Optoelectronics with single layer group-VIB transition metal dichalcogenides

1 Introduction

Atomically thin materials have attracted a great deal of attention since they exhibit rich and intriguing properties that have been impossible to extract from their bulk counterparts. A typical layered material consists of a stack of planes of atoms held together through strong in-plane covalent and weak out-of-plane van der Waals forces [1]. The weak out-of-plane van der Waals forces enable the separation and isolation of stable atomically thin layers, the so-called 2D materials. An important property of 2D materials is their large surface-to-volume ratio that makes them very sensitive to external perturbations. It has been shown that the band gap in atomically thin materials can be tuned by changing the number of layers [2], [3] or as a function of strain [4], a property that is highly desirable in modern-day electronics industry. Secondly, the Fermi level and the charge density can be tuned as a function of external gate voltage, which enables the strong modulation of the electrical and optical properties of SL materials. Also, it has been shown that the charge density in doped 2D semiconductors is more stable against thermal fluctuations [5].

Graphene, a zero band gap semimetal, was the first 2D material that has been mechanically exfoliated [6] and is one of the most extensively studied 2D materials due to its wide range of promising applications in science and technology. However, graphene is not the best material for every application since its zero band gap hinders its application in high-performance field effect transistors and in optoelectronics. Soon after the discovery of graphene, the search for other 2D materials began, leading to the discovery of a slew of 2D materials [7], [8], such as hexagonal boron nitride, phosphorene, and single-layer (SL) transition metal dichalcogenides (TMDCs). TMDCs have the chemical composition of MX₂; M= transition metal such as Mo, W, Nb, Ta, Re, Zr, and X= S, Se, Te. They exhibit a wide range of properties. For example, NbS₂, NbSe₂, TaS₂, and TaSe₂ are metals in bulk form [9], [10], while ReS₂, ReSe₂, MoS₂, MoSe₂, WS₂, WSe₂, MoTe₂, and WTe₂ are semiconductors [11], [12], [13], [14], [15]. Among the various SL TMDCs, group-VIB ones (M=Mo, W; X=S, Se) have been most extensively studied, because they are found to be stable in air at room temperature [16], [17]. Also, we are interested in the optoelectronic properties and SL group-VIB TMDCs show excellent photoabsorption in the visible and near-infrared regions. Therefore, in this short review, we limit ourselves to group-VIB TMDCs such as MoS₂, MoSe₂, WS₂, and WSe₂ and hereafter, the term TMDCs is specific for group-VIB TMDCs unless stated otherwise.

SL TMDCs are direct band gap semiconductors [3], [12] exhibiting a relatively high absorption in the visible and the near-infrared regions, which makes them ideal candidates for atomically thin photo-active materials [1], [18], [19]. Also, in contrast to graphene, SL TMDCs exhibit large intrinsic spin-orbit coupling (SOC) [13], [14], originating from the d-orbitals of the transition-metal atoms. Moreover, transport measurements such as on/off current ratio (up to 1×10⁸) and steep subthreshold swing of 70 mVdec⁻¹ [20], [21] are clearly observed in TMDCs. All these fascinating properties make SL TMDCs promising candidate materials for nanoelectronic, optoelectronic, and spintronic devices.

Large-scale synthesis of 2D materials is one of the significant issues for fabricating practical devices made of layered materials. Different fabrication techniques such as chemical vapor deposition (CVD), physical vapor deposition (PVD), and molecular beam epitaxy (MBE) have been employed to obtain microscale samples. It has been observed that samples obtained through these techniques are highly poly-crystalline and include interfaces such as edges, heterostructures, grain boundaries, and most importantly point defects. These imperfections do not always degrade the materials properties, but they often bring new physics and even useful functionality.

Defects play an important role in tailoring electronic and optical properties of semiconductors and have been the subject of intense research over the last few decades. Point defects in semiconductors can trap charge carriers and localize excitons. Excitons bound to defects, if recombining radiatively, lead to light emission at lower energies than the optical interband transition. The interaction between these defects and charge carriers becomes stronger at reduced dimensionalities, which can be understood in a simplified hydrogenic model of excitons; for example, in three dimensions, shallow defects bind electrons at ground state binding energy equal to 13.6 eV×m*/εr2, where m^* and ε_r are effective mass and relative dielectric constant. This value increases to 54.4 eV×m*/εr2[22] in 2D, simply due to the reduced dimensionality. Recently, in SL WSe₂, the exciton binding energy has been determined to be 0.37 eV, which is about an order of magnitude larger than the one observed in III–V semiconductor quantum wells, and therefore renders the exciton excited states observable even at room temperature [23].

In this short review, we first present electronic and optical properties of SL TMDCs. Then we present a survey of the hitherto reported point defects in SL TMDCs. This includes point defects, line defects, and heterostructures of different SL TMDCs. We will discuss the formation of defects and their influence on the electronic and optical properties of SL TMDCs. We will then briefly review recent progress in optoelectronics and photonics applications of defected TMDCs. Finally, we will discuss the major challenges and future opportunities in this rapidly progressing field of research.

2 Basic electronic structure and optical selection rules

Bulk TMDCs consist of stacked monolayers (MX₂) of atoms which are bonded through out-of-plane weak van der Waals forces, while each monolayer also has three stacked layers (X-M-X) of atoms held together through strong covalent bonds (Figure 1A). Bulk TMDCs have indirect band gaps (Γ and Q-points), lying below the direct gap by 0.6 eV (K and K′-point) in the case of MoS₂ (Figure 1B). However, as the number of layers decreases, strong confinement in the out-of-plane direction leads to a significant increase in the indirect gap while leaving the direct gap unaffected at the K(K′)-point (Figure 1B). This leads to a crossover to a direct-gap material in the limit of SL TMDCs. Thus, band gap engineering can be carried out by changing the number of layers in a given TMDC (Figure 1C).

Figure 1:

Electronic and optical properties of SL TMDCs.

(A) Structure of MX₂, blue spheres represents transition metal M atoms, while brown spheres represent chalcogen X atoms. (B) The energy band structure of bulk (left) and SL (right) pristine MoS₂, which shows that bulk MoS₂ is an indirect while SL MoS₂ is a direct band gap semiconductor. Basic topology of the band structure for all TMDCs is essentially the same, details are, however, different which are summarized in Table. 1. Blue curves show the valance band (VB) and conduction band (CB), while red curves show VB-1 and CB+1. (C) Photoluminescence spectra, which show an order of magnitude difference in absorption between SL and bilayer MoS₂[3]. Also it can be seen that band gap shifts as a function of thickness. (D) A schematic illustration of circular dichroism. Electronic bands around K and K′ points (for instance for MoX₂). Spin and valley degrees of freedom are locked together, which results in optical valley-dependent selection rules.

The electronic states of a crystal transform according to the irreducible representations of the symmetry group of the crystal. SL TMDCs have D_3h point group symmetry. σ_h represents the reflection about the horizontal plane, passing through the origin and perpendicular to the axis with the highest symmetry, and is a symmetry operation of D_3h. SL TMDCs are invariant with respect to the σ_h reflection about the z=0 or M-plane of atoms, where the z-axis is oriented perpendicular to the M-plane of atoms. Therefore, electronic states break down into even and odd or symmetric and antisymmetric states with respect to σ_h. This leads to the nontrivial consequence that TMDCs have two band gaps, an in-plane ε_|| and a substantially larger out-of-plane band gap ε_⊥ [24], [25]. The out-of-plane band gap ε_⊥ is a consequence of the nonzero thickness of SL TMDCs, which is usually neglected due to its substantially larger values for pristine cases. In TMDCs, d-orbitals of transition metal atoms and p-orbitals of chalcogen atoms contribute in the bonding process. From the first principle calculations, it has been shown that at band edges, contribution of the p-orbitals of the chalcogen atoms is small and the electronic structures of SL TMDCs near the band edges can be explained by considering the d-orbitals of the M atoms only [3], [9], [12], [26], especially d_xy, dx2−y2, and dz2. In the Brillouin zone, the Γ-point has the highest symmetry and is described by the D_3h point group, while the K and K′-points have lower symmetry and are described by C_3h point group. By using symmetry adopted basis, a 2-band k·p Hamiltonian [13] can be written as

where a, t, σ_i’s, and Δ are lattice constant, hopping integral, Pauli matrices, and band gap, respectively. These parameters are summarized in Table 1. τ=±1 is the valley index and 2λ_v(c₎ is the spin splitting at the valence (conduction) band edge due to the SOC. A similar Hamiltonian can also be derived by using the tight-binding model [27]. The orbital part of Hamiltonian Eq. (1) (first two terms) looks similar to the Hamiltonian of SL graphene with staggered sublattice potential, which is not surprising as both systems have essentially the same structural properties. The third term in Eq. (1) represents the SOC due to the broken inversion symmetry and due to the presence of heavy transition metal atoms (SOC~150 (400) meV for Mo(W)X₂), in contrast to graphene (SOC~1 meV), where SOC is strongly suppressed due to the inversion symmetry. Spin splitting in SL TMDCs at K and K′ points is opposite in sign due to time reversal symmetry. Therefore, the electromagnetic wave couples differently at the two valleys, i.e. right or left circularly polarized waves excite electrons in one of the valleys only, a phenomenon known as valley-dependent circular dichroism (CD) (Figure 1D). SL TMDCs provide an excellent platform to observe CD because of their hexagonal or honeycomb lattice structure, direct band gap, and large SOC. CD has been observed in MoS₂ [28], [29], [30]. An imbalance of charge carriers in the two valleys or valley polarization can be induced when charge carriers are excited by valley-selective CD. Valley polarization then leads to the so-called valley Hall effect, which has been observed for SL MoS₂[31].

It should be noted that all bands are spin split due to the intrinsic SOC, except at the time reversal invariant points Γ and M. Smaller but significant spin splitting is also present at the conduction band edge due to the intrinsic SOC [32], [33], [34]. Interestingly, depending on the transition metal atom (Mo or W), the spin splittings at the conduction band edge have opposite signs; consequently, the lowest energy transitions in MoX₂ (WX₂) are made by bright (dark) excitons [35].

3 Defects in TMDCs

One of the greatest challenges being faced today in commercializing atomically thin materials is how to produce high-quality samples, on a large scale at low cost, and in a controllable manner. The quality of samples plays crucial role as the presence of irregularities or defects in the crystalline structure have an adverse effect on the electronic and optical properties. The most common and energetically favorable irregularities are point defects, most important of them are antisite defects (Figure 2A) and vacancies (Figure 2B) [45], [46]. Various fabrication techniques such as MBE, PVD, and CVD have been employed to obtain the micrometer or wafer scale samples of SL TMDCs. Although CVD and mechanical exfoliation (ME) have been the most successful fabrication techniques to generate samples with high quality of SL TMDCs, studies show that those samples have an abundance of point defects (Figure 2C, D), most importantly S vacancies in MoS₂ due to the lowest formation energy of these defects. In contrast to MoS₂, scanning tunneling microscope (STM) images show that defects in CVD-grown WSe₂ reside on W sites instead of selenium sites [47]. This shows that the Se vacancy is not the most favorable defect in CVD-grown WSe₂. In addition to the S vacancy, S₂ and Mo vacancies have also been observed in MoS₂[45].

Figure 2:

Growth related point defects, taken from Ref. [45].

(A) Antisite defects in PVD MoS₂ monolayer. Scale bar, 1 nm. (B) Vacancies including V_S and VS2 observed in ME monlayers, similar to that observed for CVD sample. scale bar 1 nm. (C, D) Histogram of various point defects in PVD, CVD, and ME monolayers. ME data are in green, PVD data are in red, and CVD in blue.

The presence of point defects degrades material properties that may result in reduced charge carrier mobility and in poor optical response. It has been observed that experimentally attainable samples of MoS₂ have considerably lower carrier mobilities in the 0.5–3 cm²V⁻¹s⁻¹ range [6] than the theoretically predicted value of 410 cm²V⁻¹s⁻¹ [20], [21]. It has recently been suggested that this discrepancy between the predicted and observed values of carrier mobility is due to the presence of impurities created during the growth process [48], [49]. Point defects can also affect the optical properties, resulting in an inhomogeneous contribution to line width and line shapes in the radiative emission spectrum. Defects break the translational symmetry of the crystalline structure. Thus, in principle, defect states can exist anywhere in the band structure, i.e. within the valence or conduction bands with regular electronic states or most importantly with in the band gap region. Within the dilute limit, localized defect states behave as trapping centers for charge carriers. Excitons bound to these defect states, if recombine radiatively leads to the light emission at the subband gap energies. An important factor characterizing the defect transition from band to band transition is the lifetime of excitons, which is inversely related to the width of the transition peak in the PL spectrum.

Although crystalline imperfections degrade materials properties, but often they can bring new physics and even useful functionalities. Point defects are sometimes required and therefore can be created artificially, e.g. high-energy α-particle irradiation or by thermal annealing of SL TMDCs [50]. Photoluminescence measurements on the MoS₂ samples irradiated by high-energy α-particles show appearence of absorption peaks at subband gap frequencies. More importantly, an increase in the radiation dose, in the nitrogen-rich environment, leads to enhancement of PL intensity of both bound and free excitons.

It has been shown that oxygen irradiation of MoS₂ samples in the presence of S-vacancy leads to an enhancement of PL intensity as a function of exposure time (see Figure 3[51]). The detailed structural analysis shows that an oxygen molecule (O₂) resides at the S-vacancy site in MoS₂. The adsorption of the oxygen molecule at the S-vacancy site exhibits strong bonding, introduces p-type doping in MoS₂, and hence a conversion from trion to exciton. Meanwhile, it has also been observed that the strong oxygen plasma treatment of mechanically exfoliated MoS₂ flakes can be used to tune the PL intensity from very high to complete quenching as a function of exposure time (see Figure 4). This behavior is a result of direct-to-indirect band transformation due to the formation of disordered regions made of MoO₃ in the MoS₂ flakes upon exposure to oxygen plasma [52].

Figure 3:

Defect-activated photoluminescence, taken from Ref. [50].

(A) Nano-Auger spectrum taken on MoS₂ before and after irrediation with α particles at dose 8×10¹³ cm⁻². (B) Raman spectrum of the same. (C) PL spectrum for pristine and irradiated monolayers MoS₂ at shown irradiation doses. The PL was taken at 77 K in N₂ (50 Torr) environment with a constant laser excitation power. The irradiation-caused enhancement in bound exciton (X_235.B) and free exciton (X₀) emission intensity are indicated. (D) Calculated band structure of monolayer MoS₂ in the presence of di-S vacancies. Red levels with in the band gap are levels appearing when S vacancies are introduced. Blue levels appear when N₂ molecules interact with S vacancies. Right panel: Total density of states of the monolayer MoS₂ with S vacancies in the poresence of N₂. Here the modeled vacancy density is 7×10¹³ cm⁻².

Figure 4:

Photoluminescence quenching, taken from Ref. [51], [52].

Left, PL spectra of SL MoS₂ after oxygen plasma irradiation with different durations. The change of PL intensities with plasma irradiation times is shown in the inset. (Right, (A)) Time-dependent photoluminescence (PL) spectra of plasma-treated SL MoS₂. Inset shows the PL intensity of A₁ and B₁ peaks (corresponding to SOC) with respect to the plasma exposure time from pristine SL MoS₂ (red curve) to t₆ (purple curve). (B) Corresponding to A₁ and B₁ excitons, the peaks in PL spectra were fitted with Lorentzian functions.

Presence of certain vacancy defects such as S₂, Mo, and 3×MoS₂ leads to the sharp optical transition in the in-plane and out-of-plane component of the electric susceptibility [24], [25] in MoS₂ in the subband gap region. Specifically, even and odd characters (with respect to σ_h symmetry) of defect states play an important role in determining the in-plane or out-of-plane resonances in the susceptibility tensor [25]. Also, the odd states are essential to understand the out-of-plane optical transitions. It should be noted that odd states are usually neglected for pristine cases.

In addition to point defects, each exfoliated sample naturally has an edge, which breaks the periodicity of the crystalline structure. The chemical properties of edges are very much different from the internal structure. Also the electronic properties are very sensitive to the type of edges, i.e. armchair (AC) and zigzag (ZZ). DFT calculations show that MoS₂ nanoribbon with an AC edge has a very small band gap of 0.56eV, which can be further increased to 0.67eV with hydrogen adsorption [53]. These values are an order of magnitude smaller than the band gap 1.9 eV of MoS₂. An armchair edge has never been found experimentally, instead ZZ edges are always observed for TMDCs because of their considerably low formation energy [54], [55]. Theoretical calculations show that an MoS₂ nanoribbon with ZZ edge is always metallic [53], [56], which is contrary to the experimental findings, which shows strong photoluminescence absorption near the edge of the TMDC flake [51]. Later, it has been explained that oxygen adsorption at the (Mo) ZZ edges, along with complex reconstructions, can lead to a band opening of 1.27eV [57].

Recently, there has been a growing interest [58], [59] in the fabrication of in-plane and out-of-plane or lateral and vertical heterostructures of different SL TMDCs. SL TMDCs are free of dangling bonds; therefore, they can be stacked vertically to form vertical heterostructure, which are bonded through van der Waal’s interaction. Optical properties of few layer TMDCs are substantially different from the SL TMDCs; therefore, vertical heterostructure can be considered as a planar defect, which brings new opportunities for tuning the optical properties of the SL TMDCs. In Ref. [58], realization of one such heterostructure MoS₂/WSe₂ has been reported. The PL measurements show that the resultant bilayer MoS₂/WSe₂ system has a band gap of 1.50 eV, lower than the band gap energy of MoS₂ (~1.9 eV) and WSe₂ (~1.75 eV) (Figure 5). It should be noted that for vertical heterojuction, there is always an indirect transition due to the type-II band alignment. However, the PL intensity of the indirect transition is considerably strong, which shows the strong interlayer coupling of charge carriers. Controllable epitaxial growth enables the formation of seamless and sharp in-plane heterojunctions of two different SL TMDCs, such as MoS₂/WS₂[59]. These heterojuctions are particularly interesting as the properties of these materials change drastically at the atomic level. The lateral heterostructure WS₂/MoS₂ shows photovoltaic effects under illumination due to the formation of a pn-junction in the SL materials in the absence of external gating. Strong and localized PL enhancement has been observed at the lateral interfaces due to the strong built-in electric field originating from the type-II band alignment [59].

Figure 5:

Photoluminescence and absorption from WSe₂/MoS₂ heterobilayers, taken from Ref. [58].

(A) PL spectra of single-layer WSe₂, MoS₂, and the corresponding heterobilayer. (B) Normalized PL (solid lines) and absorbance (dashed lines) spectra of single-layer WSe₂, MoS₂, and the corresponding heterobilayer, where the spectra are normalized to the height of the strongest PL/absorbance peak. (C) Band diagram of WSe₂/MoS₂ heterobilayer under photoexcitation, depicting (1) absorption and exciton generation in WSe₂ and MoS₂ single layers, (2) relaxation of excitons at the MoS₂/WSe₂ interface driven by the band offset, and (3) radiative recombination of spatially indirect excitons. (D) An atomistic illustration of the heterostructure of SL WSe₂/SL MoS2 with few-layer h-BN spacer in the vdW gap. (E) Normalized PL spectra from SL WSe₂/SL MoS2 heterostructure with n layers of hexagonal boron nitrite (n=0, 1, and 3).

4 Optoelectronics with TMDCs

Optoelectronics deals with the emission and detection of photons in a controllable manner. Nanomaterials such as carbon nanotubes and semiconductor quantum dots have been studied for the use in optoelectronic applications such as solar cells, lasers, LEDs, and photodetectors. Optoelectronic devices that are made of transparent materials are gaining immense interest due to their use in solar cells and transparent displays. The electronic band structure of materials is directly related to the absorption or emission of light. Graphene, the first 2D material, is a semimetal and is therefore not a suitable material to absorb light in the pristine form. Certain experimental techniques are devised such as doping and graphene patterning or chemical treatment [60], [61] to induce band gaps and enhance the optical response of graphene, which is necessary for optoelectronic applications. However, all these approaches require complex engineering, which is sometimes hard to accomplish.

In contrast to graphene, SL TMDCs are intrinsically direct band gap semiconductors and therefore provide a more straightforward platform for the realization of optoelectronic devices. SL TMDCs exhibit strong light–matter interaction [62], [63], arising from the band nesting and van Hove singularities in the density of states, in turn resulting in considerable absorption of light. Several studies show that TMDCs absorb 5–10% of the incident light, which is an order of magnitude higher than the conventional 3D semiconductors such as GaAs and Si. Furthermore, strong exciton and trion binding energies which are even visible at room temperature make them promising for the fabrication of high-performance atomically thin optoelectronic devices. The low dimensionality of 2D semiconductors makes them easily tunable, as the electrical, optical, and mechanical properties can all be controlled using multiple modulation methods. Such flexibility offers the potential for device applications such as tunable excitonic optoelectronic devices. The atomically thin van der Waal nature provides a platform for reducing the form factor and substrate-free assembly, compared to semiconductors requiring epitaxial substrates.

4.1 Light-mitting diodes (LEDs)

In LEDs, electrons and holes across a pn-junction recombine radiatively. The efficiency of an LED depends on the radiative recombination rate as there can be nonradiative pathways as well such as Auger recombination. SL materials-based LEDs have advantages over the conventional 3D semiconductor-based LEDs, given that the quantum efficiency of LEDs can be enhanced by avoiding the total internal reflections. An SL TMDCs-based atomically thin LED has been realized in vertical hetrojunction of p-type SL WSe₂ and n-type SL MoS₂. In contrast to conventional LEDs, light emission is not just confined to lateral hetrojuctions, but fast photoresponse is observed throughout the entire WSe₂/MoS₂ overlapping region. However, charge transfer takes place due to the type-II confinement at the vertical interface, which leads to the low quantum efficiency [64]. Also, the indirect band gap formation due to the mismatch of conduction electrons (n-type MoS₂) and valence band holes (p-type WSe₂) leads to low quantum yield of radiative recombination [64]. To enhance the quantum efficiency of SL TMDCs-based LEDs, a band engineering scheme has been devised such that the charge carriers remain confined within the emitting layer. A schematic diagram is shown in Figure 6, consisting of a stack of SL materials. Graphene layers are used for injecting holes and electrons [65]. The injected electrons and holes have to overcome the potential barrier given by hexgonal boron nitrite to reach to an SL of TMDC. This multiple quantum well device has an efficiency of 10%, which is comparable to the efficiency of modern-day organic LEDs [65], [66].

Figure 6:

Van der Waals light-emitting diodes, taken from Ref. [62].

Schematic of a van der Waal’s heterostructure operating as an LED. When a bias voltage V_b is applied to the graphene (G) electrodes, the injected charges recombine radiatively in the monolayer TMDC. (B) Band diagram of the LED with an applied bias. Electrons (e⁻) and holes (h⁺) from the graphene electrodes tunnel through the hexagonal Boron Nitrite dielectric layers (red and blue arrows) and reach the monolayer TMDC.

5 Single quantum emitters

Single photon emission lies at the heart of many applications in photonics, including optical quantum computing [69] and quantum cryptography [70], [71], [72]. To make these applications practical, solid-state single photon sources (SPSs) which can generate well separated (in time) and indistinguishable photons are required. Efforts have been made to develop single photon sources based on single molecules [73], quantum dots, and one-dimensional materials [74].

SPS based on SL WSe₂ have been reported recently in a series of publications [75], [76], [77], [78]. It has been shown that long-lived localized exciton states bound to defects [75] or impurities [77] in SL WSe₂ decay radiatively, thereby emitting antibunched single photons at constant time intervals due to the Pauli exclusion principle Figure 7. In contrast to the free exciton emission, which has a broad line widths of (10 meV), the line width of the emitted single photons is very narrow (0.1 meV) (Figure 7B). The narrow line width of the emission spectra is a signature of the localized nature of the emission states. In a defect-free monolayer of WSe₂, light illumination creates delocalized excited states. However, the presence of defects in SL WSe₂ leads to localized excitonic states that can decay radiatively on a much slower rate, emitting one photon at a time. An atomically thin layer of WSe₂ has potential advantages over traditional solid-state emitters like nitrogen vacancy defects in diamond. For example, defects can exist naturally or can easily be introduced by irradiating the sample with a beam of high-energy particles. Photons from localized emitters in 3D materials have to travel through host media with a high refractive index, whereas in the case of a SL TMDC, photons can be immediately utilized with a strongly enhanced photon extraction efficiency. It may also be easier to integrate them into electronic devices than it is for other solid-state SPSs.

Figure 7:

Emergence of narrow spectral lines and observation of photon antibunching, taken from Ref. [75].

(A) PL spectrum of localized emitters. The left inset is a high-resolution spectrum of the highest intensity peak (single quantum emitter). The right inset is a zoom-in of the SL valley exciton emission, integrated for 60 seconds. The emission of the localized emitters exhibits a red shift and much sharper spectral lines. (B) A histogram comparison of the linewidth between the emission from the delocalized valley exciton and 92 localized emitters, grouped in 23.3 μeV bins. (C) Second-order correlation measurement of the PL from SQE under a 6.8 μW CW laser excitation at 637 nm. The red line is a fit to the data, from which we extract g² (0)=0.14±0.04. (D) Intensity correlation measurement under 9.5 μW (average) pulsed excitation at 696.3 nm with a repetition rate of 82 MHz, and a pulse width of ~3 ps, which revealed g² (0)=0.21±0.06.

6 Outlook and conclusion

In this brief review, we present an overview of the basic electronic and optical properties of pure and defected SL TMDCs, with emphasis on their possible applications in future optoelectronic and photonic devices. SL TMDCs exhibit a number of exciting properties such as high optical response due to the direct band gap, band gap tunability, and strong SOC together with valley structure leading to circular dichroism, which are important both for fundamental research and for industrial applications. The presence of defects in SL TMDCs may lead to an enhancement of optical response at the subband gap energies, which are advantageous in making tunable light detectors. In addition, certain types of point defects in SL TMDCs enable the quantum emission of electrons at regular time intervals, giving rise to the single photon sources. Although new device concepts have shown the potential of SL TMDCs, many challenges have remained to be overcome and new opportunities are needed to be explored. Heterostructures of different SL TMDCs and with recently discovered atomically thin semiconductors such as phophorene [2], [4] provide new opportunities for band gap engineering to achieve desirable functionalities. However, there are certain challenges that should be met and overcome to fully appreciate the technological potential of SL TMDCs. One of the biggest challenges faced by the research community is the fabrication of high-quality, micrometer-sized samples of SL TMDCs to avoid the effects of grain boundaries. For solid-state samples, crystal growth methods need to be improved to achieve large areas, large grain sizes, uniformity, and control on the number of layers. The presence of defects significantly alters the materials’ electronic and optical properties; however, it is encouraging that the effects of defects on SL TMDCs are not necessarily detrimental as long as it is possible to control the creation of the defects. At present, atomic scale synthesis and manipulation of defects remains a challenge. It is also a challenge for experimentalists to introduce specific types of defects into desired locations in controlled concentrations.