Transition Metal Dichalcogenides for the Application of Pollution Reduction: A Review

The material characteristics and properties of transition metal dichalcogenide (TMDCs) have gained research interest in various fields, such as electronics, catalytic, and energy storage. In particular, many researchers have been focusing on the applications of TMDCs in dealing with environmental pollution. TMDCs provide a unique opportunity to develop higher-value applications related to environmental matters. This work highlights the applications of TMDCs contributing to pollution reduction in (i) gas sensing technology, (ii) gas adsorption and removal, (iii) wastewater treatment, (iv) fuel cleaning, and (v) carbon dioxide valorization and conversion. Overall, the applications of TMDCs have successfully demonstrated the advantages of contributing to environmental conversation due to their special properties. The challenges and bottlenecks of implementing TMDCs in the actual industry are also highlighted. More efforts need to be devoted to overcoming the hurdles to maximize the potential of TMDCs implementation in the industry.


Introduction
Transition metal dichalcogenides (TMDCs) are a large family of two-dimensional (2D) layered materials, which are scientifically interesting and industrially important. These materials have attracted tremendous attention because of the unique structural features and interesting properties, such as optoelectronics, electronics, mechanical, optical, catalytical, energy-storage, thermal, and superconductivity properties [1][2][3][4][5][6][7]. TMDCs are the compounds of the chemical formula MX 2 , where M is a transition metal element of groups IV-VII B (Mo, W, V, Nb, Ta, Ti, Zr, Hf, Tc, and Re) and X is a chalcogen element (S, Se, and Te). The X-M-X unit layer consists of three atomic layers, in which one centre atom layer (M) is sandwiched between two chalcogen atom layers (X). TMDCs occupy the layered structures, which resemble that of graphite. The interlayers are stacked by weak van der Waals force, leading to the formation of monolayers or nanolayers from the bulk materials via exfoliation [8].
Nanomaterials 2020, 10 Different stacking of the layers along c-axis determines polymorphic crystal structures in TMDCs, and the common phases are 1T, 2H, 3R, and Td phases (T-trigonal, H-hexagonal, R-rhombohedral, and Td-distorted octahedral) [9]. There are more than 40 different TMDC types, including metals (such as TiS 2 and VSe 2 ), superconductors (such as TaS 2 and NbS 2 ), semimetals (such as MoTe 2 and WTe 2 ), and semiconductors (MoS 2 , MoSe 2 , WS 2 , and WSe 2 ). TMDCs exhibit interesting band structures with tunable bandgaps. The bandgap is one of the most important factors in 2D materials for determining the properties and applications. For instance, graphene is a semimetal with zero bandgap, which limits its applications in electronics and photo-electronics. TMDCs exhibit variable bandgaps from 0 to 3 eV, which can be tuned by thickness [10], defects [11], dopants [12], and mechanical deformations (by applying the tensile strain or compressive strain) [13,14]. The most studied semiconducting TMDCs (e.g., MoS 2 , MoSe 2 , WS 2 , and WSe 2 ) have shown typical features in electronic structures. The bandgap increases with the decreasing thickness and it possesses the transition from indirect in the bulk crystals to direct in the monolayers [10,15]. For instance, the indirect bandgap of −1.29 eV will be changed to a direct bandgap of −1.8 eV when bulk MoS 2 is down to a monolayer [16].
Benefiting from their unique crystal structures and electronic structures, TMDCs have shown great potential in various fields, including electronics/optoelectronics [1,17], catalysis [18], energy storage [19] and conversion [20], sensing [21,22], and so on. The application of TMDCs in pollution reduction is a compelling research topic. The increasing environmental pollution issue has been one of the most serious problems on Earth. Enormous efforts have been made to search the efficient and low-cost methods for addressing the environmental pollution issue. TMDCs may be a kind of promising materials for tackling these problems with several advantages. Firstly, TMDCs have a high surface-to-volume ratio. They offer more effective active sites on the surface, as well as abundant unsaturated surface sites. Thus, the layered TMDCs are excellent platforms for the anchor of semiconductor nanoparticles in various photocatalytic applications [23,24]. Due to their high surface-to-volume ratio, TMDCs are extremely sensitive to the surrounding atmosphere and can be utilized in toxic gas sensing and adsorption. Secondly, TMDCs have tunable bandgaps, which enhances the photocatalytic performance in nanocomposite by offering appropriate bandgap and band alignment [25]. Thirdly, defect engineering can be easy to implement in 2D materials, which have been confirmed to be an efficient method for intensifying the catalytic activities in TMDCs [26][27][28][29]. Lastly but importantly, there is a large variety for TMDCs (about 40 kinds) and they have an abundant amount in nature or can be synthesized [9]. So far, MoS 2 , WS 2 , MoSe 2 , and ReS 2 have been naturally found [30][31][32]. Specifically, MoS 2 exists as molybdenite in nature and is the main source of molybdenum with a large amount [33]. The main metals (W and Mo) in TMDCs are both abundant, cheap, and widely used in industry [34]. TMDCs can be prepared by using various techniques, such as chemical vapor deposition (CVD) [35,36], chemical vapor transport (CVT) [37,38], flux growth method [39], hydrothermal synthesis [40], Langmuir−Schaefer deposition [41], etc. In addition, the top-down exfoliated method can be also used to fabricate few-layer TMDCs from bulk crystals, e.g., mechanical exfoliations and liquid phase exfoliations [42][43][44]. With increasing interests in TMDCs for applications, we aim to prepare an overview of the recent progress of TMDCs in reducing the environmental pollution. We will summarize the representative efforts, including gas adsorption and removal, gas sensing, wastewater treatment, fuel cleaning, CO 2 valorization, and conversion.

Gas Adsorption and Removal
In recent years, TMDCs have been used as gas removal (via adsorption) in pollution reduction. Figure 1 presents the number of research articles that involved the use of TMDCs in gas adsorption applications. The increasing trend (the total number of related articles published in the year 2019 is more than three-fold than that of the year 2015) indicates the potential of TMDCs in gas removal processes. Based on the bibliology search, MoS 2 is still the most widely studied TMDCs among the TMDCs considered in this review. However, researchers started to realize that there are other TMDCs Nanomaterials 2020, 10, 1012 3 of 33 (e.g., MoSe 2 and WS 2 ) that offer better performances as compared to MoS 2 (e.g., greater charge storage ability [45]). This further led to a gradual increase in research articles that studied the use of other potential TMDCs, since the year 2018. Figure 2 outlines the schematic diagrams of each gas adsorption process. In general, the mechanism of these adsorption processes is mainly driven by the charge transfer process between the adsorbates and the TMDCs-based adsorbent, where the charge movement is dependent on the nature of the gas adsorbate (i.e., oxidizing or reducing) [46]. For instance, carbon monoxide (CO) in Figure 2a is an example of reducing gas. Due to the existence of lone pair on the carbon atom, CO will donate electrons to the TMDC surface, which further cause the CO to be chemisorbed on the surface. Whereas, oxidizing gas such as nitrogen dioxide (NO 2 ) in Figure 2c, will uptake the electron from the surface instead (mainly due to the existence of unpaired electron on the nitrogen atom). To note, such electron movement will lead to the deviation in electrical conductivity of the TMDC materials [47]. The following subsections discuss various gas adsorption processes. Nanomaterials 2020, 10, x FOR PEER REVIEW  3 of 34 potential TMDCs, since the year 2018. Figure 2 outlines the schematic diagrams of each gas adsorption process. In general, the mechanism of these adsorption processes is mainly driven by the charge transfer process between the adsorbates and the TMDCs-based adsorbent, where the charge movement is dependent on the nature of the gas adsorbate (i.e., oxidizing or reducing) [46]. For instance, carbon monoxide (CO) in Figure 2a is an example of reducing gas. Due to the existence of lone pair on the carbon atom, CO will donate electrons to the TMDC surface, which further cause the CO to be chemisorbed on the surface. Whereas, oxidizing gas such as nitrogen dioxide (NO2) in Figure  2c, will uptake the electron from the surface instead (mainly due to the existence of unpaired electron on the nitrogen atom). To note, such electron movement will lead to the deviation in electrical conductivity of the TMDC materials [47]. The following subsections discuss various gas adsorption processes.   Nanomaterials 2020, 10, x FOR PEER REVIEW  3 of 34 potential TMDCs, since the year 2018. Figure 2 outlines the schematic diagrams of each gas adsorption process. In general, the mechanism of these adsorption processes is mainly driven by the charge transfer process between the adsorbates and the TMDCs-based adsorbent, where the charge movement is dependent on the nature of the gas adsorbate (i.e., oxidizing or reducing) [46]. For instance, carbon monoxide (CO) in Figure 2a is an example of reducing gas. Due to the existence of lone pair on the carbon atom, CO will donate electrons to the TMDC surface, which further cause the CO to be chemisorbed on the surface. Whereas, oxidizing gas such as nitrogen dioxide (NO2) in Figure  2c, will uptake the electron from the surface instead (mainly due to the existence of unpaired electron on the nitrogen atom). To note, such electron movement will lead to the deviation in electrical conductivity of the TMDC materials [47]. The following subsections discuss various gas adsorption processes.

Adsorption of Carbon Monoxide (CO)
Carbon monoxide (CO) is by far the most hazardous greenhouse gases, which is 210 times easier to bind with hemoglobin as compared to the oxygen [49]. To date, numerous studies have discovered the use of TMDCs as a catalyst to adsorb CO and further convert it into other products via catalytic reaction. For instance, Li et al. [50] proposed the use of aluminum oxide-doped MoS 2 as the nanocatalysts to promote the CO methanation process (CO + 3H 2 → CH 4 + H 2 O) and enhance the stability of this catalytic reaction (MoS 2 with 25.6% of Al 2 O 3 provides the greatest methanation stability). The schematic diagram of this process is shown in Figure 2a. The incorporation of metal-based promoters, in this case, Al 2 O 3 powder has effectively reduced the aggregation effect between MoS 2 (lower tendency to pore blockage) [50]. Aside from that, researchers also adopted the density functional theory (DFT) calculation to explore the theoretical potential of each TMDC as the alternative catalyst for CO oxidation (CO + O 2 → CO 2 + O) and CO dissociation. For instances, studies showed that metal-doped MoS 2 (e.g., Au 29 , Cu, Ag, Co, Rh, Ni, Ir, and Fe [51][52][53]) can significantly enhance the O 2 dissociation (which is the essential step for CO oxidation). On the other hand, the latter five mentioned metals possess the greatest potential as they favor CO dissociation [52]. More recently, application of other TMDCs (e.g., Pt and Au nanoclusters with WSe 2 , where the carbon atom in CO will be strongly adsorbed on the Pt/Au decorated WSe 2 monolayer [54]; Rh-doped MoSe 2 [55]), in CO adsorption have also been carried out. Since CO dissociation is the first step of the Fischer-Tropsch process [56], the aforementioned TMDC-based nanocatalysts can be served as the substituent cheap and stable catalyst to convert CO into other valuable liquid hydrocarbons.

Adsorption of Water Vapor (H 2 O)
Water vapor (H 2 O) is the largest contributor to the current global warming issue (i.e., roughly accounted for 60% of the entire warming effect [57]). Numerous works have proposed the use of TMDC-based nanocatalysts to convert H 2 O into clean hydrogen (H 2 ) via hydrogen evolution reaction (HER) (see Figure 2b). In general, hydrogen atoms in H 2 O will be adsorbed on the active site of the HER nanocatalysts. With the aid of the doped metal, the dissociation of the O-H bond in H 2 O is enhanced. The gas is then desorbed as H 2 [58,59]. This can mitigate the warming effect attributed to the water vapor content in the atmosphere, and at the same time, serves as an alternative greenway for hydrogen production. The past two decades ago, MoS 2 was referred to as one of the most prominent alternatives to substitute the conventional HER catalysts [60]. Nevertheless, the commercialization of TMDC-based HER nanocatalysts is still hindered due to various technical challenges (e.g., poor HER stability under acidic conditions [61,62] and weak intrinsic conductivity [63]). Numerous works have discovered ways to enhance the HER performance of MoS 2 . Just to name a few, these works proposed to improve the HER activity by (i) coating the MoS 2 with metals (e.g., palladium [64], aluminum [65], gold [66], platinum [67], etc.); (ii) using a novel electrochemical approach to deposit the MoS 2 nanosheet [68]; (iii) introducing the use promising support for the MoS 2 nanoparticles (e.g., multi-wall carbon nanotubes [69]); (iv) varying the choice of dispersion media used [70]; and (v) coupling with other electrocatalysts (e.g., molybdenum carbide [71]). On the other hand, some researchers also proposed the use of other TMDCs in HER applications. For instance, Wang et al. [72] proposed to use a low-cost MoSe 2 nanosheet, which offers decent active sites for HER activity and better HER performance as compared to the aforementioned MoS 2 -based catalysts. On the other hand, Seok et al. [73] have reported the intrinsic activity for HER on MoTe 2 nanosheet via the first-principle calculation and scanning tunneling microscopy (STM) study. Besides, numerous researchers also successfully developed promising WS 2 -based and WSe 2 -based electrocatalysts recently (e.g., Te-WS 2 nanosheet by Pan et al. [74]; and Ni-WSe 2 nanosheet by Kadam et al. [75]).

Adsorption of Carbon Dioxide (CO 2 )
Aside from water vapor, carbon dioxide (CO 2 ) is another key contributor to climate change issues (accounting for more than 75% of the total emitted greenhouse gases [76]). To address this issue, some researchers have attempted to apply MoS 2 nanosheets as an effective membrane [77] and adsorbent [78,79] to separate CO 2 from the gaseous mixture. Sun et al. [78] reported that the adsorption force between CO 2 and MoS 2 varies according to the strength of the applied electric field. Generally, CO 2 will adsorb on the surface when an electric field is applied and desorb when the electric field is relieved (see Figure 2d). This unique feature makes it become a potential carbon capture media. In addition to the carbon capture process, TMDCs can be modified so that the captured CO 2 can be catalytically converted into other CO 2 -reduction products. Shi et al. [80] were the first to discover the potential of copper modified MoS 2 nanosheets in the CO 2 reduction process. The incorporation of Cu nanoparticles has enhanced the adsorption capability as CO 2 can now be adsorbed not only on the MoS 2 nanosheets but also on the doped metal nanoparticles [80] (see Figure 2d). This work found that most of the CO 2 was converted into CO (i.e., faradaic efficiency (FE) = 33-41%), followed by methane (FE = 7-17%) and with a trace amount of ethylene (FE = 2-3.5%) [80]. More recently, a research project in Korea has discovered the potential of a composite catalyst that encompasses MoS 2 nanosheets and n-type Bi 2 S 3 , in the CO 2 -to-CO photoreduction process [81]. DFT calculations of various modified-TMDC nanosheets (e.g., SnO 2 -loaded MoS 2 nanosheet [82]; TiO 2 -doped MoS 2 nanosheets [83]) for the CO 2 reduction process were also conducted recently. Nevertheless, the research on the potential of other TMDCs in CO 2 adsorption is still considered scarce. To date, only a few have discovered their potential for the CO 2 sensor [84,85].

Sulphur Content Removal
Based on the air quality data observed by the NASA Ozone Monitoring Instrument (OMI) satellite, about 60% of the total sulphur emissions in 2018 were anthropogenic emissions [86]. Despite the recent advancement in renewable energy and air treatment technologies have reduced the global sulphur emissions (sulphur emissions in 2015 is about 30% less as compared to that of 1990 [87]), the health impact of sulphur content in the atmosphere can still be severe. Thus, the sulphur emissions (including SO 2 , H 2 S, etc.) must still be monitored and controlled closely. Wei et al. [56], based on the DFT simulation, have explored the adsorption performance of SO 2 and H 2 S on the Ni-doped MoS 2 monolayer material. To note, due to the large number of free electrons offered by the doped metals, the overall adsorption capability of the TMDC-based adsorbents are, therefore, gradually enhanced [88]. This study is then extended by other researchers, by testing the effect of other doped metal atoms (e.g., palladium [89], platinum [90], gold [89], and copper [91]). All these studies show that the metal-doped MoS 2 monolayer materials can offer promising adsorption properties to SO 2 and H 2 S gases, where both gases are chemisorbed on the surface with strong interactive forces. Similarly, various DFT studies have also been conducted to study the potential of using metal-doped MoSe 2 nanosheets (e.g., rhodium-doped MoSe 2 monolayer material [55] and palladium-doped MoSe 2 monolayer material [92]) and WSe 2 nanosheets (e.g., N-doped TiO 2 /WSe 2 nanocomposite [93] and silver-doped WSe 2 monolayer material [94]) in sulphur content removal applications. Dan et al. [95], on the other hand, studied the photocatalytic H 2 S splitting (H 2 S → H 2 + S) on the proposed novel composite catalysts, MnS/In 2 S 3 -MoS 2 based on first-principles calculations. This is an attractive and cost-effective way to generate clean H 2 that can substitute conventional coal-derived hydrogen production [96]. In other words, this also indirectly contributes to pollution mitigation. The schematic diagram for the aforementioned sulphur removal process is shown in Figure 2e, where the desorption process can be conducted by altering the operating conditions (e.g., temperature and pressure).

Nitrogen Oxide (NO x ) Removal
Nitrogen oxide (NO x ) has adverse effects on human health and ecosystems. To date, numerous works have revealed the potential of applying TMDCs for NO x removal purposes. Notably, a novel Cu 2 O-anchored MoS 2 nanocomposite developed by Yuan et al. [97], can achieve up to 53% NO x removal. In another research conducted in China, a similar NO removal rate (about 51%) can also be obtained using the MoS 2 -g-C 3 N 4 nanocomposite [98]. In the same year, Xiong et al. [99] managed to develop a novel (BiO) 2 CO 3 /MoS 2 photocatalysts that could achieve 50% NO removal for five consecutive cycles. The study was extended by incorporating the use of carbon nanofibers into nanocomposites fabrication [100]. The developed Bi 2 O 2 CO 3 -MoS 2 -CNFs nanocomposites can attain almost 70% of NO removal rate. Under visible light irradiation, the oxidative power offered by the nanocomposites is sufficient to oxidize the NO x into NO 3− (see Figure 2c). Besides, first-principle studies were also conducted to determine the metal-doping effects on the adsorption capabilities on MoS 2 (e.g., (vanadium, niobium, tantalum)-doped MoS 2 [101]), MoSe 2 (e.g., palladium-doped MoSe 2 [92] and rhodium-doped MoSe 2 [55]), and WSe 2 [102] monolayer materials to NO x gases.

Gas Sensing for Pollution Reduction
Gas pollution is commonly caused by direct greenhouse gases (CO, CO 2 , N 2 O, CH 4 , Fluorinated Gas, etc.) [103], indirect greenhouse gases (NH 3 , NO x , H 2 S, etc.) [104], and other traces of toxic gases. These gas pollutants may cause climate change, ozone pollution, and even threaten food security [105] Further adverse effects that are caused by these problems include reduced photosynthesis performances [106], increased chances of respiratory problems [107], and acid rain [108].
Gas sensing can be deployed in multiple locations to act as an environmental monitoring system [109]. More recent applications of gas sensing products include gas monitoring using drones [110], wearable gas sensing devices [111], and the internet of things (IoT) multi-gas sensing modules [112]. For gas sensing TMDC-material in the device, the most common working principle is for adsorption of gas particles towards the TMDCs for a charge transfer, giving a change of resistivity in material, then desorption from the TMDCs [113]. In this field, researchers are constantly searching for TMDC materials that can provide a very low limit of detection (LOD), high sensitivity, short response and recovery time [114][115][116].

NO x Detection
NO x detection by TMDCs has received much research interest due to its effectiveness and applicability. The measurement of NO x in environmental monitoring is a common requirement [117] while the design of a perfect NO x sensor poses some challenges. Traditional SnO 2 sensors have problems with the dual response towards oxidizing gases and reducing gases, giving low sensitivity when performing measurements in a multi-gas environment [118]. The selectivity of the gas between NO 2 and NO is also a crucial challenge for NO x sensors [119].
A popular TMDC material for NO 2 detection is MoS 2 [113,120,121]. Earlier works [121] used CVD-grown MoS 2 to detect NO 2 gas at room temperature with LOD down to parts per billion (ppb) ranges, however, the recovery time was long. Consequently, a higher temperature was used to accelerate the desorption kinetics of the NO 2 gas molecule [113,120]. Other researchers also created nanocomposites of TMDCs with materials such as graphene aerogel [122] and reduced graphene oxide (rGO) [123] to improve sensitivity and lower LOD. Pham et al. [115] used a red-light illumination to provide photon energy, which matches the bandgap of the MoSe 2 sensor. The technique resulted in low LOD (25 ppb) in room temperature conditions with high sensitivity. Moreover, Liu et al. [114] synthesized a flower-like porous SnS 2 NSs with edge exposed MoS 2 nanosphere, which resulted in a very fast response time of 2 seconds. Group-10 noble TMDCs such as PtSe 2 has also received much research attention due to their widely tunable bandgap and excellent performance in gas sensing [124]. From a first-principle study, Sajjad et al. [125] discovered that monolayer PtSe 2 exhibited Nanomaterials 2020, 10, 1012 7 of 33 lower adsorption energy than MoS 2 and graphene. For this, Yim et al. [126] used thin films of PtSe 2 stacked on Si to achieve NO 2 detection down to 9 ppb. Other TMDC materials such as WS 2 [122] and MoTe 2 [127] also showed that they have excellent NO 2 sensing properties, and more research work is to be expected for different transition metals. A collection of NO 2 sensing TMDC materials is tabulated in Table 1. Although NO and NO 2 are sometimes measured as total NO x , many academic studies require differentiation between NO from NO 2 [133,134]. A distinct property between the two gases is that NO is a colorless gas while NO 2 is reddish-brown in color [135]. Due to its reduced electron valence in NO, first-principle studies demonstrated that the adsorption energy of NO to be slightly higher than that of NO 2 [136]. Nevertheless, the binding of NO with MoS 2 is still considered one of the strongest in many gas molecules including CO, CO 2 , NH 3 , CH 4 , H 2 O, N 2 , O 2 , and SO 2 [136]. This computational finding was verified by a few experiments that successfully detected NO with MoS 2 materials down to the ppm level [137][138][139]. Although there is not much research work that focuses on NO detection, it is demonstrated to give promising performances using TMDC material (see Table 2).

Ammonia Detection
Ammonia is an indirect greenhouse gas, as it is quickly soluble by water vapor or rain in the atmosphere and goes down to the land to be readily used as fertilizers by plants [141]. The lifespan Nanomaterials 2020, 10, 1012 8 of 33 of this alkaline and reactive gas in the atmosphere is short, around 1 day [142]. Nevertheless, direct exposure to ammonia gas from its emission source can still cause health issues to humans [143]. Gases such as NO 2 are electron-accepting, while NH 3 donates electrons to the TMDCs surface [113] to give a change of resistivity in the material (see Figure 3a,b). Due to their different adsorption mechanism, NO 2 and NH 3 are two flagship gases to study the generic effect of gas sensing [113,144,145].

Ammonia Detection
Ammonia is an indirect greenhouse gas, as it is quickly soluble by water vapor or rain in the atmosphere and goes down to the land to be readily used as fertilizers by plants [141]. The lifespan of this alkaline and reactive gas in the atmosphere is short, around 1 day [142]. Nevertheless, direct exposure to ammonia gas from its emission source can still cause health issues to humans [143]. Gases such as NO2 are electron-accepting, while NH3 donates electrons to the TMDCs surface [113] to give a change of resistivity in the material (see Figure 3a,b). Due to their different adsorption mechanism, NO2 and NH3 are two flagship gases to study the generic effect of gas sensing [113,144,145]. Many works have also indicated that TMDC materials have excellent properties to act as excellent ammonia gas sensors [145]. For example, Cho et al. [146] used a 2D graphene/MoS2 heterogeneous structure to improve the sensitivity of the sensor towards gas molecules. Burman et al. [139] studied the effect of vacancy sites on MoS2 sensors and proposed a UV-treated sensor that can be synthesized from powder. Other TMDC materials such as MoTe2 [147] and WS2 [148] also showed performances of being able to detect ammonia gases down to ppm levels. Even artificial neural networks were used to post-process signals from TMDC ammonia gas sensors to improve gas concentration predictions [144]. A few of the significant works of applying TMDC materials as ammonia gas sensors are found in Table 3.  Many works have also indicated that TMDC materials have excellent properties to act as excellent ammonia gas sensors [145]. For example, Cho et al. [146] used a 2D graphene/MoS 2 heterogeneous structure to improve the sensitivity of the sensor towards gas molecules. Burman et al. [139] studied the effect of vacancy sites on MoS 2 sensors and proposed a UV-treated sensor that can be synthesized from powder. Other TMDC materials such as MoTe 2 [147] and WS 2 [148] also showed performances of being able to detect ammonia gases down to ppm levels. Even artificial neural networks were used to post-process signals from TMDC ammonia gas sensors to improve gas concentration predictions [144]. A few of the significant works of applying TMDC materials as ammonia gas sensors are found in Table 3.

Volatile Organic Compound (VOC) Detection
Volatile organic compounds (VOC), commonly known as aromatic hydrocarbons fractions, are organic chemicals that have high vapor pressure in atmospheric conditions. VOC are studied to be Nanomaterials 2020, 10, 1012 9 of 33 emitted in landfills [153], newly renovated buildings [154], industrial refineries [155], and other leakage sources. Long-term exposure to VOC can lead to dysfunction of central nervous systems, memory loss, and cause congenital anomalies for reproduction [156]. VOC sensors are important for environmental detection [154], industrial emission control [157], and even for cancer diagnosis [158].
The challenge of designing sensors for VOC is that VOC is not a single gas molecule but can consist of multiple molecules (such as benzene, ethylene glycol, formaldehyde, etc.) with slightly different properties. Barzegar et al. [159] represented the VOC molecule behavior by using xylene isomers and methanol, showing high potential for adsorption between VOC molecules and a Ni-decorated MoS 2 sensor. A thiolated ligand conjugate MoS 2 sensor was also shown to exhibit excellent sensitivity down to a concentration of 1 ppm [160]. The work represented VOC by studying a combination of toluene, hexane, ethanol, propanal, and acetone gas molecules while demonstrating the conjugation of thiolated ligand improves the charge carrier density of the sensor (see Figure 4). A recent work from Tomer et al. [161] demonstrated that a cubic Ag(0)-MoS 2 loaded g-CN 3D porous hybrid had multifunctional abilities towards sensing VOC by studying n-butanol, isopropanol, benzene, and xylene gas molecules. Short response time and recovery time of 7-15 s and 6-9.5 s respectively were obtained for all VOC molecules at 175 • C. Zhao et al. [162] presented an optical VOC microsensor using a photonic crystal cavity integrated with MoS 2 . VOC molecules of acetone, methanol, ester, dichloromethane, and methylbenzene were studied and the optical microsensor resulted in a LOD of 2.7 ppm, response time of 0.3 s, and recovery time of 100 s.

Volatile Organic Compound (VOC) detection
Volatile organic compounds (VOC), commonly known as aromatic hydrocarbons fractions, are organic chemicals that have high vapor pressure in atmospheric conditions. VOC are studied to be emitted in landfills [153], newly renovated buildings [154], industrial refineries [155], and other leakage sources. Long-term exposure to VOC can lead to dysfunction of central nervous systems, memory loss, and cause congenital anomalies for reproduction [156]. VOC sensors are important for environmental detection [154] , industrial emission control [157], and even for cancer diagnosis [158].
The challenge of designing sensors for VOC is that VOC is not a single gas molecule but can consist of multiple molecules (such as benzene, ethylene glycol, formaldehyde, etc.) with slightly different properties. Barzegar et al. [159] represented the VOC molecule behavior by using xylene isomers and methanol, showing high potential for adsorption between VOC molecules and a Nidecorated MoS2 sensor. A thiolated ligand conjugate MoS2 sensor was also shown to exhibit excellent sensitivity down to a concentration of 1 ppm [160]. The work represented VOC by studying a combination of toluene, hexane, ethanol, propanal, and acetone gas molecules while demonstrating the conjugation of thiolated ligand improves the charge carrier density of the sensor (see Figure 4). A recent work from Tomer et al. [161] demonstrated that a cubic Ag(0)-MoS2 loaded g-CN 3D porous hybrid had multifunctional abilities towards sensing VOC by studying n-butanol, isopropanol, benzene, and xylene gas molecules. Short response time and recovery time of 7-15 s and 6-9.5 s respectively were obtained for all VOC molecules at 175 °C. Zhao et al. [162] presented an optical VOC microsensor using a photonic crystal cavity integrated with MoS2. VOC molecules of acetone, methanol, ester, dichloromethane, and methylbenzene were studied and the optical microsensor resulted in a LOD of 2.7 ppm, response time of 0.3 s, and recovery time of 100 s.

Detection of Sulphur Gases and Other Gases
The unique properties of TMDC materials also showed many preliminary potentials for the detection of sulphur gases and other emissions. Chen et al. [163] demonstrated from density functional theory (DFT) that MoS 2 monolayers exhibit great adsorption energy towards SF 6 decomposition (which includes SO 2 , SOF 2 , SO 2 F 2 , H 2 S, and HF). The adsorption of SF 6 decomposition by monolayer PtSe 2 was also studied by DFT methods and found to give excellent adsorption energies [164]. Park et al. [165] experimentally demonstrated that a Pt nanoparticle decorated MoS 2 gas sensor can achieve high sensitivity detection of H 2 S down to 30 ppm. Recent work from Yang et al. [166] demonstrated that Ni-and Cu-embedded MoS 2 monolayers can improve the adsorption energy of SO x and O 3 molecules compared to that of the pristine MoS 2 .
TMDCs also demonstrate adsorption and sensing ability for CO gas. The work of Ma et al. [167] studied the potential of sensing CO and NO gas molecules by doping metal particles on MoS 2 monolayer. The doping of Au, Pt, Pd, and Ni has shown to alter the transport property of MoS 2 , giving better adsorption energy for CO and NO gas detection. Recently, Yang et al. [91] presented a strategy of using Ti doping and the application of an electric field to improve adsorption energies in MoSe 2 and MoS 2 materials. The work concluded that Ti doping strategy improved the sensitivity of both materials towards CO and NO gas detection. Another recent work from Shen et al. [168] provided a study of using a borophene/MoS 2 heterostructure to detect small gas molecules (CO, NO, NO 2 , and NH 3 ), which exhibited different resistive properties towards these molecules when changing voltage direction (see Figure 5). Modulation of MoS 2 by antisite doping and strain also showed improved gas detection sensitivity [169]. Ma et al. [170] studied the effects of defects on WSe 2 monolayer sensor and found that Se vacancies can improve sensing ability of H 2 O and N 2 molecules. Further interest gas detection also extends towards Cl 2 , PH 3 , AsH 3 , BBr 3 , and SF 4 gas molecules on MoS 2 [171]. More materials will be expected to be discovered for gas detection applications in the future.
detection of sulphur gases and other emissions. Chen et al. [163] demonstrated from density functional theory (DFT) that MoS2 monolayers exhibit great adsorption energy towards SF6 decomposition (which includes SO2, SOF2, SO2F2, H2S, and HF). The adsorption of SF6 decomposition by monolayer PtSe2 was also studied by DFT methods and found to give excellent adsorption energies [164]. Park et al. [165] experimentally demonstrated that a Pt nanoparticle decorated MoS2 gas sensor can achieve high sensitivity detection of H2S down to 30 ppm. Recent work from Yang et al. [166] demonstrated that Ni-and Cu-embedded MoS2 monolayers can improve the adsorption energy of SOx and O3 molecules compared to that of the pristine MoS2.
TMDCs also demonstrate adsorption and sensing ability for CO gas. The work of Ma et al. [167] studied the potential of sensing CO and NO gas molecules by doping metal particles on MoS2 monolayer. The doping of Au, Pt, Pd, and Ni has shown to alter the transport property of MoS2, giving better adsorption energy for CO and NO gas detection. Recently, Yang et al. [91] presented a strategy of using Ti doping and the application of an electric field to improve adsorption energies in MoSe2 and MoS2 materials. The work concluded that Ti doping strategy improved the sensitivity of both materials towards CO and NO gas detection. Another recent work from Shen et al. [168] provided a study of using a borophene/MoS2 heterostructure to detect small gas molecules (CO, NO, NO2, and NH3), which exhibited different resistive properties towards these molecules when changing voltage direction (see Figure 5). Modulation of MoS2 by antisite doping and strain also showed improved gas detection sensitivity [169]. Ma et al. [170] studied the effects of defects on WSe2 monolayer sensor and found that Se vacancies can improve sensing ability of H2O and N2 molecules. Further interest gas detection also extends towards Cl2, PH3, AsH3, BBr3, and SF4 gas molecules on MoS2 [171]. More materials will be expected to be discovered for gas detection applications in the future.

TMDC Materials for Wastewater Treatment
The United Nations has highlighted that 80% of wastewater is released back to the environment without sufficient treatment [172]. The effect of industrial development and human activities has released many pollutants into the water. The major water pollution sources come from various source such as industry waste [173], sewage and wastewater [174], oil leakage [175], chemical fertilizer and

TMDC Materials for Wastewater Treatment
The United Nations has highlighted that 80% of wastewater is released back to the environment without sufficient treatment [172]. The effect of industrial development and human activities has released many pollutants into the water. The major water pollution sources come from various source such as industry waste [173], sewage and wastewater [174], oil leakage [175], chemical fertilizer and pesticides [176], and mining activities [177]. These pollutants require high oxygen for oxidation decomposition, which reduces the dissolved oxygen level in the water that will damage the aquatic ecosystem [178]. Wang et al. [179] reviewed that many technologies including adsorption, ion exchange, membrane filtration, chemical precipitation, and electrochemistry have been widely used in water treatment. Despite the application of various technologies, quality water supply remains unsustainable [180]. Many researchers have reported promising outcomes with the application of nanomaterial such as TMDC-based material in water treatment technology [181].

Adsorption for Wastewater Treatment
Among wastewater treatment technologies, the adsorption is considered the most inexpensive, fast and simple operation method [182]. Recently, researchers are focusing on the application of TMDC-based material in adsorption technology. That MoS 2 nanosheet is used in the adsorption process mainly due to its large surface area, excellent chemical and thermal stability and environmentally friendly [183]. Many application of TMDC-based material has been successfully demonstrated with an effective outcome including the removal of methyl orange [184], Rhodamine B (RhB) [185], and Congo Red [186]. Li et al. [187] stated that the application of MoS 2 adsorbent can increase the adsorption capacity in removing organic dye. An alternative facile oxidation strategy was proposed by Li et al. [182] to synthesize tungsten disulphide/tungsten trioxide (WS 2 /WO 3 ) heterostructures to remove RhB molecules from wastewater (see Figure 6a). This strategy has contributed to higher adsorption capacity through manipulating the surface property. Massey et al. [188] further added that the hierarchical microsphere of MoS 2 nanosheets has demonstrated high adsorption capacity with 297, 216, and 183 mg/g of methylene blue, rhodamine 6G, and fuchsin acid dye respectively as compare to conventional absorbent such as activated carbon. Most importantly, the hierarchical microsphere of MoS 2 nanosheets can be regenerated effectively without affecting the performance of adsorption capacity. decomposition, which reduces the dissolved oxygen level in the water that will damage the aquatic ecosystem [178]. Wang et al. [179] reviewed that many technologies including adsorption, ion exchange, membrane filtration, chemical precipitation, and electrochemistry have been widely used in water treatment. Despite the application of various technologies, quality water supply remains unsustainable [180]. Many researchers have reported promising outcomes with the application of nanomaterial such as TMDC-based material in water treatment technology [181].

Adsorption for Wastewater Treatment
Among wastewater treatment technologies, the adsorption is considered the most inexpensive, fast and simple operation method [182]. Recently, researchers are focusing on the application of TMDC-based material in adsorption technology. That MoS2 nanosheet is used in the adsorption process mainly due to its large surface area, excellent chemical and thermal stability and environmentally friendly [183]. Many application of TMDC-based material has been successfully demonstrated with an effective outcome including the removal of methyl orange [184], Rhodamine B (RhB) [185], and Congo Red [186]. Li et al. [187] stated that the application of MoS2 adsorbent can increase the adsorption capacity in removing organic dye. An alternative facile oxidation strategy was proposed by Li et al. [182] to synthesize tungsten disulphide/tungsten trioxide (WS2/WO3) heterostructures to remove RhB molecules from wastewater (see Figure 6a). This strategy has contributed to higher adsorption capacity through manipulating the surface property. Massey et al. [188] further added that the hierarchical microsphere of MoS2 nanosheets has demonstrated high adsorption capacity with 297, 216, and 183 mg/g of methylene blue, rhodamine 6G, and fuchsin acid dye respectively as compare to conventional absorbent such as activated carbon. Most importantly, the hierarchical microsphere of MoS2 nanosheets can be regenerated effectively without affecting the performance of adsorption capacity.  Kumar et al. [189] concluded that the synthesis of MoS 2 with magnetic nanoparticles can be effective adsorption material especially in removing heavy metal such as chromium. The efficiency of chromium removal is highly dependent on the pH of the solution and the adsorption of Cr(VII) and Cr(III) can be selected by changing the pH (see Figure 6b). Besides that, the regeneration of absorbent does not reduce the efficiency of chromium uptake significantly.
On top of that, the CeO 2 -MoS 2 hybrid magnetic biochar (CMMB) also exhibits a strong magnetic ability to remove lead(II) (Pb(II)) and humate from water treatment [191]. The CMMB can remove >99% of Pb(II) and humate within 6 h. The application of MoS 2 /thiol-functionalized multiwalled carbon nanotube (SH-MWCNT) has also displayed high adsorption capacity for heavy metal removal [177]. However, the spent adsorbent can be further used for photocatalytic and electrochemical applications.
Furthermore, the synthesized MoS 2 -coated melamine-formaldehyde (MF@MoS 2 ) sponges that exhibit a superhydrophobic and superhydrophilic characteristic (see Figure 6c) demonstrated a highly selective adsorption capacity [190]. The MF@MoS 2 sponges have high absorption performance for oil and organic solvent and water-soluble dye. It also exhibits high discoloration efficiency of 98% methyl orange within 10 min. Wan et al. [192] added that the MF@MoS 2 can be modified from room temperature vulcanized silicon rubber.

Membrane Technology in Wastewater Treatment
In water treatment systems, the typical microporous membrane pore size is about 0.1-5 µm, which limits membrane technology on water purification [179]. The development of nanoporous membranes exhibited high filtering performance in dealing with most of the pollutants inducing microbes, organic molecules, heavy metal, and salts.
Heiranian et al. [193] highlighted that the single-layer MoS 2 membrane with a full Mo pore (see Figure 7a) exhibited 88% of ion rejection. The work also demonstrated that the water flux was two to five orders of magnitude greater than other known nanoporous membranes. This technology is crucial to replacing the reverse osmosis (RO) membrane, especially in water desalination processes. Later, Kou et al. [194] discussed that a 0.74 nm nanopores in monolayer MoS 2 membranes should be compatible with Debye screening length for the electrostatic interaction and lesser than the mean free path of molecules in water. Recent work from Kozubek et al. [195] proposed using irradiation with highly charged ions (HCIs) to create pores in MoS 2 . They found that pore creation efficiency has a linear relationship with potential energies for pore radius of 0.55-2.65 nm. Kozubek et al. [195] added that the HCI method has parallel writing capabilities, giving a high potential for mass production. In the application of water desalination, Ma et al. [196] stated that the synthesis of MoS2/GO membrane resulted in an enhanced water flux (from 8.83 to 48.27 L·m −2 ·h −1 ) with improved salts removal capacity (from 54.32% to 96.85%). The MoS2/GO membrane also exhibited different mechanism for dyes and ions (see Figure 7b) where dyes are physically sieved while ions have a  [193]. Copyright 2015 Springer Nature Limited CC BY 4.0). (b) MoS 2 /GO nanosheets intercalated membrane for removal of dyes and salts (Reproduced with permission from [196]. Copyright 2020 Elsevier B.V.).
In the application of water desalination, Ma et al. [196] stated that the synthesis of MoS 2 /GO membrane resulted in an enhanced water flux (from 8.83 to 48.27 L·m −2 ·h −1 ) with improved salts removal capacity (from 54.32% to 96.85%). The MoS 2 /GO membrane also exhibited different mechanism for dyes and ions (see Figure 7b) where dyes are physically sieved while ions have a Donnan effect due to existing trapped ions within interlayer spacing [196]. Gao et al. [197] demonstrated the performance of MoS 2 membrane's performance in rejecting 100% of methylene blue dye. The work highlighted that the increase in nanosheets to the membrane will reduce permeability and increase dye rejection rate.

Photocatalyst Technology in Wastewater Treatment
As the majority of water treatment facilities are located outdoors, the availability of solar energy can be useful for photocatalytic technology in water treatment. Chu et al. [198] reported that the development of hexagonal 2H-MoSe 2 photocatalyst exhibits outstanding photo-absorption and photocatalytic reaction in reducing hexavalent chromium (Cr(VI)) under ultraviolet (UV) light. The 2H-MoSe 2 yields 99% Cr(VI) reduction rate with the presence of UV light. Mittal and Khanuja [199] revealed that MoSe 2 is a good photocatalyst not only for Cr(VI) but also for methylene blue (MB) and RhB. The MoSe 2 /strontium titanate (SrTiO 3 ) heterostructure also showed great potential as a wastewater treatment photocatalyst with a degradation rate of methyl orange at 99.46% under the optimum loading weight of 0.1 wt.% under UV light [200]. It is also found that there is no significant loss in the performance of MoSe 2 /SrTiO 3 after reuse for 6 times. The working principle of MoSe 2 /SrTiO 3 photocatalyst is demonstrated in Figure 8.  [193]. Copyright 2015 Springer Nature Limited CC BY 4.0). (b) MoS2/GO nanosheets intercalated membrane for removal of dyes and salts (Reproduced with permission from [196]. Copyright 2020 Elsevier B.V.).
In the application of water desalination, Ma et al. [196] stated that the synthesis of MoS2/GO membrane resulted in an enhanced water flux (from 8.83 to 48.27 L·m −2 •h −1 ) with improved salts removal capacity (from 54.32% to 96.85%). The MoS2/GO membrane also exhibited different mechanism for dyes and ions (see Figure 7b) where dyes are physically sieved while ions have a Donnan effect due to existing trapped ions within interlayer spacing [196]. Gao et al. [197] demonstrated the performance of MoS2 membrane's performance in rejecting 100% of methylene blue dye. The work highlighted that the increase in nanosheets to the membrane will reduce permeability and increase dye rejection rate.

Photocatalyst Technology in Wastewater Treatment
As the majority of water treatment facilities are located outdoors, the availability of solar energy can be useful for photocatalytic technology in water treatment. Chu et al. [198] reported that the development of hexagonal 2H-MoSe2 photocatalyst exhibits outstanding photo-absorption and photocatalytic reaction in reducing hexavalent chromium (Cr(VI)) under ultraviolet (UV) light. The 2H-MoSe2 yields 99% Cr(VI) reduction rate with the presence of UV light. Mittal and Khanuja [199] revealed that MoSe2 is a good photocatalyst not only for Cr(VI) but also for methylene blue (MB) and RhB. The MoSe2/strontium titanate (SrTiO3) heterostructure also showed great potential as a wastewater treatment photocatalyst with a degradation rate of methyl orange at 99.46% under the optimum loading weight of 0.1 wt.% under UV light [200]. It is also found that there is no significant loss in the performance of MoSe2/SrTiO3 after reuse for 6 times. The working principle of MoSe2/SrTiO3 photocatalyst is demonstrated in Figure 8.

Fuel Cleaning
Fuels are means of transfer for energy in our ecosystems. Most fuels have many imperfections related to conversion inefficiencies [201] and fuel impurity (including sulphur, nitrogen, aromatic, etc.) [202]. Direct usage of crude fuels with such impurity content will lead to an increase in gas emissions [203,204]. Most refineries for oil-based fuel carry out fuel cleaning processes such as desulphurization, dearomatization, denitrogenation, and deoxygenation [202,205,206] to remove such impurities before the fuel is being converted. The use of impurity removal technology generally depends on the techno-economics and fuel costs during application [207]). Fuels with a high impurity

Fuel Cleaning
Fuels are means of transfer for energy in our ecosystems. Most fuels have many imperfections related to conversion inefficiencies [201] and fuel impurity (including sulphur, nitrogen, aromatic, etc.) [202]. Direct usage of crude fuels with such impurity content will lead to an increase in gas emissions [203,204]. Most refineries for oil-based fuel carry out fuel cleaning processes such as desulphurization, dearomatization, denitrogenation, and deoxygenation [202,205,206] to remove such impurities before the fuel is being converted. The use of impurity removal technology generally depends on the techno-economics and fuel costs during application [207]). Fuels with a high impurity such as sulphur content can potentially cause an elevated particulate emission [208] while causing catalyst poisoning in subsequent systems [209].

Fuel Hydrodesulfurization
Hydrodesulfurization is the process of reduction of sulphur content from fuel such as diesel [202] by catalytically reacting them with hydrogen. Paul et al. [210] studied the mechanism of vacancy formation on the MoS 2 catalyst in the hydrodesulfurization process, demonstrating the activation energy of 0.5 eV.
TMDC materials such as Co/Ni promoted MoS 2 were popular for the use of hydrodesulfurization due to its high activity, good selectivity, resistance to deactivation, and regeneration ability [211]. Commonly, TMDC-based catalysts are used for hydrodesulfurization in the oil and gas industry for high-efficiency sulphur removal from naphtha [212]. For the hydrodesulfurization of naphtha, a recent work from Mahmoudabadi et al. [213] achieved 100% conversion efficiency using MoS 2 quantum dots nanocatalyst under the pressure of 15 bars, temperature of 280 • C and liquid hourly space velocity (LHSV) of 4 h −1 . The direct use of unsupported Ni/MoS 2 catalyst was reported [214] to remove furfurylamine (FA) and dibenzothiophene (DBT). Liu et al. [215] demonstrated that for a MoS 2 /NiMo catalyst, the higher the number of MoS 2 slabs being stacked, the better the hydrodesulfurization selectivity for DBT. Additionally, Rangarajan et al. [216] studied the preferred active sites of a metal-promoted MoS 2 and found that organosulfur and organonitrogen compounds bind weaker on sites with exposed metal on the corner and the sulphur edge of MoS 2 . Hydrodesulfurization of DBT by various morphologies of MoS 2 catalyst was also studied by Tye and Smith [217]. This work showed that exfoliated MoS 2 has the highest conversion, compared to the commercially crystalline MoS 2 powder and MoS 2 derived from soluble Mo precursors. While selectivity was correlated to the edge sites of MoS 2 . A 3D NiS-MoS 2 /Graphene nanohybrid was also shown to give a high DBT conversion rate of 82.6%, showing potential for 3D composites. Recent work from Abbasi et al. [218] demonstrated that a cobalt-promoted MoS 2 achieved more than 80% conversion for the application of diesel, showing potential for the transfer of high-performance TMDC catalysts from oil and gas to diesel/biodiesel industries.
Besides hydrodesulfurization to remove DBT, the removal of thiophene was also demonstrated to be feasible with MoS 2 catalysis by DFT methods [219]. Kaluza et al. [220] synthesized a MoS 2 catalyst supported on mesoporous alumina that gave a thiophene conversion of more than 60% at around 400 • C. For the removal of carbonyl sulphide (COS), Liu et al. [221] used a microwave activated MoS 2 /graphene catalyst that achieved over 90% conversion for temperatures over 280 • C. Later, the use of a monolayer MoS 2 anchored on reduced graphene oxides [222] has shown to give over 90% conversion and a reduced temperature at about 180 • C.

Fuel Hydrodeoxygenation
Hydrodeoxygenation is the removal of oxygenated components from the fuel by reacting it with hydrogen to form water [206]. Liu et al. [223] found that a MoS 2 monolayer doped with an isolated Co atom can effectively reduce hydrodeoxygenation of 4-methylphenol to toluene from 300 • C to 180 • C while maintaining high conversion and selectivity of 97.6% and 98.4% respectively. From a DFT perspective, Li et al. [224] demonstrate a generic structure of Co atom vacancy in MoS 2 , claiming that the catalyst can activate hydrodeoxygenation reactions at lowered temperatures. Recent work from Wu et al. [225] adsorbed Co oxide at the edge of MoS 2 with sulphur defects and formed a Co-MoS 2−x catalyst that converts lignin-derived phenols (4-methylphenol) to arenes with 97.4% conversion and 99.6% toluene selectivity at a mild temperature of 120 • C (see Figure 9a). For direct industrial application, palm kernel oil was converted with a yield over 90% to jet fuel-like hydrocarbon (see Figure 9b) via a Ni-MoS 2 /γ-Al 2 O 3 hydrodeoxygenation catalyst [206]. The work showed high potential for converting biomass-based fuel to high-value fuels using TMDC-based catalysts. Moreover, Co-and Nipromoted MoS 2 catalyst was also used to co-process diesel and vegetable oil via hydrodeoxygenation reaction [226]. Alvarez-Galvan et al. [227] demonstrated that transition metals phosphides were useful as a hydrodeoxygenation catalyst for waste cooking oil to green biodiesel. More research work is required to explore the possibility of TMDC materials in this field. and bio-oil yield of 86.24%. Later, work from Zhou et al. [229] synthesized a series of MoSe2 catalysts for the conversion of alkali lignin into bio-oil. The work resulted in 96.46% conversion of lignin and 93.68% yield of bio-oil. A recent review by Porsin et al. [230] revealed that sulphide catalysts such as MoS2 can also convert fatty acid triglycerides to motor oil by hydrodeoxygenation. As for wider fuel cleaning applications, TMDC materials have also been reported to purify gas fuels such as hydrogen and methane [231] showing high potential in diverse fuel applications in the future.

CO2 Valorization and Conversion
Valorization of carbon dioxide (CO2) to raw chemical materials and clean fuels is an opportunity for the artificial carbon cycle, which contributes to the mitigation of global warming and alleviates the usage of fossil [232]. Over the decades, enormous efforts in searching of alternative technologies to mitigate the CO2 emissions through CO2 capture from concentrated industrial exhausts [233,234], sequestration of CO2 in the underground [235,236] and conversion of CO2 to energy-rich fuels powered by renewable energy resources [237] have been discovered. Throughout all these methods, CO2 molecules are not solely can be removed from the atmosphere but also can be converted into value-added chemicals such as methanol, formic acid, methane, and syngas [238][239][240]. Recently, electrochemical conversion of CO2 to value-added chemicals through the principal of CO2 electrochemical reduction reaction (CO2ERR) as shown in Figure 10a [241], has emerged as a comparative alternative to its counterparts such as biochemical and thermochemical technologies [242][243][244]. Alternatively, semiconductor-based photocatalysis of CO2 reduction has been another frontier in CO2 conversion [245]. When a light source with appropriate photoenergy is illuminated on this photocatalyst, electron-hole pairs are formed [246,247]. With the migration of the electronhole pair to the surface, the release of energy from within can reduce surface adsorbed CO2 [246]. Another interesting type of photocatalyst is the Z-scheme photocatalyst, which performs charge transfer between a reduction semiconductor catalyst and an oxidation semiconductor catalyst, potentially giving higher performances, efficiency, and synthesis possibilities [248]. The general working principle of photocatalyst for CO2 conversion is illustrated in Figure 10b. For catalytic deoxygenation of alkali lignin into bio-oil, Li et al. [228] proposed the use of a flower-like hierarchical MoS 2 -based composite catalyst that achieved a lignin conversion of 91.26% and bio-oil yield of 86.24%. Later, work from Zhou et al. [229] synthesized a series of MoSe 2 catalysts for the conversion of alkali lignin into bio-oil. The work resulted in 96.46% conversion of lignin and 93.68% yield of bio-oil. A recent review by Porsin et al. [230] revealed that sulphide catalysts such as MoS 2 can also convert fatty acid triglycerides to motor oil by hydrodeoxygenation. As for wider fuel cleaning applications, TMDC materials have also been reported to purify gas fuels such as hydrogen and methane [231] showing high potential in diverse fuel applications in the future.

CO 2 Valorization and Conversion
Valorization of carbon dioxide (CO 2 ) to raw chemical materials and clean fuels is an opportunity for the artificial carbon cycle, which contributes to the mitigation of global warming and alleviates the usage of fossil [232]. Over the decades, enormous efforts in searching of alternative technologies to mitigate the CO 2 emissions through CO 2 capture from concentrated industrial exhausts [233,234], sequestration of CO 2 in the underground [235,236] and conversion of CO 2 to energy-rich fuels powered by renewable energy resources [237] have been discovered. Throughout all these methods, CO 2 molecules are not solely can be removed from the atmosphere but also can be converted into value-added chemicals such as methanol, formic acid, methane, and syngas [238][239][240]. Recently, electrochemical conversion of CO 2 to value-added chemicals through the principal of CO 2 electrochemical reduction reaction (CO 2 ERR) as shown in Figure 10a [241], has emerged as a comparative alternative to its counterparts such as biochemical and thermochemical technologies [242][243][244]. Alternatively, semiconductor-based photocatalysis of CO 2 reduction has been another frontier in CO 2 conversion [245]. When a light source with appropriate photoenergy is illuminated on this photocatalyst, electron-hole pairs are formed [246,247]. With the migration of the electron-hole pair to the surface, the release of energy from within can reduce surface adsorbed CO 2 [246]. Another interesting type of photocatalyst is the Z-scheme photocatalyst, which performs charge transfer between a reduction semiconductor catalyst and an oxidation semiconductor catalyst, potentially giving higher performances, efficiency, and synthesis possibilities [248]. The general working principle of photocatalyst for CO 2 conversion is illustrated in Figure 10b

Conversion of CO2 to Syngas and Other Gases
To achieve this attractive blueprint, the key issue is to develop electrocatalysts with high activity, superior selectivity, highly durable, environmentally friendly as well as low cost [250][251][252]. In 2014, the first pioneering work of elucidating the TMDCs materials as advanced electrocatalysts for CO2 reduction was by Asadi and coauthors [253]. They have claimed that MoS2 material serves as a highly efficient electrocatalyst for the conversion of CO2 to syngas. The combination of the edge states of MoS2 in contact with ionic liquid solvent electrolytes has served as a new paradigm for CO2 reduction, providing the advantage of favorable electronic properties of MoS2 and an electrolyte that transfer the CO2 molecules to the active sites for reaction. This study is a breakthrough in this field, in which the performance of the catalytic activity for CO2 reduction reaction is far exceeding than other conventional catalysts such as carbon nanotubes, graphene, noble metal carbides, and transition metals.
Recent research work of TMDCs in converting CO2 to interesting gas products using photo-and electro-catalysis methods are shown in Table 4. Up to date, most reports of TMDCs in this field are focusing on the conversion of CO2 to CO, suggesting that there is still a huge potential of different electrochemical reaction pathways using TDMCs that can be explored in the near future such as the conversion of CO2 to CO and H2 to CH4. Wang et al. [254] synthesized a marigold-like SiC@MoS2 nanoflower for the conversion of CO2 and H2O to CH4 and O2 using no sacrificial agents while operating within the visible light spectrum (see Figure 11a). The work reported production rates of 323 and 621 µ L•g -1 •h -1 for CH4 and O2 while maintaining stable characteristics for 40 hours. Asadi et al. [255] also have synthesized a series of TMDCs (e.g. WSe2, WS2, MoSe2, and MoS2) using the chemical vapor transport (CVT) growth technique followed by liquid exfoliation for electrochemical CO2 conversion using ionic liquid electrolyte (EMIM-BF4). By benchmarking with the bulk Ag and Ag NPs, the current density of TMDCs are more than tenfold (130-330 mA cm −2 ) of the current density of Ag (3.3 mA cm −2 ) and Ag NPs (10 mA cm −2 ) with 90% CO. Overall, WSe2 NFs gave the best performance and exhibited a current density of 18.95 mA cm −2 with a high turn-over frequency of CO of 0.28 s −1 at an overpotential of 54 mV. This can be related to the intrinsic properties of WSe2 (very low work function and high-volume surface area).

Conversion of CO 2 to Syngas and Other Gases
To achieve this attractive blueprint, the key issue is to develop electrocatalysts with high activity, superior selectivity, highly durable, environmentally friendly as well as low cost [250][251][252]. In 2014, the first pioneering work of elucidating the TMDCs materials as advanced electrocatalysts for CO 2 reduction was by Asadi and coauthors [253]. They have claimed that MoS 2 material serves as a highly efficient electrocatalyst for the conversion of CO 2 to syngas. The combination of the edge states of MoS 2 in contact with ionic liquid solvent electrolytes has served as a new paradigm for CO 2 reduction, providing the advantage of favorable electronic properties of MoS 2 and an electrolyte that transfer the CO 2 molecules to the active sites for reaction. This study is a breakthrough in this field, in which the performance of the catalytic activity for CO 2 reduction reaction is far exceeding than other conventional catalysts such as carbon nanotubes, graphene, noble metal carbides, and transition metals.
Recent research work of TMDCs in converting CO 2 to interesting gas products using photo-and electro-catalysis methods are shown in Table 4. Up to date, most reports of TMDCs in this field are focusing on the conversion of CO 2 to CO, suggesting that there is still a huge potential of different electrochemical reaction pathways using TDMCs that can be explored in the near future such as the conversion of CO 2 to CO and H 2 to CH 4 . Wang et al. [254] synthesized a marigold-like SiC@MoS 2 nanoflower for the conversion of CO 2 and H 2 O to CH 4 and O 2 using no sacrificial agents while operating within the visible light spectrum (see Figure 11a). The work reported production rates of 323 and 621 µL·g -1 ·h -1 for CH 4 and O 2 while maintaining stable characteristics for 40 h. Asadi et al. [255] also have synthesized a series of TMDCs (e.g. WSe 2 , WS 2 , MoSe 2 , and MoS 2 ) using the chemical vapor transport (CVT) growth technique followed by liquid exfoliation for electrochemical CO 2 conversion using ionic liquid electrolyte (EMIM-BF4). By benchmarking with the bulk Ag and Ag NPs, the current density of TMDCs are more than tenfold (130-330 mA cm −2 ) of the current density of Ag (3.3 mA cm −2 ) and Ag NPs (10 mA cm −2 ) with 90% CO. Overall, WSe 2 NFs gave the best performance and exhibited a current density of 18.95 mA cm −2 with a high turn-over frequency of CO of 0.28 s −1 at an overpotential of 54 mV. This can be related to the intrinsic properties of WSe 2 (very low work function and high-volume surface area). (a) (b) Figure 11. (a) Photochemical reaction pathway for the conversion of CO2 and H2O to CH4 and O2 under visible light using SiC@MoS2 nanoflower (Reproduced with permission from [254]. Copyright 2018 American Chemical Society) and (b) schematic of the WSe2-based artificial leaf cell design for the conversion of CO2 and H2O to CO and H2 (Reproduced with permission from [255]. Copyright 2016 The American Association for the Advancement of Science).

Direct Conversion of CO2 to High-Value Chemicals
Direct conversion of CO2 to high-value chemicals (such as intermediate chemicals, fuels, and alcohols) has a high potential for commercial value [269,270]. TMDCs have demonstrated properties for catalyzing CO2 into important alcohol, especially methanol [85,271]. Bolivar et al. [272] studied the use of a Pt/MoOx/MoSe2 electrocatalyst and proposed that oxygen-containing surface species lowers the electric potential for methanol production. Moreover, Tu et al. [273] have demonstrated the methanol production of TiO2 and 0.5 wt.% MoS2 nanocomposites was higher than Pt/TiO2, Au/TiO2, and Ag/TiO2 for the photocatalytic reduction of CO2. For this TiO2/MoS2 composites, Xu et al. [274] showed that electrons are transferred from MoS2 to TiO2 upon contact, therefore promoting the separation of charge carriers upon photoexcitation. In 2018, Francis et al. [275] have reported on In addition, a lab-scale synthetic custom-builds wireless setup for photochemical studies of WSe 2 /IL is also being developed by this group [255]. The build-up of the artificial leaf that performed the photosynthesis process is investigated to further confirm the CO 2 conversion rate efficiency (see Figure 11b). The system comprises of three main components: (i) harvest light using two amorphous silicon triple-junction photovoltaic cells in series, (ii) CO 2 reduction using WSe 2 /IL on the cathode and lastly, and (iii) evolution of oxygen using cobalt (CoII) oxide/hydroxide in potassium phosphate electrolyte [265,266]. Surprisingly, the cell managed to function continuously for more than 4 h before corrosion happens at the transparent indium tin oxide layer on the anode, proving that this theoretical experiment is workable and insightful, which could be extended to the technology development scale in near future. Further studies on the defect engineering for CO 2 reduction catalysts were also discussed by Wang et al. [267] to improve functionality by defect engineering (such as holes, doping, vacancies, phase change, edge, grain boundary, lattices distortion, and other methods). For this, MoS 2 with substitutional defects on the edge was studied [253] and showed to give better current density than Au, Ag, Cu nanopowder, and the polycrystalline catalyst. Ji et al. [268] suggested that group 10 TMDCs have a high potential for CO 2 reduction as they do not suffer from OH poisoning. The work proposed that the reaction energies for CO 2 reduction could be tuned by vacancy densities.

Direct Conversion of CO 2 to High-Value Chemicals
Direct conversion of CO 2 to high-value chemicals (such as intermediate chemicals, fuels, and alcohols) has a high potential for commercial value [269,270]. TMDCs have demonstrated properties for catalyzing CO 2 into important alcohol, especially methanol [85,271]. Bolivar et al. [272] studied the use of a Pt/MoO x /MoSe 2 electrocatalyst and proposed that oxygen-containing surface species lowers the electric potential for methanol production. Moreover, Tu et al. [273] have demonstrated the methanol production of TiO 2 and 0.5 wt.% MoS 2 nanocomposites was higher than Pt/TiO 2 , Au/TiO 2 , and Ag/TiO 2 for the photocatalytic reduction of CO 2 . For this TiO 2 /MoS 2 composites, Xu et al. [274] showed that electrons are transferred from MoS 2 to TiO 2 upon contact, therefore promoting the separation of charge carriers upon photoexcitation. In 2018, Francis et al. [275] have reported on the activity of single crystals and thin films of MoS 2 for the reduction of CO 2 dissolved in an aqueous electrolyte to yield 1-propanol, ethylene glycol, and t-butanol and hydrogen. At an applied potential of −0.59 V, the Faradaic efficiencies for the reduction of CO 2 to 1-propanol are ca. 3.5% for MoS 2 single crystals and 1% for thin films with low edge-site densities. WS 2 quantum dots doped Bi 2 S 3 nanotubes also demonstrated high (38.2 µmol·g −1 ·h −1 ) production of methanol while co-producing ethanol [85]. Many researchers have also performed the conversion of CO 2 to high-value products such as methanol, acetaldehyde [249] and ethanol [276] by using TMDC catalysts (see Table 5).

Challenges and Future Prospects
The application of TMDCs in pollution reduction is a breakthrough as has high availability [30][31][32], high economic potential [280][281][282], and excellent efficiency [254,276] for the application of pollution reduction. For TMDCs availability and efficiency, we have already discussed them in Sections 1 and 2, Sections 3-6 (depending on application) respectively. In terms of commercial techno-economic analysis, a few works have also pointed out the advantage of using TMDC catalysts. Valle et al. [280] studied the economics of converting biomass-derived syngas to ethanol using an indirect circulating fluidized bed gasification (iCFBG) system via the KCoMoS 2 catalyst. The work highlighted that the iCFBG system could reach a minimum ethanol selling price (MESP) of 0.74 $/L (USD) while technologies such as entrained-flow gasification have that at 0.98$/L. This MoS 2 -based catalyst was also found to be more commercially viable than the Rh-Mn/SiO 2 catalyst due to their cheaper economical costs, although the latter has better energy efficiency [281]. Phillips [282] also concluded that the use of Co/MoS 2 has potential to lower down ethanol MESP down to 0.267 $/L (USD) highlighting that Ru or Rh-based catalysts are economically expensive even at low concentrations. Nevertheless, authors acknowledge that the techno-economic analysis of cutting-edge TMDCs technologies is still generally unknown. Thus, more TMDCs research studies should include consideration of economic analysis. Furthermore, despite the unique physiochemical properties offered by the TMDCs (e.g., the large surface area, which acts as a two-edged sword [283], tunable absorption heterostructure [200,284], flexible, diversiform, and functional properties [146]), more exploration on the operating conditions, kinetic, and thermodynamic parameters should be carried out to further confirm its ability [285]. This is because most of the reviewed works are conducted under a controlled environment with a specially designed flow. Under this context, whether obtained simulation and experimental results reflect the actual situation remains uncertain. The main limitations of commercializing of TMDCs-based technology are: (i) Most research works are focusing on the development of the materials as an effective pollutant degradation catalyst during oxidation, neglecting that in reality, the number of organic pollutants is in a mixture that may affect the performance of the engineered material [286,287]. (ii) The rate of product formation for TMDC-based technology, such as syngas and methanol from the conversion of CO 2 is still far from the industrial scale, which is not sufficient to accommodate the global demand [286,287]. More research effort is required to improve efficiencies. (iii) High preparation/fabrication cost and poor uniformity of the materials in large-scale production [288]. More effort is required in sustainable synthesis and fabrication of TMDC-based technologies. (iv) The larger surface area offered by the TMDCs can enhance the adsorption capability, but also cause higher effects of environmental disturbance [283]. Precise defect engineering to improve material performances can also be carried out [225,267]. Optimization and data analytics on the design dilemma and defect engineering should be highlighted and properly studied.
As a future prospect, commercial validation on the performance on a larger scale needs to be carried out. The results obtained from the experiment can be validated by conducting techno-economic analysis, such as Monte Carlo analysis for economic and risk feasibility [289,290]. Furthermore, advanced machine learning methods can also be used to accelerate experimental designs [291,292], obtain optimal performances [293,294], and for the discovery of new TMDCs composite structures [295,296].

Conclusions
The unique and exclusive features of TMDCs (e.g., layered structure, tunable bandgap, unique optical, thermal and electrical properties, etc.) have been the main driving force that drives the researchers' attention on exploring the potential of the 2D materials in pollution reduction applications. This work summarized the state-of-the-art applications of various TMDCs under the context of pollution mitigation (including (i) gas adsorption and removal, (ii) gas sensing, (iii) wastewater treatment, (iv) flue cleaning, and (v) CO 2 valorization and conversion). In addition to the up-to-date progress of TMDCs research, this article also discussed some of the key challenges for the future commercialization of TMDC materials. Apparently, many of the reviewed research works have authenticated their substantial potential to substitute the existing pollution mitigation media. Nevertheless, the current applications are still restricted to a lab basis, where the deviation of the actual performance of TMDCs under larger-scale production remains as the research gap. To usher TMDCs into the next level of utilization (i.e., from the lab scale to the industrial scale), the following three research directions should be followed up, (i) techno-economic analysis (TEA) study, (ii) experimentation under more rigorous and realistic conditions, and (iii) experimental optimization for application purpose. The authors sincerely hope this review can serve as an insightful guideline that inspires more researchers to venture into this new and exciting cutting-edge field.

Conflicts of Interest:
The authors declare no conflict of interest.