As the installed capacity of wind and light power soars, more and more problems are gradually emerging [1,2]. Especially, the characteristics of randomness and intermittency of wind and light power result in the generations of large-scale uncontrolled low-quality power and the serious abandonment problem [3,4]. Hydrogen energy is recognized as a clean energy carrier with high energy density, high conversion efficiency, and zero carbon emissions [5,6]. The conversion of wind and light energy into usable hydrogen energy by electrochemical water splitting becomes an attractive and profitable way to the maximized and multi-channel utilizations of renewable energy. However, the benchmark electrocatalysts for hydrogen evolution reaction (HER) are still noble platinum-based materials and the future industrial applications are severely discouraged by their scarcity and prohibitive cost. The development of earth-abundant alternatives with delightful performance is consequentially desirable. Moreover, considering pH fluctuations under operating conditions as well as the ease and cost of electrocatalyst production, the HER electrocatalysts that are versatile for different pH scenarios could be more important [7-9]. Therefore, the physical and electronic structure of electrocatalyst materials should be elaborately designed to simultaneously fulfill the adsorption/desorption/dissociation processes of reactant (H₂O)/intermediate (OH⁻ and H*) species involved in electrochemical water splitting, which is still challenging.

Transition metal sulfides (TMSs) are found as the key structural/component motif in nature nitrogenase and hydrogenase [10-12]. The corresponding exploitations of low-cost TMSs-based electrocatalyst have been extensively expanded. The performance improvement strategies including intrinsic and extrinsic interface/surface/composition modulation and morphology control have been executed for layered WS₂ and MoS₂ as well as nonlayered Zn/Co/Cu/Fe/Ni_xS_y [13-22]. However, the inherent shortcomings of TMSs in mediating the HER should be overcame with purpose. First, metallic conductivity is a prerequisite for electrical energy utilization. However, most of TMSs exhibited semiconductor-like high resistance in electrochemical application due to the presence of a decent band gap [23-25]. Second, taking layered WS₂/MoS₂ as examples, the theoretical and experimental results demonstrated that the edge (basal plane) of WS₂/MoS₂ are (not) active sites for HER [1,26-28]. Unfortunately, the percentage of the edge sites in the whole WS₂/MoS₂ layer is extremely limited. Thus, the cooperative manipulations of the accessibility of the edge sites and the activation of the basal plane sites are effective pathways to promote the HER activity, which are highly dependent on the structure design of TMSs. Third, the HER electrocatalysts with high tolerance to the pH of electrolyte are desirable from the perspectives of technological cost, efficiency, and working condition [29,30]. However, it is usually assumed that TMSs do not have the ability to dissociate water in neutral and alkaline media, which may explain why TMSs have been rarely reported for HER among all pH [31-34]. Moreover, following the widely established criterion, i.e., ΔG_H* (the free energy of H adsorption) [35], the pure TMS generally showed a poor ΔG_H* far from the ideal zero value of the baseline Pt [36-39]. Taken together, the issues identified above pose systemic challenges for the physical and chemical modifications of TMSs aiming at developing high-performance and multi-functional HER electrocatalysts.

With these points in mind, we elaborately manipulate the sequential hydrolysis of W and Ni sources to construct super-hybrid metal sulfide nanoarray comprising P doped WS₂ nanosheets, NiS and Ni₃S₄ nanoparticles. The super-hybrid nanosheet arrays exhibited abundant plane- and edge-type WS_2nullNiS and WS_2nullNi₃S₄ heterointerfaces, which were proved to be robust for electrocatalytic HER under all-pH electrolytes. The density functional theory (DFT) result indicated that WS₂-plane-related heterointerfaces could be more active for water dissociation and hydrogen adsorption, thus leading to the activity promotion. Moreover, the open super-hybrid nanosheet array and the P doping enable the unobstructed active heterointerfaces and facilitate the electron transfer as well as the mass transport during HER. This study presents an effective strategy for arming TMS-based materials in electrocatalysis application.

The schematic illustration for the synthesis procedure was given in Fig. 1. Recently, many instances have reported that POMs with precise cluster structures and convenient preparation are promising precursors for the preparation of functional materials at an atomic/molecular scale. POMs can not only supply abundant transition metals and the inherent heteroatoms doping, but also display controllable dissociation kinetics for constructing functional materials. The synergy of POMs and other metal precursors is highly valuable and provides an emerging vision for advanced catalyst materials. The super-hybrid transition metal sulfide nanoarrays of NiS nanoparticle/WS₂ nanosheet/Ni₃S₄ nanoparticle (Super-NiS/WS₂/Ni₃S₄) with high spatial ordering and abundant plane- and edge-type WS_2nullNiS and WS_2nullNi₃S₄ heterointerfaces were elaborately constructed though one-pot hydrothermal sulfidation. Typically, Ni foam was applied as both Ni source and three-dimensional hard template. Polyoxometalates (phosphotungstic acid, PW₁₂) functioned as W and P precursors. The synthetic kinetics were modulated by the optimized reaction concentration and temperature. As a result, the Ni and W sources were sequentially dissociated to form WS₂ nanosheet-based array and NiS/Ni₃S₄ nanoparticle.

The XRD pattern of Super-NiS/WS₂/Ni₃S₄ is given in Fig. 2a. The signal peaks of initial nickel foam disappear (Fig. S1 in Supporting information), indicating the total transformation of nickel foam into nickel sulfide after hydrothermal reaction. The peaks at 2θ = 14.3°, 33.4°, and 39.5° indicated the presence of (002), (101), and (103) planes in hexagonal WS₂ (JCPDS card No. 02–0331). The observable signals at 2θ = 32.3°, 35.9°, 40.6°, and 48.9° are well matched with the (300), (021), (211), and (131) planes of NiS (JCPDS card No. 02–1443). The diffraction peaks presented at 2θ = 31.3°, 37.9 ^o, 49.9°, and 54.8° are readily indexed to the (113), (004), (115) and (044) planes, respectively, of cubic Ni₃S₄ (JCPDS card No. 43–1469). The sample prepared without PW₁₂ was used as a reference to analyze the role of WS₂, and its XRD pattern indicated the co-existence of NiS/Ni₃S₄ (Fig. S2 in Supporting information). In order to compare the structural merits, the normal NiS/WS₂/Ni₃S₄ was also prepared with Na₂WO₄ as W source, as indicated in Fig. S3 (Supporting information). The highly uniform surface structure of Super-NiS/WS₂/Ni₃S₄ was clearly recorded by scanning electron microscopy (SEM) images (Figs. 2b–f). Strictly speaking, the Super-NiS/WS₂/Ni₃S₄ presented nanosheet-based 'rod-like' array with uniform spatial distribution throughout the three-dimension framework (Figs. 2b and c). Each 'rod' is composed of countless vertically growing WS₂ nanosheets and exhibits a size of 1.0 µm in width and 10 µm in height (Fig. 2d). The soft feature of WS₂ nanosheets tends to form porosity, which can provide abundant accessibility for mass transport in reactions. High-magnification SEM image demonstrated the existence of spreading NiS/Ni₃S₄ nanoparticles decorated on the surface of WS₂ nanosheets (Figs. 2e and f). The visible scene implies the strong coupling among the WS₂ nanosheets. The intimate conjunction between WS₂ nanosheets and the NiS/Ni₃S₄ is expected. In addition, this 'rod-like' nanosheet array also contribute the complete exposure of active planes and edges in Super-NiS/WS₂/Ni₃S. The transmission electron microscopy (TEM) image was used to give subtle structure of Super-NiS/WS₂/Ni₃S₄. The low-magnification TEM image (Fig. 2g) verified the composition of hierarchical cross-linked nanosheets and the surface-modified nanoparticles in targeted sample. The high-resolution TEM images indicated the presence of nanoparticles on the plane and the edge of nanosheet (Fig. 2h). The nanosheet has a lattice distance of 0.62 nm and about 5 layers in thickness, verifying the feature of laminar WS₂. The lattice spacing of 0.33 nm, 0.29 nm and 0.24 nm are well matched with the (022), (101) and (220) facets of Ni₃S₄ and NiS, respectively. The scanning transmission electron microscopy (STEM) image (Fig. 2i) further confirmed the 'rod-like' nanosheet assembly structure. The elemental mapping image from energy-dispersive X-ray (EDX) spectroscopy discloses the distribution of W, Ni, S, and P elements in Super-NiS/WS₂/Ni₃S₄. Specifically, a close observation of EDX mapping in Fig. 2j indicated that the mapping of P and W elements are matched well. The result implies that P element from phosphotungstic acid (PW₁₂) should be retained and mainly doped in WS₂ nanosheet with 0.29 wt% content (Fig. S4 in Supporting information). The promotion of intrinsic conductivity of WS₂ nanosheet can be predicted owing to the P doping. The aforementioned images demonstrate that the constructed Super-NiS/WS₂/Ni₃S₄ has the characteristic of elaborate heterogeneous interfaces comprising Ni₃S₄ and NiS nanoparticles-decorated plane and edge of hierarchical WS₂ nanosheet. The observations are also consistent the surface texture observed in the SEM image (Fig. 2c) and the XRD signals. The reference NiS/Ni₃S₄ is only evolved from nickel foam and shows unromantic porous skeleton morphology (Fig. S5 in Supporting information). The NiS/WS₂/Ni₃S₄ prepared with Na₂WO₄ as W source exhibited flower-like nanosheet structure (Fig. S6 in Supporting information). By contrast, the PW₁₂ as precursor demonstrated a special role in modulating the formation, growth, and assembly behaviors of WS₂ nanosheets, thus mediating the sequential construction of super-hybrid NiS nanoparticle/WS₂ nanosheet/Ni₃S₄ nanoparticle structure.

The compositions and chemical valences of Super-NiS/WS₂/Ni₃S₄ (shorted as Super-NiWSP), NiS/WS₂/Ni₃S₄ (shorted as NiWS), and NiS/Ni₃S₄ (shorted as NiS) were analyzed by X-ray photoelectron spectroscopy (XPS). The XPS spectra of Ni 2p_3/2 in Super-NiWSP, NiWS, and NiS were given in Fig. 3a. The peaks located at 853.2, 856.9, and 862.0 eV are consistent with Ni²⁺, Ni³⁺, and satellite signals, respectively [40-42]. The S 2p spectrum in Fig. 3b displayed two spin-orbit doublets of 2p_3/2 and 2p_1/2 at 161.8 eV and 163.5 eV, respectively, corresponding to the Ni-S and W-S chemical bonds [43-45]. As shown in Fig. 3c, multiplet peaks are clearly observed for XPS spectrum of W 4f between 30 eV and 40 eV The W 4f signal can be deconvoluted into WS₂ (31.8 and 33.9 eV), WS₃ (35.7 eV) and WO₃ (37.9 eV) peaks [43,46,47]. The existence of P doping in Super-NiWSP was evidenced by the signal in Fig. 3d [48-50]. The XPS result indicated P content of 2.85 at%. The EDX and XPS analyses suggested the P doping in the resultant Super-NiWSP. The XPS peak positions among Super-NiWSP and the reference NiWS and NiS are analyzed in order to obtain important information about the interface interaction. The peak positions of Ni 2p, W 4f, and S 2p in Super-NiWSP and NiWS showed obscure shift, indicating that the retained P doping in Super-NiWSP leaded to negligible change in chemical valences of Ni, W, and S in comparison with NiWS. Therefore, the performance differences between Super-NiWSP and NiWS described later should origin from the differences in intrinsic structures. Compared with pure NiS sample, the binding energy of Ni 2p_3/2 in Super-NiWSP is up-shifted by about 0.9 eV The metals with higher valence were supposed to be active for water dissociation [1]. On the contrary, the S peak (161.8 eV) in Super-NiWSP exhibited a blue shift by 0.5 eV in comparison with 162.3 eV of NiS. The finding suggested that the existences of notable electron transfer from NiS/Ni₃S₄ side to WS₂ and the strong electronic interactions between WS₂ and NiS/Ni₃S₄, implying highly coupled interface in Super-NiWSP. The frontier orbital states of TMS are expected to be well manipulated towards favorable adsorption modulation for reaction intermediates [1,4].

The electrocatalytic HER activities of super-NiWSP, NiWSP, and NiS were estimated under all pH (pH 0, 7, and 14) electrolytes. The linear sweep voltammetry (LSV) curves in alkaline, acid, and neutral media were provided in Figs. 4a–c, respectively. The benchmark Pt/C displayed the lowest overpotential of 26/37/68 mV to actuate the current density of 10 mA/cm² for hydrogen evolution in alkaline/acid/neutral electrolytes. The reference NiWS and NiS needed corresponding 115/141/221 mV and 152/185/318 mV overpotentials at 10 mA/cm² to realize HER. By contrast, the Super-NiWSP exhibited a dramatic promotion for HER activity, which only requires 57/95/151 mV overpotentials at 10 mA/cm² to drive hydrogen evolution under alkaline/acid/neutral environments. Moreover, the activity of Super-NiWSP is also comparable to some of the recently reported data from TMSs-based electrocatalysts [1,4] in alkaline electrolyte (η@10 mA/cm², edge-rich WS₂–136 mV, Ni₃S₂–223 mV, NiS₂–148 mV, CoS₂–193 mV, Ni_xCo_3-xS_4nullNi₃S₂–136 mV, MoS₂/Ni₃S₂–110 mV, Mo_(1-x)W_xS₂@Ni₃S₂–98 mV, and Super-Co₃S₄/P-WS₂/Co₉S₈–58 mV), in acid electrolyte (η@10 mA/cm², N-doped WS₂–150 mV, defect-rich WS₂–145 mV, CoS₂–145 mV, Zn_0.3Co_2.7S₄–80 mV, Fe_4.5Ni_4.5S₈–146 mV, NiCo₂S₄/Pd-87 mV, FeNiS-105 mV, Super-Co₃S₄/P-WS₂/Co₉S₈–70 mV), and in neutral electrolyte (high-index faceted Ni₃S₂/nickle foam-170 mV, CoS_x/F-doped tin oxide(FTO)−168 mV, CoMoNiS/nickle foam-117 mV, Co₃S₄–420 mV, S-NiFe₂O₄/nickle foam-197 mV, NiSe₂/nickle foam-169 mV, CoNi₂S₄/WS₂/Co₉S₈–146 mV, and Super-Co₃S₄/P-WS₂/Co₉S₈–129 mV). The Tafel slopes in alkaline/acid/neutral media were given in Figs. 4d–f to analyze the hydrogen-evolution kinetics of Super-NiWSP, NiWSP, and NiS. Apart from the outstanding Pt/C, the Tafel slope values of Super-NiWSP were fitted to be 99/101/119 mV/dec for alkaline/acid/neutral HER, which is smaller than the corresponding 135/150/146 mV/dec of the typical reference NiS. The smaller Tafel slopes of Super-NiWSP demonstrate favorable hydrogen-evolving kinetics under all pH environments. Moreover, as indicated by the smaller charge transfer resistance (R_ct) from electrochemical impedance spectroscopy (EIS) tests (Fig. S7 in Supporting information), the NiS/Ni₃S₄ activated plane and edge of hierarchical WS₂ nanosheet and the doping of P element lead to improved intrinsic conductivity in Super-NiWSP [1,13,51]. In addition, the electrochemical surface areas (ECSA) of samples were assessed by electrochemical double layer capacitance (C_dl) tests, as exhibited in Figs. S8–S10 (Supporting information). Super-NiWSP provides 65/71/45 mF/cm² of ECSA for alkaline (Fig. 4g)/acid (Fig. 4h)/neutral (Fig. 4i) HER, which is larger than 41/48/32 mF/cm² of NiWS and 18/23/13 mF cm⁻² of NiS, respectively (). The vertically aligned nanosheet-based array enables the maximum exposure of the crosslinked active WS₂-plane/edge-type heterostructures, thus accounting for the promotion of electrochemical surface areas in Super-NiWSP. Furthermore, the chronoamperometric measurements (i-t) plots (Figs. 4j–l) of Super-NiWSP remained well at about 30 mA/cm² for 24 h under different pH scenarios and exhibited negligible activity attenuations, verifying the satisfying electrocatalytic stability. Taking Super-NiWSP after stability test under alkaline condition as example, the XRD pattern (Fig. S11 in Supporting information) indicates a slight decrease of the signal intensity but a good reservation of WS₂, NiS, and Ni₃S₄ crystalline phases. The SEM image (Fig. S12 in Supporting information) demonstrates that although some minor degradation is observed in surface structure, the vertically aligned nanosheet-based array structure is well remained. Taken together, the electrochemical tests confirmed the superior HER activity and stability of non-precious Super-NiWSP electrocatalysts and the favorable versatility for electrolyte condition change.

To disclose the activity origin of Super-NiS/WS₂/Ni₃S₄ for electrocatalytic HER, the additional density functional theory (DFT) calculations were provided. The models of WS₂-plane-related and WS₂-edge-related WS_2nullNiS and WS_2nullNi₃S₄ heterointerfaces were established and shown in Fig. S13 (Supporting information). The free energy calculations of H₂O* and H* on WS₂-plane-related and WS₂-edge-related WS_2nullNiS and WS_2nullNi₃S₄ heterointerfaces were exhibited in Figs. 5a and b, respectively. According to the computational results, the H₂O molecules exhibit lower adsorption energies on the WS₂ plane-related heterointerfaces (Fig. 5a) in comparison with the WS₂ edge-related heterointerfaces (Fig. 5b), implying that the H₂O was preferentially adsorbed on WS₂-plane-type heterointerfaces. Moreover, the negative changes in free energy for water dissociation indicate that the dissociation process is thermodynamically favorable for WS₂-plane-related heterointerfaces. In terms of the HER step, a lower free energy absolute value indicates a greater likelihood for hydrogen desorption [52-54]. For the WS₂-plane-NiS interface (Fig. 5a), the free energy changes for the hydrogen evolution step are 0.15 eV (Tafel) and −0.36 eV (Heyrovsky), which are closest to zero among all the catalysts studied. Thermodynamically speaking, the WS₂-plane-NiS interface exhibits the optimal hydrogen desorption behavior. The calculations indicate that WS₂-plane-related heterointerfaces own the advantages of the preferential adsorption and the thermodynamically favorable dissociation for water molecules [55-58]. Furthermore, the hydrogen desorption step at the WS₂-plane-NiS interface is thermodynamically optimal. Furthermore, the electronic properties were further explored aiming at unveiling the intrinsic activation mechanism of WS₂-plane-NiS interface. The calculated electron density difference plot at WS₂-plane-NiS heterointerfaces was given in Fig. 5c, where the yellow section is related to the electron accumulation while green region means electron depletion. It can be seen that the electrons are transferred from NiS to WS₂, which is also consistent with the XPS analysis. Thus, the holes are accumulated on NiS side and facilitate the dissociation of H₂O. The electrons enriched on WS₂ will benefit for the hydrogen evolution [59-63]. In addition, the partial density of states (DOS) of S on NiS at pristine NiS and WS₂-plane-NiS interface were demonstrated in Fig. 5d, where the p band center for up- and down-spins was calculated as well. As clearly shown, the formation of WS₂-plane-NiS interface leaded to an upraised p band center of S, remarkable band for up-spin situation. The upraised p-band implies the enhanced interactions between NiS and WS₂ in WS₂-plane-NiS interface and the favorable dissociation/adsorption/desorption processes of H₂O/H* species involved in electrochemical water splitting, thus leading to the superior electrocatalytic activity.

In summary, we successfully constructed super-hybrid transition metal sulfide nanoarrays of NiS/WS₂/Ni₃S₄ with high spatial ordering and abundant plane- and edge-type WS_2nullNiS and WS_2nullNi₃S₄ hetrointerfaces. The polyoxometalate (PW₁₂) exhibited special hydrolysis kinetics for the programmed growth and assembly of WS₂ nanosheets and NiS/Ni₃S₄ nanoparticles in one-pot synthesis. The Super-NiS/WS₂/Ni₃S₄ demonstrated superior electrocatalytic HER activity and favorable versatility under different electrolytes conditions, which only required 57/95/151 mV overpotentials to actuate 10 mA/cm² in alkaline/acid/neutral media. The theoretical and experimental results indicated that WS₂-plane-related heterointerfaces exhibited the favorable adsorption and dissociation of water molecules. The WS₂-plane-NiS interface is thermodynamically optimal for hydrogen evolution judging from the Gibbs free energy (ΔG_H*) values. The collaborations of the open nanosheet-based vertical array, the abundant plane- and edge-type active interfaces, and the phosphorus doping in Super-NiS/WS₂/Ni₃S₄ not only provide maximum exposure of active sites, but also strengthen mass transport and electron transfer in electrocatalysis, thereby systematically promoting water-splitting hydrogen production. Our recent findings suggest that the polyoxometalates-based precursors are versatile for synthesizing ordered TMSs aiming at energy-related applications.