Properties, preparation, and application of tungsten disulfide: a review

The structure of WS₂ comprises stacked triple layers formed by a transition metal layer (W atom) sandwiched between two S atom layers, each of which has a hexagonal lattice structure. In the three-layer stack, W atoms and S atoms are bonded by strong ionic-covalent bonds. The structure formed by these three layers is held together by weak van der Waals interactions, which allow mechanical peeling of WS₂ layers. In the bulk phase, polymorphisms are unique features of TMDs. The atomic structure of WS₂ depends on the filling of the d-orbitals in the transition metal [7]. There are three possible stacking structures of WS₂ (figure 1), namely, 1T, 2H and 3R. The 1T structure is formed by tetragonal symmetry and octahedral coordination. The d-orbitals of transition metals are divided into two energy levels: dxy, yz, zx and d_x ² _−y ² _,z ². The 2H structure is formed by hexagonal close packing and coordination of triangular prisms, and 3R is a rhombic symmetrical structure [8]. In the 2H and 3R phases, the d-orbitals of transition metals are divided into three energy levels: d_z ², d_x ² _−y ² _{, xy} and d_{xy, yz} . For a single-layer WS₂ film, the 2H structure is the most stable and the most common structure. In addition to the more common 1T, 2H and 3R stacked structures mentioned in this article, WS₂ also has a 1T' structure, which has also attracted widespread attention owing to its excellent topological insulation characteristics [9].

Zoom In Zoom Out Reset image size

To change the atomic structure of WS₂, the lithium-ion intercalation method and the nonintercalation phase transformation method are often used. Lithium-ion intercalation is a reversible process. When an excess charge is added, the 2H phase transition becomes unstable and changes to the 1T phase; however, when the lithium ion is removed from WS₂, the excess charge is removed. Subsequently, the WS₂ in the 1T phase is recovered to the most stable 2H phase [10]. The intercalation-free phase transformation method includes stimulation through infrared laser [11] and electron beam irradiation [12]. The reactions are irreversible. These reactions produce irreversible chalcogenide vacancies, which are the driving forces for the transition from 2H to 1T. However, this method is destructive and cannot reverse the phase transition from 2H to 1T. These phase transitions can change the properties of WS₂ to between those of a semiconductor and a metal [13].

Owing to the influence of quantum confinement and interlayer interactions, indirect-to-direct bandgap transitions will occur when the thickness of WS₂ is reduced to a single layer [14]. In general, with the change in the WS₂ thickness, the direct bandgap at the K point is constant. This is because the corresponding valence and conduction band states at K are only related to the transition metal state and are not affected by the interlayer interactions. However, when the thickness of WS₂ changes from two layers to one layer, the indirect bandgap from Γ to K is larger than the direct bandgap. This occurs because the valence band state at Γ is related to the $d_z^2$ orbital of the metal and the P_Z orbit of the sulfur group element. For single-layer WS₂, there is no Coulomb repulsion between the P_Z orbitals of chalcogenide elements in the adjacent layers, resulting in the stability of the valence band state of Γ, which causes indirect-to-direct gap transitions [15].

WS₂ can grow in horizontal and vertical directions. In the horizontal growth process, the WS₂ layer grows along the plane direction and the film is parallel to the growth substrate [16]. In the vertical growth process, the WS₂ layer grows vertically on the growth substrate, exposing the edge position of the film [17, 18]. These two different morphologies lead to different material properties and have different applications. The horizontal growth of WS₂ is mainly used in optoelectronic technology owing to its indirect-to-direct bandgap transition and large on/off ratio as a transistor. Because of the chemically reactive edge position of WS₂ grown vertically, several electrochemical studies have been conducted [19]. The main reason for the two types of morphologies is the deposition thickness of the precursor. With the decrease in the precursor layer thickness, the WS₂ thin films transition from vertical growth to horizontal growth [20].

Physical methods include pulsed magnetron sputtering and pulsed laser deposition (PLD). The magnetron sputtering growth method requires the use of charged particles to bombard the surface of the target under vacuum conditions to cause the target to sputter. In this approach, the neutral target atoms or molecules are deposited on a substrate to form a thin film. Usually, an inert gas is glow discharged in a low-pressure environment to generate incident ions. For WS₂ deposition, a cathode target made of WS₂ is sputtered under a high negative cathode voltage of 1–3 kV DC or a 13.56 MHz radiofrequency voltage. In addition, 0.1–10 Pa argon gas is passed into the vacuum chamber to generate glow discharge. Under the action of an electric field, Ar ions are accelerated toward the WS₂ target and bombard the target surface with high energy, which removes small particles of the target material that deposit on the substrate to form a WS₂ film [21]. To improve the adhesion of WS₂ coatings, Rodrigues et al [22] deposited a Cr intermediate layer with a thickness of approximately 400 nm on a Si substrate and then deposited a WS₂–CF coating (as shown in figure 2(a)). The important advantages of this method are the simple equipment, the low operating temperature of the substrate, large coating area and the ability to produce dense and well-adhered films [21].

Zoom In Zoom Out Reset image size

In contrast to magnetron sputtering, the equipment used for PLD is complex and its operating principle is relatively complicated. When a high-energy pulsed laser impacts the solid WS₂ target, a laser pulse with a sufficiently high energy flux and short pulse width will be generated and the target is exposed to this pulse. The incident laser energy is absorbed by the target, resulting in a rapid temperature increase that causes some of the target to evaporate and form a localized high-temperature and high-density plasma. In accordance with the law of gas dynamics, the emitted plasma plume preferentially moves toward the substrate, thus forming a WS₂ film [24]. In 2018, Tian et al [23] used PLD technology to grow controllable layers of WS₂ thin films on sapphire substrates. They found that the thickness of the 2D WS₂ films could be accurately controlled by controlling the number of incident pulses (monolayer and multilayer films were successfully prepared). Figure 2(b) shows a schematic of a deposition chamber for the growth of WS₂ films using PLD. The most significant advantage of this method is that the thickness of the film can be controlled. The film is uniform and of good quality, but it has the disadvantages of high cost and complex operation.

Two common methods for preparing WS₂ films by chemical means are chemical vapor deposition (CVD) and growing single-crystal WS₂ from an aqueous solution by the hydrothermal method under high-temperature and high-pressure conditions. CVD is the most common method used to prepare WS₂.

The CVD method involves a reaction process in which a gaseous precursor undergoes a chemical reaction on a solid surface to generate a solid deposit. The preparation of WS₂ by CVD can be achieved using either a one- or a two-step method. The one-step method involves the direct heating of sulfur and tungsten sources in a CVD furnace, followed by control of the carrier gas flow and then deposition of the WS₂ film on the target substrate (as shown in figure 3(a)) [25]. A common tungsten source used is WO₃, where the reaction 2WO₃ + 7 S → WS₂ + 3 SO₂ occurs [26]. The most commonly used substrates are silicon, copper and quartz glass. The two-step method involves the deposition of a tungsten or tungsten compound film on the target substrate, which is then vulcanized in a CVD furnace to synthesize the WS₂ film [27]. Usually, the deposition is carried out on the substrate by electron-beam evaporation or magnetron sputtering. For example, Carlo et al [28] prepared W films by magnetron sputtering and then sulfurized them. Their experimental results showed that the number of layers can be highly controlled by controlling the initial deposition thickness of W films, and WS₂ thin films can be prepared over a large area using this method. CVD can accurately control the thickness and size of WS₂ films by controlling the changes in temperature and carrier gas flow. High-quality WS₂ films can be synthesized using CVD, with good optical and electrical properties. Moreover, the two-step method can also enable the preparation of large-scale films, thus overcoming the drawback of the one-step method. However, this method still has disadvantages such as the need for a high substrate temperature and relatively expensive equipment.

Zoom In Zoom Out Reset image size

The hydrothermal method is a typical wet chemical synthesis method. For preparing WS₂ particles, the following representative process can be used [29]. First, 10 mmol of Na₂WO₄ 2 H₂O and 10 mmol of Na₂S are dissolved in 50 ml of deionized water and continuously stirred for 30 min to form a clear solution. The solution is then transferred to a hydrothermal reactor at 220 °C for 5 h. The recovered solution is then centrifuged and rinsed with ethanol. Finally, WS₂ particles are collected after vacuum drying at 75 °C (figure 3(b)). This method has been widely used in the preparation of 2D nanomaterials, especially WS₂ particles. The advantages of the hydrothermal method are high purity, the crystallinity of the formed particles and facile control of the operating parameters [30]. However, the thickness of the films cannot be controlled when used in devices, and the reaction process is not visible.

Bulk WS₂ can be stripped by physical and chemical methods, which are classified into mechanical exfoliation, liquid exfoliation and lithium-ion intercalation methods. In recent years, to obtain large-area and high-quality single-layer WS₂ films, researchers have tried to grow single-layer WS₂ films on single-crystal substrates and then peel them off through atomic or molecular intercalation.

The mechanical stripping method is a simple physical method used to produce WS₂ thin films. First, transparent adhesive tape is applied to the WS₂ bulk laminated sheet to be stripped. The bulk material is then repeatedly peeled using the tape to thin the laminated sheet. Finally, the thin WS₂ film is transferred from the tape to the target substrate (e.g. SiO₂/Si), and the tape is slowly peeled off after a rest period to ensure that the WS₂ film has adhered well to the target substrate (as shown in figure 4(a)) [31]. The prepared single-layer TMD materials have few defects and are highly crystalline, resulting in good photoelectric properties, such as high luminous efficiency and high carrier mobility. However, it is difficult to peel off a large-area film [32], which limits their practical applications, and the repetition rate is low.

Zoom In Zoom Out Reset image size

The liquid-phase stripping method is based on the use of ultrasonic waves to generate cavitation bubbles or a shear force to separate the layered materials from each other. Then, the material is dispersed in a solvent to obtain 2D single-layer and few-layer materials [35]. The choice of solvent is the most critical factor in the liquid-phase stripping method. The surface tension of WS₂ is estimated to be 40 mJ m⁻² [36], so a suitable solvent must be selected to successfully peel the WS₂ material. The most commonly used solvents are propanol, ethanol and water. To shorten the ultrasonic treatment time and improve the stability of the suspension, Adilbekova et al [33] used ammonia as the solvent for liquid-phase stripping. Figure 4(b) shows a schematic describing the preparation of WS₂ films by the liquid-phase stripping method. This method can be used to prepare WS₂ in large quantities, but the thickness of the WS₂ film produced using this method is not uniform and the purity of the film is low.

The lithium-ion intercalation method uses lithium-containing solvents (such as n-butyl lithium) that intercalate into the WS₂ layered nanomaterials to form numerous intermediate compounds. These intermediate compounds increase the interlayer distance of WS₂ and weaken the van der Waals forces between the layers, resulting in exfoliation. Then, the material is washed and dried to obtain single-layer or multilayer WS₂ nanosheets. Ghorai et al [34] intercalated WS₂ at room temperature in the presence of an inert solvent, n-hexane or toluene, using lithium halide as the lithium source. Figure 4(c) shows a schematic of the peeling of WS₂ nanosheets. The intercalation of lithium ions into 2D layered materials results in interlayer separation, which is an electrochemical stripping method. The greatest advantages of this method are the high yield of WS₂ thin films and the large film area. However, the formation of defects, high cost, long process time and high sensitivity to environmental changes are some of the disadvantages of this process.

In recent years, the epitaxial growth of WS₂ thin films on single-crystal substrates [37] has attracted considerable attention. An important advantage of this method is that the strong interaction between the monolayer and the substrate can enhance the orientation of the layer [38]. However, the same interaction can also lead to the hybridization effect, resulting in the deterioration of the intrinsic properties of the monolayers [39]. Many researchers have tried to reduce this hybridization effect by inserting atoms or molecules between layers to peel off single WS₂ films. For example, Mahatha et al [40] inserted Bi ions between layers to effectively reduce the interaction between the WS₂ layer and the Ag substrate. This method not only achieves effective separation of single-layer WS₂ from the substrate but also produces high-quality single-layer WS₂ films.

Layered WS₂ can be used as an electrocatalyst and a photocatalyst owing to its unique bandgap characteristics, inherent vacancy defects and low conductivity [41, 42]. Photocatalysis and electrocatalysis are essential in our daily life, and they have been widely used in environmental protection and the generation of clean energy [43, 44].

4.1.1. Photocatalysis.

The intrinsic properties of materials, such as the morphology, crystal structure and specific surface area determine the photocatalytic efficiency of the materials. On the nanoscale, higher specific surface areas provide a greater number of active centers that promote the charge transfer of photo-produced electron–hole pairs [45]. The bandgap and semiconducting properties of WS₂ are excellent for photocatalysis [46]. In particular, stripped WS₂ nanosheets have a bandgap of 1.9 eV and a wide absorption spectrum in visible light, exhibiting good performance in several photocatalytic reactions, including degradation of dyes and rhodamine B, production of H₂, and reduction of nitrophenol.

However, because of the rapid recombination of excitons, the photocatalytic activity of pure WS₂ materials is very weak and this hinders surface charge transfer [47]. Therefore, researchers have tried to improve the catalytic activity of WS₂ by compounding WS₂ with other substances. Xu et al [48] used an improved liquid exfoliation method to prepare one– and two-layer semiconductor WS₂ nanosheets. They mixed WS₂ and CdS in different mass ratios to form WS₂/CdS hybrids. Subsequently, in the photocatalytic hydrogen production process using lactic acid as the sacrificial electron donor, compared to pure CdS, the H₂ release rate of the WS₂/CdS hybrid system was 26 times higher, and the optimal photocatalytic quantum yield was greater than 60%. Chen et al [49] prepared a three-dimensional composite of graphene oxide/WS₂ nanosheets/Mg-doped ZnO nanohybrid (rGOWMZ). They optimized the interface charge transfer of the rGOWMZ and the effective synergistic effect of the components of this three-dimensional composite resulted in enhanced photocatalytic activity. Figure 5 shows the photocatalytic decomposition results of rhodamine B using different photocatalysts (blank, graphene oxide/Mg-doped ZnO composite (rGOMZ) and rGOWMZ) illuminated using ultraviolet light or a xenon arc lamp.

Zoom In Zoom Out Reset image size

4.1.2. Electrocatalysis.

Recently, researchers have discovered that owing to the quantum confinement effect, WS₂ can be converted into a direct semiconductor with a bandgap of 2.1 eV by controlling the chemical composition and the number of layers [57]. In addition, the saturation absorption characteristics of WS₂ are better than those of graphene and carbon nanotubes in the near-infrared and mid-infrared bands [58]. Owing to these excellent properties of WS₂, it is increasingly being applied to fiber lasers and solid-state lasers as an SA.

4.2.1. SAs for fiber lasers.

4.2.2. SAs for solid-state lasers.

Currently, there is strong interest in developing new energy production and storage systems. As a 2D layered material, WS₂ has been widely used in lithium-ion batteries, sodium-ion batteries and solar cells owing to its excellent properties.

4.3.1. Lithium-ion and sodium-ion batteries.

Lithium-ion batteries have long been important energy storage devices, but Li resources are limited and, hence, expensive. Sodium-ion batteries have received widespread attention as viable energy storage devices in recent years owing to the abundant availability of sodium resources [71]. Large-capacity, high-rate anode materials are the key to producing viable lithium- and sodium-ion batteries [72]. Currently, owing to their large interlayer spacing and weak van der Waals forces, 2D crystal materials, especially TMDs, have excellent theoretical electrochemical intercalation performance (conducive for Li⁺ and Na⁺ intercalation and delamination). They can be used as anode materials for sodium- and lithium-ion batteries [73]. In 2017, Lei et al [74] used polar WS₂ nanosheets as C@WS₂ independent electrodes in Li–S batteries. As shown in figure 6, the Li–S battery prepared using the C@WS₂/S composite electrode retained ∼90% of the specific capacity after 1500 cycles at a rate of 2 °C and exhibited excellent cycle stability. Wang et al [75] prepared hollow spherical WS₂ structures composed of porous nanosheets with a layer spacing of 1.02 nm by a solvothermal method. The prepared WS₂ was used as the anode of a sodium-ion battery, giving an output of 353.2 mA h g⁻¹ after 80 cycles under a specific current of 0.2 A g⁻¹. The porous nanosheets substantially reduced the diffusion length of the sodium ions, whereas the hollow spherical structure provided a large surface area for the volume expansion occurring during sodium-ion insertion/extraction.

Zoom In Zoom Out Reset image size

The use of WS₂ as an anode material still has the following disadvantages: (a) large volume expansion and rapid capacity decay; and (b) low ion diffusion coefficient and conductivity response of the pure metal sulfide [76, 77]. One optimization method is the use of nanostructured WS₂ materials that have a high specific surface area and short ion diffusion length [78]. Liu et al [79] synthesized WS₂ nanowires with increased interlayer spacing by a simple solvothermal method. When the specific current was 100 mA g⁻¹, the initial capacity of the WS₂ nanowires was 605.3 mA h g⁻¹. After 50 cycles, only 20% of the capacity was lost. Compared to bulk WS₂, the layered structure and specific surface area of the WS₂ nanowires considerably reduce the decay in the battery capacity. Another optimization method is to introduce WS₂ into a carbon-containing matrix (e.g. carbon nanotubes, graphene or pyrolytic carbon), which can buffer large volume changes and improve the conductivity of the electrode [80]. Li et al [72] prepared WS₂/C composites for application in sodium- and lithium-ion batteries. Carbon provides a good conductive substrate and buffers large changes in the volume of the WS₂ material during the sodium-/lithium-ion intercalation/delamination process. Compared to the bare WS₂, the synthesized WS₂/C composite had a higher reversible specific capacity and improved cycle performance.

4.3.2. Solar cells.

Photodetectors are sensitive to changes in their external environment and so they can be applied to flame detection, measurement control and video imaging. Presently, semiconductor materials (such as Si, GaN and InGaAs) are widely used in photodetectors for sensing spectral changes in the ultraviolet to infrared regions. However, traditional semiconductor thin-film materials, silicon nanomaterials, graphene and other functional materials have several inherent disadvantages. For example, the growth and processing of traditional semiconductor thin-film materials require high temperatures, which are unsuitable for the flexible polymeric substrates used for flexible photodetectors. In addition, the zero bandgap and low light absorption of graphene make it unsuitable for use in photodetectors. WS₂ has a suitable bandgap and good thermal stability; hence, it is widely used in photodetectors.

Perea-López et al [88] used the CVD method for growing layered WS₂ on a silicon substrate as a light sensor. The responsiveness of the device was only 97 μA W⁻¹. Lan et al [89] synthesized WS₂ monolayer films using an enhanced CVD method. However, the preparation and transmission processes of this enhanced method are very complicated. In addition, the area of the WS₂ films produced is relatively small, which hinders their application in large-area photodetectors. To overcome these challenges, Li et al [90] used WS₂ nanosheets to prepare novel large-area flexible photodetectors on filter membranes using a simple process. The photodetectors can respond to broadband wavelengths from 532 nm to 1064 nm. The bending resistance of the flexible photodetector was significant, as shown in figure 7, which shows the light response in a bending state. After 200 bending cycles, the photocurrent was 80% of the initial value.

Zoom In Zoom Out Reset image size

Micro- and nano-WS₂ are excellent lubricating anti-friction materials, which can be used to solve the dynamic friction problems in the mechanical manufacturing of WS₂. As the WS₂ layers are connected by weak van der Waals forces, interlayer sliding easily occurs and the friction coefficient is low [91]. In 2018, Wu et al [91] compared the lubrication performance of a base oil, an ionic liquid (C₇H₁₁F₃N₂O₃S) and WS₂. Figure 8 shows the changes in the friction coefficients over time at a speed of 1450 rpm and under a constant load of 392 N. With increasing friction times, all friction coefficients increased, although those of the samples containing ionic liquids tended to decrease after peaking. In addition, the samples with WS₂ and ionic liquids maintained their initial friction coefficient for a longer time compared to the base oil. Moreover, the addition of WS₂ can significantly reduce the friction coefficient. This is because WS₂ can be used as a solid or liquid lubricant. The micro-/nano-WS₂ materials can be used in harsh working conditions, such as those involving high pressures, temperatures and loads, and in corrosive and radiative environments. Therefore, these materials are widely used in high-tech fields such as aviation and aerospace [92].

Zoom In Zoom Out Reset image size

In the automotive industry, efficient lubricants can increase the service life of various mechanical components. Aldana et al [93] studied the role of nano-WS₂ as a lubricant additive for automotive gearboxes. After chemical analyses and transmission tests, they only observed W and S friction films on the gear surface; however, WS₂ nanoparticles can enter the cracks and act as crack sealants, consequently reducing the propagation of cracks and preventing premature failure of mechanical parts. Huang et al [94] prepared self-lubricating WC–Ni–Cr-based composites containing WS₂ using pulse current sintering at different temperatures and pressures. They observed that when the pressure exceeds 250 MPa and the sintering temperature is as low as 950 °C, the sintered composite material can be densified while preventing the decomposition of the disulfide. The tribological performance of the composites under sliding wear in air showed that the friction coefficient of high-pressure-sintered WS₂ composites was lower than that of WC–Ni–Cr cermets without WS₂. They inferred that wear debris accumulated in the low-lying areas of the wear surface and formed a discontinuity owing to self-lubrication by the WS₂ layer, thereby reducing the friction coefficient.

With the increasingly serious environmental problems, researchers are focusing on monitoring indoor and outdoor air quality. This has led to a considerable increase in research on gas sensors. However, fabricating gas sensors that are continuous and reliable, convenient and cheap is difficult. Previously, metal oxide semiconductors were typically used as sensing materials. Although their effect is good, most of the sensing needed to be carried out in high-temperature environments with a large amount of power, because they consume a significant amount of power. People are seeking new materials that can operate at low temperatures. WS₂, an inorganic layered material whose structure is held together by van der Waals forces, is an appropriate material because of the following two reasons: (a) WS₂ has a high specific surface area, and because the van der Waals forces are very weak, it has a large crystal spacing, which makes this structure suitable for a variety of interactions with the environment; (b) WS₂ has a low bandgap and good conductivity; consequently, its working temperature is low [95].

In 2014, O'Brien et al [96] fabricated a WS₂ gas sensor by sulfurizing a WO₃ film at 500 °C using H₂S plasma. Their study was the first to prove the gas-sensitive response of WS₂ films and it is also an important attempt at low temperature detection. Subsequently, Järvinen et al [95] deposited a 20 nm thick metallic W film on the chip by sputtering and then sulfurized it in a quartz tube. At room temperature, the gas sensor was used to monitor H₂, H₂S, CO and NH₃ in the air. WS₂ had excellent selectivity to NH₃ and high sensitivity (0.10 ± 0.02 ppm⁻¹) Moreover. the sensor can respond to a stimulus quickly and recover quickly after the stimulus is removed.

Because of its high switching current ratio, high thermal stability, electrostatic integrity [97] and the absence of dangling bonds, WS₂ has been widely used in field-effect transistors. However, the electron mobility of single-layer WS₂ is low, and researchers have proposed several methods to overcome this drawback. Iqbal et al [98] tried to use CVD to sandwich single-layer WS₂ between hexagonal BN films and fabricated metal electrodes with Al and Au to achieve an ohmic contact. This scheme shows a very high mobility of 214 cm₂ V s⁻¹ at room temperature and it also avoids the hysteresis phenomenon of single-layer WS₂. The hexagonal boron nitride film grown by CVD not only provides a stable platform for WS₂ field-effect transistors but also acts as a protective layer against the external environment. In 2020, Zheng et al [99] reported for the first time a series of vertical heterostructures with high quality and a large area of WS_{2 (1−x)} Se_2x /SnS₂ by two-step CVD. Based on the obtained structure, they successfully designed a dual-channel-type tunable field-effect transistor and the transition from a pure n-type unipolar transistor to a bipolar transistor was realized in this system.