Abstract

In this study, a new and facile process was developed for the preparation of composite catalysts based on tungsten oxide (WO3) by batch reactor routes. The structures, morphologies, compositions, and characteristics of synthesized materials were investigated and confirmed. Using batch reactor processes, WO3 nanorods (WO3 NR), heterostructures of WS2/WO3 nanobricks (WS2/WO3 NB), and WS2/WO3 nanorods (WS2/WO3 NR) were successfully prepared. The prepared materials were then employed for hydrogen evolution reaction (HER) to investigate their catalytic performance. The results indicated that the electrocatalytic activities of WS2/WO3 NR are significantly improved compared to those of WO3 NR and WS2/WO3 NB. This improvement could be attributed to the formation of heterostructure between WS2 and WO3 elements in highly uniform materials, which could create the synergistic effect and further improve the catalytic activities of the catalyst. The data shows that the Tafel slope of WS2/WO3 NR (82.7 mV dec−1) is significantly lower than that of WO3 NR (112.5 mV dec−1) and WS2/WO3 NB (195.5 mV dec−1). Furthermore, the resistance of WS2/WO3 NR (397.7 Ω) is markedly decreased compared to those of WO3 NR (1816 Ω) and WS2/WO3 NB (3597 Ω). The results indicate that WS2/WO3 NR could be a great catalyst for electrochemical applications.

1. Introduction

Over the last decades, the excessive use of fossil fuels has become a critical issue that must be addressed because of greenhouse gas emissions and global warming [15]. New materials and breakthrough technologies are increasingly necessary to conserve energy and develop environmentally friendly energy sources. Tremendous efforts have been made to overcome the emerging environmental problems [613]. With the advent of cutting-edge technologies, hydrogen is one of the most prominent candidates as an ecosystem-friendly and reusable energy source [1421]. Different materials have been investigated to enhance reaction performance and produce hydrogen gas [2228]. Thus far, noble metals such as Pt group metals have demonstrated higher active electrocatalyst performances for the hydrogen evolution reaction (HER) than other materials [2932]. However, the scarcity and high cost of those catalysts are their major drawbacks. Therefore, the design of abundant, low-cost materials with excellent catalytic performance for HER applications remains a considerable challenge for researchers. Transition metals and their compounds are promising candidates for the preparation of excellent catalytic materials for the HER. Recently, transition metals and their compounds have been extensively investigated for HER performance because of their benefits for electrocatalyst materials [3335]. Among many materials, the well-known transition metal W and its derivatives have distinctive electronic characteristics [3638]. In particular, WO3 has been extensively investigated for various applications such as electrochromic devices [3941], sensors [42, 43], solar cells [44, 45], photocatalysts [46, 47], and HER applications [4851].

Catalysts based on WO3 exhibit poor catalytic properties because the adsorption energy of atomic hydrogen on WO3 molecules is inadequate, resulting in the low activity of WO3 for the HER in the electrolyte [52, 53]. This can be attributed to the low conductivity and intrinsic inactivity of WO3. In recent decades, several approaches have been developed to improve the HER performance of WO3, including metallic and non-metallic doping to form hybrid materials [54, 55]. Noble metal doping or hybrid materials may create synergistic effects between different elements that can tune the electronic structure, increase the number of catalytic active sites, and enhance the HER activity. However, noble metal doping has several limitations, including complicated processes and expensive materials [56, 57]. Therefore, numerous studies have been conducted to develop facile, scalable, and low-cost processes for the synthesis of heterostructure catalysts based on WO3 materials [5860]. In previous researches, the performance of different catalysts based on various structures of WO3 and/or WS2 is listed in Table S1.

Herein, a simple, low-cost, and scalable method is introduced to synthesize different structures of WO3 and its derivatives using a batch reactor. Various measurements were conducted to confirm the formation and morphology of the synthesized materials. Different materials including WO3 nanorods (WO3 NR) and heterostructures of WS2/WO3 nanobricks (WS2/WO3 NB) and WS2/WO3 nanorods (WS2/WO3 NR) were prepared. Electrochemical studies demonstrated that the HER activity of WS2/WO3 NR was higher than those of WO3 NR and WS2/WO3 NB. The highly uniform and synergistic effect of WS2 and WO3 elements in the synthesized WS2/WO3 NR could be responsible for its excellent HER performance. Based on the catalytic performance, the synthesized WS2/WO3 NR is a promising material for electrochemical applications.

2. Experimental Section

The experiments were conducted in a closed batch reactor. The different mechanisms of procedures are briefly illustrated in Section 2.6.

2.1. Materials

All materials were used as received, without further purification. Ammonium metatungstate hydrate (NH4)6H2W12O40.xH2O) (AMT), thioacetamide (TAA) (C2H5NS, 99%), and hydrochloric acid (HCl, ACS reagent, 37%) were purchased from Sigma–Aldrich. Dimethylformamide (DMF) was supplied by Alfa Aesar. Deionized (DI) water (18.3 MΩ cm−1) was obtained from Millipore Milli-Q.

2.2. Synthesis of WO3 NR

TAA (3.75 g) was dissolved and stirred for 30 min in DI water (25 mL) in a 100 mL Teflon beaker. Subsequently, AMT (4 g) was added to the prepared solution, followed by stirring continuously for 30 min. Then, the Teflon beaker was placed in a batch reactor system and heated at 180°C for 24 h. After cooling to room temperature, the precipitate was formed at the bottom of the Teflon beaker, and it was collected and centrifuged thrice with DI water. The obtained powder, WO3 NR, was dried in a vacuum dryer at 90°C for 12 h.

2.3. Synthesis of WS2/WO3 NB Composite

The WS2/WO3 NB composite was also synthesized via the process used for WO3 NR, with slight modifications. Initially, TAA (3.75 g) was dissolved and stirred for 30 min in DMF (25 mL) in a 100 mL Teflon beaker. Subsequently, AMT (4 g) was added to the prepared solution, followed by stirring continuously for 30 min. Then, the Teflon beaker was placed in a batch reactor system and heated at 180°C for 24 h. After cooling to room temperature, the precipitate was formed at the bottom of the Teflon beaker, and it was collected and centrifuged thrice with DI water. The obtained powder, WS2/WO3 NB, was dried in a vacuum dryer at 90°C for 12 h.

2.4. Synthesis of WS2/WO3 NR Heterostructure

The WS2/WO3 NR heterostructure was also synthesized via the technique used for WO3 NR and WS2/WO3 NB, with slight modifications. Firstly, TAA (3.75 g) was dissolved and stirred for 30 min in DI water (25 mL) in a 100 mL Teflon beaker. Subsequently, AMT (4 g) was added to the prepared solution, followed by stirring continuously for 30 min. After that, 37% HCl (3 mL) was added to the Teflon beaker, followed by stirring continuously for 30 min. Then, the Teflon beaker was placed in a batch reactor system and heated at 180°C for 24 h. After cooling to room temperature, the precipitate was formed at the bottom of the Teflon beaker, and it was collected and centrifuged thrice with DI water. The obtained powder, WS2/WO3 NR, was dried in a vacuum dryer at 90°C for 12 h.

2.5. Electrochemical Measurements

The HER performance of different materials was evaluated by a three-electrode system (Ivium potentiostat V55630) using 0.5 M H2SO4 electrolyte solution. Graphite rod, saturated calomel, and catalyst-coated glassy carbon electrodes with diameters of 3 mm were used as the counter, reference, and working electrodes, respectively. The catalyst inks were prepared by mixing 1 mg of each powder with 1 mL of DMF water and 50 μL of Nafion (5%) which works as the stabilizer. The prepared inks were then drop coated on glassy carbon, followed by drying at 90°C for 30 min. Linear sweep voltammetry (LSV) was conducted to measure the HER performance (scan rate of 10 mV s−1). To calculate the double layer capacitance (), the cyclic voltammetry measurements were investigated from 0 to 0.2 V at various scan rates of 10, 20, 30, 40, and 50 mV s−1. Electrochemical impedance spectroscopy (EIS) was studied at a potential of 280 mV and frequencies ranging from 100 kHz to 0.1 Hz.

2.6. Material Characterization

The crystallinity of the synthesized materials was confirmed using X-ray diffraction (XRD, D8-Advance/Bruker-AXS). Additionally, Raman spectroscopy (LabRAM HR, Horiba Jobin Yvon) was also studied to confirm the structures of prepared materials. After that, the morphologies, sizes, and shapes of the synthesized materials were analyzed using field-emission scanning electron microscopy (FE-SEM, SIGMA/Carl Zeiss). The chemical compositions as well as the oxidation states of the constituent elements of WO3 NR, WS2/WO3 NB, and WS2/WO3 NR were then investigated by using X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific, K-Alpha, USA).

2.7. Synthesis Process and Proposed Mechanisms

The proposed reaction mechanism of the synthesis process is shown in Figure 1. (1)Synthesis of WO3 NR. Source materials (AMT, TAA, and ).(2)Synthesis of WS2/WO3 NB Composite. Source materials (AMT, TAA, and DMF).(3)Synthesis of WS2/WO3 NR Heterostructure. Source materials (AMT, TAA, HCl, and H2O).

When hydrothermal or solvothermal processes are conducted to prepare materials in closed system, the reactions inside the batch reactor could take place and be highly complicated. Based on the source materials and their properties, the above-suggested mechanisms (1-13) could take place in batch reactor, depending on the source materials. Those processes take place in a closed system at a high temperature (180°C) for a long time (24 h). Using AMT, TAA, and different agents (H2O, DMF, and H2O+HCl), various gases could be released (H2S, NH3, SO2, or even water vapour) in a small volume of reactor, which could significantly change the pH of the solution, the pressure inside of the batch reactor, and/or produce new ions. Those factors are the main reason which leads to the form of various structures and morphologies of synthesized materials [23, 61].

3. Results and Discussion

The crystallinity and structure of the synthesized materials were well investigated by using XRD measurement. In Figure 2, it is clear that all the peaks in the XRD pattern of WO3 NR can be indexed to the monoclinic WO3 (JCPDS Card No. 83-0950) and the hexagonal WO3 (JCPDS Card No. 85-2460), as confirmed in a previous study [62, 63]. No other peaks could be observed in the XRD pattern of WO3 NR which indicated that the high purity of WO3 NR material was successfully synthesized. To confirm the appearance of WS2 in WS2/WO3 NB and WS2/WO3 NR, the pure WS2 was also synthesized by Teflon line autoclave as previous process [64]. The XRD pattern of pure WS2 is provided in Figure S1. In the XRD pattern of WS2/WO3 NB, there are three peaks located at 29° (004), 32° (101), and 35° (102) which could be ascribed to the hexagonal phase of WS2 (JCPDS Card No. 08-0237) [24]. The XRD intensity of WO3 peaks in WS2/WO3 NB is lower than those of WO3 NR, and those peaks could be ascribed to the orthorhombic phase of tungsten oxide hydrate WO3.H2O which was confirmed in the previous study [48]. The XRD result indicates the mixed phase of WS2 and WO3, suggesting the coexistence of WS2 and WO3 in WS2/WO3 NB. In the XRD pattern of WS2/WO3 NR, there are also various peaks of WO3 which are in line with the peaks of WO3 NR as mentioned above. Besides that, there are two peaks of WS2 located at 29° (004) and 35° (102). The XRD data indicated the successful synthesis of different catalysts including WO3 NR, WS2/WO3 NB, and WS2/WO3 NR.

Figure 3 shows the Raman spectra of the synthesized WO3 NR, WS2/WO3 NB, and WS2/WO3 NR. In the Raman spectrum of WO3 NR, the strongest peak centered at 786 cm−1 corresponds to the stretching vibration of O–W–O bonds. The weak peak at 292 cm−1 is attributed to the bending vibration of W–O–W bonds [65]. The vibrational modes centered at 103 cm−1 correspond to the lattice modes of hexagonal WO3 [66]. Furthermore, a shoulder peak appearing at 935 cm−1 is attributed to the –W=O bonds of the hexagonal WO3 crystal [67, 68]. The Raman spectra of WS2/WO3 NB and WS2/WO3 NR reveal that both materials exhibit characteristic peaks at 264, 350, 418, 705, and 805 cm−1. The peaks located at 264 cm−1 are assigned to the bending vibration of bridging oxygen of W–O–W bonds. The peaks located at approximately 705 and 805 cm−1 correspond to the stretching vibrations of the O–W–O bonds. The Raman peaks of WS2 appear in the spectra of WS2/WO3 NB and WS2/WO3 NR. The peak position of the in-plane mode is located at approximately 350 cm−1. For the interlayer vibration mode A1g, the peaks are located at approximately 418 cm−1, as reported in previous studies [69, 70]. The Raman spectra indicate that all the materials, namely, WO3 NR, WS2/WO3 NB, and WS2/WO3 NR, were effectively synthesized.

XPS was used to study the chemical compositions of the material surfaces. Figure 4(a) shows the wide XPS spectra of WO3 NR, WS2/WO3 NB, and WS2/WO3 NR. Peaks of W, S, O, and N are observed in all the synthesized materials, except for the absence of the S peak in the WO3 NR spectrum. This result confirms that WO3 NR, WS2/WO3 NB, and WS2/WO3 NR were successfully synthesized. The coincident appearance of nitrogen elements could be caused by using of AMT as a source of W which contains nitrogen elements. The presence of in-situ doped N2 elements in catalysts could increase the contact of catalysts with electrolyte and improve the intrinsic conductivity of materials [71]. Figures 4(b), 4(d), and 4(f) depict the high-resolution fitted peaks of W 4f in WO3 NR, WS2/WO3 NB, and WS2/WO3 NR, respectively. In Figure 4(b), the XPS spectrum of W 4f in WO3 NR is deconvoluted into two main states, namely, W 4f7/2 and W 4f5/2, located at approximately 35.2 and 37.4 eV, respectively. The XPS data indicates that pure WO3 NR was synthesized. In contrast, the XPS spectra of W 4f in WS2/WO3 NB and WS2/WO3 NR can be fitted into four different peaks, as shown in Figures 4(d) and 4(f), respectively. The peaks centered at approximately 31.8 and 33.9 eV are assigned to W (IV) of W–S bonding (WS2), whereas those located at approximately 35.7 and 38 eV can be indexed to W (VI) of W–O bonding (WO3). In addition, S peaks are observed only for the synthesized WS2/WO3 NB and WS2/WO3 NR. The high-resolution XPS peaks of S in the synthesized WS2/WO3 NB and WS2/WO3 NR are presented in Figures 4(c) and 4(e), respectively. In Figures 4(c) and 4(e), the main doublet of the binding energies of 161.6 and 162.8 eV in the high-resolution XPS peaks of S 2p is ascribed to the S 2p3/2 and S 2p1/2 states of the W–S bond in WS2, respectively. Moreover, the peak located at approximately 168.7 eV corresponds to the S–O bond (SO2), which is attributed to the inevitable oxidation of the composite in air [72]. The presence of WO3 and WS2 in XPS data confirms the successful synthesis of WS2/WO3 NB and WS2/WO3 NR.

Figure S2 shows the high-resolution XPS peaks of O 1 s in the synthesized materials including WO3 NR, WS2/WO3 NB, and WS2/WO3 NR. The deconvoluted peaks of O 1 s in the synthesized materials are considerably similar, with slightly shifted peaks after fitting, which could be assigned to the different synthesized structures. All the O 1 s peaks can be fitted to two main peaks, wherein the higher peaks are assigned to the W–O–W bonding and the lower peaks correspond to the –OH groups owing to contamination or crystal water [73]. All the XPS data confirm that different materials such as WO3 NR, WS2/WO3 NB, and WS2/WO3 NR were successfully synthesized.

The morphologies, sizes, and shapes of the prepared materials were investigated using FE-SEM at various scales. Figure 5 presents the FE-SEM images of WO3 NR, WS2/WO3 NB, and WS2/WO3 NR. The sizes, shapes, and morphologies of the synthesized materials are highly uniform. In Figures 5(a) and 5(b), the length of WO3 NR is approximately 200–300 nm with a diameter of approximately 30–50 nm. In contrast, in Figures 5(c) and 5(d), the shape of WS2/WO3 NB appears in cubic shape, with dimensions of approximately . As can be seen from Figures 5(e) and 5(f), it seems that the WS2 layer covers the surface of the WO3 NR material. The synthesized WS2/WO3 NR is highly uniform, with a length range of approximately 100–200 nm and a diameter range of approximately 15–30 nm. The FE-SEM data indicate that different morphologies of catalysts based on WO3 have been well prepared.

During the HER, H2 is released when protons (H+) in the electrolyte receive electrons from an applied voltage on the electrode surface. The HER in acidic media involves the following steps:

The efficiency of hydrogen evolution depends strongly on the electrode properties because hydrogen gas could be released on the cathode surface. Therefore, the more conductive and larger the active surface of the catalysts, the higher the catalytic performance. Figure 6 shows the electrochemical measurements of all samples with reference to the HER performance. The LSV data clearly indicate the poor catalytic behavior of WS2/WO3 NB. This could be explained by the big size of the WS2/WO3 NB particles, as revealed by the FE-SEM results. The large size of the WS2/WO3 NB is the primary reason for its low conductivity and small active surface. In contrast, the WO3 NR and WS2/WO3 NR samples exhibit good HER activities and higher current densities at lower overpotentials. The WO3 NR and WS2/WO3 NB samples achieve a current density of 10 mA cm−2 at overpotentials of 284 and 394 mV, respectively. The WS2/WO3 NR sample attains the current density of 10 mA cm−2 at an overpotential of 224 mV, implying that the HER activity of WS2/WO3 NR is enhanced compared with that of WO3 NR and WS2/WO3 NB. These results suggest that the heterostructure of WS2 and WO3 could play a crucial role in improving the HER activity in acidic media. The binary structure of WS2 and WO3 in the sample may create a synergistic effect that consequently increases the number of active sites and enhances the conductivity, thereby improving the HER performance [64, 7477].

The Tafel slope is another important parameter for evaluating the catalytic activity of catalysts and is strongly related to the catalytic activity of the materials. The reaction mechanism and exchange current density were interpreted based on the calculated Tafel slopes. Figure 6(b) depicts the Tafel slopes of the synthesized materials. The Tafel slope of WS2/WO3 NR (82.7 mV dec−1) is considerably smaller than that of WO3 NR (112.5 mV dec−1) and WS2/WO3 NB (195.5 mV dec−1), which suggests that the HER kinetics of WS2/WO3 NR are faster than those of WO3 NR and WS2/WO3 NB. In Figure 6(c), the EIS results derived from the recorded Nyquist plots reveal that WS2/WO3 NR presents a smaller semicircle diameter than that of WS2/WO3 NB and WO3 NR, indicating the higher conductivity and charge-transfer rate and consequently the faster HER kinetics of WS2/WO3 NR. The equivalent circuit in the inset is composed of constant-phase elements and charge-transfer resistances. The fitted values are listed in Table 1. Notably, the charge-transfer resistance of WS2/WO3 NR (397.7 Ω) is considerably lower than that of WS2/WO3 NB (3597 Ω) and WO3 NR (1816 Ω). Therefore, the electron conduction on the surface of WS2/WO3 NR is superior to that on WS2/WO3 NB and WO3 NR. of WS2/WO3 NR was also calculated by using cyclic voltammetry (CV) tests at different scan rates. The CV test results of WS2/WO3 NR at various scan rates are shown in the inset of Figure 6(d). is calculated to be approximately 1.327 mF cm−2, which is comparable to that of WS2 hollow spheres in a previous study [50]. The stability of WS2/WO3 NR was also investigated by i-t measurement for 12 h which is provided in Figure S3.

4. Conclusions

In summary, WO3 NR, WS2/WO3 NB, and WS2/WO3 NR heterostructures were successfully synthesized using batch reactor routes. The structure, morphology, composition, and characteristics of the synthesized materials were completely confirmed. Subsequently, the HER performances of the prepared materials were thoroughly investigated. The catalytic activities of WS2/WO3 NR have been considerably improved compared with that of WO3 NR and WS2/WO3 NB. The improved performance of WS2/WO3 NR heterostructures could be attributed to the coexistence of WS2 and WO3 materials, which could create a synergistic effect between the two materials and further improve the conductivity and intrinsic HER activity. The Tafel slope of the WS2/WO3 NR (82.7 mV dec−1) is considerably lower than that of pure WO3 NR (112.5 mV dec−1) or WS2/WO3 NB (195.5 mV dec−1). Besides that, the long-time stability of prepared WS2/WO3 NR heterostructures was also confirmed. This study provides a prominent strategy for designing heterostructures of transition-metal sulfides/oxides based on WO3 and WS2 to prepare an efficient catalyst for electrochemical processes or energy-storage applications.

Data Availability

The data used to support the findings of this study are included within the manuscript and the supplementary information files.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Authors’ Contributions

Tuan Van Nguyen and Kim Anh Huynh contributed equally to this work and co-first authors.

Acknowledgments

This research was supported by the NRF funded by the Korean government (2022M3H4A1A01012712 and 2022M3H4A1A04096380).

Supplementary Materials

Figure S1: XRD pattern of WS2 material. Figure S2: (a) High-resolution XPS profiles of O 1 s in WO3 NR, (b) WS2/WO3 NB, and (c) WS2/WO3 NR. Figure S3: stability test of WS2/WO3 nanorod catalyst by i-t measurement for 12 hours. Table S1: different catalysts based on WO3 and/or WS2 for HER. (Supplementary Materials)