Two-Dimensional Doped Materials

Abstract

The recently intensified research in atomically thick two-dimensional (2D) materials has been motivated by their unique properties and the possibility of updating the future electronic and optoelectronic technologies. Doping can change the band structure of a semiconductor and regulate its physical and chemical properties. Doping has a significant effect on the electronic structure of 2D materials due to their atomic thickness. Here, we present a tutorial review of 2D doped materials (except graphene), including various doping types and theoretical calculations, the preparation and characterization methods, and its multifunctional application. Finally, we will summarize by stating the current challenges and future opportunities in the development of 2D doped materials.

1. Introduction

In 2004, A. K. Geim and K. S. Novoselov discovered graphene from bulk graphite, opening the door to the study of two-dimensional materials (2DMs) [1]. Two-dimensional materials possess an atomic thickness and their intralayer atoms are combined by covalent bonds in nature, whereas the interlayers are coupled by weak van der Waals forces, thus allowing the crystal to readily cleave along the layer’s surface [2,3,4,5]. A variety of 2DMs have been found experimentally, including hexagonal boron nitride (hBN), transition metal dichalcogenides (TMDs), and layered metal oxides [6,7,8,9,10,11]. It is reported that there should be more than 1000 kinds of 2DMs in the world by high-throughput calculations [12]. The diverse families of 2DMs display versatile properties with potential applications in field effect transistors, optoelectronic devices, energy storage, and conversion, catalysis, and biomedicine [13,14,15]. In order to maximize the application of 2DMs, it is necessary to develop several approaches to modulate their properties, including constructing a heterostructure, fabricating a novel device, and doping [16,17,18,19].

In classical semiconductor theory, doping is the intentional introduction of impurities into a material, which is fundamental to controlling the properties of bulk semiconductors [20]. A doping semiconductor can change its band structure, including the size of the band gap, the type of the carrier, and the location of the Fermi level, and thus regulate its electrical, optical, electrochemical, and magnetic properties [21,22,23]. The development of modern semiconductor technology is based on the theoretical research and technological innovation of semiconductor doping. Compared with bulk materials, 2DMs are very sensitive to the external factors for the unique structure where almost all the atoms are exposed to the surface. Thus, the band structures of 2DMs can be more validly tuned by various doping approaches [24,25,26,27,28].

Herein, we present a forward-looking review article that discusses the state-of-the-art in 2D doped materials. Initially, we will outline the different types and theoretical calculations of 2D doped materials. Additionally, we will discuss the various strategies to doping 2DMs and how to characterize them. Finally, we will highlight the optical, electronic, magnetic, and electrochemical properties of 2D doped materials and address future directions which have yet to be explored for novel 2D doped materials intended for various practical applications.

2. Classic Semiconductor Doping Type

Doping can tune the electrical property of a semiconductor, which greatly promotes the progress of semiconductor integration technology [29]. There are two main doping technologies for a semiconductor: thermal diffusion and ion implantation. The former is to heat the semiconductor, which intensifies the thermal motion of the atom, so that the intrinsic atom can obtain enough energy to leave a vacancy away from the lattice or make the impurity atoms diffuse in the intrinsic semiconductor. The latter is to ionize impurity atoms into ions and accelerate them with a strong electric field to obtain high kinetic energy, then directly bombard the intrinsic semiconductor to achieve an impurity injection. Here, we will give a brief analysis of the three doping types, including substitutional doping, vacancy, and interstitial doping [30].

2.1. Substitutional Doping

Substitutional doping, which is where one atom replaces another one to form a new semiconductor, is a common type of doping in semiconductor integrated process (Figure 1a). Substitutional doping often occurs in the location where there are vacancies in the material, rather than exchanging positions with the host material. At the same time, the size of the impurity atom and host atom is similar.

2.2. Vacancy

Many semiconductor materials are grown at relatively high temperatures, and there will be more vacancies in the body at high temperatures (Figure 1b). At a certain temperature, there is a certain equilibrium value. When the crystal cools down at a fast rate, some vacancies are frozen to form nonequilibrium vacancies, which affects the physical properties of the semiconductor.

2.3. Interstitial Doping

In the process of a thermal diffusion and ion implantation, a part of the atoms get enough energy to overcome the binding of the surrounding atoms and squeeze into the gap between the atoms to form interstitial atoms (Figure 1c), which will have a great impact on the electrical, optical, and electrochemical properties of semiconductors.

The doping mechanism of n-type and p-type semiconductors are shown in Figure 1d,e, respectively. When the majority carriers of a semiconductor are electrons, they are called n-type semiconductors. When the majority carriers are holes, they are called p-type semiconductors. An n-type semiconductor can be obtained by doping phosphorus in monocrystalline silicon, while a p-type semiconductor can be formed by a doping boron. N-type and p-type semiconductors are the cornerstones of integrated circuits.

After doping in intrinsic semiconductors, the donor and acceptor impurity levels will be introduced. Since E_D is very close to E_C, the electron at the donor level can transit from the donor-bound state to the conduction band to become a conducting electron, thus increasing the electron concentration and forming an n-type semiconductor (Figure 1d). Similar to this, since E_A is very close to E_V, the hole at the acceptor level can transit from the acceptor bound state to the valence band to become a conducting hole, thus increasing the hole concentration and forming a p-type semiconductor (Figure 1e).

3. Doping Types in 2D Doped Materials

Due to the unique structure of 2DMs, almost all atoms are exposed on the surface. Compared with the bulk materials, we conclude six doping types in 2DMs (Figure 2). Because 2D doped TMDs have been widely reported, we will take TMDs as an example to show the various doping types, which includes the transition elements and chalcogenide elements substitution, vacancy, intercalation, and the physical and chemical adsorption.

3.1. Substitutional Doping

Two-dimensional alloys have been reported early in MX₂ (M = Mo, W, Re, Hf, Sn; X = S, Se, Te) [31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48]. It is a substitutional doping type and can be classified into a transition elements substitution (M_xM’_1-xX₂, x represents the content of M) (Figure 2a) and chalcogenide elements substitution (MX_2xX’_2(1-x), x is the content of X) (Figure 2b). The doping concentration can be controlled by tuning the growth parameters, and the distribution of the doped atoms are random [33,49,50].

3.2. Vacancy

It is well known that vacancy defects either naturally formed or artificially made have a great influence on the electronic structure of 2DMs (Figure 2c). A theoretical calculation has shown that S vacancies in monolayer MoS₂ will form donor-like levels and lead to n-type conductivities, while Mo vacancies bring acceptor-like levels and p-type conductivities [51]. Thus, vacancy is a common doping form and can determine the carrier type of 2D semiconductors.

3.3. Intercalation

The layers of 2DMs are combined by a van der Waals force. Therefore, the atoms and molecules can be inserted into the van der Waals layers by chemical methods [52]. These intercalated ones can change the electronic properties of the 2DMs, including the transformation of 2DMs from n-type to p-type [53], tuning the magnetic property [54,55], or bringing them into a phase transformation [56]. In our opinion, this intercalation should belong to the one of doping, and some of the doping will lead to a change in the space distance between the van der Waals layers (Figure 2d).

3.4. Adsorption

In addition to the atomic intercalation, the adsorption can also be an important doping type. When 2DMs exist in air or are specially treated, molecules and particles can adsorb on the surface of 2DMs. These molecules and particles will transfer the electron with 2DMs [57,58,59,60,61]. As the specific surface area of the 2DMs is large, the adsorption will seriously affect their electronic structure, thus we also classify this situation as a type of doping. Generally speaking, adsorption can be divided into two types: physical adsorption (Figure 2e) and chemical adsorption (Figure 2f). Physical adsorption refers to the formation of a van der Waals force between the adsorption and 2DMs, while the chemical adsorption means the formation of a covalent bond between the adsorption and 2DMs. In the experiment, we can choose different adsorbents to enhance their influence on 2DMs, which will affect the electrical transport and photoelectric properties of 2DMs [61].

4. Theoretical Investigation of 2D Doped Materials

Up to now, researchers have studied the effects of various kinds of doping in 2DMs by a theoretical calculation, including substitutional doping, intercalation, and adsorption. A theoretical calculation can analyze the formation energy of the doping to determine the possibility of doping. At the same time, it can study the influence of doping on the band structure of 2D materials to predict the influence of doping on the electrical, optical, and magnetic properties of 2D materials. Compared with the experiment, the theoretical calculation is relatively simple, and the results obtained may not be completely consistent with the experimental results, but it will have a certain guiding significance for the experiment. Kang et al. investigated the stability of 2D alloys MX_2(1-x)X’_2x (M = Mo, W, and X, X’ = S, Se, Te) by employing the cluster expansion method and the special quasi-random structure approach (Figure 3a) [62]. The stability of the alloys at 0 K can be evaluated from their formation enthalpies (ΔH) which depends on the specific atomic arrangement in the alloys. The results show that for many configurations of this alloy, ΔH is negative, which means that the ordered alloy can form spontaneously (Figure 3b). There are three stable ground states, with concentration x equal to 1/3, 1/2, and 2/3 (Figure 3a). In these ground states, the clustering of the S or Se atoms is not stable, and the S (Se) atoms prefer to occupy the neighbor sites of the Se (S) atoms. These results show that the 2D alloy is thermally stable in theory [62]. Because of the high temperature in the synthesis of 2D alloys, it is difficult to observe the ordered situation in the experiment.

W atoms can be doped arbitrarily in MoS₂ to form an alloy, while other metal atoms, such as Mn and Co, cannot be doped in a high concentration due to a lattice mismatch, thus only a small amount can be done. Cheng et al. investigated various atoms doped on the MoS₂ monolayer by density functional theory [64]. To check the stability of different dopants, they have calculated the binding energies as summarized in Figure 3d. The binding energy is defined by the equation:

$$E_b=E_v+\mu -E_d$$

(1)

where E_v is the energy of a relaxed MoS₂ monolayer with one Mo vacancy, μ is the energy of the doped transition metal atom, and E_d is the energy of a relaxed MoS₂ monolayer with one Mo atom replaced by the transition metal. The result demonstrated that E_b of different dopants varies from 0.2 to 16.0 eV. The binding energy at IIIB to VIII10 is relatively high (Figure 3d). It is concluded that the substitution doping of the transition metal atoms from the IIIB to VIII 10 groups is stable [64]. Doping transition metal atoms from the IB and IIB groups is difficult because of their low binding energy (especially for Hg). For magnetic atoms, such as Mn, Fe, and Co, the doping of MoS₂ can be a promising approach to achieve 2D diluted magnetic semiconductors [63,64,67,68]. Figure 3c shows the long-range interaction of Mn_Mo defects through the Se 4p state in the spin-polar isosurface diagram [63]. Two Mn_Mo defects were obtained, separated by a distance of ~6.5Å in MoSe₂. The result shows that the coupling between the Mn spins and Se spins in the AFM resulted from the hybridization between the localized Mn 3d and the delocalized Se 4p states. More importantly, every time the Se spins come across a Mn_Mo defect, the AFM alignment between the Se and Mn leads to an effective FM alignment of all the Mn_Mo spins. Because the Se 4p states are delocalized, the FM alignment between the Mn_Mo spins can also be expected to act over a long range. The ferromagnetism will be more stable and the Curie temperature will be higher with the increase in the Mn doping concentration [63].

It has been reported that the electronic and phase structure of 2DMs can be tuned by inserting atoms and ions into the van der Waals gap [69]. The mechanical exfoliation method always leads to a 2D trigonal prismatic phase (labeled as H-MoS₂), which is found to be a semiconductor with a direct band gap between the occupied d_z2 and the empty d_{x2−y2, xy} orbitals [70]. The space group of the H-MoS₂ phase is P6/mmc, which is considered to be a stable configuration under normal conditions [71]. However, the solvent-based exfoliation method results in different octahedral coordinated phases with each Mo atom bonded with six S atoms, which is referred to as T-MoS₂ [72,73]. The structure of MoS₂, which is similar to WTe₂ with zigzag Mo−Mo chains rather than that of TiS₂, is labeled as ZT-MoS₂. Lithium-inserted MoS₂ possesses a distorted octahedral coordinated structure with a rhombus shape connected Mo–Mo chains, which is referred to as DT-MoS₂ [74]. O-MoS₂ is used to describe all three octahedral coordinated MoS₂ regardless of the detail of the molybdenum clusterization. M. Kan et al. discussed the possible transition paths of different phases of MoS₂. As the concentration of the lithium atoms is increased, H-MoS₂ becomes O-MoS₂ and then becomes DT-MoS₂. When all the Li atoms are extracted from the system, the structure is ZT-MoS₂. By heating or aging, ZT-MoS₂ will transform into the most stable phase H-MoS₂ (Figure 3e) [65].

Up to now, MoS₂ transistors have shown ultrahigh gas sensing capabilities for NO and NH₃. This is mainly because the adsorbed gas changes the electronic structure of MoS₂ by a charge transfer mechanism. The adsorption energy of various gas molecules on the MoS₂ monolayer was determined by five methods (Figure 3f) [66]. For the same molecule, a different calculate method results in a different adsorption energy, while the trend of the adsorption energy is consistent with that of the gas molecules. The adsorption energy of NO₂ and NO are high, which means that they can be easy to adsorb on MoS₂ (Figure 3f). For all the molecules, no chemical bonding is found between the adsorbate and MoS₂. To further elucidate the charge transfer between the gas adsorbate and the MoS₂ monolayer, Figure 3g provides an isosurface map of the difference in the electron charge density of these gas molecules. There is a charge accumulation on the surface between the MoS₂ and NO adsorbate, demonstrating the charge-donor characteristics of the molecules, which is also reflected by a high adsorption energy between them (Figure 3f) [66].

5. Synthesis of 2D Doped Materials

Compared with pure 2DMs, the synthesis of 2D doped materials is more challenging. Here, we summarized the methods to synthesize 2D doped materials, including a mechanical exfoliation, surface functionalization, vapor phase deposition, plasma technology, absorption, the electrochemical method, and thermal evaporation.

5.1. Mechanical Exfoliation Method

The mechanical exfoliation method was well known after graphene was exfoliated through tape in 2004 [1]. Because of the weak van der Waals force between the layers of layered materials, ultrathin and high-quality 2DMs can be obtained by this method (Figure 4a). The bulk counterparts of 2D doped crystals can be obtained by a chemical vapor transport (CVT) method where the source material is placed in the quartz tube by vacuum sealing technology and the samples are grown under an appropriate temperature. This method can obtain high-quality 2D substitution-doped materials with a low cost, but the layer thickness of 2D materials is uncontrollable, the efficiency is low, and large-scale production is difficult to achieve. Xie et al. have firstly exfoliated the Mo_1−xW_xS₂ monolayer alloy from its bulk counterparts and shows its good optical performance [31].

5.2. Surface Functionalization

Surface functionalization is an adsorption-doped method and is possible to integrate various compositions of 2D functional units on single few-layered 2DMs. This method is simple to operate, but the stability needs to be improved because the dopants are on the surface and the interaction with the 2D materials is weak. Many 2DMs, such as MoS₂, MoSe₂, and InSe, have chalcogenide atoms with one bond occupied by lone pair electrons. According to the Pauli exclusion principle, the fully occupied orbital cannot accept extra electrons to form a chemical bond, resulting in an inert 2D surface. However, Lewis acids, featuring empty electron orbitals, can accept these lone pair electrons and form stable coordinate covalent bonds [75]. Ajayan et al. made use of the lone pair electrons found in most of 2D TMDCs and reported a functionalization method to tune their electronic transport via a Lewis acid–base reaction that does not alter the host structure (Figure 4b) [75].

5.3. Vapor Phase Deposition Method

Vapor phase deposition, including physical vapor deposition (PVD) and chemical vapor deposition (CVD), is a very common method to obtain 2D doped materials [21]. This method can control the thickness and size of 2D substitution-doped materials and can achieve the growth of large-area and high-quality 2D materials. Pan et al. developed a PVD method to grow a 2D alloy with a controlled concentration [76]. The precursor is usually the corresponding powder of a chalcogenide semiconductor. The two boats in the system are linked together and can be pushed optionally during the growth process. The vapor concentration of the two materials can be controlled precisely to modulate the chemical composition of the alloy [35,36,76]. Until now, there are mainly three CVD methods for synthesizing the 2D doped materials: (1) a direct synthesis from various source materials by a chemical reaction. In the synthesis of Co-doped MoS₂, Co₃O₄, MoO₃, and S are used as the source materials in a CVD system [81]. (2) Metal alloys were obtained by a vacuum evaporation on the substrate, and then vulcanized in the CVD system. Er-doped Mo thin films were deposited on SiO₂/Si substrates by sputtering high purity Er and Mo targets with a different power to achieve Er doping into the films. Then, Er-doped MoS₂ layered nanosheets were synthesized by the CVD method [77]. (3) The CVD approach combined with a solution-processed precursor deposition. The MoO₃ and WO₃ solution were prepared by adding MoO₃ and WO₃ powder into NH₄OH and then annealed to form films, respectively. The films can be used as source materials to form a 2D alloy by the CVD approach [82].

5.4. Plasma Treatment

Plasma technology is used for doping 2DMs because of its simplicity and universality. It is well known that in chemically grown 2DMs, vacancy defects are particularly noticeable due to imperfections in the growth process. It is an effective approach to obtain 2DMs with a vacancy defect by an oxygen plasma exposure and hydrogen treatment (Figure 4e) [78]. Plasma can impact 2DMs to form vacancy defects, and H₂ reacts with 2DMs to form a vacancy. The concentration of the vacancy increases with the increase in the dose and exposure time in the oxygen plasma’s exposure and temperature, and the hydrogen concentration in the hydrogen treatment [78].

5.5. Molecular Absorption Method

A molecular physical and chemical absorption method are extensively reported to achieve both n- and p-doping in 2DMs utilizing the surface charge transfer mechanisms which are aided. This method is simple to operate, but the stability needs to be improved because the dopants are on the surface and the interaction with the 2D materials is weak. Figure 4f shows that NO₂ is located on WSe₂ by physical and chemical absorption, and the sample is placed inside an evacuated Pyrex glass flask with an inert gas flow for the gas exchange [60,83]. Subsequently, a constant NO₂ gas flow is introduced through the flask and maintained throughout the entire doping procedure, exhausting an external oil bath. Simultaneously, the flask is placed on a hot plate with the sample temperature maintained at a certain temperature and cooled down to room temperature [60].

5.6. Electrochemical Intercalation Method

Electrochemical intercalation has been recently used for doping 2DMs and obtaining a 2D superlattice. This method can obtain 2D intercalated materials with novel structures and physical properties, but it is difficult to achieve a mass preparation. Wang et al. used a home-designed electrochemical-optical measurement platform for the in situ monitoring of the intercalation dynamics and the corresponding evolution of the electronic and optical properties (Figure 4g) [79]. This method can be used for several different 2D atomic crystals, such as black phosphorus, molybdenum disulfide, and tungsten diselenide, to produce a broad class of superlattices with tailored molecular structures, interlayer distances, phase compositions, electronic, and optical properties [79].

5.7. Thermal Evaporation Method

The thermal evaporation method is carried out to product Au-decorated WSe₂ with a p-type doping [80]. This method can achieve adsorption doping, but its ability to precisely control the physical properties is weak. A more p-type behavior of the decorated WSe₂ field effect transistor is realized by the electron transfer from the WSe₂ to the gold (Au) decorated on the WSe₂ surfaces. This simple and scalable method is considered to be a promising method to introduce p-doping to other 2DMs (Figure 4h) [80].

6. Characterization of 2D Doped Materials

There are mainly seven methods to characterize 2D doped materials, including optical microscopy, X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), high angle annual dark field scanning transmission electron microscopy (HAADF-STEM), aberration-corrected transmission electron microscopy (TEM), and Raman and photoluminescence (PL) spectra.

6.1. Optical Microscopy

Two-dimensional doped materials with a different electronic structure and band gap may show a different color optical image. It is reported that the colors of Cu-SnS₂ and Co-SnS₂ become more opaque and turn dark blue and violet-red, respectively, from the light blue color of SnS₂ (Figure 5a) [53].

6.2. Scanning Transmission Electron Microscopy

High angle annual dark field scanning transmission electron microscopy (HAADF-STEM) is one direct way to characterize the atomic arrangement in 2D substitution-doped materials, which has an atomic resolution and Z-contrast capability (contrast proportional to ~Z^1.5−1.9) [86,87]. The HAADF-STEM image has been carried on 2D alloys [31,33,41,47,88,89], Mn-MoS₂ [90], Fe-MoS₂ [91], Fe-SnS₂ [92], Nb-MoS₂ [84], and Ti-InSe [75]. In the MoS_2(1−x)Se_2x alloy, the HAADF-STEM image is used to quantify the concentration of the Se local atom on the nanometer scale and map the atomic distribution of the Se dopants, where a single Se substitution site and a double substitution site are indicated in bright green and white, respectively (Figure 5b,c) [47].

6.3. Raman Spectroscopy

Raman spectroscopy has been widely used in 2D substitution-doped materials to characterize the phonon structure, including the crystallinity, composition, and doping. Modified random element isodisplacement (MREI) has been successfully used to explain the Raman behavior of the 2D alloys. The MREI model is based on the assumptions of isodisplacement and randomness: (1) the same kind of atoms vibrate in the same amplitude and phase and (2) all the substituted atoms are randomly distributed [32]. The MREI model can predict the mode behaviors and fit the composition-dependent Raman frequencies of A’₁and E’ modes in a 2D alloy [32]. In the Raman spectra of Mo_1-xW_xSe₂, when the W component x is increased from 0 to 1, the A_1g mode continuously changes from 240.1 to 249.3 cm⁻¹, which is close to the linear dependence on the W component (Figure 5d). In addition, as the W component changes, the second-order Raman peak also shifts and exhibits a dual-mode behavior [48]. The photoluminescence spectroscopy (PL) measurements are a non-destructive measurement method that quickly and easily characterizes the defects, impurities, and luminescence properties of 2DMs. In the PL spectrum of Mo_1-xW_xSe₂, as the W component increases, the PL emission exhibits a bowing effect (Figure 5e). The emission energy is first red-shifted to a minimum at x = 0.2–0.3, and then monotonically blue-shifted [48].

6.4. X-ray Photoelectron Spectroscopy

XPS is an effective characterization for understanding the chemical states and composition of elements in 2D substitution-doped materials. The binding energy of the XPS spectra is referenced to the Fermi level in the material; the red/blue shifts of the binding energy can be interpreted by the move down/up of the Fermi level in the semiconductor, demonstrating p/n doping, respectively [93,94,95]. The XPS has been used to prove the doping concentration in Nb-MoS₂ and it is found that the core-level peaks of Mo and S in the doped sample show a uniform shift toward the lower binding energies compared to those of the undoped one, demonstrating p doping (Figure 5f) [84].

6.5. Aberration-Corrected Transmission Electron Microscopy

Aberration-corrected transmission electron microscopy (TEM) is used to characterize the vacancy defect in 2DMs. A typical atomic image of a single-layer MoS₂ showed the expected hexagonal symmetry, and the defects can be clearly distinguished by analyzing the intensity profile (Figure 5g) [85].

6.6. X-ray Diffraction Pattern

XRD is a powerful tool for analyzing the crystal structure of materials and is powerful to characterize 2D intercalated materials. The crystal plane spacing can be calculated from the XRD pattern by the Bragg equation. An intercalation in 2DMs will result in the increasing of the (001) crystal plane spacing, which can be characterized by XRD (Figure 5h) [79,96].

7. Properties and Applications of 2D Doped Materials

A wide variety of 2DMs have been used as the active material to fabricate field-effect transistors (FETs) [97,98,99], a photodetector [100], and hydrogen evolution catalysts [101]. Doping 2DMs can regulate their properties to meet the practical application [102]. We will focus on the applications of 2D doped materials in electrical, optoelectronic, catalytic, and magnetic fields.

7.1. Electrical Transport and Devices

To reduce the channel length in the current silicon-based transistor, the thickness of the semiconductor must also be reduced to ensure that 2D electrostatic screening effects are minimized [103]. Two-dimensional semiconductors with an atomic thickness are suitable for field effective transistors (FETs) because of their compatibility and the lack of short-channel effects. The mobility (μ) and current ON/OFF ratio (I_ON/I_OFF) are performance indicators of FETs. μ is an important physical parameter to indicate the speed of the carrier electron movement under an electric field. A high mobility can reduce the power consumption of devices and improve the current carrying capacity and operating frequency of devices. The mobility is decided by the mean-free time (τ) and effective mass (m*) of the electron as follows:

$$\mathsf{\mu }=\frac{q\tau }{m^{*}}$$

(2)

where m* is related to the band structure. Because doping tends to increase the scattering and τ, it is a feasible method to reduce the effective quality by selecting specific dopants. On the other hand, doping can also achieve a low contact resistance and improve the mobility of 2DMs by adjusting the Fermi energy level to adjust the work function.

A large ON/OFF ratio can contribute to a reduced power dissipation and static leakage current. The ON current (I_ON) is related to the carrier mobility, saturation velocity, and metal–semiconductor contact resistance, and has no direct relation with the bandgap [104]. The OFF current (I_OFF) of the transistor is strongly dependent on the bandgap (E_g), according to the equation:

$${\mathrm{I}}_{\mathrm{OFF}}\propto exp\frac{E_g}{\alpha KT}$$

(3)

where K is the Boltzmann constant, T is the temperature, and α is a constant.

Doping can change the band’s structure and affect the electrical properties of 2DMs. In a typical n-type 2D semiconductor, doping can introduce more electrons or holes, resulting in the Fermi level closing to the conduction band or valence band (Figure 6a). It can even turn an n-type semiconductor into a p-type. So, more n-type and p-type materials can be obtained by doping, which is the cornerstone of the whole integrated circuit. There are some doping methods to control the conductive type of 2D semiconductors: substitution, surface functionalization, and intercalation.

Natural MoS₂ show n-type. Niobium (Nb) has one less valence electron than Mo and is suggested as the suitable acceptor [106]. It was reported that 0.5%Nb substitutional doped MoS₂ show p-type with a degenerate hole density of ~3 × 10¹⁹ cm⁻³ (Figure 6b) [84]. Two-dimensional TMDs have a van der Waals gap between two neighboring chalcogenide atoms layers and the chalcogen ions on the 2D surfaces are capped with lone pair electron orbitals [75]. On the basis of the Pauli exclusion principle, the fully occupied orbital cannot accept extra electrons to form a chemical bond, resulting in an inert 2D surface. However, Lewis acids (e.g., TiCl₄), possessing empty electron orbitals of the metallic ions, can accept these lone pair electrons of chalcogenide atoms and form stable coordinate covalent bonds [75]. Thus, some n-type 2D semiconductors can be changed into the p-type by the surface functionalization doping method via a Lewis acid–base reaction (Figure 6c) [76]. There are van der Waal gaps in 2D layered materials, and heteroatom can be intercalated into the gap by a solvent-based method. The electrons of intercalated heteroatom will couple with those of 2DMs, thus changing the band structure of the 2DMs. Gong et al. reported that pristine bilayer SnS₂ show the n-type, the Cu intercalated bilayer SnS₂ displays a hole field-effect mobility of ~40 cm² V⁻¹ s⁻¹, and the Co intercalated bilayer SnS₂ exhibits a metal behavior with a sheet resistance comparable to that of few-layer graphene (Figure 6d) [53]. Chalcogenide atoms in 2D TMD can easily form vacancies, which can be occupied by a molecular doping. A new type of 2D doped material, such as Cl-doped MoS₂ (Figure 6e), can shorten the width of the Schottky barrier between MoS₂ and metal, and thus reduce its contact resistance (Figure 6f) [61]. The 2D vertical superlattice can be formed by inserting organic molecules into 2DMs by the electrochemical method [107]; it can increase the interlayer spacing and change the electronic structure of materials and make the multi-layer 2DMs exhibit single-layer properties (Figure 6g). At the same time, the stability of 2DMs in air can be improved (Figure 6h) [79]. The electronic structure of 2DMs can be changed by the charge transfer between the organic compound and 2DMs by placing some electron donor organic compound on the surface of 2DMs by chemical method. When benzyl viologen (BV), one of the highest reduction potentials of all electron donor organic compounds, as a surface charge transfer donor for MoS₂ flakes can obtain a high electron sheet density of ~1.2 × 10¹³ cm⁻², which corresponds to the degenerate doping limit for MoS₂ (Figure 6i) [105].

7.2. Optoelectronics, Catalysis and Magnetism

In an optoelectronic device based on 2DMs, the photocurrent (I_ph) is defined by the equation:

$$I_{ph}=\left(\frac{Pet\alpha }{hv}\right){\tau }_l/{\tau }_t$$

(4)

where P is the incident power, e is the electron charge, t is the thickness of 2DMs, α is the absorption coefficient of 2DMs at the photon energy hv, τ_l is the minority carrier lifetime, and τ_t is the transit time of the electron in 2DMs. The photoresponsivity R of the device can be described by the equation:

$$R=I_{ph}/P=\left(\frac{et\alpha }{hv}\right){\tau }_l/{\tau }_t$$

(5)

When the illumination reaches the sample, the electron–hole pairs are generated [108,109]. The photoconductivity is enhanced by decreasing the recombination of these electron–hole pairs via the trap states to capture the minority carriers to prolong their τ_l, and at the same time, most carriers are allowed to transmit effectively multiple times over the channel between the source and drain the electrode.

Doping often leads to an increase in the defects in 2DMs, which leads to the increase in the trapped states in the band structures, τ_l, and the photoresponsivity [110,111,112]. On the other hand, the defect trap states will prolong the photo response time. Doping can lead to the introduction of impurity levels or a changing of the band structure, which can improve the optical absorption coefficient. Two-dimensional SnSSe as a phototransistor is demonstrated to exhibit a high photoresponsivity of about 6000 A W⁻¹ due to the presence of Van Hove singularities in the band structure introduced by doping (Figure 7a) [44]. CoMoS and NiMoS formed by Co and Ni doping MoS₂ have excellent hydrogen evolution reaction (HER) performances. For an HER, the catalytic performance is closely related to the adsorption energy of hydrogen and the kinetic energy barrier of the hydrogen generation pathway. The introduction of a foreign metal element (e.g., Co, Ni, or Fe) in the MoS₂ lattice affords the opportunity to engineer the electronic and/or surface structures for improving the HERs performances. The optimized Co-doped MoS₂ catalyst shows a superior HER performance with a high exchange current density of 0.03 mA·cm⁻², a low onset potential of 90 mV, and a small Tafel slope of 50 mV·dec⁻¹, which also exhibits the excellent stability of 10,000 cycles with a negligible loss of the cathodic current (Figure 7b) [113]. Lithium intercalation in 2D layered materials can be used to tune the phase. The 2H-1T phase of MoS₂ transform during the intercalation reaction, which is driven by an electron transfer from Li to MoS₂, causing a change in the electron count from d² to d³, leading to the destabilization of the original crystal structure [114]. The plasma-assisted approach (SF₆, CHF₃, CF₄, and O₂ plasmas) is an effective method to make a change from n-type MoS₂ to p-type MoS₂ by doping (Figure 7d). Because of the relatively strong electronegativity of the F and O atoms, the excess electrons are preferentially transferred from the MoS₂ layers onto the F and O atoms, resulting in p-doping in the MoS₂ layers [115,116,117]. Charge doping is used to tune the phase of TMDs by modulating the Fermi level and changing the electronic properties. The excitation of the plasmon modes in Au nanoparticles deposited on a MoS₂ monolayer may induce a transient reversible 2H to 1T phase transition in MoS₂ by hot electron doping (Figure 7e). The doping of the 2H phase first leads to the destabilization of the lattice, and then, through the population of the Mo 4d orbitals, into a structural transition to the 1T phase. The Raman spectra taken with and without Au NPs further confirms this phase transition (Figure 7f) [118]. The magnetic atoms doped in TMDs may induce a ferromagnetism by p–d antiferromagnetic coupling (Figure 7g). It is reported that undoped SnS₂ show no magnetism (Figure 7h), but Fe-doped SnS₂ with a Curie temperature of 31 K show a ferromagnetism under 2 K (Figure 7i) [92,119].

8. Conclusions and Outlook

Doping is an important technology in the field of semiconductor science, which is also very important for the future device application research of 2DMs. In this review, we systematically introduced 2D doped materials, including various doping types and theoretical calculations, preparation and characterization methods, and a multifunctional application.

Although in the past few years great progress has been achieved in the field of 2D doped materials, there still exist a number of significant issues that need to be addressed and studied. First, the device size of 2DMs will continue to shrink, and the doped atoms will have a great impact on the device’s performance. Therefore, it is urgent to develop a 2DM doping method with a large-scale, universal, and atomic accurate. Second, the electrical, optoelectronic, catalytic, and magnetic properties of 2DMs have been widely studied. How to select appropriate doped atoms to significantly improve these performance needs further research, including theoretical and experimental research. Third, most of the 2D doped materials are still focused on basic research. In order to further promote their practical application, they need to be constructed into multi-functional devices, such as the p-n junction formed by 2D p-type doped and n-type doped materials, and magnetic doped semiconductors used in semiconductor spintronic devices.