Introduction
The discovery of graphene and its fascinating properties has stimulated serious research on other materials which have interesting properties when it comes to a two dimensional (2D) structure in comparison with their bulk materials [1–5]. Two dimensional transition metal dichalogenides (TMDs) is one of these group of materials which consists of the transition metal (M) and chalcogen (X) with the MX2 formula [6–9]. MoS2 is known as one of the very common transition metal dichalcogenides (TMDs) [6,7,10–13], which have the weakly interacting layered character of graphite and providing novel properties when it has a two dimensional structure [1,14]. The 2D layered structure MoS2 could be found in enormous quantities in nature [8,9]. In the MoS2 structure, two S atomic layers are attached to the Mo atomic layer by covalent bonds and make a sandwich structure with a thickness of 6–7 Å while the weak van der Waals bonds keep the two adjacent MoS2 layers together [9,15–17]. There are a vast variety of applications for MoS2, such as advanced energy storage and conversion [18], electrochemical catalysis [19,20], and sensing [21–23] due to its unique properties of the anisotropy. The structure of MoS2 plays a very important role in the electrical conductivity of this material [24–27]. The metal coordination of a single layer MoS2 could be observed as an octahedral or trigonal prismatic, as shown in Fig. 1 [28]. Hence, a number of these single layers could be made to produce various polymorphic structures which could be defined by the type of symmetry or the number of stacking layers in its unit cell (1, 2, or 3). Common polymorphs of MoS2 could have tetragonal, hexagonal, and rhombohedral symmetries that are known as 1T, 2H and 3R structures, respectively. The 1T form shows a metallic behavior with a better conductivity in comparison to the semiconducting behavior of 2H and 3R forms. In other words, the 1T form has a better conductivity than the other two semiconducting forms. It has also been proved that the catalytic properties of these three forms are different, where the 1T phase has a better catalytic properties than 2H phase [29].
In this review, new approaches and various synthesis methods to synthesize MoS2 is presented. This review is segmented into three main parts: preparation, which is described as top-bottom and bottom-top methods, such as liquid-based ultrasonic and mechanical exfoliations, chemical exfoliation via lithium ion intercalation, chemical vapor deposition, and hydrothermal-solvothermal approach; parameters affecting MoS2 properties, which are discussed as structural strain, temperature, impurity doping, core-shell and structural defects; and the applications of MoS2, which is discussed in the area of solar thermal water purification, photocatalytic process and in photo electrocatalytic HER. It is arranged in a manner to facilitate the reader to grasp the in-depth idea about the technological developments in MoS2 material for energy conversion applications and how MoS2 stands out as a promising alternative 2D material for solar energy conversion.
Synthesis methods
Because of the attractive and important properties and the potential applications of MoS2 nanosheets, various preparation methods have been demonstrated, such as mechanical exfoliation [30–34], electrochemical Li-intercalation and exfoliation [35,36], direct sonication in solvents [37,38], and chemical vapor deposition (CVD) [39,40]. The synthesis methods can be categorized into two general types: the top-down methods and the bottom-up methods. The top-down methods mechanically, ultrasonically, or chemically exfoliates bulk materials by overcoming the weak interlayer binding force, such as van der Waals, while the bottom-up methods assemble MoS2 nanosheets using individual atoms.
Top-down methods
Mechanical exfoliation by Scotch tape can generate large, defect-limited, and electronic-grade MoS2 nanosheets for fundamental studies [41,42]. However, this low yield, uncontrollable approach has limited usefulness in practical application, which usually requires a large quantity of nanomaterials with good quality. In contrast, liquid-based ultrasonic exfoliation can easily produce bulk dispersion of monolayer or few layer MoS2 nanosheets or their mixtures [37,43]. Another advantage is that ultrasonic exfoliation does not induce any structural distortion and thus maintains the semiconducting 2H phase, unlike the phase conversion during the chemical exfoliation. However, the major limitation of ultrasonic exfoliation is that the raw product is mostly in a multilayer form.
To further increase the yield, chemical exfoliation has been used to produce MoS2 monolayer via lithium ion intercalation [44]. In 2014, Voiry and coworkers functionalized TMD materials including MoS2, WS2, and MoSe2, where the functionalization reaction is enabled by transferring electrons between the electron-rich metallic 1T phase and an organohalide reactant, contributing to functional groups that are covalently involved to the chalcogen atoms of the transition metal dichalcogenide [45]. Studies have shown that using butyllithium intercalation as the chemical exfoliation would result in crystal phases with dissimilar electronic properties in the nanosheets [46,47]. Take, for instance, the thermodynamically stable 2H phase in the trigonal prismatic MoS2, where the butyllithium intercalation contributes to partial transformation of the crystal structure from the 2H phase to the octahedral 1T metallic phase caused by the electron transferring from the butyl group to the MoS2 sheets, as shown in Figs. 2(a) and 2(b) [44,46,48].
Bottom-up methods
MoS2 nanosheets can also be produced by first decomposing Mo- and S-containing precursors and then assembling Mo and S atoms, such as chemical vapor deposition (CVD). CVD is a widely used technique for growing 2D nanomaterials in a controllable manner [40,49–52]. It can be classified into three options: thermolysis of precursors containing Mo and S atoms; vaporization and decomposition of Mo and S precursors and subsequent formation of MoS2 layers on a growth substrates; and direct sulfurization of Mo-based films, such as Mo metal or MoO3. Another widely explored bottom-up method is to use a hydrothermal-solvothermal reaction, which occurs in a sealed autoclave at high temperature and pressure [53–59]. This method usually produces hybrid 1T/2H phase in the as-prepared MoS2 nanosheets, while pure 2H-MoS2 can be obtained by post-annealing. Besides its simplicity and wide applicability, the hydrothermal-solvothermal approach is very attractive owing to its facile hybridization with other functional nanomaterials [60].
Parameters affecting MoS2 properties
MoS2 is a semiconductor with an indirect band gap and consists of S-Mo-S sheets that are held together by van der Waals interactions through a hexagonal structure [61]. However, the single layer MoS2 has a direct gap with a band-gap energy of about 1.9 eV [42,62,63], hence it has become an exciting candidate for electronic, optoelectronic, and photovoltaic applications. In this review, the description of different parameters for tuning MoS2 properties and various applications are mainly focused on. There are several parameters that affect the band gap and electrical properties of MoS2.
Structural and strain effects
In 2012, Scalise et al. studied the electronic and vibrational properties of 2D honeycomb structures of MoS2 by applying tensile and compressive biaxial strains to tune the in-plane XY orbital interactions, and therefore, increased the interlayer atomic distance [61]. They observed that by increasing the level of applied strain, the shrinkage of energy band gap occurred. In other words, both the bottom of the valence band and the top of the conduction band would cross the fermi level and it is predictable to detect a transition limit of the system from semiconducting to metallic, while the strain range is about 8%–10%. There is not any significant difference observed in the case of Mo-S bond lengths and S-S distances with respect to the bulk structure for the multilayer MoS2 structures with 1, 2 and 3 S-Mo-S. The nature of the gap and the electronic properties of layered MoS2 could be affected by the variations in the charge interactions along the Z-direction (out-of-plane). They reported that the monolayer MoS2 was tuned to an indirect gap semiconductor, even by applying less than 1% of tensile strain.
In 2014, Ebnonnasir et al. found out the band gap of MoS2 could be affected by changing the orientation of graphene in MoS2 heterostructures, and outlined the physical basis of this outcome to the highly sensitive dependence of the band gap value on the thickness of the MoS2 layer [64]. The band structure would be changed due to any fluctuations through the band length that can happen by interfacial electronic transfer owing to applied changes in registry between graphene and MoS2. They also showed that graphene-MoS2 hetero-structures are delightfully appropriate for photovoltaic devices, in which the MoS2 could form the exciton pair that would be collected at graphene layers situated on either side of the MoS2 layer.
A number of other studies were conducted to investigate the effect of structure and strain on the band gap of TMDs [65–67]. Peelares et al. used the first-principles hybrid density functional theory to examine the electronic structure of MoS2 and explore strain effects corresponding to experimentally accessible uniaxial and biaxial stress conditions [65]. The band structure could affect the tensile uniaxial strain, hence the transition from the bulk to the monolayer band structure can be monitored. Moreover, Shi et al. reported that WS2 possessed the lightest effective mass at the same strain value compared to Mo(S,Se) and W(S,Se) monolayers [68].
Temperature effects
The crossover of band gap from direct to indirect band gap could occur by changing the temperature as well as structural and strain effects. In 2012, Tongay et al. explored the thermally driven crossover from indirect to direct band gap in two dimensional semiconductors by comparing MoSe2 and MoS2 [69]. In a few layer MoSe2 that has a good thermal stability and a direct band gap of 1.55 eV, the indirect band gap and direct band gap are nearly degenerate, despite of the well-explored MoS2. That is to say increasing the temperature contributes to the interlayer thermal expansion and reducing the coupling between the neighboring layers which results in the fact that the system would be driven toward the quasi-2D limit. In contrast, not only MoS2 has a higher band gap than MoSe2, but also the direct and indirect band gaps are well-separated in energy and consequently far from degeneration. As shown in Fig. 3(a), exfoliated few-layer flakes have the characteristic of A1g (out-of-plane) and E2g1 (in-plane) Raman modes at two locations as 243.0 and 283.7 cm-1 for MoSe2 and 408.7 and 383.7 cm-1 for MoS2, respectively. The observed A1g mode is at a higher frequency than the E2g mode for MoSe2. Additionally, the peak position of these Raman modes has a slight affiliation to the thickness of layers. Hence, for the single layer, the A1g Raman mode relaxes to 241.2 (406.1) cm-1 as the E2g1 mode strengthens to 287.3 (384.7) cm-1 for MoSe2. The out-of-plane A1g mode is expected to unstiffen owing to the decrease in the restoring forces ascending from the lack of interlayer coupling, when the interlayer coupling is lacking in the single layer limit. It is of great importance to mention that this model does not account for the strengthening of the in-plane E2g1 mode. More captivatingly, the intensity ratio between the A1g and E2g1 modes (I A1g /I E2g1) increases from 4.9 for few-layer (~10 layers) to 23.1 for the single-layer MoSe2, while the ratio remains nearly a constant (~1.2) in the MoS2 case. In addition, the bulk MoSe2 has an indirect band gap with a value of 1.1 eV [70], and therefore the band gap PL is anticipated to be relatively weak. However, the few layer MoSe2 flakes display ongoing improvement in PL greatness at around 1.5 - 1.6 eV, and the PL peak intensity reaches its maximum value for a single-layer MoSe2 as presented in Fig. 3(b). The temperature reliance of PL dignified on single- and few-layer samples of MoSe2 and MoS2 is displayed in Figs. 3c–3f. The rate of the indirect-to-direct band gap crossover varies considerably between MoS2 and MoSe2. Even though single layer MoSe2 is a direct band gap semiconductor (1.34 eV), the indirect band gap value (1.50 eV) lies near the direct band gap. This dissimilarity value of 0.16 eV is much slighter than the difference of 0.35 eV between the direct (1.54 eV) and indirect (1.89 eV) band gap for the single-layer MoS2. By increasing the number of layers, the quantum confinement in the perpendicular direction is undisturbed, and consequently the indirect band gap tends to be reduced, while the direct band gap value remains largely unchanged, owing to the heavier effective mass associated with the K symmetry point.
Impurity doping
By comparing the bulk semiconductors with nanocrystals, the latter have a various and increasing range of factors that can control the electronic band gap of these materials as can be observed in Fig. 4, including size, shape, and composition. Quantum confinement can move the band gap of most semiconductors by over 1 eV, giving a huge range of continuous tunability through the size and shape for a single material composition [71]. The use of quantum confined structures permits the independent tuning of size and band gap through the operation of homogeneously alloyed materials such as CdSeyTe1–y and CdxZn1–xS [71]. Through a chemical alloying process by impurity doping, an intraband electronic energy level would be formed, which allows lower energy light emission to the ground state from the defect state. Doped nanocrystals have exciting properties for biolabeling and device applications, including large stokes shifts, paramagnetic properties, and improved lasing. However, due to the chemical contrasts between the dopants and their crystalline matrices, the creation of these doped nanocrystals is still a challenge.
Zhang et al. theoretically studied the influence of impurities on the electrical and optical properties of monolayer MoS2 and concluded that the band gap would be dominated by donor impurities in 2017 [72]. They used the VB and VIIB transition metal atoms such as V, Nb, Ta, Mn, Tc, and Re and reported that the n-type or p-type doping could tune the Fermi level into the conduction band or the valence band, respectively. Therefore this leads to the degenerate semiconductor while the compensatory doped systems where the amount of valence electrons is not alerted keep on direct band gap ranging from 0.958 eV to 1.414 eV. By analyzing the densities of states, it has found that the LUMO orbitals of donor impurities play a critical part in band gap tuning. Owing to band gap reduction, doped MoS2 has a minor threshold energy of photon absorption and an improved absorption in near infrared region. These outcomes deliver a noteworthy direction for the design of new 2D optoelectronic materials based on MoS2 and other TMDs.
Core-shell (Type-II quantum dots)
In 2003, Bawendi et al. observed the separation of the electron and the hole between the core and shell materials through (core) shell semiconductor heterostructures, in which the conduction and valence bands of the core and shell material are staggered [73]. Li and coworkers reported that the flower-like MoS2/BiVO4 composite with heterojunction showed outstanding performance for photodegradation of methylene blue due to the staggered band alignment shaped between MoS2 and BiVO4 [74,75]. Based on these studies, the 2D-layered MoS2 is found to have immense potential to be used for photocatalysis applications. It is of great importance to indicate that MoS2 has some disadvantages, such as inadequate charge segregation and underprivileged charge mobility [76]. Both of them will lead to low photocatalytic behavior. Figure 5(a) is the UV absorption spectra of BiVO4, MoS2, and BiVO4@MoS2 where the bare MoS2 nanosheets express noteworthy absorption both in the ultraviolet and visible regions. Meanwhile, the absorption wavelength for pure BiVO4 lies around 500 nm and attributes to the intrinsic band gap absorption. By depositing MoS2 nanosheets on the surface of BiVO4, a better visible-light photocatalytic performance is observed for the composite in comparison with the pure BiVO4. Moreover, the formula of (ahn)n = A (hn-Eg) is being used to compute the band gap energy (Eg) of BiVO4 and MoS2, where h, n, Eg, and A are the absorption coefficient, Planck’s constant, light frequency, band gap, and a constant, respectively. In addition, the index n would be determined based on the electronic transition of the semiconductor. For instance n = 2 for direct-gap semiconductor and n = 0.5 for indirect-gap semiconductor would be considered. Therefore, the value of index n is equal to 2 for MoS2 and BiVO4. The band gap energy can be projected from the intercept of the tangent to the plot of (ahn)2 versus the radiation energy (hn), as can be seen in Figs. 5b–5f. They reported the band gap energy for BiVO4 and nanosheet MoS2 as 2.50 eV and 1.62 eV, respectively [77], while the bulk MoS2 has shown a band gap energy of 1.23 eV [78], can be assigned to the strong quantum confinement effect of the thin nanosheets, which also makes the MoS2 nanosheets an effective visible-light photocatalyst [79]. The values of band gap energy are reported as 2.19, 2.09 and 1.90 eV for BiVO4@MoS2 (2 wt%), BiVO4@MoS2 (5 wt%), and BiVO4@MoS2 (10 wt%), respectively. Since the photocatalytic property of the photocatalyst is contributed to its band structure, there is a simple approach to determine the band edge positions of both the conduction band (ECB). Besides, the valence band (EVB) at the point of zero charge (pHZPC), can be calculated by ECB= X-Ee- 0.5Eg [80]. The X is the absolute electronegativity of the semiconductor, Ee is the energy of free electrons on the hydrogen scale (4.5 eV), Eg is the band gap, and the valence band edge (EVB) can be determined by EVB = ECB + Eg. Therefore, the ECB values of BiVO4 and MoS2 are determined to be 0.29 and 0.03 eV, respectively, where the EVB values of BiVO4 and MoS2 are estimated to be 2.79 and 1.65 eV, respectively. Figure 6 demonstrates the photocatalysis mechanism of p-n heterojunction photocatalyst and the schematic drawing of the electron-hole separation process.
Structural defects
A number of studies have established that grain boundaries [81–83], the size of different constituents [84,85], and the interface with other monolayers [86] can greatly affect the electronic properties of two-dimensional semiconductors. In 2012, Zou and coworkers theoretically considered predicting dislocations and grain boundaries in TMDs from the first principles [83]. As Fig. 7 demonstrates, the defects of extended grain boundaries (GB) take about significant changes to the electronic structures for MoS2, where the direct band gap is calculated at K point as 1.8 eV for perfect MoS2. Figures 8a-8d illustrate that the plots of local density of states (LDOS) and the presence of localized states could be observed with the energies intensely in the middle of the pristine MoS2 gap which has been enhanced by the GB. The partial charge density distributions for these localized states have been computed and it is detected that they mainly initiate from the d orbitals of metal atoms in the Mo- and S-rich GBs, or from localized p states of S atoms in 6|8s. By reducing the carriers, these deep states may cut down the performance of devices based on polycrystalline MoS2, A-GB’s of other tilt angles. The local bands turn out to be dispersive for a greater tilt a where the cores are situated denser and relate with each other more tenaciously, and the discrete metallic behavior is emerged at a = 60° (from AC) GB. The states delocalized in one dimension associate with the metallic stripe bands overpassing the Fermi level. The inset of Fig. 7(d) shows the metallic behavior through the partial charge density during the LDOS contribution and proves that the wave functions are certainly corresponding to the GB and can be identified by scanning tunneling microscopy (STM). This study has shown that such metallic stripes could indeed introduce novel and valuable functionality brought by carefully engineered GB, in contrast to their more often detrimental role. Figure 9 shows the basic edge dislocations shaped by removal of shaded atoms from the lattice.
Applications of MoS2
Solar thermal water purification
Due to the increase in global population and demand for clean fresh water, major developments for water treatment are taken into consideration. Solar steam generation based on photothermal effect has attracted great attention as a new water treatment technique because of the present of abundant and sustainable sunlight. Until now for efficient light absorption, carbon-based materials like graphene, graphene oxide and carbon nanotubes have been used for the treatment. The results indicate that by using the carbon-based material, the solar energy conversion efficiency is low, which needs to be rectified for better results. The synthesis of these materials requires many steps and high temperature procedures by which the cost effectiveness issue comes into consideration as it is an important factor for solar thermal treatment. By considering the above factors, the need for cost effective, scalable and highly efficient photothermal materials is required. Chemically exfoliated MoS2 is used for the generation of fresh water, as MoS2 is highly efficient, scalable, and environmentally benign for solar evaporators. It was also reported that phase transition of MoS2 from 2H phase which is a trigonal prismatic coordination to 1 T phase which is a octahedral coordination during the exfoliated process could enhance the light absorption of the ce-MoS2 [60,87].
Jun et al. first investigated the earth-abundant and economical MoS2 as a suitable material for solar steam generation [88]. Bulk MoS2 particles showed multiple-stacked sheets with large lateral dimensions as shown in Fig. 8(a). The exfoliated MoS2 nanosheets had an average thickness of 1.5 nm with lateral dimensions of 200–800 nm, as revealed by atomic force microscopy (AFM) images (Fig. 8(d)). X-ray photoelectron spectroscopy (XPS) was used to analyze the phase composition, as shown in Fig. 8(e). The spectrum of bulk MoS2 exhibited only 2H peaks at 229.4 and 232.6 eV, suggesting that natural MoS2 consists of only the 2H phase. In contrast, the peaks corresponding to ce-MoS2 shifted to lower binding energies, and 1T phase peaks could be identified at 228.7 and 231.8 eV upon deconvoluting the spectrum using Gaussian-Lorentzian curve-fitting. During the Li intercalation process, the phase transition from 2H to 1T drastically increased the evaporation rates of chemically exfoliated (ce)-MoS2/bacterial nanocellulose (BNC), because of the effective heat localization and good light absorption at the water/air interface. The cost effective, efficient steam generation of ce-MoS2 nanosheets, along with their low cytotoxicity, promise their utility as a photothermal material for solar steam generation, water desalination and sterilization, and photothermal therapy.
Photocatalytic process
An effective photocatalyst relies on its band edge positions, which determines the redox potentials of photo-generated electrons and holes. In general, as the potential of conduction band (or valence band) becomes smaller (or larger), the photogenerated electrons (or holes) have a stronger reductive (or oxidative) capability [89]. The appropriate band structure makes MoS2 nanosheets one of the most promising photocatalyst candidates.
Hydrogen has been considered as a clean energy resource that can be easily stored and used without producing any greenhouse gases. Since the emergence of photocatalytic water splitting in 1972 [90], hydrogen evolution reaction (HER) via photocatalysis has attracted enormous attention because it only requires water and solar energy. However, producing the conventional photocatalyst involves the utilization of noble metals or alloys, such as Pt, which hinders their large-scale application [91]. Thus, the development of competitive alternatives to noble metals is pursued. MoS2 has recently been regarded as a promising candidate of noble-metal-free co-catalyst as it is an earth-abundant and visible light-responsive photocatalyst with unique physical and chemical properties [92,93]. Chang et al. systematically investigated the relationship between MoS2 layer number and photocatalytic hydrogen generation activity [94]. MoS2 loaded CdS (MoS2/CdS) with MoS2 layer numbers varying from 1 to 112 were prepared in their study. The corresponding hydrogen production activity in Na2S-Na2SO3 and lactic acid solutions were explored, as shown in Fig. 10. It was demonstrated that the photocatalytic activity is increasing with decreasing MoS2 layer number and highest H2 production rate was achieved when the MoS2/CdS had a single layer (SL) MoS2. This layer-number-dependent photocatalytic activity is contributed to three major reasons. The first one is that the unsaturated S atoms of the exposed edge sites are the active sites of hydrogen generation. With the decrease of MoS2 layer number, more active S atoms are exposed compared to the bulk material. The second one is that reducing the MoS2 layer number is beneficial to the separation of charge carriers. It is easier for electrons to transport from CdS to SL MoS2 than the bulk material. Finally, the conduction band minimum (CBM) of SL MoS2 is not only more negative than H+/H2 potential but also negative versus the CBM of bulk form, thereby more electrons in the conduction band can reduce H+ to H2.
The metallic 1T-MoS2 has a tremendous photo-induced catalytic activity because of the tunable band gap and the fact that both the edges and the basal plane of the 1T-MoS2 are catalytically active. The metallic 1T- MoS2 nanosheets can also act as an ideal support and co-catalyst for various composite catalyst which gives further enhancements in the performance in the presence of a photocatalyst. These characteristics make 1T-MoS2 nanosheets and their composites a promising substitute to noble catalyst-based photocathodes for solar light-driven catalytic applications. Han et al. proposed that few layers MoS2 coating on TiO2 nanosheets (MST) will increase the photocatalytic activity for hydrogen production [95]. The report from the experimental and theoretical calculations stated that MST as a photocatalyst showed a significant low-charge recombination rate and excellent long-term durability. The authors also reported that there was a significant increase of 31.9% in photocatalytic activity for the hydrogen production when compared to TiO2. Hsiao et al. developed a one-pot synthesis method for the preparation of 1T-MoS2 for photocatalytic hydrogen evolution (HCS) [96]. He stated that the amount of hydrogen evolved over 1T-MoS2@HCS reached 143 mmol in two hours which was 3.6 times higher when compared to that of 2H-MoS2@HCS. Furthermore, 1T- MoS2 could also be used as a counter electrode material for dye-sensitized solar cells (DSSCs), which had a higher conversion efficiency when compared to 2H- MoS2@HCS counter electrode.
Over the past few years, MoS2 has also been identified as efficient photocatalysts to degrade organic-pollutants in wastewater. MoS2 is a nontoxic environment friendly material which has shown a good stability against photo corrosion and a good absorbent of the solar spectrum [97–100]. Since pure MoS2 has a low photocatalytic performance, it has been combined with other semiconductor materials [101]. This combination allows a rapid charge separation due to different energy level forms of each semiconductor [102]. Several studies have been developing various combinations such as MoS2/RGO composites [103], MoS2 nanosheets supported by TiO2/g-C3N4 photocatalysts [104], ZnO-MoS2-RGO heterostructure [105], MoS2/BiOBr microspheres [106], and flower like MoS2/CdS heterostructures [107].
Extensive research has been conducted for the photocatalytic inactivation of microbial pollutant due to the superior inactivation ability of photocatalytic semiconductors compared with the conventional disinfection methods [108–110]. The mechanism of photocatalytic disinfection involves reactive oxygen species [110], which can trigger the breakdown of cell membranes and then promote the internalization of the semiconductor photocatalyst, ultimately leading to cell death [111,112]. Therefore, improving the separation of photogenerated electrons and holes is in favor of photocatalytic disinfection efficiency. Awasthi et al. attached flower-like ZnO on the surface of layered MoS2 via the hydrothermal method [113]. An enhanced antibacterial activity was observed by using gram negative E. coli bacteria in comparison with pristine ZnO and MoS2. It is not only attributed to the efficient separation rate of electron and hole pairs in the ZnO/MoS2 nanocomposite, but also the increased surface area induced by the layered MoS2. Liu et al. prepared a ternary photocatalyst (carbon nanotubes(CNTs)-MoS2-Ag) and discovered its antibacterial activity against the Staphylococcus aureus and Escherichia coli [114]. It is reported that the antibacterial mechanism of this material not only includes the microorganism killing effect of Ag particles, but also relies on light absorption.
Photoelectrocatalytic HER
Utilization of semiconductor thin films as photoelectrode in photoelectrocatalytic water splitting application has been considered as a clean and reproducible energy [115,116]. The photoelectrocatalytic HER relies on the catalyst material having two properties: the material should be active for the electrocatalytic HER, and it should absorb sunlight, giving an electron hole pair with the right potential to drive the HER reaction. Recently, sizeable amount of highly active photocatalysts have been reported where they all possess one identical weakness in respective developments. That is, their wide band gaps have caused a major stumbling block in the refinement of their photocatalytic properties [117].
MoS2 is a semiconductor with a direct band gap around 1.7 eV, indirect band gap at around 1.2 eV [118], and the band position slightly more positive than that for the HER [119]. This makes MoS2 a promising candidate for use in a tandem approach for the HER side if the conduction band could be pushed to slightly more genitive values (Fig. 11(a)). This means that MoS2 will not evolve hydrogen when illuminated [120] unless a negative bias (Fig. 11(b)) or quantum confinement is introduced [121]. The second way to utilize MoS2 as a photoelectrocatalyst is to use the electrocatalytic HER activity together with a good photon absorber (exemplified for Si in Fig. 11(c)). In these systems, the distinct difference to pure MoS2 is clearly seen, as in these systems the photon absorber may be freely chosen to provide the sufficient overpotential to drive the HER. A novel MoS2 nanomaterial comprising a narrow-but-targeted direct band gap has been produced [121] and MoS2 has proven to have an excellent electrocatalytic hydrogen evolution activity and active edge defect sites [122–124]. He et al. reported that 1T-MoS2/TiO2 nanotubes and 1T- MoS2/Si-doped TiO2 nanotubes had a high conductivity and an excellent catalytic activity for HER [125]. MoS2 decoration provided high light absorbance property for these hybrids and the highly interface-induced effect between nanotubes and MoS2. The results illustrated that the hybrids showed higher photocatalytic and photoelectrocatalytic activities than TiO2. Li et al. reported that when 1T-MoS2 were coupled with inorganic-lig and stabilized CdSe/ZnS quantum dots through the process of self-assembly approach, it could be used for various applications in solar to fuel conversions [126]. Under the optimum condition, hydrogen gas is produced at a rate of 155 mmol/h per mg. As a result of improved light harvesting, facilitated interfacial charge transfer and excellent proton reduction ability, it has a high activity for solar hydrogen evolution. Xu et al. developed a metallic 1T-MoS2/O-g-C3N4 heterostructure using the in situ growth method which exhibited a relatively greater performance when compared to 2H-MoS2/O-g-C3N4 material [127]. The major reason for the greater performance is the more exposed edge sites, active basal planes, and closed contact layers between the two components. This close contact is more favorable to transfer photo-generated electrons from the conduction band of O-g-C3N4 to the surface of 1T-MoS2, which leads to the greater performance of the 1T-MoS2/ O-g-C3N4 system. Du et al. also reported a novel photocatalyst 1T-MoS2/CdS hybrid for photocatalytic hydrogen evolution. The optimized hybrid shows 35-folds improvements in photocatalytic hydrogen evolution in visible light radiation when compared to pure CdS nanosheets [128]. Jin et al. reported a unique heterostructure which consists of exfoliated 1T-MoS2 on planar p-Si and behaves as the efficient and robust photocathode for solar driven HER [129]. This unique heterostructure exhibits an excellent onset of photocurrent and a high current density for planar p-Si photocathodes which uses non-noble metal catalyst. From the electrochemical impedance spectroscopy measurements demonstration, it is inferred that the excellent performance of the CVD grown 1T-MoS2 /p-Si photocathode is due to the smaller charge-transfer resistance across the interfaces of semiconductor and catalyst or catalyst and electrolyte. The electrochemical impedance spectroscopy measurements also showed slow carrier recombination dynamics and efficient charge carrier separation. Another research developed a new novel heterostructure ammoniated with MoS2 with a unique 1T/2H phase by using the facile hydrothermal approach [130]. The enhancement of the electronic conductivity increase in density of the active sites and improved efficiency in both electro transfer and mass transport are resulted from the presence of the 1T phase in MoS2. Due to the presence of both the 1T/2H phases of MoS2, the photo electrocatalytic properties are improved.
Fig.11 Band gap position and redox levels for the HER for (a) MoS2 at the flat band potential; (b) MoS2 at large negative bias; (c) Si at flat band potential; and (d) Si at negative bias (From the figure it is seen that the HER does not occur spontaneously on bulk MoS2, but requires additional quantum confinement, or high p-type donor concentrations. In (b) the MoS2 is biased enough that the HER occurs, the exact potential cannot be defined. On Si the HER occur readily even at slightly positive bias vs. RHE, until the flat band potential (c) at more negative potentials the HER occurs even faster (d) (Reproduced from Ref. [121] with permission from Royal Society of Chemical) |
Carbon materials play important roles in improving the interfacial charge transfer between MoS2 catalysts and the electrode substrate. Xiang et al. synthesized a layered MoS2/graphene hybrid and incorporated it with TiO2 to form a new composite material [131]. As shown in Fig. 12, under light irradiation, electrons transfer from the valence band of TiO2 to the conduction band of TiO2. Subsequently, one part of the electrons is directly delivered to MoS2, while the other part of the electrons flows to MoS2 through graphene. The edge sites of MoS2 can effectively reduce water to H2. The introduction of graphene material can remarkably improve the charge transport and hinder the recombination of electrons with holes.
Fig.12 Schematic illustration of the charge transfer in TiO2/MG composites (The proposed mechanism for the enhanced electron transfer in the TiO2/MG system under irradiation assumes that the photo excited electrons are transferred from the CB of TiO2 not only to the MoS2 nanosheets but also to the C atoms in the graphene sheets, which can effectively reduce H+ to produce H2 (Reproduced from Ref. [131] with permission from American Chemical Society) |
Conclusions and outlook
As a repressive layered transition metal dichalcogenides (TMDs), MoS2 has been widely invested for solar energy conversion applications. In this review, the recent progress of MoS2 for its applications in the fields of solar thermal water purification, photocatalytic process, and photoelectrocatalytic HER has been summarized. In particular, the recent advances of structure control, elemental doping and synthesis methods for preparing MoS2 have been described. Besides, their potential advantages and limitations have also been emphasized, such as the rapid recombination of photo generated carriers, limited active edge sites and difficulties in photocatalyst recycling that hinder the practical application of MoS2-based photocatalysts in solar energy conversion. In conclusion, MoS2 has been experiencing a renaissance in recent years, largely owing to the demand for renewable energy conversion and storage. The unique properties of MoS2-based photocatalysts and their good performance in different applications suggest that it is a promising earth-abundant photocatalyst. Although experimental results on a laboratory scale have been reported, the real application of MoS2 as catalytic or electrode materials is still challenging. This means that more encouraging studies are highly desirable in this field.