npsm 새물리 New Physics : Sae Mulli

pISSN 0374-4914 eISSN 2289-0041
Qrcode

Article

Research Paper

New Phys.: Sae Mulli 2023; 73: 403-412

Published online May 31, 2023 https://doi.org/10.3938/NPSM.73.403

Copyright © New Physics: Sae Mulli.

Synthesis of Tungsten Disulfide for Electrocatalysts

Keshab Pandey1, Hae Kyung Jeong1,2*

1Department of Physics, Daegu University, Gyeongsan 38453, Korea
2Department of Materials-Energy Science and Engineering, Daegu University, Gyeongsan 38453, Korea

Correspondence to:*E-mail: outron@gmail.com

Received: January 3, 2023; Revised: February 1, 2023; Accepted: February 23, 2023

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License(http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Transition-metal dichalcogenides, including tungsten disulfide (WS2), are promising alternatives to noble-metal catalysts in an electrocatalytic hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). However, the low density of active sites and low electrical conductivity in their natural forms severely limit their electrocatalytic performance. Here we synthesize WS2 by a simple hydrothermal method to improve the HER and OER activities. The WS2 synthesized at 200 °C for 24 h exhibits low overpotentials of -200 mV at -10 mA/cm2 and 190 mV at 10 mA/cm2 with low Tafel slopes of 92 mV/dec and 112 mV/dec for the HER and OER, respectively. The enhanced performance of the HER and OER is due to the higher electrochemically active surface area and lower impedance than those of commercial WS2. Hence, this work emphasizes the simple hydrothermal method for developing WS2, which has a high catalytic activity, a long lifetime, and low cost, for a variety of electrocatalytic applications.

Keywords: Tungsten disulfide, Electrocatalysts, Hydrogen evolution reaction, Oxygen evolution reaction

To solve the energy crisis and environmental problems, global efforts are being made to discover renewable energy sources that can replace fossil fuels[1,2]. Hydrogen is considered one of the most eco-friendly renewable energy sources, serving as an alternative to fossil fuels[3]. Water electrolysis is the most efficient method for producing pure hydrogen and oxygen[3,4]. The development of electrocatalysts with considerably high efficiency, low cost, and good durability and stability for electrocatalytic water splitting is critical for the large-scale commercialization of water electrolysis[5]. Water splitting comprises two half-cell reactions[6], which are the hydrogen evolution reaction (HER) and the oxygen evolution reaction (OER). Both processes are extremely slow without a suitable catalyst. In addition to water electrolysis, HER/OER catalysts are directly coupled with several renewable energy conversion and storage systems[7], such as metal–air batteries, fuel cells, supercapacitors, and solar cells. Platinum, iridium, and ruthenium dioxides are frequently used as the best catalysts to accelerate HER and OER because of their low overpotential and high stability and catalytic activity[8,9]. However, scarcity, high cost, low ionic conductivity, and long-term instability are common drawbacks to their practical applications. Therefore, it is highly desirable to investigate robust and efficient new alternative catalysts capable of accelerating sluggish kinetics through fast electron transfer with a highly scalable performance at the lowest possible cost. Catalysts with bifunctional properties may play an important role in the production of hydrogen and oxygen for clean and sustainable energy.

Several strategies, including the use of metal sulfides[10-13], have recently been developed to realize highly efficient catalysts for HER and OER. The transition-metal dichalcogenides from Group VI elements have recently sparked intense research as efficient HER and OER catalysts because of the catalytically active S atoms on their edge sites[14]. They are two-dimensional layered compounds linked by van der Waals forces[15]. With such a layered structure, ions can move freely through the channels, thereby increasing the charge-transfer rate and enhancing electrochemical reactions.

In comparison, the hydrothermal method is deemed to be the most simple and straightforward method for mass-producing tungsten disulfide (WS2) nanostructures. The hydrothermal method is used to successfully synthesize and investigate low-dimensional WS2 nanostructures[16], the hyacinth flower-like architectures of WS2 nanorods[17], and 1T phase WS2 nanosheets[18]. The various morphologies of WS2 are formed by adding surfactants, such as cetyltrimethylammonium bromide (CTAB) and polyethylene glycol[16], as a template by self-assembly to form the various nanostructure materials[19]. The synthesized WS2 can be used in electrocatalysts; however, more research is required to replace noble-metal-based catalysts. In practical applications, the best catalysts have demonstrated low overpotential, excellent catalytic activity, and good stability.

Here, we have successfully synthesized time-dependent WS2 through a simple and easy hydrothermal method for bifunctional electrocatalysts for HER and OER. The effect of time on the morphologies of the WS2 nanostructures is studied and discussed. At 200 °C for 24 h, the synthesized WS2 has a low overpotential of -200 mV at -10 mA/cm2 and 190 mV at 10 mA/cm2, with low Tafel slopes of 92 mV/dec and 112 mV/dec for HER and OER, respectively. More significantly, this demonstration provides a typical illustration of the process of designing transition-metal dichalcogenide materials devoid of noble metals to function as co-catalysts in electrocatalytic HER and OER applications.

1. Sample preparation

Sodium tungstate dihydrate (Na2WO4. 2H2O, ≥99%), WS2 (powder, 2 μm, 99%), thiourea (CH4N2S, ≥99%), CTAB (≥98%), hydroxylamine hydrochloride (NH2OH.HCl, 99%), sulfuric acid (H2SO4, ≥98%), and potassium hydroxide (KOH, ≥85%) were purchased from Merck. Ethanol and hydrochloric acid (HCl, 35% – 37%) were obtained from Samchun Pure Chemical Co., Ltd. (Seoul, Republic of Korea). Aqueous solutions with distilled deionized (DI) water were used.

First, 30 mL of DI water was mixed with 1.6 g of sodium tungstate dihydrate (0.005 mol), 2.2 g of thiourea (0.03 mol), and 0.7 g of hydroxylamine hydrochloride (0.01 mol). Thereafter, the surfactant, CTAB, (0.24 g) was added to the solution during the magnetic stirring process. The pH of the mixture was corrected to 6 (near neutral) by adding 2 mol/L of HCl and continuously stirring for 60 min. The resulting solution was added to a 50 mL Teflon-lined hydrothermal autoclave reactor made of stainless steel, which was tightly sealed and treated at 200 °C for 6, 12, and 24 h. The obtained samples were blackish-gray. Before characterization, the obtained materials were dried in an oven (HAN BAEK SCIENTIFIC CO) at 65 °C for 12 h. The samples after the 6, 12, and 24 h of hydrothermal treatment were marked as Syn WS2-1, Syn WS2-2, and Syn WS2-3, respectively. For comparison, the commercial WS2 was named WS2 and compared with the other synthesized WS2 samples. Figure 1 illustrates a schematic representation of the sample preparation: the process of WS2 formation and growth, which includes self-assembly, Ostwald ripening in an aqueous solution using CTAB as a shape controller, and an oxidation–reduction reaction process described elsewhere[16].

Figure 1. (Color online) Schematic illustration of the hydrothermal synthesis of WS2.

2. Physicochemical characterization

The surface morphologies of the samples were examined by field-emission scanning electron microscopy (Hitachi, Japan). Energy-dispersive X-ray spectroscopy (EDS, JEOL, S-4300) was used to map, configure, and determine the elemental composition of the samples. The lattice parameters, crystallinity, and phase nature were determined using an X-ray diffraction (XRD) spectrometer (D/MAX-2500/PC, Japan) at 40 kV with Cu Kα (λ = 1.54 Å). Raman spectroscopy was used to assess the phase and vibration modes of the materials (NANOBASE 100X XperRAM C, Korea, with a laser excitation of 532 nm).

3. Electrochemical characterization

The electrochemical performance of the samples was investigated at room temperature using Bio-Logic, SP-150 (France). In 1 M KOH electrolyte, the reference electrode was Ag/AgCl (3M KCl saturated) and the counter electrode was platinum wire. The working electrodes were made as follows: 5 mg of each sample was individually deposited in 2 mL of isopropyl alcohol and sonicated to create a homogeneous mixture. Further, 5 μL of the sonicated homogeneous suspension was dispersed on a glassy carbon electrode (outer diameter: 6 mm, inner diameter: 3 mm). In the working electrodes, the loaded active mass was 2.5 g/m2. The samples were electrochemically investigated by cyclic voltammetry (CV), chronocoulometry (CC), potential electrochemical impedance spectroscopy (PEIS), and linear sweep voltammetry (LSV). To investigate the capacitive behavior and determine the reversibility of an electrochemical reaction of the samples, CV measurements were performed within a potential range of 0–0.8 V at scan rates of 25, 50, 100, 150, and 200 mV/s. As previously described, CC was used to calculate the electrochemically active surface area[20,21]. PEIS was used to characterize electron transfer and recombination movements in the frequency range from 100 MHz to 500 kHz and investigate the electrochemical impedance properties. The potentials were normalized to the reference hydrogen electrode following the Nernst equation[22] after the polarization curves (j–V plots) were studied by LSV at a sweep rate of 10 mV/s. The equation is as follows:

ERHE=EAg/AgCl+0.059 pH+0.098 V.

The overpotential (η) was obtained using Eq. (2).

η(V)=ERHE1.23

All measurements were made in relation to the Ag/AgCl reference electrode. Tafel slopes were calculated using polarization curves to calculate the effectiveness and rate of an electrochemical reaction to the overpotential. The relationship between the overpotential (η) and generated current was demonstrated using the Tafel equation (η=blogj/j0, where b is the Tafel slope, j denotes the current density, and j0 is the exchange current density)[20]. The CV, CC, PEIS, and LSV measurements were evaluated using the three-electrode system.

Figure 2 shows the SEM images of WS2, Syn WS2-1, Syn WS2-2, and Syn WS2-3. WS2 has a two-dimensional layered, flat, and tightly stacked structure, as seen in Fig. 2(a–c). However, Syn WS2-1, treated at 200 °C for 6 h, as shown in Fig. 2(d–f), has a cone-shaped elongated structure with visible nodes and a smooth surface. Additionally, these structures provide additional gaps to mitigate radial and vertical strains, allowing the ion host materials to remain intact while cycling. When the growth time was extended to 12 h, the smooth surface was changed to a spikey surface, which was composed of nanosheets, and the morphology of Syn-WS2-2 is shown in Fig. 2(g–i). Furthermore, Syn-WS2-3, with a growth time of 24 h, demonstrates that the nanosheets have an open pore structure, as shown in Fig. 2(j–l). The porous structure could provide a relatively large surface area and rapid transportation of electrons, resulting in the improvement of electrochemical performance. Figure 3 shows the EDS mapping and elemental composition of WS2 and Syn WS2-3, confirming that WS2 consists of tungsten and sulfur. Table 1 shows the elemental distribution of the samples from the EDS analysis. The atomic composition of WS2 is as follows: W, 9.0%; S, 18.0%; C, 61.0%; and O, 12.0%, and the atomic composition for Syn WS2-3 is as follows: W, 10.0%; S, 19.5%; C, 58.5%; and O, 12.0%, confirming that the synthesized WS2 has the atomic ratio of 1:2 for W and S within the specified purity and EDS resolution.

Table 1 Comparison of the elemental distribution of the samples from EDS.

SampleElementwt%at%
WS2W49.09.0
S19.018.0
C25.061.0
O7.012.0
Syn WS2-3W52.010.0
S20.019.5
C22.058.5
O6.012.0


Figure 2. (Color online) SEM results of the (a–c) commercial WS2, (d–f) Syn WS2-1, (g–i) Syn WS2-2, and (j–l) Syn WS2-3.

Figure 3. (Color online) EDS mapping results of the (a) commercial WS2 (WS2) and (b) synthesized WS2 (Syn WS2-3).

The Raman results of the samples are exhibited in Fig. 4(a), confirming that the E2g1 mode, located at approximately 352.2 cm-1, correlates with the S atoms oscillating in the opposite direction from the W atoms in the same plane. However, the A1g mode at 421.2 cm-1 refers to the out-of-plane directions that occur from the S atoms[23]. The peaks of the synthesized WS2 are less intense and have a higher width than commercial WS2 because of the decrease in the van der Waals interaction between the interlayers. These findings show that the crystal structures and phases of the synthesized WS2 and commercial WS2 are comparable. The XRD results of the commercial WS2 and synthesized WS2 are shown in Fig. 4(b), showing that the crystal structures of Syn-WS2-1, Syn-WS2-2, and Syn-WS2-3 are similar to that of WS2. The hexagonal phase of WS2 is well represented by the most intense (002) peak, and other rather weak peaks of (004), (101), (103), and (110) planes are well presented. They emerged at 14.4°, 28.9°, 33.6°, 39.5°, and 58.3°, respectively, following the standard values (JCPDS no. 84-1398) of 2H-WS2[24]. Additionally, the interlayer d-spacing of Syn-WS2-3 was calculated from peaks (002), (004), (101), (103), and (110) and was 6.1, 3.0, 2.6, 2.2, and 1.5 Å, respectively[25]. The results from the Raman spectroscopy and XRD are consistent.

Figure 4. (Color online) (a) Raman spectroscopy and (b) XRD results of the samples.

Figure 5 shows the CV results of the samples. The scan rate-dependent CV results of WS2, Syn WS2-1, Syn WS2-2, and Syn WS2-3 exhibit quasi-rectangular-shaped CV curves at scan rates from 25 to 200 mV/s. The Syn WS2-3 sample shown in Fig. 5(d) displayed a more rectangular shape with a large integrated area, demonstrating a high-rate capability. Resultantly, Syn WS2-3 is expected to have higher catalytic activity than the others.

Figure 5. (Color online) CV results of (a) WS2, (b) Syn WS2-1, (c) Syn WS2-2, and (d) Syn WS2-3
at the various scan rates.

Furthermore, at a scan rate of 100 mV/s, the Syn WS2-3 sample, as shown in Fig. 6(a), exhibits the highest current response compared to the other samples, indicating that Syn WS2-3 might have a high specific surface area, high electric conductivity, and high capacitance. The specific capacitance , which is proportional to the surface area, could be obtained from the CV results[21,26]. At a scan rate of 100 mV/s, Syn WS2-3 had the highest capacitance of 153 F/g, approximately 3.7 times that of WS2, 41 F/g. The capacitances of Syn WS2-1 and Syn WS2-2 were 72 and 106 F/g, respectively, at the scan rate of 100 mV/s. As a function of the scan rate, the resulting specific capacitance is shown in Fig. 6(b). The high capacitance of the Syn WS2-3 sample indicates that it has a large surface area and a low charge-transfer resistance.

Figure 6. (Color online) (a) CV results of the samples at 100 mV/s with (b) the corresponding specific capacitance, (c) impedance results, and (d) CC results of the samples.

Internal resistance and charge-transfer resistance were calculated using Z-fitting software based on the equivalent Randle circuit, as previously described[27,28]. Figure 6(c) displays the PEIS results, which are Nyquist plots (-Im |Z| as a function of Re|Z|), of the samples. An internal resistance (Rs), a charge-transfer resistance (RCT), a Warburg component (W), and a double-layer capacitance (CDL) are shown in the inset of Fig. 6(c). The RCT is indicated by the diameter of the semicircle. Internal resistance and charge-transfer resistance were calculated to be 22 and 50 Ω for WS2, 13 and 28 Ω for Syn WS2-1, 8 and 17 Ω for Syn WS2-2, and 4 and 11 Ω for Syn WS2-3, respectively. It was observed that Syn WS2-3 had the lowest internal and charge-transfer resistances, implying that it will provide better catalytic performance than commercial WS2.

The electrochemically active surface area was calculated from the CC measurement, as illustrated in Fig. 6(d), where the slope is proportional to the active surface area[21]. The surface areas of WS2, Syn WS2-1, Syn WS2-2, and Syn WS2-3 were 2.7 ×105,3.5×105,5.2×105, and 5.9 × 10-5 cm2, respectively. The Syn WS2-3 sample had the largest surface area, and it was 2.2 times larger than that of WS2. The Syn WS2-3 sample was expected to have better catalytic performance than the others because of its large surface area and low resistance, which are key characteristics of catalysts.

With a three-electrode setup in 1M H2SO4 within the scan rate of 10 mV/s, the electrocatalytic behavior of the HER of the samples was further investigated. The representative LSV polarization curves are shown in Fig. 7(a), and the Syn WS2-3 sample (red) had the lowest overpotential of -200 mV at -10 mA/cm2, followed by WS2 (-530 mV), Syn WS2-1 (-420 mV), and Syn WS2-2 (-302 mV). Syn WS2-3 exhibited higher electrocatalytic activity toward HER than the WS2 sample because of its high electrical conductivity and active site for hydrogen absorption, suggesting that the porous structure of WS2, given by the synthesis, might be advantageous for the HER. Further, the commercial Pt/C reference was compared, and the overpotential of Pt/C was found to be -0.08 V at -10 mA/cm2 under the same conditions as the samples.

Figure 7. (Color online) (a) LSV results for the HER with (b) the corresponding Tafel plot; (c) LSV results for OER with (d) the corresponding Tafel plot.

The Tafel slope measures the rate of the electrochemical reaction to overpotential, and it is useful for estimating the efficiency of the catalytic reaction. The Tafel slopes obtained for WS2, Syn WS2-1, Syn WS2-2, and Syn WS2-3 are plotted in Fig. 7(b) and are 254 mV/dec, 206 mV/dec, 136 mV/dec, and 92 mV/dec, respectively. In comparison to the other synthesized samples, the Syn WS2-3 sample had the lowest slope, which indicated the highest electron-transfer rate and rapid hydrogen generation. For comparison, the corresponding Tafel slope of Pt/C was investigated under identical conditions and found to be 42 mV/dec.

To analyze the catalytic activity of different samples toward OER, LSV was employed in a 1M KOH electrolyte at a scan rate of 10 mV/s, as shown in Fig. 7(c). The Syn WS2-3 sample exhibited significantly lower onset potential and higher current density than the other synthesized samples. When compared to the other synthesized materials in this work, Syn WS2-3 demonstrated superior OER activity, with an onset potential of 1.42 V and a low overpotential of 190 mV at 10 mA/cm2. Resultantly, Syn WS2-1 and Syn WS2-2 had onset potentials of 1.56 V and 1.47 V, respectively, which were lower than the onset potential of WS2 (1.61 V) but higher than the onset potential of Syn WS2-3. By presenting an effective active site and a synergistic effect, the catalyst can be enhanced by providing an equal concentration. The corresponding Tafel plots are shown in Fig. 7(d). Tafel slopes of 234, 177, 144, and 112 mV/dec were obtained for the WS2, Syn WS2-1, Syn WS2-2, and Syn WS2-3 samples, respectively. The Tafel slope of the Syn WS2-3 sample was approximately 2.1 times lower than that of the WS2 sample, indicating that Syn WS2-3 had the highest OER activity. It was determined that WS2 at 200 °C for 24 h provided the best HER and OER catalytic performance. Under identical conditions, the OER activities of the commercial RuO2 reference were investigated for comparison, and the overpotential of the commercial RuO2 reference was observed to be 1.31 V at 10 mA/cm2 with a Tafel slope of 81 mV/dec.

In addition to the requirement for high activity, electrocatalysts must be durable and stable. Continuous LSV measurements for HER and OER were performed at a scanning rate of 10 mV/s to test the long-term durability and stability of Syn WS2-3. The consistent durability of HER was observed, as shown in Fig. 8(a), where the polarization curves remained nearly identical after 1,000 and 3,000 cycles. Likewise, Fig. 8(b) depicts the dependable durability of OER after 1,000 and 3,000 continuous cycles at a scanning rate of 10 mV/s.

Figure 8. (Color online) Durability test for the (a) HER and (b) OER of the Syn WS2-3 sample.

In this study, hydrothermal synthesis was used to develop WS2 as a cost-effective, highly efficient, and long-lasting novel electrocatalyst for HER and OER. The optimal Syn WS2-3 catalyst demonstrated exceptional HER and OER activities, with extremely low overpotentials of -200 mV at -10 mA/cm2 and 190 mV at 10 mA/cm2, as well as low Tafel slopes of 92 mV/dec and 112 mV/dec, respectively. The WS2 synthesized for 24 h at 200 °C revealed an open pore structure, which could increase the active surface area and enhance the electrocatalytic activity toward HER and OER performances. The high catalytic behavior was attributed to the large surface area, high specific capacitance, and low resistance, which are all important factors for enhanced HER and OER performances. Further, the Syn WS2-3 sample showed good stability and reliable durability after 1,000 and 3,000 cycles. These studies make it easier to manufacture commercial water-splitting catalysts on a large scale and may encourage the development of electrocatalysts for energy-harvesting applications.

  1. L. Wu et al, Phys. Chem. Chem. Phys. 21, 6071 (2019).
    Pubmed CrossRef
  2. J. Yang et al, J. Power Sources 545, 231948 (2022).
    CrossRef
  3. B. H. R. Suryanto et al, Nat. Commun. 10, 5599 (2019).
    Pubmed KoreaMed CrossRef
  4. S. Wang, A. Lu and C. J. Zhong, Nano Converg. 8, 4 (2021).
    Pubmed KoreaMed CrossRef
  5. B. You and Y. Sun, Acc. Chem. Res. 51, 1571 (2018).
    Pubmed CrossRef
  6. Z. P. Ifkovits et al, Energy Environ. Sci. 14, 4740 (2021).
    CrossRef
  7. S. T. Rahman, K. Y. Rhee and S. J. Park, Nanotechnol. Rev. 10, 137 (2021).
    CrossRef
  8. N. Mamaca et al, Appl. Catal. B 111-112, 376 (2012).
    CrossRef
  9. K. Pandey and H. K. Jeong, Chem. Phys. Impact 6, 100175 (2023).
    CrossRef
  10. A. P. Tiwari, D. Kim, Y. Kim and H. Lee, Adv. Energy Mater. 7, 1602217 (2017).
    CrossRef
  11. M. Wang, L. Zhang, Y. He and H. Zhu, J. Mater. Chem. A 9, 5320 (2021).
    CrossRef
  12. Z. Thiehmed, A. Shakoor and T. Altahtamouni, Catalysis 11, 1283 (2021).
    CrossRef
  13. G. Ghanashyam and H. K. Jeong, Chem. Phys. Lett. 794, 139492 (2022).
    CrossRef
  14. D. Rojas et al, Electrochem. Commun. 115, 106718 (2020).
    CrossRef
  15. R. Bian et al, Natl. Sci. Rev. 9, 164 (2022).
  16. S. Cao et al, Mater. Lett. 129, 205 (2014).
    CrossRef
  17. Y. Wu et al, Mater. Lett. 189, 282 (2017).
    CrossRef
  18. M. Piao et al, Nanotechnology 29, 025705 (2018).
    Pubmed CrossRef
  19. S. Y. Moon, T. Kusunose and T. Sekino, Mater. Lett. 63, 2038 (2009).
    CrossRef
  20. G. Ghanashyam and H. K. Jeong, J. Electrochem. Sci. Technol. 13, 120 (2022).
    CrossRef
  21. G. Ghanashyam and H. K. Jeong, J. Energy Storage 26, 100923 (2019).
    CrossRef
  22. T. Niyitanga and H. K. Kim, Chem. Eng. J. 451, 138649 (2023).
    CrossRef
  23. F. Wang et al, 2D Mater. 4, 015007 (2016).
    CrossRef
  24. Y. Liu et al, Front. Chem. 6, 236 (2018).
    CrossRef
  25. G. Ghanashyam and H. K. Jeong, New Phys.: Sae Mulli 68, 181 (2018).
    CrossRef
  26. K. Pandey and H. K. Jeong, Chem. Phys. Lett. 809, 140173 (2022).
    CrossRef
  27. G. P. Ojha et al, Energy 188, 116066 (2019).
    CrossRef
  28. K. Pandey and H. K. Jeong, New Phys.: Sae Mulli 72, 274 (2022).
    CrossRef

Stats or Metrics

Share this article on :

Related articles in NPSM