25 Introduction, Synthesis and Characterizations of Transition metal Dichalcogenides (TMDs)
Dr. Anchal Srivastava
Contents of this Unit
1. Introduction
2. Crystal Structure of TMDs
3. Synthesis of TMDs
4. Transfer of TMDs to different substrates
5. Characterizations
5.1 Optical Microscopy
5.2 Raman Microscopy
5.3 Atomic Force Microscopy (AFM)
5.4 Transmission Electron Microscopy (TEM)
6. Summary
Learning Outcomes
- After studying this module, you shall be able to
- Know about various 2D materials which overcome the limitation of graphene zero band nature which limits its application in semiconducting devices needing high on-off ratio.
- Learn various synthesis approaches for synthesizing 2D materials such as mechanical exfoliation, liquid phase exfoliation, chemical vapour deposition etc.
- Learn about different characterization techniques for the confirmation of the presence of mono and few layer TMDs.
1. Introduction
Graphene, while being fundamentally and technologically interesting for a variety of applications, is chemically inert and can only be made active by functionalization with desired molecules, which in turn results in the loss of some of its exotic properties. In contrast, single-layered 2D transition metal dichalcogenides (TMDs), whose generalized formula is MX2, where M is a transition metal of groups 4–10 and X is a chalcogen (Fig. 1) exhibit versatile chemistry. TMDs with the chemical formula MX2 (X = S, Se, Te; M = transition metal) Represent a particularly interesting class of 2D materials comprising both semiconductors and metals. For example, MoS2, MoSe2, WS2, and WSe2 were shown to undergo a transition from indirect to direct band gap materials when their thickness is thinned down to a single layer. Together with their strong interaction with light and relatively high charge carrier mobilites, this has opened up the possibility of using few layer TMDs in a range of applications including ultrathin field effect transistors, photodetectors, light emitting diodes and solar cells. To date, optoelectronic research in monolayer TMDs has mainly focused on the Mo-and W-based compounds which have (optical) band gaps in the range of 1.6−2.0 eV.
TMDs are fundamentally and technologically intriguing due to their unique electronic and optical properties. The behaviors of TMDs are varied, ranging from insulating, semiconducting, truly metallic to superconducting. Reducing the stacking of TMDs to atomic layers leads to new characteristics induced by quantum confinement effects. The properties of bulk TMDs are diverse ranging from insulators such as HfS2, semiconductors such as MoS2 and WS2, semimetals such as WTe2 and TiSe2, to true metals such as NbS2 and VSe2. A few bulk TMDs such as NbSe2 and TaS2 exhibit low-temperature phenomena including superconductivity, charge density wave (CDW, a periodic distortion of the crystal lattice) and Mott transition (metal to non-metal transition). Exfoliation of these materials into mono or few layers largely preserves their properties, and also leads to additional characteristics due to confinement effects. The chemistry of MX2 compounds thus offers opportunities for going beyond graphene and opening up new fundamental and technological pathways for inorganic 2D materials.
2. Crystal Structure of TMDs
Many TMDs crystallize in a graphite-like layered structure that leads to strong anisotropy in their electrical, chemical, mechanical and thermal properties. Group 4-7 TMDs in Fig. 1.a are predominantly layered, whereas some of group 8-10 TMDs are commonly found in non-layered structures. In layered structures, each layer typically has a thickness of 6~7 Å, which consists of a hexagonally packed layer of metal atoms sandwiched between two layers of chalcogen atoms. The interlayer M-X bonds are predominantly covalent in nature, whereas the sandwich layers are coupled by weak van der Waals forces thus allowing the crystal to readily cleave along the layer surface. As suggested like graphene, single layers of TMDs are stabilized by development of a ripple structure.
The metal atoms provide four electrons to fill the bonding states of TMDs such that the oxidation states of the metal (M) and chalcogen (X) atoms are +4 and –2, respectively. The lone-pair electrons of the chalcogen atoms terminate the surfaces of the layers, and the absence of dangling bonds renders those layers stable against reactions with environmental species. The M–M bond length varies between 3.15 Å and 4.03 Å, depending on the size of the metal and chalcogen ions.
Fig. 1. About 40 different layered TMD compounds exist. The transition metals and the three chalcogen elements that predominantly crystallize in those layered structure are highlighted in the periodic table. Partial highlights for Co, Rh, Ir and Ni indicate that only some of the dichalcogenides form layered structures. b,c, c-Axis and section view of single-layer TMD with trigonal prismatic (b) and octahedral (c) coordinations. Atom colour code: purple, metal; yellow, chalcogen.
The metal coordination of layered TMDs can be either trigonal prismatic or octahedral (typically distorted and sometimes referred to as trigonal-antiprismatic) as shown in Fig. 1b and c, respectively. Depending on the combination of the metal and chalcogen elements, one of the two coordination modes is thermodynamically preferred.
In contrast to graphite, bulk TMDs exhibit a wide variety of polymorphs and stacking polytypes (a specific case of polymorphism) because an individual MX2 monolayer, which itself contains three layers of atoms (X–M–X), can be in either one of the two phases. Most commonly encountered polymorphs are 1T, 2H and 3R where the letters stand for trigonal, hexagonal and rhombohedral, respectively, and the digit indicates the number of X–M–X units in the unit cell (that is, the number of layers in the stacking sequence). There are three different polytypes (that is, three different stacking sequences) for 2H polymorphs. A single TMD can be found in multiple polymorphs or polytypes, depending on the history of its formation. For example, natural MoS2 is commonly found in the ‘2H phase’ where the stacking sequence is AbA BaB (The capital and lower case letters denote chalcogen and metal atoms, respectively). Synthetic MoS2, however, often contains the 3R phase where the stacking sequence is AbA CaC BcB. In both cases, the metal coordination is trigonal prismatic. Group 4 TMDs such as TiS2 assume the 1T phase where the stacking sequence is AbC AbC and the coordination of the metal is octahedral. For the sake of simplicity, we will focus our attention on monolayer TMDs in the discussions below.
It should be highlighted that monolayer TMDs exhibit only two polymorphs: trigonal prismatic and octahedral phases. The former belongs to the D3h point group whereas the latter belongs to the D3d group. In the following discussion, they are referred to as monolayer 1H (or D3h)- and 1T (or D3d)- MX2, respectively.
3. Synthesis of TMDs
To realize the applications of TMDs a consistent, reliable, simple and inexpensive method of growing high-quality, uniform, and continuous, single and few-layer TMDs films is a prerequisite. Two-dimensional TMDCs can be fabricated using two types of approaches: the top-down approach, where the bulk forms are exfoliated into a few-layer structures and monolayers (MLs), and the bottom-up approach using growth methods such as chemical vapour deposition (CVD) or molecular epitaxy. Work already focusing on scale-up to large area arrays is in progress
3.1. Top Down Approach:
3.1.1. Mechanical Exfoliation:
To date, mechanical exfoliation is the most efficient way to produce the cleanest, highly crystalline and atomically thin nanosheets of layered materials. In a typical mechanical exfoliation process, appropriate thin TMDC crystals are first peeled off from the bulk crystal by using an adhesive Scotch tape (Fig.2, top panel). These freshly cleaved thin crystals on the Scotch tape are brought into contact with a target substrate and rubbed to further cleave them. After the Scotch tape is removed, singlelayer and multilayer TMDC nanosheets are left on the substrate (Fig. 2, bottom panels a–d).
While this method produces single-crystal flakes of high purity and cleanliness that are suitable for fundamental characterization and even for fabrication of individual devices, it is not scalable and does not allow systematic control of flake thickness and size. Therefore, this technique is not feasible for large-scale production of TMDC
Fig. 2. Top: Schematic of micromechanical cleavage technique (the Scotch tape method) for producing few-layer structures. Top row adhesive tape is used to cleave the top few layers from a bulk crystal. Bottom left the tape with removed flakes is then pressed against the substrate of choice. Bottom right some flakes stay on the substrate, even on removal of the tape. Bottom: Mechanically exfoliated singleand few-layer MoS2 nanosheets on 300 nm SiO2/Si. Optical microscopy (a–d).
3.1.2. Liquid-phase Exfoliation:
Mechanical exfoliation using scotch-tape method yields the highest quality monolayered samples, which are ideal for demonstration of high performance devices. However liquid exfoliation methods are likely to be better suited for fundamental and proof-of-concept demonstrations in applications where large quantities of materials are required, such as electrochemical energy storage, catalysis, sensing or fillers for composites. Liquid exfoliation by direct sonication in commonly used solvents such as dimethylformamide and N-methyl-2pyrrolidone has been used to disperse graphene. The similar method can be applied to fabricate single-layer and multilayer nanosheets of a number of layered inorganic compounds, such as MoS2, WS2, MoSe2, NbSe2, TaSe2, NiTe2, MoTe2 etc. a number of layered crystals including BN, TMDs and transition metal oxides were successfully exfoliated in water. The nanosheets were dispersed by sonication in an aqueous solution of the surfactant sodium cholate, which coats the sheets, preventing their re-aggregation. These direct sonication techniques rely on the solvent or surfactant to overcome the cohesive energy between the neighbouring layers; which means the solvents must be chosen to have surface energies that are comparable to those of the material. The key challenge of these methods is to enhance the yield of the single layers (as opposed to few-layers) and to maintain the lateral dimensions of the exfoliated sheets. One of the most effective methods for mass production of fully exfoliated TMD nanosheets is the ultrasound-promoted hydration of lithium-intercalated compounds. the preparation of single-layer MoS2 with n-butyl lithium dissolved in hexane as the intercalation agent. An important step in the lithium intercalation process is the formation of LixXS2 compound and this reaction can be tuned to control the yield of monolayers. The degree of lithiation also has implications on the amount of 1T phase present in MX2. The lithiated solid product can then be retrieved by filtration and washed with hexane to remove excess lithium and organic residues from n-butyl lithium. The extracted product can be readily exfoliated by sonication in water.
The yield of this method is very high. Although the yield of the lithium intercalation method for obtaining single-layer TMDs is nearly 100%, some challenges remain. The first is that the experiment is carried out at high temperature (for example, 100 °C) for long durations (for example, three days). Also, the lithium intercalation must be carefully controlled to obtain complete exfoliation while preventing the formation of metal nanoparticles and precipitation of Li2S.
3.2. Bottom Up Approach:
Whilst the mechanical and liquid-phase exfoliation processes are both top-down processes, efforts have been made since the beginning of graphene research to produce graphene in a bottom-up process instead. The aim is to produce large single crystals of TMDs on substrates suitable for further study in a process scalable for commercial use. The mechanical and liquid-phase exfoliation processes both fail to achieve this.
3.2.1. Synthesis via metal chalcogenisation:
An example of a two-step process is shown in Fig. 3a where a nm-thick Mo layer was first deposited onto SiO2/Si substrates using e-beam evaporation, and then vapor phase sulphurised at 750 ◦C in the CVD system. The reaction mechanism can be understood as a direct chemical reaction. Therefore, the size and thickness of the pre-deposited Mo film determine the thickness and size of the obtained MoS2 thin film. In this way, large-area MoS2 with only a few layers in thickness could be synthesized, with the sample size only limited by the size of the growth substrate. However, because of the high melting point (2610 ◦C) of Mo, the migration of Mo atoms is effectively suppressed at the growth temperature.
Fig. 3 CVD growth of MoS2. a Sulphurisation of a Mo thin film was pre-deposited on a SiO2 substrate. b Schematics for the synthesis and cleavage of MoS2.
3.2.2. Chemical Vapor Deposition (CVD):
Chemical vapor deposition of graphene on copper has been a major breakthrough that has enabled the preparation of large-area graphene. Very recently, synthesis of large-area ultrathin MoS2 layers using CVD has been demonstrated using several approaches. Most of the current CVD research has focused on MoS2; we therefore introduce the details of MoS2 growth by CVD, then discuss strategies for extending the methodology to other single-layered TMD materials.
A two-step thermolysis process shown in Fig. 4 a was reported for deposition of three-layered MoS2 sheets by dip-coating in ammonium thiomolybdates [(NH4)2MoS4] and converting to MoS2 by annealing at 500 °C followed by sulfurization at 1,000 °C in sulfur vapour. The chemical reaction leading to the formation of the MoS2 layers is (NH4)2MoS4 + H2 → 2NH3 + 2H2S +MoS2.
A different strategy is also reported for deposition of single-layer MoS2 is based on the sulfurization of Mo metal thin films (Fig. 4c). Adsorption of sulfur on the Mo film to form MoS2 has been studied since the 1970s58 and it has been demonstrated through low-energy electron diffraction, auger electron spectroscopy and thermal desorption spectroscopy that sulfur atoms form ordered phases on the Mo crystal face.
Large-area MoS2 monolayer flakes using the gas-phase reaction of MoO3 and S powders (Fig. 4d). It has been observed that treatment of substrates with aromatic molecules such as reduced graphene oxide, perylene-3,4,9,10-tetracarboxylic acid tetrapotassium salt (TPAS) and perylene-3,4,9,10-tetracarboxylic dianhydride (PTCDA) prior to deposition assists in the growth. However, full coverage of the substrate is a challenge using this method. Few other reports shows that wafer-scale deposition can be achieved using the same chemistry, where the few-layer MoS2 was obtained after direct sulfurization of MoO3 thin-films on sapphire substrates (Fig. 4e). These previous techniques can be classified into three categories: (i) vaporization of metal and chalcogen precursors and their decomposition, followed by deposition of TMD on a substrate, (ii) direct sulfurization (or selenization) of metal film, and (iii) conversion of MO3 (metal oxide) to MS2 (metal disulfide) by sulfurization. Precursors can be fed into the reaction zone using the experimental schemes shown in Fig 4f,g.
Fig. 4 Chemical vapour deposition of ultrathin TMDs. a, Schematic of MoS2 layer deposited by two-step thermolysis, and the films obtained on a sapphire and silica sustrate. b, MoS2 nanosheets on CVD graphene/Cu substrates through van der Waals epitaxy. c, Schematic illustration of single- to few-layered MoS2 by sulfurization of Mo thin film. d, Alternative method for the growth of a MoS2 monolayer, also on a substrate (sample) from MoO3 and S powders through a gas-phase reaction. The substrate has been treated with aromatic molecules to assist the growth of larger flakes, as shown by the optical micrograph and atomic force microscopic images of the triangular nanosheets that are obtained. The red circles represent the heating reaction chamber. e, Wafer-scale deposition of mono-to few-layered MoS2 films obtained by direct sulfurization of MoO3 thin films, and their transfer onto arbitrary substrates. f,g, CVD of ultrathin TMDs by vaporization and decomposition of a singleprecursor in solution (f) and by vaporization and decomposition of metal and chalcogen precursors in solid forms (g).
4. Transfer of TMDs to different substrates
When a thin TMDC layer is synthesized it is important for fundamental and applied research for it to be transferred to an arbitrary substrate As the growth temperature of TMDC monolayers are relatively high, temperature sensitive substrates (such as polymer-based substrates) cannot be used in the synthetic process, while their use is essential for flexible electronics. It is thus essential to develop a transfer technique to implement large-area TMDC on different substrates. The as-grown MoS2 sample was cut into three pieces and treated for 30 s with DI water, isopropyl alcohol, and acetone, respectively. The surface of the as-grown monolayer is hydrophobic, so that isopropyl alcohol and acetone spreaded out on MoS2, whereas water remained as a droplet. During the 30 s, the as-grown MoS2 monolayer started to break into small pieces and floating on the water droplet, which demonstrated that the as-grown MoS2 monolayer can be easily removed from the substrate with DI water. a surface-energy-assisted process has been developed that allowed the authors to perfectly transfer centimeter-scale monolayer and few-layer TMDC films from original growth substrates onto arbitrary substrates with no observable wrinkles, cracks, or polymer residues. The unique strategies used in this process included leveraging the penetration of water between hydrophobic TMDC films and hydrophilic growth substrates to lift off the films and dry transferring the film after the lift off. The whole process is schematically illustrated in Fig. 5
Fig. 5 Illustration of the surface-energy-assisted transfer process. a–h Typical images of the transfer process. The arrows in e–g point toward the MoS2 film for visual convenience.
5. Characterization:
Probing the molecular and vibrational structure of 2D materials is inherently challenging due to the small sample size. However, numerous methods have been developed to enable their identification and characterization. In this section, we will provide a standard how-to guide for studying single-layer-thick materials.
5.1. Optical Microscopy:
Optical microscopy is initially the most powerful high-throughput method for initially identifying single- and multiple-layer flakes. Dielectric-coated SiO2/Si substrates are the most common substrates used to visualize and locate single and few layers (Figure 6a,e,f). The color of the dielectric-coated wafer depends on an interference effect from reflection off of the two surfaces of the dielectric. Single-and few-layer flakes on the surface of the dielectric modify the interference and create a color contrast between the flake and the substrate. For optimal contrast, the thickness of the dielectric coating needs to be within 5 nm of the ideal value. However, since the index of refraction of many novel materials is unknown, it is often necessary to first experimentally exfoliate onto a range of substrates having different dielectric thicknesses to initially determine the optimal thickness.
Fig. 6 a–m Color optical images of 1L–15L MoS2 on 300 nm SiO2/Si. The scale bars are 5 µm for images a–l and 10µm for image m. The digital is shown in a–m indicate the layer numbers of the corresponding MoS2 nanosheets. Plots of (n) CD values and (o) CDR, CDG, and CDB values of 1L–15L MoS2 on 300 nm SiO2/Si
5.2. Raman Spectroscopy of TMDs:
Raman Spectroscopy is a useful method for fingerprinting a material and layer-dependent changes of the vibrational structure (Fig. 7c). It is necessary to use low power (e.g., 16 kW/cm2 for Bi2Te3,131 or 40 kW/cm2 for GeH) when obtaining Raman spectra of single layers to prevent sample decomposition. The Stokes/anti-Stokes peak ratio can be used to determine the local temperature. Raman spectra can also detect vibrational modes that are active due to symmetry breaking in single-layer films and enhanced vibrations in out-of-plane modes when the single layer is suspended. When varying the layer number between bulk, few, and single layers, the Raman spectra differ in spectral width and intensity due to differing interlayer interactions. Substrates that have vibrational modes that overlap with those in the material of interest should be avoided. Vibrational spectroscopy, in general, and Raman spectroscopy, in particular, could be used to detect isotopic-enriched 2D materials to study the growth mechanism as demonstrated and thus be used to study the growth mechanism.
Fig. 7 (a) Optical micrograph of thin films of MoS2 (b) AFM of MoS2.(c) Raman spectra of single-layer, few-layer, and bulk MoS2 films.
5.3. Thickness determination using AFM:
AFM is a powerful technique to determine layer thickness (Fig. 8b,d) with a precision of 5%. However, discrepancies arise from differences in the interactions of the tip with the sample and substrate. For example, the thickness of a single layer can be better determined by measuring the height of a second layer on a first layer, rather than the height between a single layer and the substrate because in the former the tip layer interactions are constant. To control these differences and obtain the most accurate height profile of a single layer, one must first optimize the tip surface distance to exclude any hysteretic artifacts. AFM image and line profile of different TMDs on SiO2/Si substrate has been shown in Fig. 8. Height difference for various TMDs is different due to the atomic size and bond length present in the material.
Fig. 8 Atomic force microscope images on SiO2 substrates. The insets of the AFM images represent height profles from the substrate onto the nanosheets. The height of the step at the edge indicates the thickness of the nanosheets.
5.4. Transmission Electron Microscopy (TEM):
Transmission Electron Microscopy (TEM) can provide detailed information on the nature of crystallinity, layer sizes, interlayer stacking relationships, and elemental composition (Figure 9a,b). In the case of graphene, selected area electron diffraction can distinguish between monolayer and multilayer graphene when the intensity ratio of the (100) to (110) diffraction spots is larger or smaller than 1, respectively. This deviation in diffraction intensity can also distinguish between single- and multilayer flakes of other 2D materials when there is a change in interlayer registration.
Fig. 9 (a) Low-resolution TEM and (b) high-resolution TEM of a 2D WS2 TEM island folded.Inset: Fast Fourier transform pattern.
6. Summary:
- In this module, you study
- The crystal structures of TMDs.
- Various synthesis techniques for TMDs growth. Advantages and disadvantage of these techniques.
- Characterization of TMDs from different techniques such as Optical microscopy, Raman spectroscopy, Atomic force microscopy, and High resolution transmission electron microscopy.
Suggested Reading
- Text book of “Two-dimensional transition-metal dichalcogenides” by Alexander V. Kolobov
- Manish Chhowalla et. al., The chemistry of two-dimensional layered transition metal dichalcogenide nanosheets, Nature Chemistry, 5, 2013.
- Sheneve Z. Butler et. al., Progress, challenges, and opportunities in two-dimensional materials beyond graphene, ACS Nano, 7, (4), 2898-2926, 2013.
- Filip A. Rasmussen et. al., Computational 2D materials database: Electronic Structure of Transition-Metal Dichalcogenides and Oxides, J. Phys. Chem. Chem. C, 119, 13169-13183, 2015
you can view video on Introduction, Synthesis and Characterizations of Transition metal Dichalcogenides (TMDs) |