Overview of Molybdenum Disulfide

Two dimensional (2D) material was not widely studied until the first experimental demonstration done by Novoselov et.al in 2004[1]. Novoselov exfoliated graphene from bulk graphite by using a scotch tape and they received the Nobel Prize in 2010. This is a useful method to fabricate a few atomic layered thick materials which are held together by weak Van der Waals force (40~70 meV). From then on, various 2D materials such as molybdenum disulfide (MoS2) and hexagonal boron nitride (h-BN) have been successfully fabricated by mechanical exfoliation method of 2D material from bulk crystal by scotch tape[2, 3]. Thanks to the discovery of graphene, it opened a window to the entire family of 2D materials[4].

However, although graphene has high mobility and low resistance, gapless property of pristine graphene makes it difficult to turn off when used in field effect transistors(FET) so that it has poor on/off current ratio (~10²)[5, 6]. Because of its poor on/off current ratio, it is unsuitable to switching application. In 2011, the first single layered molybdenum disulfide (MoS2) transistor was fabricated by Andras Kis group[7]. Single layered MoS2

is a kind of transition metal dichalcogenides (TMDCs) which has higher energy bandgap

Recently, many researchers become interested in researching this TMDCs due to their properties complementary to those in graphene[8-10]. TMDCs are atomically thin semiconductors of the type MX2, where M is a transition metal of groups 4-10 (except group 8) and X is a chalcogen as shown in Fig. 1.1. One layer of M atoms is sandwiched between two layers of X atoms and each layer is usually 6~7 Å thick as shown in Fig 1.2.

TMDCs are held together by weak Van der Waals force between layers. Within each layer, the M-X bond is a covalent bond.

The crystal structure of single layer MoS2 is similar to that of the graphene honeycomb lattice. However, each carbon atom in graphene is replaced by Mo or a pair of S atoms for MoS2. In nature, MoS2 has three polytypes which are 2H, 3R and 1T phases[11-13]. Both 2H and 3R phases are semiconducting and have trigonal prismatic coordination[13]. However, 1T phase is metallic[14] and has octahedral coordination[13].

The 2H-MoS2, which is two layers per unit cell in the hexagonal symmetry, stacking sequence is AbA BaB (the capital represents chalcogen and the small letter represents metal in TMDCs). The 3R-MoS2, which is three layers per unit cell in the rhombohedral symmetry, stacking sequence is AbA BcB CaC. The 1T-MoS2, which is one layers per unit cell in tetragonal symmetry, stacking sequence is AbC [15] as shown in Fig. 1.3. The metal coordination and top view of this three phases are shown in Fig. 1.4.

Nature Nature chemistry vol 5 April 2013

Fig. 1.1 The transition metals and the three chalcogen elements are shown in the periodic table

Nature Nanotechnology, 2011, 6, 147-150

Fig. 1.2 Three dimensional representation of the structure of MoS2

Nature Nanotechnology, 2012, 7, 699-712 

Fig. 1.3 Three polytypes of MoS2: 2H (two layers per unit cell), 3R (three layers per unit cell) and 1T (one layer per unit cell).

Chem Commun., 2017, 53, 3054-3057

Fig. 1.4 The metal coordination and top view of 1T, 2H, and 3R phase.

The band structures of MoS2 with different thickness, calculated by density functional theory (DFT), are shown in Fig. 1.5[16]. The lowest energy transitions are labeled by the solid arrows. For bulk MoS2 crystal, the valence band maximum is located at the Γ-point, while the conduction band minimum is located almost halfway along the Γ-K range. Therefore, bulk MoS2 is an indirect bandgap material and its bandgap is 1.29eV. When the layer number decreases, the conduction minimum moves upward and the valence band maximum (Γ-point) moves downward. However, the direct transition at K-point barely changes when the layer number decreases. Finally, MoS2 becomes a direct bandgap (~1.9eV)[2, 17, 18] material when it is a single layer. This phenomena could be explained as follows[16]: The conduction band states at the K-point are mainly composed of strongly localized d orbitals at Mo atom sites. Mo atoms are located in the middle of the S-Mo-S sandwich structure, so the interlayer coupling is weak. In contrast, states near the Γ-point and the point of indirect bandgap mainly originate from a linear combination of d orbitals at Mo atoms and antibonding pz orbitals at S atoms. Strong interlayer coupling results in the energy dependence on layer thickness. Similar changes in the electronic band structure as a function of sample thickness have been predicted for other semiconducting 2D materials[19, 20]. The MoS2 band gap energy versus layer number is

summarized in Fig 1.6[2].   

Nano letter, 2010, 10, 1271‐1275 

Fig. 1.5 Calculated band structures of (a) bulk (b) quadrilayer (c) bilayer, and (d) monolayer MoS2.

Phys. Rev. Letter. 2010, 105, 136805. 

Fig. 1.6 Band gap energy of MoS2 as function of layer number for N= 1~6.

Because MoS2 is a semiconductor with band gap from 1.29eV to 1.9eV, it could complement graphene which is gapless in pristine condition in many electronic and photonic applications, especially for transistors[21, 22]. Furthermore, MoS2 is very sensitive to environment due to its high surface-to-volume ratio[23]. Therefore, MoS2 can easily absorb molecules on the surface. According to the charge transfer theory, charged molecules absorbed on MoS2 affect the conduction properties of the MoS2 film changing its electric characteristics[24]. MoS2 can be applied to fabricate gas sensor such as NH3, NO, biosensors for proteins or pH sensor by using these electric characteristics[25-28].

Other applications such as memory devices[29-31] and photodetectors[32] have also been reported. Some group also have made p-n diode and solar cells by using WSe2/MoS2

heterostructures[33-35]. Many applications using MoS2 have been reported widely.