Well-defined CoSe₂@MoSe₂ hollow heterostructured nanocubes with enhanced dissociation kinetics for overall water splitting

Abstract

Hollow heterostructures have tremendous advantages in electrochemical energy storage and conversion areas due to their unique structure and composition characteristics. Here, we report the controlled synthesis of hollow CoSe₂ nanocubes decorated with ultrathin MoSe₂ nanosheets (CoSe₂@MoSe₂) as an efficient and robust bifunctional electrocatalyst for overall water splitting in a wide pH range. It is found that integrating ultrathin MoS₂ nanosheets with hollow CoSe₂ nanocubes can provide abundant active sites, promote electron/mass transfer and bubble release and facilitate the migration of charge carriers. Additionally, the surface electron coupling in the heterostructures enables it to serve as a source of sites for H⁺ and/or OH⁻ adsorption, thus reducing the activation barrier for water molecules adsorption and dissociation. As a result, the title compound, CoSe₂@MoSe₂ hollow heterostructures, exhibits an overpotential of 183 mV and 309 mV at a current density of 10 mA cm⁻² toward hydrogen evolution reactions and oxygen evolution reactions in 1.0 M KOH, respectively. When applied as both cathode and anode for overall water splitting, a low battery voltage of 1.524 V is achieved along with excellent stability for at least 12 h. This work provides a new idea for the design and synthesis of high-performance catalysts for electrochemical energy storage and conversion.

---

Introduction

Global environmental pollution and the energy crisis have become a bottleneck restricting the development of international economy and society, which forces people to accelerate the development and utilization of sustainable clean new energy.^1–3 Hydrogen is widely considered as one of the most promising energy carriers to traditional petroleum energy by virtue of its clean emissions and low cost-efficiency.⁴ Electrocatalytic water splitting is a reliable and promising technology for the preparation of high purity hydrogen and oxygen.⁵ The theoretical thermodynamic potential to drive water splitting is 1.23 V. However, due to the existence of dynamic overpotentials of hydrogen evolution reactions (HER) and oxygen evolution reactions (OER), the efficiency of electrolysis is seriously hindered. In this regard, efficient electrocatalysts are urgently needed to lower the activation barrier and accelerate the catalytic reaction process.⁶ At present, noble metal-based materials, such as Pt/C, Pd, RuO₂ and IrO₂, are acknowledged to be the most effective electrocatalysts owing to their low overpotentials and high reaction rates.⁷ However, their large-scale commercial application is greatly impeded by their low reserves, high cost and easier deactivation. Therefore, it is highly desirable to develop low-cost, efficient and robust alternative catalysts for water splitting both in acid and alkaline media.

Recently, the design and synthesis of cost-effective and stable electrocatalysts for electrochemical water splitting have been achieved by optimizing the composition and structure of non-noble metal catalysts, including metal alloys,⁸ carbon materials^9,10 and transition metal compounds.^11–13 Among them, cobalt-based chalcogenides with different stoichiometric ratios, such as Co₉S₈,¹⁴ CoS₂,^15,16 CoSe₂,^17,18 and CoS_xSe_2−x,¹⁹ have attracted more and more attention in view of many advantages including low toxicity, rich structure chemistry, low overpotential and fast kinetics. For example, Liu et al. prepared atomically thick CoSe₂ nanosheets as highly efficient OER catalysts by exfoliating CoSe₂-based lamellar compounds.²⁰ It was found that the vacancies (V_Co) existing on the surface of CoSe₂ nanosheets served as efficient active sites to catalyze the OER, manifesting a low overpotential of 320 mV at 10 mA cm⁻² in alkaline media. Liu and co-workers reported that the Co-doped NiSe₂ nanoparticle films electrodeposited on conductive Ti plates (Co_0.13Ni_0.87Se₂/Ti) exhibited good HER electrocatalytic activity in 1.0 M KOH, and showed a low overpotential of 64 mV to reach 10 mA cm⁻² with a small Tafel slope of 63 mV dec⁻¹.²¹ Given these facts, cobalt-based selenides are promising electrocatalysts for water splitting. Furthermore, it is widely accepted that a large specific surface area is favorable for catalytic reactions. As an important and useful morphology, a hollow micro-/nanostructure can provide many advantages for constructing highly catalytically active electrocatalysts, such as large exposed surfaces, rich active sites and rapid electron/mass transfer, which are favorable for catalytic reactions. For instance, Yu and co-workers found that Ni and Co incorporated MoS₂ nanoboxes manifested impressive HER activity with an onset potential of 125 mV and a Tafel slope of 51 mV dec⁻¹.²² Therefore, the controlled synthesis of hollow nanostructures is a promising way to significantly enhance the catalytic activity of cobalt-based selenides.

Although cobalt selenides, especially CoSe₂, have demonstrated promising capabilities as OER catalysts, pure CoSe₂ still exhibited unsatisfactory catalytic activity. Besides, pure CoSe₂ is not stable in acid electrolytes, implying that it can hardly serve as a bifunctional electrocatalyst for overall water splitting. Coincidentally, recent studies have shown that the construction of heterostructured electrocatalysts via combining HER and OER active nanocomponents together in one system is an effective way to obtain more outstanding electrocatalytic performance. The delicate heterostructures with refined electron distribution not only provide sufficient adsorption sites for the intermediates but also improve electron transfer efficiency at their interface.^23,24 These advantages contribute significantly to enhanced HER and OER kinetics, making them efficient bifunctional catalysts for electrochemical water splitting. For example, Yu's group found that the CoSe₂/MoS₂ heterostructures manifested better HER performance than pure MoS₂ in 0.5 M H₂SO₄ solution with a low overpotential of 68 mV at 10 mA cm⁻².²⁵ Lin et al. reported that heterogeneous MoS₂/NiS₂ nanosheets on carbon cloth displayed enhanced catalytic performance for water splitting, which delivered an overpotential of 62 mV and 278 mV at the current density of 10 mA cm⁻² for the HER and OER, respectively.²⁶ Based on these facts, CoSe₂-based hollow heterostructures are reasonably expected to be efficient and robust electrocatalysts for overall water splitting in a wide pH range. It is well known that MoS₂ and MoSe₂ with abundant edge sites are powerful HER catalysts, but they often display limited OER performance. Thus, integration of the CoSe₂ hollow nanocubes with an active and stable HER catalyst such as MoSe₂ nanosheets into one system is expected to display outstanding catalytic performance for overall water splitting. However, to the best of our knowledge, the controlled synthesis of CoSe₂/MoSe₂ nanocubes with a well-defined hollow core–shell nanostructure has never been reported.

Herein, in the present work, we reported the design and synthesis of hollow CoSe₂ nanocubes decorated with ultrathin MoSe₂ nanosheets as a cost-effective electrocatalyst for electrochemical overall water splitting. The CoSe₂@MoSe₂ heterostructures were synthesized by two-step hydrothermal reactions. Firstly, the hollow CoSe₂ nanocubes were prepared via a first hydrothermal reaction based on ionic exchange reactions between Co-based Prussian blue analogues (Co₃[Co(CN)₆]₂, denoted as Co–Co PBA) and Na₂SeO₃. Subsequently, the CoSe₂@MoSe₂ hollow heterostructures were synthesized through the in situ growth of ultrathin MoSe₂ nanosheets on the surface of CoSe₂ nanocubes. Benefiting from the merits of structure and compositions, such a delicate structure can provide abundant well-exposed active sites, promote electron/mass transfer and facilitate the migration of charge carriers. As a result, the CoSe₂@MoSe₂ heterostructures displayed impressive catalytic performance, with an overpotential of 187 mV and 309 mV at a current density of 10 mA cm⁻² toward the HER and OER in 1.0 M KOH, respectively. When applied as both cathode and anode for overall water splitting, a low battery voltage of 1.524 V is achieved along with excellent stability for at least 12 h. The results presented here demonstrated that the CoSe₂@MoSe₂ complex nanocube is a cost-effective and robust bifunctional catalyst for overall water splitting.

Experimental

Synthesis of Co–Co PBA nanocubes

Co–Co PBA nanocubes were synthesized as reported in our previous work.²⁷ In a typical synthesis, 2.0 mmol of K₃[Co(CN)₆] were dissolved in 100 mL of H₂O under vigorous stirring to form a clear solution A. Then, the solution A was uniformly mixed with 100 mL of solution B containing 3.0 mmol C₄H₆O₄Co·4H₂O and 4.5 mmol Na₃C₆H₅O₇·2H₂O. The obtained mixed solution was allowed to age for 20 h at room temperature. Finally, the product was collected by centrifugation and washed with water and ethanol several times before drying at 60 °C overnight.

Synthesis of hollow CoSe₂ nanocubes

CoSe₂ hollow nanocubes were prepared following Fang's procedure with a few modifications.²⁸ Typically, 30 mg of Co–Co PBA nanocubes were uniformly dispersed in 10 mL of H₂O under ultrasound. At the same time, 15 mL of a solution in which 60 mg of Na₂SeO₃·2H₂O was dissolved was poured into the above Co–Co PBA suspension and magnetically stirred for 30 min. Then, 1.0 mL of N₂H₄ was added to the above mixed solution, and stirred for 30 min until the solution was completely mixed. Finally, the mixture was transferred to a Telfon-lined stainless steel autoclave and kept in an oven at 160 °C for 8 h. After the reaction was completed, it was cooled to room temperature. The final product was collected by centrifugation, washed several times with water and ethanol, and then dried under vacuum at 60 °C for 12 h.

Synthesis of CoSe₂@ hollow MoSe₂ heterostructures

CoSe₂@MoSe₂ hollow heterostructures were obtained by using a facile hydrothermal reaction process. Typically, 50 mg of the as-obtained CoSe₂ hollow nanocubes, 65 mg of Na₂SeO₃·2H₂O, 100 mg of Na₂MoO₃·2H₂O were dissolved in 50 mL of N,N-dimethylformamide (DMF) to form a homogeneous suspension with the assistance of ultrasonication for 30 min. The resulting mixture was transferred into a Telfon-lined stainless-steel autoclave and kept at 200 °C for 15 h in an electric oven and then cooled down to room temperature naturally. The final product was collected by centrifugation, washed several times with water and ethanol, and dried overnight under vacuum at 60 °C. The preparation process of pure MoSe₂ particles is similar to the above method, except for the addition of CoSe₂ hollow nanocubes.

Characterization

The phase composition and crystal structure of the obtained samples were analyzed by X-ray diffraction (XRD) on a Rigaku D/MAX RINT-2000 X-ray diffractometer. The morphology and microstructure were examined by field emission scanning electron microscopy (FESEM; JEOL-7600F) and transmission electron microscopy (TEM; JEOL, JEM2100F). The composition and elemental mapping of the samples were analyzed by using an energy dispersive X-ray spectrometer (EDX) attached to a SEM instrument. An X-ray photoelectron spectrometer (XPS, VG Microtech ESCA2000) was used to analyze the compositions and chemical state of the elements in the synthesized sample. The specific surface areas and pore size distribution of the sample were determined by nitrogen adsorption/desorption measurement using a Quantachrome Autosorb-iQ instrument.

Electrochemical measurement

The electrocatalytic performance was tested using a standard three-electrode system of an electrochemical workstation (CHI 660e). The acidic and alkaline electrolyte solutions were 0.5 M H₂SO₄ and 1.0 M KOH, respectively. A platinum mesh and a saturated calomel electrode (SCE) (or a standard Hg/HgO electrode) were used as the counter and reference electrodes in 0.5 M H₂SO₄ (or 1.0 M KOH). To prepare a working electrode, 3 mg of CoSe₂@MoSe₂ was dispersed in a mixed solution of 55 μL of ethanol, 165 μL of water and 20 μL of naphthol, and sonicated to obtain a uniform dispersion. Finally, 3 μL or 48 μL of the above liquid was dropped on a glassy carbon electrode (diameter: 3 mm) or nickel foam (1 × 1 cm²) to obtain a CoSe₂@MoSe₂ modified electrode. Electrodes of CoSe₂ hollow nanocubes, MoSe₂ microspheres, and 20 wt% Pt/C catalysts were also prepared using the same method for comparison. The polarization curves of OER and HER activity were characterized by linear sweep voltammetry (LSV) with a scan rate of 5 mV s⁻¹. Prior to LSV, each electrode will be activated for 40 cycles at a scan rate of 50 mV s⁻¹. In an acidic and alkaline medium (0.5 M H₂SO₄ or 1.0 M KOH), all graphs shown are calibrated to the reversible hydrogen electrode (RHE) based on the equation (E_RHE = E_SCE + 0.059 × pH + 0.241 or E_RHE = E_Hg/HgO + 0.059 × pH + 0.098) and corrected for iR compensation. The overpotential (η) of the OER is calculated according to the following formula: η(V) = E_RHE − 1.23 V. The AC impedance test (EIS) was performed at open circuit voltage (10⁻² to 10⁴ Hz) with an AC voltage of 5 mV. The electric double layer capacitance (C_dl) of the electrode is measured by cyclic voltammetry (CV), and the illegal pull-over overvoltage range is 0.1 to 0.2 (1.3 to 1.4 V) (vs. RHE at 0.5 M H₂SO₄ or 1.0 M KOH) with a sweep rate of 20–200 mV s⁻¹.

Results and discussion

The schematic diagram for preparing CoSe₂@MoSe₂ hollow heterostructures is shown in Fig. 1. First, highly uniform Co–Co PBA nanocubes were synthesized as templates through a facile coprecipitation method. Then, these solid Co–Co PBA nanocrystals were successfully converted into nanoplate-assembled CoSe₂ hollow nanoboxes via a simple anion-exchange reaction between Co–Co PBA nanocubes and Na₂SeO₃ at elevated temperatures. Afterwards, the CoSe₂@MoSe₂ hollow heterostructures were prepared by in situ growth of ultrathin MoSe₂ nanosheets on the surface of hollow CoSe₂ nanocubes via a facile hydrothermal method.

Fig. 1 Schematic illustration of the synthesis of hollow CoSe₂@MoSe₂ heterostructures via a two-step hydrothermal procedure (I: selenization, II: in situ coating of MoSe₂ on the CoSe₂ nanocube surface).

The typical scanning electron microscopy (SEM) image in Fig. 2a showed that the as-synthesized Co–Co PBA nanocubes were uniform and the average size was ca. 1 μm. The corresponding transmission electron microscopy (TEM) image in Fig. 2b indicated that these Co–Co PBA nanocubes had a solid texture with a smooth surface. The Co–Co PBA nanocubes were then used as sacrifice templates to prepare CoSe₂ hollow nanoboxes. It can be seen that the CoSe₂ nanocubes still maintained the cubic structure of the Co–Co PBA nanocubes (Fig. 2d), but its surface became very rough. The obvious contrast differences in the TEM image demonstrated that the CoSe₂ nanoboxes had a distinct hollow structure (Fig. 2e), which was favorable to promote electron/mass transfer and bubble release. The magnified TEM image of one single CoSe₂ nanocube indicated that the shell was composed of randomly assembled nanoplates, and the thickness of the shell was ∼150 nm (Fig. 2f). This result indicated that the morphology of Co–Co PBA changed significantly during the selenization process, which agreed well with Fang's results.²⁸ The formation of hollow CoSe₂ nanocubes can be explained by the Kirkendall effect. In the early stages of selenization, the Se²⁻ ion reacted with CoCo PBA nanocubes, resulting in the formation of a thin layer of CoSe₂ nanocrystallites on the surface. The size of Se²⁻ is larger than that of Co²⁺, thus the inward diffusion of Se²⁻ ions was slower than the outward diffusion of Co²⁺ ions, leading to the reactions mainly taking place on the preformed CoSe₂ shell and finally the formation of hollow CoSe₂ nanocubes.²⁹ Form the high-resolution TEM (HRTEM) image of a CoSe₂ layer, we can infer that the thickness of the CoSe₂ nanoplate is about 12.8 nm (Fig. 2g). The lattice spacing of the plane was ca. 0.29 nm, which could be indexed to the (101) face of CoSe₂ (Fig. 2h). To explore structure changes during the synthesis processes, the phase composition and crystal structure of the products were characterized by X-ray diffraction (XRD) measurements (Fig. 2i). It can be seen that all characteristic peaks of the as-prepared Co–Co PBA nanocubes are basically consistent with the previous work.³⁰ After the selenidation reaction, new strong diffraction peaks can be clearly observed, illustrating the high crystallinity of the product. Additionally, all the diffraction peaks corresponded to the orthorhombic structure of CoSe₂ (JCPDS no. 53-0449). No diffraction peaks of impurities could be observed, indicating that the Co–Co PBA nanocubes were completely transformed into hollow CoSe₂ nanocubes.

Fig. 2 SEM image (a), TEM image (b) and crystal structure (c) of the as-prepared Co–Co PBA nanocubes; SEM (d), TEM (e and f) and HRTEM (g and h) images of the CoSe₂ hollow nanocubes. XRD patterns (i) of the corresponding products.

The CoSe₂@MoSe₂ hollow heterostructures were controllably synthesized by the in situ growth of ultrathin MoSe₂ nanosheets on the surface of hollow CoSe₂ nanoboxes. The composition and crystal structure of the CoSe₂@MoSe₂ heterostructures were firstly investigated by XRD analysis. As shown in Fig. 3a, the diffraction peaks of CoSe₂ can still be observed. In addition, some new peaks marked with clubs at 13.7°, 31.4°, 37.9° and 55.9° could be attributed to the (002), (100), (103) and (110) planes of hexagonal phase MoSe₂ (JCPDS No. 29-0914, a = b = 3.287 Å, c = 12.925 Å), respectively.³¹ This result indicates that the as-obtained sample consists of CoSe₂ and MoSe₂. X-ray photoelectron spectroscopy (XPS) was then performed to determine the chemical state of the elements in the CoSe₂@MoSe₂ sample. The survey spectrum in Fig. S1† indicates the presence of Co, Mo, and Se in the sample. Fig. 3b–d present the high-resolution XPS spectrum of Co 2p, Mo 3d and Se 3d, respectively. As shown in Fig. 3b, the peak of Co 2p_3/2 could be deconvoluted into two peaks at 778.3 and 780.3 eV, and the peak of Co 2p_1/2 could be divided into two peaks at 793.2 and 796.4 eV.³² The peaks at 778.3 and 793.2 eV were attributed to the Co–Se bonding configuration, and the peaks at 780.3 and 796.4 eV implied the presence of a Co–O bonding configuration. Compared to that of pure CoSe₂, the peak of the Co 2p spectrum displays a positive shift of nearly 0.2 eV, suggesting the existence of strong electronic interactions between CoSe₂ and MoSe₂ (Fig. S2†).³³ For the Mo 3d spectrum, the strong peaks located at the binding energies of 228.4 and 231.7 eV correspond to Mo 3d_3/2 and Mo 3d_5/2 of Mo⁴⁺ species, respectively.³⁴ Meanwhile, the two small peaks at 232.3 and 235.3 could be attributed to Mo⁶⁺ from Na₂SeO₄ or MoO₃ formed on the surface of MoSe₂. The other peak at 229.5 eV could be assigned to Se 3s of MoSe₂. Fig. 3d displays a high resolution Se 3d spectrum, the binding energies of 54.1 and 54.9 eV corresponding to the Se 3d_5/2 and Se 3d_3/2 could be assigned to the Se anion in pyrite CoSe₂.³⁵ The observed peak of 59.4 eV could be attributed to the SeO bond due to the weak oxidation of the sample.³⁶ The XPS characterization was in agreement with the XRD result, suggesting the successful synthesis of CoSe₂@MoSe₂ hybrids.

Fig. 3 XRD pattern (a) of the as-formed CoSe₂@MoSe₂ heterostructures. High-resolution XPS spectra of Co 2p (b), Mo 3d (c) and Se 3d (d) in the CoSe₂@MoSe₂ heterostructures.

The atomic structures, morphology details and crystal quality of the as-obtained CoSe₂@MoSe₂ hollow heterostructures were further explored by SEM and TEM measurements. As shown in the SEM images presented in Fig. 4a and b, the surface of the CoSe₂@MoSe₂ heterostructures was rather rough, and the hollow interior could be clearly observed from the cracked particle. This unique hollow structure can also be confirmed by the low magnification TEM characterization shown in Fig. 4c. The magnified TEM image in Fig. 4d shows that many two-dimensional MoSe₂ nanosheets were distributed on the surface of the CoSe₂ nanosheets by forming a heterostructure. The close contact with each other could enhance the charge transfer during the catalytic reaction process.³⁷ Additionally, from the edge view shown in Fig. 4e, we can see that the thickness of MoSe₂ nanosheets was in the range of 1–4 nm. The ultrathin MoSe₂ nanosheets with maximized exposed edges were conducive to exposing the catalytic sites and accelerating the catalytic reactions.³⁸Fig. 4f displays the HRTEM image of a CoSe₂ layer, corresponding to region 1 marked in Fig. 4e. The clear lattice fringe addressed the high crystalline nature of the CoSe₂ nanoplates. The interlayer spacing of the plane was 0.26 nm, which well agreed with the (111) plane of CoSe₂. The HRTEM image in Fig. 4g, corresponding to region 2 displayed in Fig. 4e, displayed a typical edge view of 4-layer MoSe₂ nanosheets which were curly and deviated from the vertical direction. The lattice spacing was about 0.65 nm, which could be indexed to the (002) face of the 2H-MoSe₂ phase. From the edge view, we can see that the MoSe₂ nanosheets are open structures with edge-termination. This unique structure meant that the MoSe₂ nanosheets were defect-rich and mainly dominated by edges with a high density of exposed edge sites and catalytically active sites, which could enhance the catalytic activity of the HER and OER.³⁹ The corresponding annular selected area electron diffraction (SAED) patterns can be indexed to the (210), (400) and (105) planes of CoSe₂ and MoSe₂, which was consistent with the HRTEM results.⁴⁰ Energy dispersive X-ray spectroscopy (EDX) showed the Co : Mo : Se atomic ratio to be 3.3 : 1 : 6.7 (Fig. S3†). The elemental mapping images confirmed that these elements were uniformly distributed in a single nanocube (Fig. 4i). Based on the above analysis, it can be concluded that the resulting product consisted of CoSe₂ and MoSe₂. The CoSe₂@MoSe₂ hollow heterostructure possessed a specific surface area of 59.124 m² g⁻¹ with a narrow pore size distribution centered at 38.2 nm, exhibiting an obvious mesoporous structure (Fig. S4†).

Fig. 4 SEM (a and b), TEM (c) and magnified TEM (d and e) images of the as-synthesized CoSe₂@MoSe₂ heterostructures; (f) region 1 (light blue cycle in (e)): HRTEM image of CoSe₂ nanosheets; (g) region 2 (red cycle in (e)): edge view of vertical few-layered MoSe₂ nanosheets. SAED patterns (h) and elemental mapping (i) of the CoSe₂@MoSe₂ heterostructures.

HER performances of the as-prepared electrocatalysts

The as-obtained catalysts were firstly dripped on glassy carbon electrodes for the linear sweep voltammetry (LSV) electrochemical test in 0.5 M H₂SO₄ solution to evaluate their HER performance. All the plots were corrected by 90% iR-compensation. As shown in Fig. 5a, the current density increased rapidly under more negative potentials, suggesting the excellent HER activity of the as-synthesized samples. Particularly, the overpotential of the CoSe₂@MoSe₂ catalyst at the current density of 10 mA cm⁻² was 183 mV, remarkably smaller than that of the CoSe₂ (220 mV). This comparison revealed that the in situ growth of MoSe₂ on CoSe₂ nanocubes to form heterostructures could effectively improve the HER electrocatalytic reactivity.⁴¹ To support this conclusion, pure MoSe₂ microspheres (Fig. S5†) and physically mixed CoSe₂ and MoSe₂ hybrids (denoted as CoSe₂/MoSe₂) were also prepared. It was found that their activities were obviously lower than that of CoSe₂@MoSe₂ hollow heterostructures. The 20 wt% Pt/C electrocatalysts exhibited the highest HER electrocatalytic activity with a low overpotential of 206 mV vs. RHE at 10 mA cm⁻². To gain insight into the rate-determining step during the HER process, the Tafel diagram was further analyzed (Fig. 5b). The specific overpotential at 10 mA cm⁻² current density (η₁₀) and the corresponding Tafel slope of different samples are displayed in Fig. 5c. The extracted Tafel slope of CoSe₂@MoSe₂ heterostructures was 43.37 mV dec⁻¹, outperforming those of CoSe₂ (49.01 mV dec⁻¹), MoSe₂ (68.26 mV dec⁻¹) and CoSe₂/MoSe₂ (48.44 mV dec⁻¹). The lower overpotential and smaller Tafel slope of CoSe₂@MoSe₂ heterostructures indicated that they had faster HER kinetics and higher catalytic performance. It is well known that the Tafel slopes of 120, 40 and 30 mV dec⁻¹ correspond to the Volmer reaction, Heyrovsky reaction and Tafel reaction for the HER catalytic process in acidic electrolytes, respectively.⁴² In the present work, the Tafel slope of CoSe₂@MoSe₂ was close to 43.37 mV dec⁻¹, suggesting that the Volmer–Heyrovsky mechanism was the rate-limiting step during the HER process.⁴³ To investigate the origin of the high electrocatalytic activity of CoSe₂@MoSe₂ heterostructures, the electrochemically active surface area (ECSA) of catalysts was measured by double-layer capacitance (C_dl) in cyclic voltammetry (CV). Fig. 5c and S4† show the CV curves of MoSe₂, CoSe₂ and CoSe₂/MoSe₂ at different scanning rates of 20–200 mV s⁻¹ in the range of 100–200 mV (vs. RHE) voltage. C_dl is determined by plotting the curve of ΔJ (ΔJ = J_anodic − J_cathodic) relative to RHE at a voltage of 150 mV, and the C_dl value of the sample is the linear slope of the curve. The C_dl value of CoSe₂@MoSe₂ was 8.86 mF cm⁻², much higher than that of MoSe₂ (5.78 mF cm⁻²), CoSe₂ (7.65 mF cm⁻²) and CoSe₂/MoSe₂ (8.22 mF cm⁻²), suggesting the existence of more active sites in the CoSe₂@MoSe₂ heterostructures.²² To further understand the intrinsic electrocatalytic activities of these electrocatalysts, the turnover frequency (TOF) at an overpotential of 300 mV was calculated. It can be seen that the CoSe₂@MoSe₂ exhibited the highest value of 0.23557 s⁻¹, larger than that of CoSe₂ (0.10354 s⁻¹), MoSe₂ (0.03918 s⁻¹) and CoSe₂/MoSe₂ (0.12826 s⁻¹). This result confirmed that the synergistic modulation of both active sites and electronic structures could effectively improve the electrocatalytic activity of catalysts. The electrocatalytic stability of electrocatalysts is one of the important parameters for evaluating their practical applications. Thus, the LSV curves were recorded between −0.45 V and 0 V vs. RHE at a scan rate of 100 mV s⁻¹ for 1500 cycles in 0.5 M H₂SO₄. As shown in Fig. 4f, the LSV curve of CoSe₂@MoSe₂ was basically the same as the initial curve after 1500 cyclic voltammetry cycles, which directly indicated that the sample had superior cyclic stability. All of these results confirmed that the ideally designed CoSe₂@MoSe₂ heterostructures had outstanding HER electrocatalytic performance in acidic solution.

Fig. 5 LSV curves (a) and corresponding Tafel plots (b) of 20%wt Pt/C, MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂ samples in 0.5 M H₂SO₄. (c) Summary of the overpotentials at a current density of 10 mA cm⁻² (η₁₀) and Tafel slopes in (a); (d) the TOF values of MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂ samples at an overpotential of 300 mV; (e) the fitted C_dl of MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂. (f) LSV curves of CoSe₂@MoSe₂ before and after 1000 cycles.

We also tested the HER performance under alkaline conditions. Fig. 6a and b show the representative iR-corrected LSV polarization curves and Tafel slopes of the as-synthesized catalysts in 1.0 M KOH with Ni foam as substrates. The η₁₀ and corresponding Tafel slopes in each case are illustrated in Fig. 6c. The CoSe₂@MoSe₂ exhibited obviously superior HER activity with a η₁₀ of 183 mV, which was lower than those of MoSe₂ (256 mV), CoSe₂ (238 mV), CoSe₂/MoSe₂ (304 mV). Although it is inferior to commercial Pt/C (58 mV), it was comparable to or better than other cobalt-based selenide catalysts in alkaline media (Table S1†). The Tafel slope of CoSe₂@MoSe₂ was 87.69 mV dec⁻¹, considerably smaller than that of MoSe₂ (144.23 mV dec⁻¹), CoSe₂ (130.14 mV dec⁻¹) and CoSe₂/MoSe₂ (99.96 mV dec⁻¹). This indicated that the CoSe₂@MoSe₂ catalyst had faster kinetics and higher HER activity even under alkaline conditions. As shown in Fig. 6c and Fig. S6,† the C_dl value of CoSe₂@MoSe₂ (38.58 mF cm⁻²) was much larger than that of bare MoSe₂ (3.44 mF cm⁻²), CoSe₂ (9.44 mF cm⁻²) and CoSe₂/MoSe₂ (28.33 mF cm⁻²). The experimental results implied that the active surface area of CoSe₂@MoSe₂ was the largest in the four catalysts. Additionally, the TOF value of CoSe₂/MoSe₂ was larger than that of CoSe₂, MoSe₂ and CoSe₂/MoSe₂ at the same overpotential, revealing the better intrinsic HER activity of CoSe₂/MoSe₂ (Fig. 6e). The polarization curve of CoSe₂@MoSe₂ after 1500 cycles was almost the same as the initial curve under alkaline conditions (Fig. 6f), implying that the CoSe₂@MoSe₂ catalyst also had good cyclic stability in alkaline solution.⁴⁴ The long-term durability test for the CoSe₂@MoSe₂ catalyst was performed by controlled potential electrolysis at a static potential of 0.206 V vs. RHE for 12 h. As shown in the inset of Fig. 6f, the current density basically remained at 10 mA cm⁻² for the initial 6 h, and then slightly boosted, which can be attributed to the activation process and the increase of the electro-active species. Afterwards, the current density barely decreased, indicating that the electrode can maintain steady HER activity over a period of 12 h.

Fig. 6 LSV curves (a) and corresponding Tafel plots (b) of 20 wt% Pt/C, MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂ in 1.0 M KOH. (c) Corresponding overpotential (η₁₀) and Tafel slopes in (a). (d) The fitted C_dl of MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂. (e) TOFs of MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂ at different overpotentials from 300 to 400 mV. (f) Durability test for CoSe₂@MoSe₂ on Ni foam before and after 1500 LSV cycles, and the inset shows the time dependence of current density under a static potential of 206 mV vs. RHE in 1 M KOH.

Evaluation of the oxygen evolution reaction

The OER activities of the MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂ samples were examined by LSVs in 1.0 M KOH solution at room temperature, and their polarization curves are displayed in Fig. 7a. The oxidation peak in the range of 1.35–1.40 V vs. RHE is attributed to the precatalytic oxidation of Co²⁺ to Co³⁺, indicating the high electrocatalytic activity of the as-fabricated samples.⁴⁵ The CoSe₂@MoSe₂ electrode showed better OER catalytic activity compared to its singe component, and the overpotential of η₁₀ was as low as 309 mV. Moreover, the CoSe₂@MoSe₂ had a lower overpotential of 382 mV at a current density of 50 mA cm⁻² (η₅₀) than the bare CoSe₂ (399 mV) and MoSe₂ (477 mV), indicating a high exposure of the active edge sites in the CoSe₂@MoSe₂ hybrids. The OER kinetics of catalysts was also explored by using Tafel plots as shown in Fig. 7b. The slope of CoSe₂@MoSe₂ of 84.04 mV dec⁻¹ was also smaller than that of MoSe₂, CoSe₂, and CoSe₂/MoSe₂, indicating a much better kinetic process (Fig. 7c). The TOF of the CoSe₂@MoSe₂ sample was much bigger than those of MoSe₂, CoSe₂, and CoSe₂/MoSe₂ (Fig. 7d). This comparison suggested an increase of the surface area, which clearly affirmed higher active surface areas and better exposure of the catalytic active sites of the CoSe₂@MoSe₂ (Fig. S7†). To investigate the reaction kinetics of the as-prepared catalysts, electrochemical impedance spectroscopy (EIS) measurements were performed in the frequency range of 0.01–10⁴ Hz. As shown in Fig. 7e (Fig. S8†), the semicircular characteristic of the EIS curve indicated that the charge transfer resistance of CoSe₂@MoSe₂ was significantly smaller than that of MoSe₂, CoSe₂ and CoSe₂/MoSe₂, implying the high catalytic activity of the CoSe₂@MoSe₂ sample.²² In addition to excellent electrocatalytic activity, CoSe₂@MoSe₂ also had good stability and durability. The LSV curves were recorded between 1.204 V and 1.574 V vs. RHE at a scan rate of 100 mV s⁻¹ for 1000 cycles in 1.0 M KOH. After 1000 cycles, the polarization curve of CoSe₂@MoSe₂ was consistent with the starting curve (Fig. 7f). Furthermore, the CoSe₂@MoSe₂ electrode maintained stable activity within 12 hours as shown in the inset of Fig. 7f.

Fig. 7 (a) LSV curves of MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂ samples for the OER in 1.0 M KOH. (b) Tafel plots of MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂ heterostructures. (c) LSV curves of CoSe₂@MoSe₂ before and after 1000 cycles. (d) TOFs of MoSe₂, CoSe₂, CoSe₂/MoSe₂ and CoSe₂@MoSe₂ at different overpotentials from 300 to 400 mV. (e) Nyquist plots and the corresponding simulated equivalent circuit diagram of the samples. (f) Time dependence of current density in 1.0 M KOH.

The above results show that the CoSe₂@MoSe₂ electrode has high HER and OER activities. We further evaluated it as a bifunctional electrocatalyst for overall water splitting applications in 1.0 M KOH solution. As shown in Fig. 8a, the CoSe₂@MoSe₂ hollow heterostructures provided a low cell voltage of 1.524 V and 1.846 V at a water splitting current density of 10 mA cm⁻² and 50 mA cm⁻², respectively. A large number of H₂ and O₂ bubbles are released from the cathode and anode during the electrolysis (Fig. 8b). This outstanding overall water splitting performance is comparable or superior to most of the previously reported non-noble metal bifunctional electrocatalysts (Fig. 8c).^46–53 In addition, the dual-function CoSe₂@MoSe₂ nanocube electrode can stably decompose water within 12 h (Fig. 8d). The current density is almost constant, revealing the favorable stability of the CoSe₂@MoSe₂ catalyst for the overall water splitting.

Fig. 8 (a) LSV curves of a two-electrode alkaline electrolyzer using different electrocatalysts/electrodes for water electrolysis at a sweep rate of 5 mV s⁻¹ in 1.0 M KOH. (b) The digital photograph of the two-electrode setup during overall water splitting. (c) The cell voltage of the overall water splitting for different electrocatalysts. (d) Time dependence of current density in 1.0 M KOH.

In the light of the above electrochemical analysis, the as-synthesized CoSe₂@MoSe₂ hollow heterostructures indeed displayed superior electrocatalytic performance due to their unique hybrid structure and composite characteristics. First, the nanoplate-assembled CoSe₂ nanocubes had larger surface areas, thus providing more active sites to catalyze the HER and OER reactions. Meanwhile, the hollow structure was beneficial for the release of H₂ and O₂ bubbles, which accelerated the mass transfer. Second, the MoSe₂ nanosheets were edge-terminated with a defect-rich structure, providing lots of exposed edges and more active sites. Third, the synergistic effects facilitated the migration of charge carriers between CoSe₂ and MoSe₂, ensuring good electron transfer over the whole electrochemical reaction. As a result, the as-designed hollow heterostructured nanocubes can yield prominent HER, and OER electrochemical activity, thus behaving as a high performance bifunctional electrode for efficient overall water splitting.

Conclusions

In summary, well-defined CoSe₂@MoSe₂ heterostructures were rationally synthesized by the in situ growth of ultrathin MoSe₂ nanosheets on the surface of hollow CoSe₂ nanocubes with Co-based Prussian blue analogues as starting materials. The CoSe₂@MoSe₂ hollow heterostructures manifested excellent electrocatalytic performance for water splitting, and delivered 10 and 20 mA cm⁻² at an overpotential of 183 mV and 348 mV towards the HER and OER with robust stability in 1.0 M KOH solution, respectively. When employed as a bifunctional electrocatalyst for overall water splitting, the alkaline electrolyzer only needs a voltage of 1.524 V to deliver a current density of 10 mA cm⁻². The enhanced catalytic performance was mainly attributed to its unique structure and composite characteristics. The results of this study would provide an effective avenue for the design and synthesis of functional heterostructure catalysts for efficient HER/OER and related energy applications.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The work was financially supported by the National Natural Science Foundation of China (21601120 and 21771124) and the Science and Technology Commission of Shanghai Municipality (17ZR1410500, 17ZR1441200, 18QA1402400, 18230743400 and 19ZR1418100).

Notes and references

1. Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. B. Chorkendorff, J. K. Norskov and T. F. Jaramillo, Science, 2017, 355, eaad4998.
2. S. Chu, Y. Cui and N. Liu, Nat. Mater., 2017, 16, 16–22.
3. Y. Jiang, D. Y. Song, J. Wu, Z. X. Wang, S. S. Huang, Y. Xu, Z. W. Chen, B. Zhao and J. J. Zhang, ACS Nano, 2019, 13, 9100–9111.
4. Y. N. Guo, T. Park, J. W. Yi, J. Henzie, J. Kim, Z. L. Wang, B. Jiang, Y. Bando, Y. Sugahara, J. Tang and Y. Yamauchi, Adv. Mater., 2019, 31, 1807134.
5. J. Jia, W. J. Zhou, Z. Q. Wei, T. L. Xiong, G. X. Li, L. L. Zhao, X. F. Zhang, H. Liu, J. Zhou and S. W. Chen, Nano Energy, 2017, 41, 749–757.
6. G. Q. Zhao, K. Rui, S. X. Dou and W. P. Sun, Adv. Funct. Mater., 2018, 28, 1803291.
7. L. L. Zhu, H. P. Lin, Y. Y. Li, F. Liao, Y. Lifshitz, M. Q. Sheng, S. T. Lee and M. W. Shao, Nat. Commun., 2016, 7, 12272.
8. H. Lv, X. Chen, D. D. Xu, Y. C. Hu, H. Q. Zheng, S. L. Suib and B. Liu, Appl. Catal., B, 2018, 238, 525–532.
9. X. P. Dai, Z. Z. Li, Y. D. Ma, M. Z. Liu, K. L. Du, H. X. Su, H. Y. Zhuo, L. Yu, H. Sun and X. Zhang, ACS Appl. Mater. Interfaces, 2016, 8, 6439–6448.
10. S. N. Wang, L. Liao, Z. P. Shi, J. J. Xiao, Q. S. Gao, Y. H. Zhang, B. H. Liu and Y. Tang, ChemElectroChem, 2016, 3, 2110–2115.
11. Y. Zhang, H. J. Zhang, Y. Y. Feng, L. Liu and Y. Wang, ACS Appl. Mater. Interfaces, 2015, 7, 26684–26690.
12. Q. F. Gong, L. Cheng, C. H. Liu, M. Zhang, Q. L. Feng, H. L. Ye, M. Zeng, L. M. Xie, Z. Liu and Y. G. Li, ACS Catal., 2015, 5, 2213–2219.
13. S. Gupta, N. Patel, R. Fernandes, R. Kadrekar, A. Dashora, A. K. Yadav, D. Bhattacharyya, S. N. Jha, A. Miotello and D. C. Kothari, Appl. Catal., B, 2016, 192, 126–133.
14. L. L. Feng, M. H. Fan, Y. Y. Wu, Y. P. Liu, G. D. Li, H. Chen, W. Chen, D. J. Wang and X. X. Zou, J. Mater. Chem. A, 2016, 4, 6860–6867.
15. C. B. Ouyang, X. Wang and S. Y. Wang, Chem. Commun., 2015, 51, 14160–14163.
16. Y. X. Li, J. Yin, L. An, M. Lu, K. Sun, Y. Q. Zhao, D. Q. Gao, F. Y. Cheng and P. X. Xi, Small, 2018, 14, 1801070.
17. I. H. Kwak, H. S. Im, D. M. Jang, Y. W. Kim, K. Park, Y. R. Lim, E. H. Cha and J. Park, ACS Appl. Mater. Interfaces, 2016, 8, 5327–5334.
18. G. Q. Zhao, P. Li, K. Rui, Y. P. Chen, S. X. Dou and W. P. Sun, Chem. – Eur. J., 2018, 24, 11158–11165.
19. Y. Li, T. Polakovic, J. Curtis, S. L. Shumlas, S. Chatterjee, S. Intikhab, D. A. Chareev, O. S. Volkova, A. N. Vasiliev, G. Karapetrov and J. Snyder, J. Catal., 2018, 366, 50–60.
20. Y. W. Liu, H. Cheng, M. J. Lyu, S. J. Fan, Q. H. Liu, W. S. Zhang, Y. D. Zhi, C. M. Wang, C. Xiao, S. Q. Wei, B. J. Ye and Y. Xie, J. Am. Chem. Soc., 2014, 136, 15670–15675.
21. T. Liu, A. M. Asiri and X. Sun, Nanoscale, 2016, 8, 3911–3915.
22. X. Y. Yu, Y. Feng, Y. Jeon, B. Guan, X. W. Lou and U. Paik, Adv. Mater., 2016, 28, 9006–9011.
23. L. Najafi, S. Bellani, R. Oropesa-Nunez, A. Ansaldo, M. Prato, A. E. D. Castillo and F. Bonaccorso, Adv. Energy Mater., 2018, 8, 1703212.
24. D. Ma, B. Hu, W. D. Wu, X. Liu, J. T. Zai, C. Shu, T. T. Tsega, L. W. Chen, X. F. Qian and T. L. Liu, Nat. Commun., 2019, 10, 3367.
25. 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.
26. J. H. Lin, P. C. Wang, H. H. Wang, C. Li, X. Q. Si, J. L. Qi, J. Cao, Z. X. Zhong, W. D. Fei and J. C. Feng, Adv. Sci., 2019, 6, 1900246.
27. S. S. Huang, H. T. Wang, S. D. Wang, Z. J. Hu, L. Zhou, Z. W. Chen, Y. Jiang and X. F. Qian, Dalton Trans., 2018, 47, 5236–5244.
28. Y. J. Fang, X. Y. Yu and X. W. Lou, Adv. Mater., 2018, 30, 1706668.
29. H. Hu, B. Y. Guan and X. W. Lou, Chem, 2016, 1, 102–113.
30. Y. Feng, X. Y. Yu and U. Paik, Chem. Commun., 2016, 52, 6269–6272.
31. B. R. Xia, T. T. Wang, W. Xiao, R. F. Zhang, P. T. Liu, J. Ding, D. Q. Gao and D. S. Xue, Sci. Rep., 2017, 7, 45307.
32. B. Wang, Z. G. Wang, X. Q. Wang, B. J. Zheng, W. L. Zhang and Y. F. Chen, J. Mater. Chem. A, 2018, 6, 12701–12707.
33. Y. Q. Yang, K. Zhang, H. L. Ling, X. Li, H. C. Chan, L. C. Yang and Q. S. Gao, ACS Catal., 2017, 7, 2357–2366.
34. X. Q. Wang, B. J. Zheng, B. Yu, B. Wang, W. Q. Hou, W. L. Zhang and Y. F. Chen, J. Mater. Chem. A, 2018, 6, 7842–7850.
35. Y. Q. Long, J. Yang, X. Gao, X. N. Xu, W. L. Fan, S. F. Hou and Y. T. Qian, ACS Appl. Mater. Interfaces, 2018, 10, 10945–10954.
36. F. E. Niu, J. Yang, N. N. Wang, D. P. Zhang, W. L. Fan, J. Yang and Y. T. Qian, Adv. Funct. Mater., 2017, 27, 1700522.
37. J. Zhang, T. Wang, D. Pohl, B. Rellinghaus, R. H. Dong, S. H. Liu, X. D. Zhuang and X. L. Feng, Angew. Chem., Int. Ed., 2016, 55, 6702–6707.
38. Y. J. Li, X. C. Sun, Z. J. Cheng, X. Xu, J. Pan, X. F. Yang, F. Tian, Y. L. Li, J. Yang and Y. T. Qian, Energy Storage Mater., 2019, 22, 275–283.
39. B. J. Zheng, Y. F. Chen, F. Qi, X. Q. Wang, W. L. Zhang, Y. R. Li and X. S. Li, 2D Mater., 2017, 4, 025092.
40. K. R. Zhou, J. R. He, X. Q. Wang, J. Lin, Y. Jing, W. L. Zhang and Y. F. Chen, Electrochim. Acta, 2017, 231, 626–631.
41. Y. N. Guo, J. Tang, Z. L. Wang, Y. M. Kang, Y. Bando and Y. Yamauchi, Nano Energy, 2018, 47, 494–502.
42. X. D. Yan, L. H. Tian, M. He and X. B. Chen, Nano Lett., 2015, 15, 6015–6021.
43. W. Liu, E. Y. Hu, H. Jiang, Y. J. Xiang, Z. Weng, M. Li, Q. Fan, X. Q. Yu, E. I. Altman and H. L. Wang, Nat. Commun., 2016, 7, 10771.
44. M. Y. Zheng, K. L. Guo, W. J. Jiang, T. Tang, X. Y. Wang, P. P. Zhou, J. Du, Y. Q. Zhao, C. L. Xu and J. S. Hu, Appl. Catal., B, 2019, 244, 1004–1012.
45. A. X. Fan, C. L. Qi, X. Zhang, X. P. Dai, Z. Dong, C. L. Luan, L. Yu, J. Q. Ge and F. Gao, ACS Sustainable Chem. Eng., 2019, 7, 2285–2295.
46. T. Chen and Y. W. Tan, Nano Res., 2018, 11, 1331–1344.
47. Y. X. Guo, Z. Y. Yao, C. S. Shang and E. K. Wang, ACS Appl. Mater. Interfaces, 2017, 9, 39312–39317.
48. L. H. He, B. B. Cui, B. Hu, J. M. Liu, K. Tian, M. H. Wang, Y. P. Song, S. M. Fang, Z. H. Zhang and Q. J. Jia, ACS Appl. Mater. Interfaces, 2018, 1, 3915–3928.
49. J. G. Hou, B. Zhang, Z. W. Li, S. Y. Cao, Y. Q. Sun, Y. Z. Wu, Z. M. Gao and L. C. Sun, ACS Catal., 2018, 8, 4612–4621.
50. X. B. Hu, Q. W. Zhou, P. F. Cheng, S. Q. Su, X. Wang, X. S. Gao, G. F. Zhou, Z. Zhang and J. M. Liu, Appl. Surf. Sci., 2019, 488, 326–334.
51. D. X. Ji, S. J. Peng, L. Fan, L. L. Li, X. H. Qin and S. Ramakrishna, J. Mater. Chem. A, 2017, 5, 23898–23908.
52. R. H. Que, G. Ji, D. S. Liu, M. L. Li, X. H. Wang and S. P. Jiang, Energy Technol., 2019, 7, 1800741.
53. S. Wan, W. Y. Jin, X. L. Guo, J. Mao, L. K. Zheng, J. L. Zhao, J. Zhang, H. Liu and C. C. Tang, ACS Sustainable Chem. Eng., 2018, 6, 15374–15382.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9nr08751f

‡ These authors contributed equally to this work.

---