An efficient WSe₂/Co_0.85Se/graphene hybrid catalyst for electrochemical hydrogen evolution reaction

Abstract

Transition metal doped layered transition metal dichalcogenides (TMDs) are regarded as promising hydrogen evolution reaction (HER) candidates due to exposed active sites at both edges and basal planes. Hydrogen absorption free energy on active sites in doped materials are adjusted to thermoneutral state for ideal bond breaking to favour the HER process after the introduction of transition metal ions in their crystalline structures. Considering the importance of the active site, charge transfer, and hydrogen adsorption free energy to HER, cobalt-containing sandwich-type polyoxometalates are used as a precursor to fabricate cobalt doped WSe₂ nanosheets via CVD selenization method. Reduced graphene oxides (rGO) are further introduced into cobalt doped WSe₂ nanosheets for solving the non-ohmic contact with current collectors, leading to enhanced exchange current density. Co–WSe₂/rGO₂ composites exhibit Tafel slope of 64 mV dec⁻¹, overpotential of 217 mV at 10 mA cm⁻², exchange current density of 15.3 × 10⁻³ mA cm⁻², charge transfer resistance of 68 Ω, and the activity is maintained after 3 h. The activated basal planes, good conductivity and adjusted hydrogen absorption free energy are attributed to enhance the HER performance. The present work opens a new avenue for the fabrication of transition metal doped TMDs using polyoxometalates with determined composition as precursors to fabricate superior HER electrocatalysts.

---

Introduction

Hydrogen energy has been regarded as a promising clean energy. Electrocatalytic hydrogen evolution reaction (HER) is one of the most effective ways to produce hydrogen.^1–3 The adsorbed hydrogen (H*) acting as an intermediate is found to be responsible for the different catalytic performances of electrocatalysts.^4–6 Platinum based catalysts show the best behavior for HER owing to their ideal hydrogen absorption free energy close to the thermoneutral state (ΔG_H* ≈ 0). However, the applications of such noble metal catalysts are obviously limited because of their prohibitive cost and low resources. Therefore, researchers have devoted tremendous efforts to designing non-precious electrocatalysts for low cost and effective HER.⁷

Very recently, layered transition metal dichalcogenides (TMDs) are found to be effective to catalyze HER due to exposed active sites at edges rather than basal planes.⁸ Large amounts of work have suggested that the exposed active sites and conductivity of TMDs are quite important to favor the HER process.^9,10 Besides shrinking the particle size or thinning the thickness down to monolayer for maximum active sites at edges, how to activate the basal planes of layered TMDs are quite important for obtaining superior HER catalysts.¹¹ Oxygen-incorporated MoS₂ ultrathin nanosheets have been developed to generate surface defects, resulting in the improvement of electrochemical hydrogen evolution via activation of basal planes.¹² In addition, incorporation of foreign elements into TMDs can effectively lead to rich active sites, which is effective to create large amounts of unsaturated sulfur atoms as active sites in basal planes and enhanced intrinsic conductivity, thus leading to improve the HER performance. For instance, Co-doped MoS₂ nanosheets or hollow structures with dominant CoMoS phase showed superior HER performance with a high exchange current density, low onset potential, and small Tafel slope.^13,14 Alloying method is another effective way for the enhancement of HER performance of TMDs. Sulfided Mo edges for MoS₂ possess too weak absorption ability for hydrogen, causing too short time for the proton–electron-transfer process. MoS_2(1−x)Se_2x alloys were capable of adjusting the hydrogen absorption free energy close to thermoneutral state and improving the electrical conductivity, showing improved HER activity.¹⁵ Besides alloying method, even hybrid catalyst of ​MoS₂/CoSe₂ can also successfully bring in more active edge sites of terminal ​S₂²⁻ and S²⁻ ions, which are HER active.¹⁶

Taken together, introduction of first-row transition metal ions such as Co into MoS₂ proves to be effective to decrease the hydrogen absorption free energy from 0.18 to 0.10 eV via the formation of S–Co bond, favorable to balance the absorption of H⁺ and bond breaking of H₂.¹⁷ WSe₂ is the first TMD material possessing both p-type and n-type behaviors in the same material. WSe₂ nanofilms on carbon fiber paper are highly efficient HER electrocatalysts, and are extremely stable in acidic solution in terms of no degradation after 15 000 continuous cycles.¹⁸ Reduced graphene oxide (rGO) was used to enhance the conductivity of WSe₂, showing improved HER activity with a Tafel slope of 64 mV per decade and good stability.¹⁹ Ternary WS_2(1−x)Se_2x nanoribbons, nanotubes, or monolayers possessing favorable ΔG_H* induced by a slight distortion of introducing bigger radius of Se into the crystalline structure of WS₂.^20–22 Considering the synergistic effects between first-row transition metal ions and TMDs in terms of exposed active sites, enhanced conductivity, and favorable ΔG_H*, introduction of Co into WSe₂ might bring in superior HER catalysts.

In the present work, as illustrated in Scheme 1, we for the first time fabricated Co-doped WSe₂ nanosheets using Co-containing sandwich-type polyoxometalates as precursors via CVD selenization method. rGO was introduced to enhance the overall conductivity to facilitate the charge transfer. The HER performance of as-fabricated catalysts was systematically evaluated, in which Co–WSe₂/rGO₂ composite exhibits Tafel slope of 64 mV dec⁻¹, overpotential of 217 mV at 10 mA cm⁻², exchange current density of 15.3 × 10⁻³ mA cm⁻², charge transfer resistance of 68 Ω, and the stable activity after 3 h under acidic conditions. For comparison, Co–WSe₂ without rGO, Co–WSe₂ prepared using inorganic salts mixture and commercial Pt/C were also carried out under similar conditions.

Scheme 1 Co-containing polyoxometalates as precursor to prepare Co–WSe₂/rGO composites.

Experimental section

Materials and methods

All starting materials were purchased from Aladdin Industrial Corporation and used as received. All solutions were prepared with triply distilled water.

Characterizations

The morphologies of composites were characterized by scanning electron microscopy (Hitachi SU-800 scanning electronic microscope) and transmission electron microscopy (FEI TECNAI High Resolution TEM at 300 kV). The Raman and PL scattering of composites were performed on a confocal microscope-based Raman spectrometer (LabRAM XploRA, incident power of 1 mW, pumping wavelength of 532 nm). The phase and lattice constant were obtained by X-ray diffraction (XRD, DIFFRACTOMETER-6000 with Cu Kα radiation (λ = 0.1542 nm)). Fourier transform infrared (FTIR) spectra of the prepared catalysts were collected on a Perkin-Elmer 1710 spectrometer using KBr pellet at room temperature. Zeta potentials of GO colloid was measured using Zeta PALS (Brookhaven, USA).

Synthesis of Co–WSe₂/rGO_x

The precursor 50 mg Bi₂W₂₂Co₂ were firstly mixed with various amounts of GO (3.786 mg mL⁻¹; the rGO mass percentage of 1%, 3.5%, 5%, 10%, 20% in composites are named from rGO₁ to rGO₅). Mixture was sonicated for 30 minutes to make a homogeneous suspension. Then the mixture were decanted into a ceramic boat and add 150 μL polyethylenimine (PEI) with concentration of 1 mg mL⁻¹ to facilitate the co-assembly of GO and Bi₂W₂₂Co₂. Before selenization, the boat loading the mixture were put into the oven at 65 °C for 6 h to obtain a dry powder. The boat containing the dried mixing precursors were put in the back zone of quartz tube, and 0.4–0.8 g selenium (99.99% Alfa) was located in the front zone. The back zone temperature raised to 450 °C at a rate of 15 °C min⁻¹, meanwhile the front zone were heated to 270–330 °C at a rate of 15 °C min⁻¹. The first conversion process was maintained for 2 hours with 12 sccm Ar gas and 8 sccm H₂ gas to guarantee the precursor completely selenization. Subsequently, the back zone temperature further raised to 935 °C at a rate of 15 °C min⁻¹, and the second conversion process was maintained for another 2 hours under same condition. After the conversion process the furnace cooled down to room temperature naturally. Finally the powder in the boat were collected for HER. Other catalysts were prepared according to similar procedure.

Electrochemical measurements

All the electronchemical measurements were performed in a three-electrode system including a glassy carbon electrode in 3 mm diameter (geometric area = 0.077 cm²) at an electrochemical station (CHI 660D). All the potentials were calibrated to a reversible hydrogen electrode (RHE). Typically, the reference electrode (Ag/AgCl) were calibrated by an OCP (open circuit potential) method in Fig. S5.† In 0.5 M H₂SO₄ E (V vs. RHE) = E (V vs. Ag/AgCl) + 0.177. 4 mg of the catalyst and 30 uL of 5% Nafion solutions were dispersed in 0.5 mL of water/ethanol (3 : 1 v/v) mixed suspension, after sonication for at least 30 minutes. 5 μL of the suspension were directly dropped onto a GCE (loading: ∼0.57 mg cm⁻²). The electrolyte (0.5 M H₂SO₄) was saturated with nitrogen for 30 minutes to eliminate the oxygen. The polarization curves were obtained at a scan rate of 5 mV s⁻¹ from 0.2 V to −0.4 V (vs. RHE). Cyclic voltammetry (CV) was measured with scan rates of 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, and 120 mV s⁻¹ respectively in potential ranging from 0.076 to 0.176 V (vs. RHE) in the investigation of electrochemical double-layer capacitances (EDLCs, C_dl). And the stability measured by a cyclic voltammetry method at a rate of 50 mV s⁻¹. All data were obtained after iR-correction. Electrochemical impedance spectroscopy (EIS) measurements were performed in the frequency range of 100 kHz to 0.1 Hz with an overpotential of −253 mV vs. RHE in 0.5 M H₂SO₄.

Results and discussion

Co-doped WSe₂ nanosheets (denoted as Co–WSe₂) were prepared via CVD selenization method using Co-containing sandwich-type polyoxometalates (abbreviated as Bi₂W₂₂Co₂) as precursors, which is synthesized according to previous work.²³ Crystal structure was drawn in Fig. S1,† wherein two cobalt ions was sandwiched between two lacunary BiW₁₁ clusters. FT-IR spectra and XRD of crystal powders both confirmed the successful synthesis of Bi₂W₂₂Co₂ in Fig. S2a and b in ESI.† Graphene oxide were synthesized according to modified Hummer's method in our previous work, which was blending with Bi₂W₂₂Co₂ with the assistance of positive charged polyelectrolytes of PEI.²⁴ GO was found to be negatively charged according to the measurement of zeta potential, which is found to be −29 mV. Taking into account the homogeneous distribution of POMs on GO sheets, positive PEI was acting as a glue to facilitate the complexing of negative POMs and negative GO. The mixture with different GO contents were placed at high temperature zone in a horizontal two-zone furnace, and Se powder the CVD technique. The experimental methods are described in detail in the ESI.† Basically, two steps are involved in the synthesis, in which first step is the synthesis of CoSe at 450 °C for 2 h, and second step is to convert precursor to WSe₂ at 935 °C for 2 h. Se precursors was located in the low temperature zone (300 °C) at the upstream. The whole synthetic procedure was illustrated in Scheme S1.† The selection of POMs offering Co and W sources instead of using common Co salts and tungstate are due to the following reasons. (1) POMs representing a well-defined library of inorganic building blocks with nano-scaled sizes, acting as ideal intercalates, fillers, or building blocks on 2D supports such as graphene.^25–27 (2) First-row transition metal ions are easily introduced into the POMs, and its single-crystal structure can be characterized, giving rise to the opportunity for the understanding of relationship between POMs structure and properties.²⁸ (3) Homogeneous distribution of POMs on GO will facilitate the thermodynamic growth of TMDs in terms of nucleation and nanoparticles merging step.²⁹ Physical mixture of Co salts and tungstate might lead to high local concentration of one component, therefore causing phase separation during CVD growth, and further resulting in the deteriorated synergistic effect between CoSe_x and WSe₂.

XRD patterns of Co–WSe₂ in Fig. 1 contain multiply peaks around 2θ = 13.6, 31.4, 37.8, 41.7, 47.4, 55.9, and 56.7, corresponding to the (002), (100), (103), (006), (105), (110), and (008) planes of WSe₂, respectively, consistent with the standard pattern of hexagonal WSe₂ (JPCDS no. 38-1388).³⁰ Meanwhile, some peaks around 2θ = 33.3, 44.7, 50.6, 60.4, 61.9, 69.9, and 71.3, correspond to the (101), (102), (110), (103), (112), (202), and (004) planes of Co_0.85Se, respectively, in good agreement with the standard pattern of hexagonal Co_0.85Se (JPCDS no. 52-1008).³¹ The low-magnification scanning electron microscopy (SEM) images of Co–WSe₂ nanosheets are presented in Fig. 2a and S3,† wherein the irregular sheet-like aggregates are obtained after 30 min sonication, and the thickness of nanosheets was ∼120 nm. rGO sheets were distributed among those rigid nanosheets, functioning as conductive additives and growth supports. The sheet-like nanostructures provide large accessible area exposing to electrolytes, favorable to the HER proceeding. High-resolution TEM image in Fig. 2b clearly indicates that the interplanar spacing of the crystal lattice in Co–WSe₂ is 0.285 nm, arising from (100) plane of hexagonal WSe₂. A closer look at the HRTEM image, two typical lattices are found easily. As indicated in Fig. 2c, honeycomb lattices is ascribing to 2H phase, hexagonal patterns in Fig. 2d means 1T phase, in good agreement with previous work.³² Metallic 1T phase possess faster charge transfer in HER process comparing to semiconductive 2H phase, thus leading to improved catalytic performance.³³ Selected area electron diffraction pattern in inset of Fig. 2b presented one set of six-fold symmetry diffraction spots with a zone axis of [001], indicating good crystallinity of as-prepared nanosheets. Energy dispersive X-ray spectroscopy (EDS) in Fig. 2d, e and S4† clearly confirmed the homogeneous distribution of W and Se throughout the whole scanning nanosheets.

Fig. 1 XRD patterns of Co–WSe₂/rGO₂ and two standard XRD profiles of WSe₂ and Co_0.85Se.

Fig. 2 (a) SEM image of Co–WSe₂/rGO₂. Red arrow in (a) indicates the thickness of nanosheet is ∼120 nm. (b) HR-TEM image of Co–WSe₂/rGO₂. Inset of (b) is the SAED pattern of Co–WSe₂/rGO₂. (c and d) Enlarged regions as marked in (b) by squares. (e) STEM image of Co–WSe₂/rGO₂. (f and g) EDS mapping of W and Se elements.

As shown in Fig. 3, the degenerated E¹_2g (in-plane) and A_1g (out-of-plane) modes for WSe₂ were overlapping to form a single sharp peak at 253 cm⁻¹. There is another shoulder peak around 260 cm⁻¹ corresponding to 2LA(M) secondary resonance mode, indicating the existence of few-layer structures among all WSe₂ nanosheets.³⁴ Besides the characteristic peak of WSe₂, two characteristic peaks at 1353 and 1597 correspond to the D and G bands of rGO, respectively, indicative of the successful combination of Co-doped WSe₂ and rGO. The increased I_D/I_G ratio of rGO in composite comparing to GO means the recovery of some part of sp² carbon via the removal of oxygen-containing groups.³⁵ The reduction of GO is led in the presence of H₂ stream at high temperature.

Fig. 3 Raman spectra of Co–WSe₂/rGO₂.

HER electrocatalytic activity of Co–WSe₂/rGO_x were carried out in a 0.5 M H₂SO₄ electrolyte with a standard three-electrode electrochemical cell setup. The LSV measurement was recorded as described in ESI.† All potential is calibrated after iR correction. Naked carbon glass electrode is catalytically inert for the HER in Fig. 4a. First of all, as shown in Fig. S5a,† the polarization curves of composites are measured with respect to different rGO content. Introduction of 1 wt% content of rGO only lower the overpotential from 262 to 260 mV at 10 mA cm⁻² comparing to Co–WSe₂ without rGO. Increasing rGO contents up to 3.5 and 5 wt% both give rise to 217 mV at 10 mA cm⁻². Keep increasing rGO content to 20 wt% will lead to the enhancement of overpotential of 290 mV. Obviously, too many rGO contents will deteriorate the catalytic performance due to the blocking of active sites of TMDs, as rGO is barely inert to proton reduction. The optimal rGO content is 3.5 or 5%, showing the lowest overpotential of 217 mV at 10 mA cm⁻². The polarization curve of physical mixture prepared from inorganic salts of Bi₂O₃, Co₂O₃, and WO₃ in the presence of 3.5 wt% rGO prepared according to the weight percentage of each component in Co–WSe₂/rGO₂ composite was measured as control groups, as depicted in Fig. 4a which showed higher overpotential at −395 mV than that of Co–WSe₂/rGO₂, as well as lower cathodic current density, confirming the superiority of POMs as precursor. The poor catalytic performance of physical mixture of each component is ascribing to the lacking of synergistic effect among all components because of the poor atomic or nanoscaled contacts among therm. There is no doubt that Pt/C illustrated the best HER activity, giving rise to overpotential of 31 mV. The catalytic activity of Co–WSe₂/rGO₂ or Co–WSe₂/rGO₃ both showed the lowest potential of 217 mV at 10 mA cm⁻² among all tested catalysts except Pt/C and will need less energy input in the HER process as well. Taken together, the introduction of metal ions and graphene will contribute to the improvement for the HER activity.

Fig. 4 (a) Polarization curves of GCE, Pt/C, Co–WSe₂, Co–WSe₂/rGO₂, physical mixture/rGO₂. (b) Corresponding Tafel slopes of j–V curves in (a).

Tafel slopes can be extracted from Tafel plots, which can provide deep insight into the catalytic activity of the electrocatalysts.³⁶ As seen in Fig. 4b and S5b,† Tafel slopes of bare Co–WSe₂, Co–WSe₂/rGO₁, Co–WSe₂/rGO₂, Co–WSe₂/rGO₃, Co–WSe₂/rGO₄, and Co–WSe₂/rGO₅, calculated based on Tafel equation are 70, 64, 64, 48, 68, 90 mV per decade, respectively. Particularly, physical mixture prepared from inorganic salts with 3.5 wt% rGO only gave 88 mV per decade, further indicative of the disadvantage of using simple blending of starting materials. As is known, a faster increment of HER velocity as potential is increased indicates a smaller Tafel slope. The Co–WSe₂/rGO₃ exhibited improved HER activity with small Tafel slope of 48 mV per decade, mainly ascribing to the introduction of Co and rGO.

For HER in acidic media, three rate-determining steps have been widely used to describe the possible mechanism according Tafel slopes. Tafel slope should be about 120 mV per decade if Volmer reaction is the rate-determining step in HER. Comparatively, the Tafel slope are about 30 and 40 mV per decade when the Heyrovsky and the Tafel reaction are a rate-determining steps, respectively.^36,37 Here, the Tafel slopes are 70, 64, and 48 mV per decade for Co–WSe₂, Co–WSe₂/rGO₂, and Co–WSe₂/rGO₃, respectively. These values fall between the Volmer mechanism and the Heyrovsky mechanism as the rate determining steps, thus the Volmer–Heyrovsky HER combination mechanism should be responsible for the HER of Co–WSe₂/rGO_x based catalysts.³⁸ The catalyst with small Tafel slopes of about 64 or 48 mV per decade herein is expected to be promising candidate for practical applications, because a slight increment of applied overpotential can generate an obvious increase of cathodic current. The exchange current densities of Co–WSe₂ and Co–WSe₂/rGO₂ are calculated to be about 10.9 × 10⁻³ and 15.3 × 10⁻³ mA cm⁻² by extrapolating the Tafel plots, respectively. In comparison with previous work in Table S1,† the exchange current density (j₀) of our Co–WSe₂/rGO₂ value is comparable to that of 3D dendritic WSe₂, about half of that of WS_2(1−x)Se_2x but showing lower overpotential at 10 mA cm⁻². j₀ values of WSe₂/W foil, WSe₂/Si NWs, and exfoliated WSe₂ are much lower than that of our catalyst. For WS_2(1−x)Se_2x nanoflakes on CF and metallic WS_2(1−x)Se_2x nanoribbons show very high j₀, mainly ascribing to improved conductivity due to direct growth on CF and metallic phase. However, considering the lower η₁₀ and Tafel slope, and higher j₀, present catalyst proved to be a promising candidate for effectively catalyzing HER.

Inspired from previous work, the effective electrochemically active area was proportional to C_dl.³⁹ As shown in Fig. 5c and 5D, CV curves of Co–WSe₂ and Co–WSe₂/rGO₂ were obtained in the potential range from 0.076 to 0.176 V without faradic current as a function of scan rates, because current response in this region was ascribed to the charging of the double layer. The plots in Fig. 5e and f were the anodic and cathodic current vs. scan rate at the centre of the scanning potential range of 0.126 V vs. RHE. The average value of the slope are the EDLCs, namely C_dl. The C_dl of Co–WSe₂/rGO₂ was ∼1.35 mF cm⁻², which is ∼2.25 times of the C_dl of Co–WSe₂. The introduction of rGO was effectively increase the active surface area, thus resulting in the increase of the density of the electrochemically active sites for proton reduction. It is assumed that the large C_dl of Co–WSe₂/rGO₂ contributed to the enhanced HER performance.

Fig. 5 Stability test for the Co–WSe₂/rGO₂ catalyst (a) by CV scanning for 1000 cycles and (b) at a constant overpotential of 217 mV. (c and d) Cyclic voltammetry of Co–WSe₂/rGO₂ and Co–WSe₂ as a function of scan rates from 10 to 120 mV s⁻¹. (e and f) The anodic (black open square) and cathodic (red open circle) charging currents of Co–WSe₂/rGO₂ and Co–WSe₂ were plotted as a function of scan rate at 0.126 V vs. RHE, respectively. The average of the absolute value of the slope of linear fitting of the plot is equal to electrochemical double-layer capacitances (EDLCs, C_dl) of each electrode.

Stability of the catalytic activity is evaluated by cycling the Co–WSe₂/rGO₂-modified GCE continuously for 1000 cycles. As shown in Fig. 5, the electrode illustrated similar j–V curve to the initial cycle at the end of the cycling procedure, showing negligible loss of the cathodic current. Meanwhile, the cathodic currents remained at 10 mA cm⁻² over 3 hour at 217 mV, indicative of good stability of Co–WSe₂/rGO₂ in acidic conditions employed for the HER process.

Taken together, the Co–WSe₂/rGO₂ showed improved catalytic performance in HER comparing with Co–WSe₂ without rGO, and Co–WSe₂ synthesized using inorganic salts as precursors. In present work, Co-doped WSe₂ might introduce more defect at basal planes, serving as new catalytically active sites for the adsorption of proton. The well-defined POMs structures have homogeneous distribution of Co or W elements in nanoscale comparing to the physical mixing of inorganic Co or W salts, resulting in easier doping pathway of Co to form close contact interfaces between Co_0.85Se and WSe₂ crystallites. The phase separation might occur under selenization process of physical mixture of Co or W salts as starting materials due to the high localized concentrations. Inspired from previous work, the edges of Co-doped WSe₂ will possess favorable hydrogen free energy close to thermalneutral value possibly via the formation of S–Co bond, which is quite important for the absorption of H⁺ and bond breaking of H₂.¹⁶ Additionally, the introduction of conductive rGO leads to tremendous decrease of charge transfer resistance between Co–WSe₂/rGO₂ and GCE, which was solidly confirmed by electrochemical impedance measurements. As shown in Fig. 6 the Nyquist plots exhibit that the charge-transfer resistances (R_ct) of Co–WSe₂ and Co–WSe₂/rGO₂ are ∼1451 and 59 Ω at 253 mV overpotential (vs. RHE), respectively. Additionally, local phase separation in physical mixture using inorganic salts results in much higher R_ct of 59 942 Ω, which is also consistent with the poor electrocatalytic HER performance. Briefly, the greatly reduced impedance facilitates faster electron transfer between Co–WSe₂/rGO₂ and the electrode, beneficial to the enhanced kinetics of the HER.

Fig. 6 Nyquist plots of Co–WSe₂/rGO₂, Co–WSe₂ and physical mixture prepared using inorganic salts of Bi₂O₃, Co₂O₃, and WO₃ in the presence of 5 wt% rGO prepared according to the weight percentage of each component in Co–WSe₂/rGO₂ composite.

Conclusions

In summary, Co–WSe₂/rGO₂ nanocomposites have successfully illustrated high activity, faster catalytic kinetics, and good stability as hydrogen evolution reaction catalysts. Co-containing POMs was selected as precursor instead of physical mixing of inorganic salts to prepare selenides via facile CVD selenization method. The activation of basal planes of WSe₂ due to the incorporation of Co, enhanced conductivity of nanocomposites by in site formation with rGO, and adjusted hydrogen free energy close to thermoneutral state, synergistically promote catalytic kinetics at catalysts interface and result in enhanced HER performances. Present work open a new avenue for the fabrication of first-row transition metal doped TMDs using polyoxometalates with determined composition as precursors to fabricate superior HER electrocatalysts.

Acknowledgements

This work is supported by the Open Project of State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (No. ES201514).

Notes and references

1. Y. Li, H. Wang, L. Xie, Y. Liang, G. Hong and H. Dai, J. Am. Chem. Soc., 2011, 133, 7296–7299.
2. D. Kong, H. Wang, Z. Lu and Y. Cui, J. Am. Chem. Soc., 2014, 136, 4897–4900.
3. J. Duan, S. Chen, M. Jaroniec and S. Z. Qiao, ACS Nano, 2015, 9, 931–940.
4. D. Voiry, H. Yamaguchi, J. Li, R. Silva, D. C. Alves, T. Fujita, M. Chen, T. Asefa, V. B. Shenoy and G. Eda, Nat. Mater., 2013, 12, 850–855.
5. P. Xiao, M. A. Sk, L. Thia, X. Ge, R. J. Lim, J.-Y. Wang, K. H. Lim and X. Wang, Energy Environ. Sci., 2014, 7, 2624–2629.
6. C. Tsai, K. Chan, F. Abild-Pedersen and J. K. Nørskov, Phys. Chem. Chem. Phys., 2014, 16, 13156–13164.
7. D. Kong, J. J. Cha, H. Wang, H. R. Lee and Y. Cui, Energy Environ. Sci., 2013, 6, 3553–3558.
8. J. Xie, H. Zhang, S. Li, R. Wang, X. Sun, M. Zhou, J. Zhou, X. W. D. Lou and Y. Xie, Adv. Mater., 2013, 25, 5807–5813.
9. G. Guan, S. Zhang, S. Liu, Y. Cai, M. Low, C. P. Teng, I. Y. Phang, Y. Cheng, K. L. Duei and B. M. Srinivasan, J. Am. Chem. Soc., 2015, 137, 6152–6155.
10. X. Zhao, X. Ma, J. Sun, D. Li and X. Yang, ACS Nano, 2016, 10, 2159–2166.
11. D. Voiry, J. Yang and M. Chhowalla, Adv. Mater., 2016 DOI:10.1002/adma.201505597.
12. Z. Chen, D. Cummins, B. N. Reinecke, E. Clark, M. K. Sunkara and T. F. Jaramillo, Nano Lett., 2011, 11, 4168–4175.
13. X. Dai, K. Du, Z. Li, M. Liu, Y. Ma, H. Sun, X. Zhang and Y. Yang, ACS Appl. Mater. Interfaces, 2015, 7, 27242–27253.
14. L. Yu, B. Y. Xia, X. Wang and X. W. Lou, Adv. Mater., 2016, 28, 92–97.
15. Q. Gong, L. Cheng, C. Liu, M. Zhang, Q. Feng, H. Ye, M. Zeng, L. Xie, Z. Liu and Y. Li, ACS Catal., 2015, 5, 2213–2219.
16. M.-R. Gao, J.-X. Liang, Y.-R. Zheng, Y.-F. Xu, J. Jiang, Q. Gao, J. Li and S.-H. Yu, Nat. Commun., 2015, 6, 5982.
17. D. Merki, H. Vrubel, L. Rovelli, S. Fierro and X. Hu, Chem. Sci., 2012, 3, 2515–2525.
18. H. Wang, D. Kong, P. Johanes, J. J. Cha, G. Zheng, K. Yan, N. Liu and Y. Cui, Nano Lett., 2013, 13, 3426–3433.
19. Y. Jung, Y. Zhou and J. Cha, Inorg. Chem. Front., 2016, 3, 452–463.
20. F. Wang, J. Li, F. Wang, T. A. Shifa, Z. Cheng, Z. Wang, K. Xu, X. Zhan, Q. Wang and Y. Huang, Adv. Funct. Mater., 2015, 25, 6077–6083.
21. Q. Fu, L. Yang, W. Wang, A. Han, J. Huang, P. Du, Z. Fan, J. Zhang and B. Xiang, Adv. Mater., 2015, 27, 4732–4738.
22. K. Xu, F. Wang, Z. Wang, X. Zhan, Q. Wang, Z. Cheng, M. Safdar and J. He, ACS Nano, 2014, 8, 8468–8476.
23. H. Liu, C. Qin, Y.-G. Wei, L. Xu, G.-G. Gao, F.-Y. Li and X.-S. Qu, Inorg. Chem., 2008, 47, 4166–4172.
24. J. Zhang, P. Hu, R. Zhang, X. Wang, B. Yang, W. Cao, Y. Li, X. He, Z. Wang and W. O'Neill, J. Mater. Chem., 2011, 22, 714–718.
25. B. Nohra, H. El Moll, L. M. Rodriguez Albelo, P. Mialane, J. Marrot, C. Mellot-Draznieks, M. O'Keeffe, R. Ngo Biboum, J. Lemaire and B. Keita, J. Am. Chem. Soc., 2011, 133, 13363–13374.
26. R. Liu, G. Zhang, H. Cao, S. Zhang, Y. Xie, A. Haider, U. Kortz, B. Chen, N. S. Dalal and Y. Zhao, Energy Environ. Sci., 2016, 9, 1012–1023.
27. J.-S. Qin, D.-Y. Du, W. Guan, X.-J. Bo, Y.-F. Li, L.-P. Guo, Z.-M. Su, Y.-Y. Wang, Y.-Q. Lan and H.-C. Zhou, J. Am. Chem. Soc., 2015, 137, 7169–7177.
28. P. Yin, T. Li, R. S. Forgan, C. Lydon, X. Zuo, Z. N. Zheng, B. Lee, D. Long, L. Cronin and T. Liu, J. Am. Chem. Soc., 2013, 135, 13425–13432.
29. R. Liu, G. Zhang, H. Cao, S. Zhang, Y. Xie, A. Haider, U. Kortz, B. Chen, N. S. Dalal and Y. Zhao, Energy Environ. Sci., 2016, 9, 1012–1023.
30. W. Jung, S. Lee, D. Yoo, S. Jeong, P. Miró, A. Kuc, T. Heine and J. Cheon, J. Am. Chem. Soc., 2015, 137, 7266–7269.
31. F. Gong, H. Wang, X. Xu, G. Zhou and Z.-S. Wang, J. Am. Chem. Soc., 2012, 134, 10953–10958.
32. A. Ambrosi, Z. Sofer and M. Pumera, Chem. Commun., 2015, 51, 8450–8453.
33. D. Voiry, M. Salehi, R. Silva, T. Fujita, M. Chen, T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nano Lett., 2013, 13, 6222–6227.
34. W. Zhao, Z. Ghorannevis, K. K. Amara, J. R. Pang, M. Toh, X. Zhang, C. Kloc, P. H. Tan and G. Eda, Nanoscale, 2013, 5, 9677–9683.
35. D. Chen, H. Feng and J. Li, Chem. Rev., 2012, 112, 6027–6053.
36. C. G. Morales-Guio, L.-A. Stern and X. Hu, Chem. Soc. Rev., 2014, 43, 6555–6569.
37. M. Zeng and Y. R. Li, J. Mater. Chem. A, 2015, 3, 14942–14962.
38. J. Zhang, S. Liu, H. Liang, R. Dong and X. Feng, Adv. Mater., 2015, 27, 7426–7431.
39. C. C. L. McCrory, S. Jung, J. C. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2013, 135, 16977–16987.

---

Footnotes

† Electronic supplementary information (ESI) available: Crystal structure of Bi₂W₂₂Co₂; FT-IR and XRD spectra; synthetic procedure scheme; EDS spectrum; polarization and Tafel curves; comparison table; CV calibration curve. See DOI: 10.1039/c6ra08618g

‡ These two authors made equal contributions.

---