W₂C nanodot-decorated CNT networks as a highly efficient and stable electrocatalyst for hydrogen evolution in acidic and alkaline media

Abstract

Although tungsten carbide (W₂C) has long been reported as an excellent platinum-like catalyst, it is still a challenge to synthesize W₂C as an electrocatalyst for a highly efficient hydrogen evolution reaction (HER) due to its high onset overpotential, inevitable aggregation, and lack of a scalable and controllable synthesis method. Herein, we synthesized W₂C nanodot-decorated CNT networks (W₂C@CNT-S) via a facile and scalable spray drying method followed by a carbonization process. It is demonstrated that this unique nanoarchitecture, constructed by ultrafine W₂C nanodots homogeneously decorated on a three-dimensional and conductive CNT skeleton, leads to the exposure of abundant catalytic sites and promotes highly efficient electron transfer and ion diffusion during the HER process. As a result, in acidic and alkaline media, the optimized W₂C@CNT-S hybrid exhibited excellent HER performance with very low onset overpotentials of only 60 and 40 mV (vs. RHE) and very small Tafel slopes of 57.4 and 56.2 mV dec⁻¹, and only needed 176 and 148 mV (vs. RHE) to obtain a current density of 10 mA cm⁻², respectively; it also showed outstanding long-term durability even after a 30-hour test in both acidic and alkaline media. This study presents an overview of a low-cost and scalable spray-drying strategy to synthesize a high-performance carbide-based electrocatalyst for hydrogen evolution.

---

1. Introduction

Hydrogen has been considered to be a sustainable and clean alternative energy source to replace traditional fossil fuels in recent years. The production of hydrogen via electrocatalytic water splitting is safe and environmentally friendly when compared with the other hydrogen generation methods, including the steam reforming of hydrocarbons, the gasification reaction, and the partial oxidation reaction of heavy oil.^1–4 Moreover, the most important reaction in water splitting is the hydrogen evolution reaction (HER), which has been drawing significant attention from researchers due to the accessible reactant and promise for large-scale production. Pt-Based catalysts have been reported as the most efficient electrocatalysts for HER; however, the high cost and scarcity of Pt seriously hinder the large-scale application of these catalysts for H₂ production.⁵ Therefore, there is an urgent need to develop non-precious transition-metal-based electrocatalysts with low-cost, large-scale, efficient HER activity and excellent long-term stability.

To date, non-precious transition-metal-based compounds, such as sulfides,^6–8 selenides,^9–12 phosphides,^13,14 nitrides,^15,16 and carbides,^17–22 have been intensively investigated to replace the Pt-based electrocatalysts towards the HER. However, there are few electrocatalysts that can simultaneously satisfy the requirements of high electrocatalytic activity and excellent long-term durability, particularly in an alkaline medium. Among these catalysts, tungsten carbides (WC and W₂C) have been proved to be a promising alternative to platinum for the HER due to their Pt-like d-band electronic density of states at the Fermi level.²³ Some efforts have been made with W₂C; however, W₂C has received far less attention than the other tungsten carbide compounds (WC) for the HER because W₂C is usually investigated as a by-product of WC. However, recent reports indicate that, compared to WC, W₂C shows better HER performance.^22–24 For example, Li's group synthesized W₂C@NWs and WC@NWs via a two-step annealing treatment; compared to WC@NWs with a Tafel slope of 71 mV dec⁻¹ in 0.5 M H₂SO₄, W₂C@NWs delivered a smaller Tafel slope of 68 mV dec⁻¹.²⁴ It was noted that the formation of W₂C required high temperature conditions over 1250 °C from the W–C phase diagram;²⁵ however, at such high temperature, serious agglomeration is inevitable and it is not easy to obtain ideal nanoarchitecture with high electrocatalytic activity. So, it is still challenging to develop a scalable synthesis method to synthesize highly efficient and stable W₂C for HER with a well-designed nanoarchitecture and less agglomeration. Although the overall thermal efficiency is low and it is not suitable for highly viscous materials, the spray-drying method has an advantage of realizing large-scale commercial production. Fortunately, Wei et al. and Yu et al. recently reported the successful synthesis of high-quality molybdenum carbide by a spray-drying process using graphene and CNTs as the carbon source and conductive framework, respectively.^26,27 The spray-drying strategy might also be effective to suppress the agglomeration of W₂C. However, to the best of our knowledge, there is no report on the synthesis of W₂C with less agglomeration by spray drying, not to mention further control of the unique nanoarchitecture and enhancement of the HER performance of W₂C in both acid and alkaline electrolytes.

To address these issues, herein, for the first time, we present a scalable and low-cost method to synthesize ultrafine W₂C nanodots-decorated CNT networks via a spray-drying approach (designated as W₂C@CNT-S) and a subsequent carbonization process. The synthesis process of W₂C@CNT-S is schematically demonstrated in Fig. 1. Due to the chemical inertness of carbon nanotube, C atom diffusion through the solid–solid interface into the tungsten lattice becomes slow and effectively limits the overgrowth of W₂C particles.²³ Moreover, the unique structure with porous CNT networks as a carbon conductivity framework can facilitate the electron transfer; it is quite obvious that the ultrafine W₂C nanodots homogenously decorated on highly conductive CNT guarantee the high conductivity and provide abundant reactive catalytic sites for the HER process. As expected, compared to the sample with W₂C particles directly loaded on CNT without spray drying (designated as W₂C/CNT), the optimized W₂C@CNT-S hybrid exhibited a more efficient HER performance with lower onset overpotentials of ∼60 and 40 mV (vs. RHE), smaller Tafel slopes of 57.4 and 56.2 mV dec⁻¹, and better stability, even after 30 h test in acidic and alkaline media, respectively.

Fig. 1 Schematic of the preparation of the W₂C@CNT-S.

2. Experimental

2.1. Synthesis of W₂C@CNT-S and W₂C/CNT

The W₂C@CNT-S hybrid was synthesized as follows: 1.478 g of ammonium metatungstate hydrate ((NH₄)₆H₂W₁₂O₄₀·xH₂O) and various amounts of CNT conductive paste (5 wt%, LB217-54, Cnano technology Ltd) were dissolved in deionized water (500 mL) and vigorously stirred for 12 h. Then, the solution was treated with sonication for 1 h to obtain an homogeneous suspension. Subsequently, the above solution was treated by a commercial spray-drying procedure (SUNYI Spray Dryer SP-1500). The experimental parameters for the spray dryer were set as follows: pumping rate was 1000 mL h⁻¹, the temperatures at the inlet and outlet of the spray dryer were 180 °C and 90 °C, respectively. The black powder was collected and further carbonized under Ar gas (100 sccm) at 800 °C for 5 h with a heating rate of 5 °C min⁻¹ to obtain the W₂C@CNT-S. For comparison, these W₂C/CNT mixtures were synthesized using a similar process followed to obtain W₂C@CNT-S, except that the mixtures were collected by oven drying instead of spray drying. In order to investigate the optimized weight ratio of W₂C and CNT on HER performance, a series of samples were prepared with various molar ratios of C/W (6 : 1, 8 : 1 and 10 : 1), with the corresponding products denoted as W₂C@CNT-S6, W₂C@CNT-S8, and W₂C@CNT-S10, respectively.

2.2. Characterization

The phases of the W₂C@CNT-S and W₂C/CNT catalysts were identified by X-ray diffraction (XRD, Rigaku Ultima IV) with Cu Kα radiation (λ = 1.54178 Å). The chemical composition of the samples was identified using X-ray photoelectron spectroscopy (XPS, Kratos XSAM 800, Al Kα radiation). Raman spectrometry (Renishaw) was performed to obtain the Raman spectra of the samples with an excitation laser line of 532 nm in ambient conditions. The surface morphology and structures of samples were examined by using scanning electron microscopy (SEM, JSM-7000F, JEOL) and transmission electron microscopy (TEM, FEI Tecnai G2 F20) with an accelerating voltage of 200 kV. The chemical compositions, HAADF image, and corresponding elemental mapping of W and C were observed with energy X-ray spectroscopy (EDX).

2.3. Electrochemical measurements

All the electrochemical experiments with the as-synthesized electrocatalysts for HER were conducted on an electrochemical workstation (CHI660D, CH Instruments Inc) with a standard three-electrode system at ambient temperature. A graphite rod and a saturated calomel electrode (SCE) were used as the counter electrode and reference electrode, respectively, in N₂-saturated 0.5 M H₂SO₄. Also, experiments were performed in 1 M KOH, with a graphite rod and Hg/HgO electrode as the counter and reference electrodes, respectively. All the potentials were corrected versus a reversible hydrogen electrode (RHE) according to E_RHE = E_SCE + 0.255 V in 0.5 M H₂SO₄, E_RHE = E_Hg/HgO + 0.93 V in 1 M KOH.¹² Typically, 4 mg of the as-prepared catalyst was added in 1 mL of mixed solution (750 μL of water and 250 μL of ethanol), and then 60 μL Nafion (5 wt%) was added into the above solution. A homogeneous ink was obtained after 30 min of ultrasonication. Later, 5 μL of the catalyst ink was dropped onto the glassy carbon electrode (GCE, catalyst loading ∼0.28 mg cm⁻²) and dried in an oven at 60 °C. Linear sweep voltammetry (LSV) was performed in 0.5 M H₂SO₄ and 1 M KOH at a rate of 5 mV s⁻¹ without iR-based correction. The double-layer capacitances (C_dl) were calculated through cyclic voltammograms (CV) at various sweep rates from 20 to 200 mV s⁻¹. Electrochemical impedance spectroscopy (EIS) was carried out in the frequency range of 10⁵ to 0.1 Hz.

3. Results and discussion

X-ray diffraction (XRD) was used to examine the chemical composition of the as-prepared W₂C@CNT-S and W₂C/CNT electrocatalysts, as shown in Fig. 2a. The characteristic diffraction peaks located at 34.5°, 37.9°, 39.5°, 52.1°, 61.8°, 69.7°, 72.8°, and 75.0° were assigned to the (100), (002), (101), (102), (110), (103), (200), and (112) crystal planes of a hexagonal W₂C (PDF#35-0776), respectively, which was consistent with the previous reports that showed superior HER activities.^23,24,28 No metallic W or WC was observed. The broad peaks at about 25.5° had relatively weak intensities. Fig. 2b shows the Raman spectra of W₂C@CNT-S and W₂C/CNT, with two distinct characteristic peaks located at 1356 cm⁻¹ and 1580 cm⁻¹, corresponding to the D band and G band of the carbon-based networks, respectively. The intensity ratio of D band and G band (I_D/I_G) can be considered as a characteristic parameter to reflect the degree of the carbon graphitization.²⁹

Fig. 2 (a) XRD patterns and (b) Raman spectra of W₂C@CNT-S and W₂C/CNT.

It is also known that the intensity ratio of the D band and G band (I_D/I_G) can be considered as a characteristic parameter to reflect the density of defects. The I_D/I_G ratio of pure spray-dried CNT (without carbonization) was 1.27 (Fig. S1†), which was higher than those of W₂C@CNT-S6 (0.994), W₂C@CNT-S8 (1.055), W₂C@CNT-S10 (1.170), and W₂C/CNT (1.194). That means, high-temperature carbonization was beneficial to decrease the density of defects; in addition, with the decreasing CNT content of W₂C@CNT-S, the density of defects decreased.

In this work, the optimized W₂C@CNT-S8 with the best HER performance was utilized in all further characterizations and discussions. The chemical environment and valence state of W₂C@CNT-S8 electrocatalyst was characterized through X-ray photoelectron spectroscopy (XPS). As observed, the XPS survey spectrum (Fig. 3a) identified the presence of the W, C, and O in the W₂C@CNT-S8. The high-resolution of W 4f spectrumcan be seen in Fig. 3b, where the peaks located at binding energy of 32.1 and 34.2 eV are attributed to tungsten carbide (W 4f_7/2 and W 4f_5/2) in general.^23,30 The peaks at binding energies of 36.0 and 38.1 eV result from the inevitable surface oxidation of tungsten carbide when exposed to air.^19,20 According to the electrocatalytic results in Fig. S2,† we confirm that the WO₃ had a negligible contribution to the HER and that the HER activity of W₂C@CNT mainly originated from the W₂C instead of WO₃.^30,31Fig. 3c displays the high-resolution C 1s XPS spectrum, with peaks at 283.2, 284.6, 286.0, and 289.0 eV, corresponding to the C–W, C=C (or C–C), C–O, and O=C–O bonds, respectively.^31–33

Fig. 3 (a) XPS survey spectrum, (b) W 4f XPS spectrum, and (c) C 1s XPS spectrum of the W₂C@CNT-S8.

The morphologies and microstructures of the as-prepared products were observed by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As can be seen in Fig. S3,† the W₂C/CNT hybrid shows a large size and the particles are not distributed homogeneously on the carbon nanotube because of the big clusters and severe aggregation during the oven-drying instead of spray-drying treatment, which indicates a small quantity of active sites for HER. The SEM images of pure CNT networks are also shown in Fig. S4,† showing a rough surface and uneven size of holes in the CNT three-dimensional (3D) networks after the spray-drying approach. These results indicate that the spray-drying approach and the carbon nanotubes as a carbon source can not only open up the possibility of more active sites and better electrical conductivity, but also provide for efficient electron transfer. As shown in Fig. 4a and b, one can clearly observe that the microsphere is composed of carbon nanotubes with a rough surface and that no large-sized particles or aggregation can be seen. Moreover, as can be seen from the high-resolution TEM images (Fig. 4c and d, and Fig. S5†), the ultrasmall W₂C nanodots are homogeneously anchored on the surface of the CNTs. The formation of W₂C nanodots through the diffusion of C atoms from the CNTs through the solid–solid interface into the W lattice is slowed down due to the chemical inertness of the multi-walled carbon nanotubes, which not only effectively limits the overgrowth of carbide nanodots by suppressing their agglomeration, but also avoids excessive carbon deposition formed on their surfaces.^23,27 Furthermore, it has been reported that the nanodot materials show excellent performance and applicable prospects in electrochemistry and energy storage.^27,32–35 The lattice fringes with an interplanar distance of 0.228 nm and 0.34 nm correspond to the (101) plane of hexagonal W₂C and (002) plane of carbon nanotubes, respectively. Fig. 4e–g show the X-ray energy dispersive spectrometry (EDX) mapping images of one carbon nanotube decorated with W₂C nanodots. From Fig. 4e–g, one can observe that the W and C are uniformly distributed into the composite, which verifies that there are homogeneous dispersion of W₂C nanodots on the CNT networks.

Fig. 4 (a, b) SEM images (inset: scale bar of 1 μm), (c, d) HRTEM images, HAADF image (e) and corresponding W and C EDX element mappings (f, g) of W₂C@CNT-S8 (with scale bar of 10 nm).

The electrocatalytic performance of the as-prepared W₂C@CNT-S and W₂C/CNT materials was first evaluated by linear sweep voltammetry (LSV) without iR-correction in 0.5 M H₂SO₄ solution. Simultaneously, the HER polarization curves of the bare GCE, commercial Pt/C (20 wt%), and the spray-dried pure CNT were also measured for comparison. As observed in Fig. 5a, the commercial Pt/C displayed the highest catalytic activity toward HER in acidic medium with nearly zero onset overpotential; whereas, the bare GCE showed very poor catalytic activity. The spray-dried pure CNT also had negligible HER activity in 0.5 M H₂SO₄. It was impressively found that the W₂C@CNT-S electrocatalysts exhibited excellent performance, much better than the W₂C/CNT catalyst. In order to investigate the optimized weight ratio of W₂C and CNT on HER performance, a series of W₂C@CNT-S samples prepared with various molar ratios of C/W (6 : 1, 8 : 1, and 10 : 1) were also tested through electrochemical measurements. According to the Fig. S6,† the optimized W₂C@CNT-S8 exhibited a much smaller onset overpotential of 60 mV (vs. RHE) than the W₂C/CNT (∼110 mV). Furthermore, it is well known that the overpotential at a current density of 10 mA cm⁻² is another indispensable parameter for the HER.^36,37 To obtain the current density of 10 mA cm⁻², only about 176 mV, 192 mV, and 220 mV were needed for W₂C@CNT-S8, W₂C@CNT-S6, and W₂C@CNT-S10, respectively, which were lower than that needed for W₂C/CNT (240 mV). These results indicate that the HER activity mainly came from the W₂C species. W₂C@CNT-S6 did not display a better performance due to the lesser content of the CNT; while W₂C@CNT-S10 containing a greater amount of CNT also did not show better performance than W₂C@CNT-S8, due to the more CNT wrapped around the W₂C nanodots, which limits exposure to the active sites. The aforementioned consequences imply that the as-synthesized W₂C@CNT-S8 was comparable to those non-noble-metal-based HER electrocatalysts in acidic media, such as W_xC@WS₂,³⁰ WS₂/WO₂,³⁸ WC,³⁹ NiWS_x,⁴⁰ CoWS_x,⁴⁰ W₁₈O₄₉@WS2,⁴¹ and WSe₂/CNT⁴² (Fig. 5e, Table S1†).

Fig. 5 (a) Polarization curves for Pt/C, bare GCE, pure CNT, W₂C/CNT-S6, W₂C/CNT-S8, W₂C/CNT-S10, and W₂C/CNT in 0.5 M H₂SO₄ with a scan rate of 5 mV s⁻¹. (b) Tafel plots and (c) estimated C_dl and ECSAs for W₂C/CNT-S6, W₂C/CNT-S8, W₂C/CNT-S10, and W₂C/CNT. (d) Nyquist plot semicircles of W₂C/CNT-S6, W₂C/CNT-S8, W₂C/CNT-S10, and W₂C/CNT at −0.17 V vs. RHE in 0.5 M H₂SO₄. The data were fitted to the equivalent circuit shown in the inset. (e) Comparison of the onset overpotentials and Tafel slopes for W₂C@CNT-S, W₂C/CNT, and other electrocatalysts in acidic media.

The Tafel slope is an inherent parameter of electrocatalysts to investigate their kinetics mechanism and intrinsic activity with the HER process. As can be seen in Fig. 5b, the Tafel slopes of Pt/C, W₂C@CNT-S6, W₂C@CNT-S8, W₂C@CNT-S10, and W₂C/CNT were calculated from linear fitting of the polarization curves by the Tafel equation (η = b log j + a, where b is the Tafel slope and j is the current density). The Tafel slope of commercial Pt/C was 30.2 mV dec⁻¹ in 0.5 M H₂SO₄, which is consistent with the reported literature.^9,11,18 It is notable that W₂C@CNT-S8 displayed a small Tafel slope of 57.4 mV dec⁻¹, which was smaller than those of W₂C@CNT-S6 (59.8 mV dec⁻¹), W₂C@CNT-S10 (68.6 mV dec⁻¹), and W₂C/CNT (72.3 mV dec⁻¹). Generally, there are two main HER mechanisms in both acidic and alkaline media.^{26,36,49–51} In acidic media, the HER mechanism includes the following three steps.

Volmer reaction

H₃O⁺ + e⁻ + catalyst → catalyst-H_ads + H₂O (1)

Heyrovsky reaction

H₃O⁺ + e⁻ + catalyst-H_ads → catalyst + H₂ + H₂O (2)

Tafel reaction

2 × Catalyst-H_ads → 2 × catalysts + H₂ (3)

Similarly, in alkaline media, the HER mechanism also includes the following three steps.

Volmer reaction

H₂O + e⁻ + catalyst → catalyst-H_ads + OH⁻ (4)

Heyrovsky reaction

H₂O + e⁻ + catalyst-H_ads → catalyst + H₂ + OH⁻ (5)

Tafel reaction

2 × Catalyst-H_ads → 2 × catalyst + H₂ (6)

The W₂C@CNT-S8 sample had the faster HER reaction kinetics owing to the smaller Tafel slope of 57.4 mV dec⁻¹, demonstrating that the reaction of hydrogen evolution on the W₂C@CNT-S8 followed the Volmer–Heyrovsky mechanism (Tafel slope range from 40 to 120 mV dec⁻¹), which is the rate-determining step.^26,36 These results indicate that the spray-drying method plays an important role in the preparation of W₂C@CNT-S networks and in attaining an outstanding HER performance.

To further investigate the intrinsic differences for HER behavior among the catalysts, the electrochemical active surface area (ECSA), which is obtained from the double-layer capacitances (C_dl), was estimated. The cyclic voltammograms (CV) were measured at various scan rates in the range 0.205–0.305 V (vs. RHE) in 0.5 M H₂SO₄ solution (Fig. S7†). Fig. 5c shows that the double-layer capacitance (C_dl) of W₂C@CNT-S8 was 24.5 mF cm⁻² in 0.5 M H₂SO₄, which was the highest among those of W₂C@CNT-S6 (16.65 mF cm⁻²), W₂C@CNT-S10 (7.97 mF cm⁻²), and W₂C/CNT (6.4 mF cm⁻²). The high C_dl value suggests the presence of additional active sites, which can enhance the transport of electrons for electrochemical processes. Furthermore, electrochemical impedance spectroscopy (EIS) was performed to investigate the favorable HER kinetics between W₂C@CNT-S and W₂C/CNT. Fig. 5d shows the corresponding Nyquist plots of these catalysts at −0.17 V vs. RHE in 0.5 M H₂SO₄. The charge-transfer resistance (R_ct) was obtained from the fitting of the semicircles via an equivalent circuit model, which is shown in the inset. The R_ct values are ranked as follows: W₂C@CNT-S8 (37.83 Ω) < W₂C@CNT-S6 (53.59 Ω) < W₂C@CNT-S10 (83.38 Ω) < W₂C/CNT (94.6 Ω), well in agreement with their Tafel slopes. The smaller R_ct value of W₂C@CNT-S8 suggests its faster electron kinetics for the HER.

The as-prepared W₂C@CNT-S and W₂C/CNT were also investigated as HER catalysts in alkaline media (1 M KOH). As demonstrated in Fig. 6, the HER performance of these samples was slightly enhanced in alkaline solution instead of acidic solution. As shown in Fig. 6a and Fig. S8,† the W₂C@CNT-S8 exhibits superior performance with a smaller onset overpotential of 40 mV (vs. RHE), while the overpotential to reach 10 mA cm⁻² was decreased to 148 mV, which was lower than that for W₂C@CNT-S6 (186 mV), W₂C@CNT-S10 (213 mV), and W₂C/CNT (248 mV). In addition, Fig. 6b displays the Tafel slopes of 56.2 mV dec⁻¹ for W₂C@CNT-S8, 60.7 mV dec⁻¹ for W₂C@CNT-S6, 63.8 mV dec⁻¹ for W₂C@CNT-S10, and 88.6 mV dec⁻¹ for W₂C/CNT. The Tafel slope suggests the W₂C@CNT-S8 catalyst proceeded through a Volmer–Heyrovsky mechanism.^36,43 The double-layer capacitances of the as-prepared samples were examined by cyclic voltammograms at various scan rates ranging from 0.03 V to −0.13 V (vs. RHE) in 1 M KOH (Fig. S9†). According to Fig. 6c, the capacitances (C_dl) of W₂C@CNT-S8 was 21.87 mF cm⁻², which was higher than those of W₂C@CNT-S6 (13.45 mF cm⁻²), W₂C@CNT-S10 (10.04 mF cm⁻²), and W₂C/CNT (6.03 mF cm⁻²), implying more active sites for the HER process. Fig. 6d shows the corresponding Nyquist plots of these catalysts by EIS measurement at −0.20 V vs. RHE. The R_ct value of W₂C@CNT-S8 (36.34 Ω) was much smaller than that of W₂C@CNT-S6 (60.61 Ω), W₂C@CNT-S10 (78.75 Ω), and W₂C/CNT (119.4 Ω), indicating the faster electron-transfer rate of the W₂C@CNT-S8 electrocatalyst for HER in alkaline solution, which was in excellent agreement with the aforementioned results of the HER activity. The W₂C@CNT-S8 catalyst had an ultralow onset overpotential of 40 mV (vs. RHE) and a smaller Tafel slope of 56.2 mV dec⁻¹ comparable to those reported for non-noble-metal-based electrocatalysts in alkaline media, such as p-WC_x NWs,⁴⁴ 1D, Co-Mo₂C,⁴⁵ WS₂/W₂C@NSPC,⁴⁶ NiS₂/MoS₂,⁴⁷ MoSe₂-CoSe₂ nanotube,¹¹ Co_0.85Se@NG₃,¹⁰ and WN NA/CC⁴⁸ (Fig. 6e, Table S1†).

Fig. 6 (a) Polarization curves for Pt/C, bare GCE, pure CNT, W₂C/CNT-S6, W₂C/CNT-S8, W₂C/CNT-S10, and W₂C/CNT in 1 M KOH with a scan rate of 5 mV s⁻¹. (b) Tafel plots and (c) estimated C_dl and ECSAs for W₂C/CNT-S6, W₂C/CNT-S8, W₂C/CNT-S10, and W₂C/CNT. (d) Nyquist plot semicircles of the W₂C/CNT-S6, W₂C/CNT-S8, W₂C/CNT-S10, and W₂C/CNT at −0.20 V vs. RHE in 1 M KOH. The data were fitted to the equivalent circuit shown in the inset. (e) Comparison of the onset overpotential and Tafel slope for W₂C@CNT-S, W₂C/CNT, and other electrocatalysts in alkaline media.

Durability is a crucial parameter for estimating the practical application of HER electrocatalysts. The long-term stability test of W₂C@CNT-S8 was assessed by 2000 CV cycles at a scan rate of 100 mV s⁻¹ in 0.5 M H₂SO₄ and 1 M KOH. As shown in Fig. 7a and c, the polarization curves of the W₂C@CNT-S8 catalyst showed negligible decay of the cathodic current after 2000 cycles of CV under both acidic and alkaline electrolytes, indicating their excellent stability. Meanwhile, chronoamperometric (CA) measurements were performed for W₂C@CNT-S8 at a constant overpotential of −180 mV and −200 mV in 0.5 M H₂SO₄ and 1 M KOH electrolyte, respectively. As shown in Fig. 7b and d, the I–t plots show that the current density exhibited negligible degradation after continuous testing for over 30 h in both 0.5 M H₂SO₄ and 1 M KOH, respectively. We further characterized the W₂C@CNT-S8 after the HER stability test in both acidic and alkaline solutions. As shown in Fig. S10–S13,† the SEM and TEM images indicate that the porous network structure of W₂C@CNT-S8 was maintained nearly unchanged even after the long-term stability tests in both alkaline and acidic solutions, which means the W₂C@CNT-S8 electrocatalyst is a promising catalyst for practical application.

Fig. 7 (a, c) Polarization curves of W₂C@CNT-S8 before and after 2000 CV cycles 0.5 M H₂SO₄ and 1.0 M KOH, respectively. (b, d) Chronoamperometry curves of W₂C@CNT-S8 at a constant overpotential of −180 mV (0.5 M H₂SO₄) and −200 mV (1.0 M KOH).

Moreover, the electrocatalytic process of W₂C@CNT-S is schematically illustrated in Fig. 8. In this process, the electrons can transfer through the CNTs to W₂C nanodots anchored on the 3D and conductive CNT skeleton. Hydrogen ions at the exposed active sites of W₂C nanodots electrocatalyst are reduced by the transferred electrons, followed by the release of hydrogen gas. As mentioned above, the excellent and efficient electrocatalytic activity and durability of W₂C@CNT-S toward HER could be attributed to the following aspects: (i) the spray-drying method can create porous CNT networks, as a carbon framework, which can lead to the uniform dispersion of W₂C nanodots, which provide more active sites to enhance the HER catalytic activity; (ii) the CNT skeleton ensures outstanding electrical conductivity and prevents the W₂C nanodots from being aggregated; (iii) due to the chemical inertness of CNT, the diffusion rate of C atom from carbon nanotubes into the W lattice is slowed down, not only effectively limiting the overgrowth of W₂C particles to ensure the formation of W₂C nanodots, but also preventing extensive carbon deposition formed on the surface of tungsten carbide;^23,27,52 (iv) the robust contact between W₂C and CNT facilitates the efficient electron-transfer rate and enables excellent durability for the hydrogen evolution reaction.

Fig. 8 Schematic illustration of the W₂C@CNT-S and the HER process.

4. Conclusions

In this work, we first developed a scalable and cost-efficient method to synthesize ultrasmall W₂C nanodots decorated on CNT networks via a spray-drying process (designated as W₂C@CNT-S). Compared to the W₂C/CNT (obtained without spray drying), the optimized W₂C@CNT-S8 exhibited better HER performance with a lower onset overpotential of 60 or 40 mV (vs. RHE), a smaller Tafel slope of 57.4 or 56.2 mV dec⁻¹, and a current density@10 mA cm⁻² of 176 or 148 mV (vs. RHE) in acidic and alkaline media, respectively. Besides, it exhibited outstanding stability after long-term operation for over 30 h in both 0.5 M H₂SO₄ and 1 M KOH. The superior HER performance in both acidic and alkaline solution could be attributed to the CNT network structure and ultrafine W₂C nanodots homogenously decorated on conductive CNT networks, which expose sufficient active sites and promote high-efficiency electron transfer for electrochemical reactions. This work presents a novel, low-cost, and scalable strategy, which can be extended to design other nonprecious-metal-based carbides with efficient activity for the hydrogen evolution reaction.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The research was supported by the National Natural Science Foundation of China (Grant No. 21773024, 51372033), and National High Technology Research and Development Program of China (Grant No. 2015AA034202).

Notes and references

1. J. A. Turner, Science, 2004, 305, 972.
2. M. S. Faber and S. Jin, Energy Environ. Sci., 2014, 7, 3519.
3. X. Zou and Y. Zhang, Chem. Soc. Rev., 2015, 44, 5148.
4. M. R. Gao, Y. F. Xu, J. Jiang and S. H. Yu, Chem. Soc. Rev., 2013, 42, 2986.
5. J. Wang, F. Xu, H. Jin, Y. Chen and Y. Wang, Adv. Mater., 2017, 29, 1605838.
6. Z. Wang, Q. Li, H. Xu, C. Dahl-Petersen, Q. Yang, D. Cheng, D. Cao, F. Besenbacher, J. V. Lauritsen, S. Helveg and M. Dong, Nano Energy, 2018, 49, 634.
7. 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.
8. S. Lu, X. Shang, L. Zhang, B. Dong, W. Gao, F. Dai, B. Liu, Y. Chai and C. Liu, Appl. Surf. Sci., 2018, 445, 445.
9. B. Yu, F. Qi, B. Zheng, W. Hou, W. Zhang, Y. Li and Y. Chen, J. Mater. Chem. A, 2018, 6, 1655.
10. H. Wang, X. Wang, D. Yang, B. Zheng and Y. Chen, J. Power Sources, 2018, 400, 232.
11. X. Wang, B. Zheng, B. Yu, B. Wang, W. Hou, W. Zhang and Y. Chen, J. Mater. Chem. A, 2018, 6, 7842.
12. B. Wang, Z. Wang, X. Wang, B. Zheng, W. Zhang and Y. Chen, J. Mater. Chem. A, 2018, 6, 12701.
13. J. Ren, L. Chen, C. Weng, G. Yuan and Z. Yuan, ACS Appl. Mater. Interfaces, 2018, 10, 33276.
14. T. Wu, M. Pi, X. Wang, W. Guo, D. Zhang and S. Chen, Appl. Surf. Sci., 2018, 427, 800.
15. H. Yan, Y. Xie, Y. Jiao, A. Wu, C. Tian, X. Zhang, L. Wang and H. Fu, Adv. Mater., 2018, 30, 1704156.
16. Y. Zhong, X. Xia, F. Shi, J. Zhan, J. Tu and H. J. Fan, Adv. Sci., 2016, 3, 1500286.
17. W. F. Chen, C. H. Wang, K. Sasaki, N. Marinkovic, W. Xu, J. T. Muckerman, Y. Zhu and R. R. Adzic, Energy Environ. Sci., 2013, 6, 943.
18. Y. Hu, D. Guan, B. Yu, W. Hou, B. Zheng, W. Zhang and Y. Chen, Electrochim. Acta, 2018, 263, 192.
19. N. Han, K. R. Yang, Z. Lu, Y. Li, W. Xu, T. Gao, Z. Cai, Y. Zhang, V. S. Batista, W. Liu and X. Sun, Nat. Commun., 2018, 9, 924.
20. Y. Xu, X. Xiao, Z. Ye, S. Zhao, R. Shen, C. He, J. Zhang, Y. Li and X. Chen, J. Am. Chem. Soc., 2017, 139, 5285.
21. Z. Zhao, F. Qin, S. Kasiraju, L. Xie, M. K. Alam, S. Chen, D. Wang, Z. Ren, Z. Wang, L. C. Grabow and J. Bao, ACS Catal., 2017, 7, 7312.
22. S. Emin, C. Altinkaya, A. Semerci, H. Okuyucu, A. Yildiz and P. Stefanov, Appl. Catal., B, 2018, 236, 147.
23. Q. Gong, Y. Wang, Q. Hu, J. Zhou, R. Feng, P. N. Duchesne, P. Zhang, F. Chen, N. Han, Y. Li, C. Jin, Y. Li and S. Lee, Nat. Commun., 2016, 7, 13216.
24. L. Zhang, Y. Ma, Z. Lang, Y. Wang, S. U. Khan, G. Yan, H. Tan, H. Zang and Y. Li, J. Mater. Chem. A, 2018, 6, 15395.
25. A. Kurlov and A. Gusev, Inorg. Mater., 2006, 42, 121.
26. H. Wei, Q. Xi, X. Chen, D. Guo, F. Ding, Z. Yang, S. Wang, J. Li and S. Huang, Adv. Sci., 2018, 1700733.
27. B. Yu, D. Yang, Y. Hu, J. He, Y. Chen and W. He, Small Methods, 2018, 1800287.
28. G. Yan, C. Wu, H. Tan, X. Feng, L. Yan, H. Zang and Y. Li, J. Mater. Chem. A, 2017, 5, 765.
29. F. Malara, S. Carallo, E. Rotunno, L. Lazzarini, E. Piperopoulos, C. Milone and A. Naldoni, ACS Catal., 2017, 7, 4786.
30. F. Wang, P. He, Y. Li, T. A. Shifa and Y. Deng, Adv. Funct. Mater., 2017, 27, 1605802.
31. I. Kim, S. Park and D. Kim, Nanoscale, 2018, 10, 21123.
32. C. Tan, Z. Luo, A. Chaturvedi, Y. Cai, Y. Du, Y. Gong, Y. Huang, Z. Lai, X. Zhang, L. Zheng, X. Qi, M. H. Goh, J. Wang, S. Han, X. Wu, L. Gu, C. Kloc and H. Zhang, Adv. Mater., 2018, 30, 1705509.
33. Z. Lai, A. Chaturvedi, Y. Wang, T. H. Tran, X. Liu, C. Tan, Z. Luo, B. Chen, Y. Huang, G. Nam, Z. Zhang, Y. Chen, Z. Hu, B. Li, S. Xi, Q. Zhang, Y. Zong, L. Gu, C. Kloc, Y. Du and H. Zhang, J. Am. Chem. Soc., 2018, 140, 8563.
34. M. Wang, C. Tang, C. Ye, J. Duan, C. Li, Y. Chen, S. Bao and M. Xu, J. Mater. Chem. A, 2018, 6, 14734.
35. K. Chang, H. Pang, X. Hai, G. Zhao, H. Zhang, L. Shi, F. Ichihara and J. Ye, Appl. Catal., B, 2018, 232, 446.
36. Y. Zhang, Y. Wang, C. Han, S. Jia, S. Zhou and J. Zang, Nanoscale, 2017, 9, 19176.
37. B. Yu, Y. Hu, F. Qi, X. Wang, B. Zheng, K. Liu, W. Zhang, Y. Li and Y. Chen, Electrochim. Acta, 2017, 242, 25.
38. J. Wang, W. Wang, Z. Wang, J. G. Chen and C. Liu, ACS Catal., 2016, 6, 6585.
39. H. Fei, Y. Yang, X. Fan, G. Wang, G. Ruan and J. M. Tour, J. Mater. Chem. A, 2015, 3, 5798.
40. P. D. Tran, S. Y. Chiam, P. P. Boix, Y. Ren, S. S. Pramana, J. Fize, V. Artero and J. Barber, Energy Environ. Sci., 2013, 6, 2452.
41. B. Seo, H. Y. Jeong, S. Y. Hong, A. Zak and S. H. Joo, Chem. Commun., 2015, 51, 8334.
42. X. Wang, Y. Chen, B. Zheng, F. Qi, J. He, P. Li and W. Zhang, Electrochim. Acta, 2016, 222, 1293.
43. J. Xiong, J. Li, J. Shi, X. Zhang, N. Suen, Z. Liu, Y. Huang, G. Xu, W. Cai, X. Lei, L. Feng, Z. Yang, L. Huang and H. Cheng, ACS Energy Lett., 2018, 3, 341.
44. B. Ren, D. Li, Q. Jin, H. Cui and C. Wang, J. Mater. Chem. A, 2017, 5, 13196.
45. J. Wan, J. Wu, X. Gao, T. Li, Z. Hu, H. Yu and L. Huang, Adv. Funct. Mater., 2017, 27, 1703933.
46. Y. Li, X. Wu, H. Zhang and J. Zhang, ACS Appl. Energy Mater., 2018, 1, 3377.
47. P. Kuang, T. Tong, K. Fan and J. Yu, ACS Catal., 2017, 7, 6179.
48. J. Shi, Z. Pu, Q. Liu, A. M. Asiri, J. Hu and X. Sun, Electrochim. Acta, 2015, 154, 345.
49. P. Kuang, B. Zhu, Y. Li, H. Liu, J. Yu and K. Fan, Nanoscale Horiz., 2018, 3, 317.
50. H. Gu, W. Fan and T. Liu, Nanoscale Horiz., 2017, 2, 277.
51. S. Zhang, X. Xiao, T. Lv, X. Lv, B. Liu, W. Wei and J. Liu, Appl. Surf. Sci., 2018, 446, 10.
52. Z. Chen, L. Duan, T. Sheng, X. Lin, Y. Chen, Y. Chu, S. Sun and W. Lin, ACS Appl. Mater. Interfaces, 2017, 9, 20594.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8nr10281c

---