared with electrocatalytic behaviors of pure MoS2 and Ni2P, the MoS2/Ni2P-50 hybrid just shows a slight decay, and thereby exhibits an excellent durability in the HER process.

HER Stability of the MoS 2 /Ni 2 P Hybrids
For practical applications, in addition to the HER activity mentioned above, the stability of electrocatalysts is another important criterion to evaluate the catalytic activity. To assess the durability of pure MoS 2 , Ni 2 P, and the MoS 2 /Ni 2 P-50 hybrid, continuous CV tests for 1000 cycles with the potential in range of −1.0 to 0.4 V were conducted at a scan rate of 50 mV/s in 1.0 mol/L KOH ( Figure 10). By comparison of polarization curves of these three samples before and after 1000 cycles, the cathodic current densities of them decrease somewhat after tests. Compared with electrocatalytic behaviors of pure MoS 2 and Ni 2 P, the MoS 2 /Ni 2 P-50 hybrid just shows a slight decay, and thereby exhibits an excellent durability in the HER process.

Conclusions
In this work, morphologies, crystallinities, and chemical components of pure MoS2, Ni2P, and three MoS2/Ni2P hybrids obtained via a hydrothermal synthesis process were characterized. Then, HER performances of them in 1.0 mol/L KOH solution were evaluated. The main conclusions are summarized as follows: (1) The result of DFT calculation is in consistence with that of experimental test. The incorporated Ni2P observably enhances the HER property of MoS2. Among the three fabricated kinds of catalysts, the HER performance of them follows the trend: Ni2P > MoS2/Ni2P > MoS2. (2) The MoS2/Ni2P-50 shows a large cathodic current density (70 mA/cm 2 at −0.48 V) and small Tafel slope (~58 mV/decade), thus exhibiting the higher HER activity than other two MoS2/Ni2P hybrids. (3) The excellent HER performance of MoS2/Ni2P-50 hybrid can be due to the desirable conductivity, the uniform morphology, large specific surface area, and favorable stability.

Conclusions
In this work, morphologies, crystallinities, and chemical components of pure MoS 2 , Ni 2 P, and three MoS 2 /Ni 2 P hybrids obtained via a hydrothermal synthesis process were characterized. Then, HER performances of them in 1.0 mol/L KOH solution were evaluated. The main conclusions are summarized as follows: (1) The result of DFT calculation is in consistence with that of experimental test. The incorporated Ni 2 P observably enhances the HER property of MoS 2 . Among the three fabricated kinds of catalysts, the HER performance of them follows the trend: Ni 2 P > MoS 2 /Ni 2 P > MoS 2 . (2) The MoS 2 /Ni 2 P-50 shows a large cathodic current density (70 mA/cm 2 at −0.48 V) and small Tafel slope (~58 mV/decade), thus exhibiting the higher HER activity than other two MoS 2 /Ni 2 P hybrids. (3) The excellent HER performance of MoS 2 /Ni 2 P-50 hybrid can be due to the desirable conductivity, the uniform morphology, large specific surface area, and favorable stability.