Advances in Two-Dimensional Magnetic Semiconductors via Substitutional Doping of Transition Metal Dichalcogenides

Transition metal dichalcogenides (TMDs) are two-dimensional (2D) materials with remarkable electrical, optical, and chemical properties. One promising strategy to tailor the properties of TMDs is to create alloys through a dopant-induced modification. Dopants can introduce additional states within the bandgap of TMDs, leading to changes in their optical, electronic, and magnetic properties. This paper overviews chemical vapor deposition (CVD) methods to introduce dopants into TMD monolayers, and discusses the advantages, limitations, and their impacts on the structural, electrical, optical, and magnetic properties of substitutionally doped TMDs. The dopants in TMDs modify the density and type of carriers in the material, thereby influencing the optical properties of the materials. The magnetic moment and circular dichroism in magnetic TMDs are also strongly affected by doping, which enhances the magnetic signal in the material. Finally, we highlight the different doping-induced magnetic properties of TMDs, including superexchange-induced ferromagnetism and valley Zeeman shift. Overall, this review paper provides a comprehensive summary of magnetic TMDs synthesized via CVD, which can guide future research on doped TMDs for various applications, such as spintronics, optoelectronics, and magnetic memory devices.


Introduction
The van der Waals crystals are two-dimensional (2D) material systems with atomically thin, layered structures. Recent discoveries in 2D materials include ferromagnetism in atomically thin layers of chromium-based alloys [1]. For example, Cr 2 Ge 2 Te 6 demonstrates a ferromagnetic order that remarkably influences the transition temperature, even with small magnetic fields [2]. Mechanically exfoliated monolayer chromium triiodide (CrI 3 ) also achieves ferromagnetic properties with out-of-plane spin orientation under 45 K [3]. While these van der Waals ferromagnets remain either metallic or insulating, spintronics and solid-state quantum information science applications benefit from semiconductors with ferromagnetic properties known as dilute magnetic semiconductors (DMSs). While these DMS materials have been extensively researched for decades in their bulk form, of which typical representatives are (Cd, Mn)Te, (Zn, Co)O, and (Pb, Eu)S [4], transitioning bulk dilute magnetic semiconductors (DMSs) into the 2D field remains challenging [5,6]. For example, MnTe has shown antiferromagnetic behavior that can be transverse to paramagnetic by reducing the thickness of the material to the few-layer regime via exfoliation [7]. Despite its promising properties in its monolayer form, achieving a monolayer has proven to be challenging due to the strong interlayer bonding between the layers [8].
Several theoretical studies have predicted that DMSs based on TMDs would exhibit ferromagnetic behaviors even at room temperature, which is a fundamental requirement for practical applications [9][10][11][12][13][14]. For example, delaminating a non-van der Waals (vdW) magnet, MnSe 2 has been reported to have long-range magnetic ordering due to the ferromagnetically coupled Mn atoms, where the Curie temperature reaches room temperature with a 5% applied strain [15,16]. First-principles calculations anticipate that Feand V-doped MoSe 2 would exhibit room-temperature out-of-plane ferromagnets at high atomic substitutions [9].
Several experimental studies have utilized ex-situ techniques to dope transition metal atoms (V, Cr, Fe, etc.) in post-processing techniques into TMDs [17]. Adsorption and intercalation of transition metals and non-metal elements have introduced para-or ferromagnetism into MoS 2 crystals for spintronic applications [18,19]. While these approaches have shown promising routes in creating novel magnetic properties in 2D materials, they resulted in a Curie temperature well below room temperature or random local clustering of magnetic precipitations. Other attempts to incorporate magnetic atoms directly in situ during growth relied upon converting bulk TMD crystals into DMSs. For example, Fe was doped in SnS 2 bulk crystal via a direct vapor phase method, while mechanical exfoliation was required to reveal the van der Waals DMS [20]. Recently, the direct in-situ growth of transition metal-doped TMDs via chemical vapor deposition (CVD) overcame some of the difficulties of post-processing conversion approaches [21]. For example, rhenium (Re), vanadium (V), or iron (Fe) were proven to substitute either molybdenum (Mo) or tungsten (W) sites in monolayer MoS 2 , WS 2 , and WSe 2 crystalline [22].
Two-dimensional DMS offers intriguing applications combining magnetism with semiconducting properties, including magneto-optoelectronic devices for spintronics, optics, and the quantum field. An Fe-doped SnS 2 homojunction device was reported to have an unsaturated magnetoresistance of 1800% [23]. A giant magnetic circular dichroism was demonstrated by the magneto-optical effects in monolayer, bilayer, and trilayer DMSs, leading to a significant MO Kerr rotation and Faraday rotation angles. [24]. Reports on device applications using 2D magnets include the magnetic tunnel junction in heterostructure devices with 2D layers of Fe 3 GeTe 2 [25], Cr 2 Ge 2 Te 6 [26], and CrI 3 [27]. Ferromagnetic properties in doped TMDs provide potential opportunities for many groundbreaking applications. Here we provide our perspective on the synthesis, characterization, and magnetic properties of magnetic 2D transition metal dichalcogenides (TMDs) formed via substitutional doping with CVD.

Effects of Magnetic Doping in 2D TMDs
Ferromagnetism in 2D limits can be induced by substituting the atoms in the d-block of the periodic table with the original transition metal atoms or chalcogen atoms, creating 2D DMS materials. Transition metal dopants and transition atoms in TMD lattices are polarized primarily by their localized 3d and 5d electrons. For instance, based on the density functional theory, a hybridization of localized transition metal-3d, delocalized Se-4p, or W-5d states results in electronic structures of spin-up and spin-down channels in the WSe 2 lattice, leading to the ferromagnetic coupling between the doped transition metal atom spins and Se or W spins [11]. Similarly, due to low formation energy, Mn dopants in MoS 2 tend to replace Mo atoms in Mn-doped MoS 2 crystals [28]. Upon substitution, a portion of the Mn dopants' valence electrons (3d 6 4s 1 ) move to the nearest S atoms, and the remaining valence electrons cause spin polarization, creating a magnetic moment [29]. Dopant concentrations also affect the magnetic-exchange coupling related to ferromagnetic and antiferromagnetic properties. Although detecting the concentration of dopants in all areas of TMDs remains challenging, first principal calculations have shown that the clustering and aggregation of dopants can significantly impact the resulting magnetic moment [9,30]. Prior research has demonstrated that isolated Mn doping at the Mo site of MoS 2 defects often cluster together [12,31], the clusters of Mn-Mo defects with low formation energies can preserve magnetic ordering, while Mn-Mn clusters lead to strong local lattice distortion [32]. When two Mn dopants are close to each other, the ferromagnetic state is preferred. However, the Mn dopants exhibit strong spin polarization and antiferromagnetic coupling with the S atoms when their separation distance is less than 16.53 Å, resulting in the same spin-polarization direction [29,33]. MoSe 2 can also be doped with two dopant atoms (Ti, V, Mn, Cr, Fe, Co, Ni), resulting in a weak magnetic/non-magnetic state, as shown in Figure 1a. Figure 1b-f illustrate the magnetically ordered state of different dopant atoms. V, Cr, Mn, Fe, and Co have magnetic moments exceeding 1.6 µB per dopant [9]. The electronic band structures also significantly change in d-orbital hybridization after Fe is substitutionally doped in monolayer MoS 2 with a further enhanced valley Zeeman splitting (Figure 1g,h) [34]. The effect of hybridization between the conduction band and d band in the transition metal dopants and the TMD lattice leads to ferromagnetism and antiferromagnetism in 2D doping TMD monolayers. In monolayer MoS 2 , Mn dopants have been found to induce ferromagnetic coupling through a double-exchange mechanism, which is mediated by the nearest neighboring sulfur atoms. Conversely, in bilayer MoS 2 , the magnetic exchange coupling between Mn dopants is influenced by the interplay of multiple interactions, including double-exchange, direct-exchange, and superexchange interactions [29]. The contribution of these interactions to the magnetic behavior in multilayer TMDs is not fully understood.

FOR PEER REVIEW
3 of 13 the same spin-polarization direction [29,33]. MoSe2 can also be doped with two dopant atoms (Ti, V, Mn, Cr, Fe, Co, Ni), resulting in a weak magnetic/non-magnetic state, as shown in Figure 1a. Figure 1b-f illustrate the magnetically ordered state of different dopant atoms. V, Cr, Mn, Fe, and Co have magnetic moments exceeding 1.6 µB per dopant [9]. The electronic band structures also significantly change in d-orbital hybridization after Fe is substitutionally doped in monolayer MoS2 with a further enhanced valley Zeeman splitting (Figure 1g,h) [34]. The effect of hybridization between the conduction band and d band in the transition metal dopants and the TMD lattice leads to ferromagnetism and antiferromagnetism in 2D doping TMD monolayers. In monolayer MoS2, Mn dopants have been found to induce ferromagnetic coupling through a double-exchange mechanism, which is mediated by the nearest neighboring sulfur atoms. Conversely, in bilayer MoS2, the magnetic exchange coupling between Mn dopants is influenced by the interplay of multiple interactions, including double-exchange, direct-exchange, and superexchange interactions [29]. The contribution of these interactions to the magnetic behavior in multilayer TMDs is not fully understood. Magnetic dopants can significantly impact the stability of TMDs in different phases (2H, 1T, 1T'). The most stable crystal phase is determined by the d-orbital electron numbers of the transition metal [35]. Most studies have investigated the impact of random doping in the 2H phase of trigonal prismatic TMDs [36]. Re-doped MoS2 nanosheets show a transformation from the 2H to 1T phase in both the Re and Mo coordinating structures, where the 1T ratios of both Mo and Re atoms were found to increase with the increasing Re-doping concentration, inducing the ferromagnetic ordering and stabilizing the metastable crystal phase [37]. In the 1T' phase, which is a metastable phase of TMDs, magnetic Magnetic dopants can significantly impact the stability of TMDs in different phases (2H, 1T, 1T'). The most stable crystal phase is determined by the d-orbital electron numbers of the transition metal [35]. Most studies have investigated the impact of random doping in the 2H phase of trigonal prismatic TMDs [36]. Re-doped MoS 2 nanosheets show a transformation from the 2H to 1T phase in both the Re and Mo coordinating structures, where the 1T ratios of both Mo and Re atoms were found to increase with the increasing Re-doping concentration, inducing the ferromagnetic ordering and stabilizing the metastable crystal phase [37]. In the 1T' phase, which is a metastable phase of TMDs, magnetic doping can alter the energy landscape of the material [38].
In 3D DMS materials, the magnetic moments of impurities are typically coupled through the Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction, which is mediated by the conduction electrons in the semiconductor host [39][40][41]. As a result, ferromagnetism is typically observed only when the Mn concentration is above a critical value, called the percolation threshold [41,42]. They form a connected network of magnetic moments that can collectively exhibit ferromagnetism with a high concentration of Mn atoms [43,44]. In Mndoped GaAs and Mn-doped InAs, the magnetic ions are randomly distributed throughout the lattice, leading to a percolative ferromagnetic phase transition as the temperature is lowered [45,46]. In contrast, some magnetically doped 2D TMDs exhibit magnetic ordering up to room temperature [9,47]. While in 2D DMS materials, the nature of the magnetic interactions can be more complex due to the reduced dimensionality of the system. This confinement leads to a more significant interaction between the dopant atoms, which enhances the formation of ferromagnetic clusters even at low doping levels [48,49].

Synthesis via Chemical Vapor Deposition (CVD)
This perspective article focuses on the CVD-grown 2D DMS. While several synthesis strategies have been demonstrated, including physical vapor deposition (PVD), atomic layer deposition (ALD), molecular beam epitaxy (MBE), and pulsed laser deposition (PLD) [50], CVD growth is commonly used to synthesize 2D TMDs [51]. Typically, the transition metal source sits in the middle of the heating zone, and the chalcogen source is placed upstream of the carrier gas (typically Ar or N 2 , usually combined with H 2 ). Heating the furnace to 700-1000 • C causes the vaporization of transition metal and chalcogen sources, leading to the formation of nanoparticles diffusing in the carrying gas, which then condenses on the substrate. Doping can be achieved by diffusing dopants into TMDs during precursor vaporization, making the CVD growth an effective means to generate TMD alloys [52,53].

Solid Source-Based Growth and Doping of TMDs
The solid source-based TMD growth requires a mixture of solid dopant powder, such as elemental substance, metal salt, and metal oxide, with a transition metal source. This growth is realized based on the simple mixing of precursors, which does not involve additional multiplex pre-operating procedures or complex pressure and flow control. For example, Fe-doped MoS 2 can be synthesized using the MoO 3 and FeS 2 powders with different Fe-to-Mo molar ratios of 0.12, 0.24, 0.48, and 0.4 (Figure 2j,k) [34]. Cr-doped WTe 2 crystals were grown via a two-step Te flux. The W and Cr powders and Te granules were mixed and heated at 1050 • C for 24 h, then Cr atoms were diffused into layered semimetal Td-Wte 2 . Figure 2f,g illustrate the corresponding elemental mapping images. [54]. Mn atoms were successfully in situ-doped into the MoS 2 lattice by introducing the Mn 2 (CO) 10 precursor upstream, as shown in Figure 2a-c, whereby the incorporation of Mn in MoS 2 was enabled by inert substrates, but not the reactive surface terminations, which led to the formation of defective MoS 2 (Figure 2d,e) [21]. Another study used Fe 3 O 4 particles cast on the substrate as the dopant source, permitting substitutional doping of Fe atoms into Mo or W sites in TMD crystals [47,55]. Figure 2h-k show the schematics and STEM images of in situ-doped monolayer Fe:MoS 2 . Intrinsic defects in doped TMD monolayers can also induce magnetic moments, and the chemical potential of the chalcogen influences the formation energy of these defects. This chemical potential can be modulated by adjusting the ratio of chalcogen to transition metal during CVD growth [56].  [21]. Schematic of the growth substrates, including (b) graphene and (c) insulating substrate (sapphire and SiO2). (d) Element analysis of Mn-doped MoS2 with (e) Mn concentrations (yellow pixels) that are equal (or higher) to those found in the MoS2 areas, which suggests that Mn may be bound to the substrate surface rather than incorporated into it [21]. (f,g) STEM-HAADF images of (f) pure WTe2 and (g) Cr0.02-WTe2 samples with their corresponding element mapping images [54].

Liquid-Assisted Growth and Doping of TMDs
The liquid-assisted growth uses a metal salt aqueous solution or metal suspension directly spun on the substrate. Although it requires more steps than using the solid source, it enables a more homogeneous mixing of precursors, contributing to more uniform doping in the growing TMD lattice than the solid source-based doping. For example, the Vdoped monolayer WSe2 demonstrated ferromagnetic domains by controlling the atomic ratio of V to W in a precursor solution from 0.1% to 40%, as illustrated in Figure 3a,b. Furthermore, in this growth, the doping concentration was controlled from 1% to a relatively high atom weight percent [57]. Lastly, the one-pot mixed-salt-intermediated CVD method was employed to synthesize single-crystal magnetic group VIII transition metaldoped MoSe2, which demonstrated excellent controllability and reproducibility ( Figure  3c), and Fe-doped MoSe2 monolayers were obtained by adjusting the FeCl3/Na2MoO4 ratios to control Fe-doping concentrations ranging from 0.93% to 6.10% [58].

MOCVD Growth and Doping
The metal-organic chemical vapor deposition (MOCVD) uses a pulsed precursor gas vapor source. It is a highly complex process for growing wafer-scale 2D crystalline layers with an excellent uniformity of deposition rates, dopant concentrations, and layer thickness [59], which are crucial for practical nanoelectronics and optoelectronics applications [60]. For example, large-area Nb-doped monolayer MoS2 was grown by employing molybdenum hexacarbonyl (Mo(CO)6) and diethyl sulfide (DES) as Mo and S precursors. Figure 3d,e show the schematic of the MOCVD process and as-grown Nb-MoS2 [60]. Multilayer V-doped WSe2 has been demonstrated on quasi-freestanding epitaxial graphene on

Liquid-Assisted Growth and Doping of TMDs
The liquid-assisted growth uses a metal salt aqueous solution or metal suspension directly spun on the substrate. Although it requires more steps than using the solid source, it enables a more homogeneous mixing of precursors, contributing to more uniform doping in the growing TMD lattice than the solid source-based doping. For example, the V-doped monolayer WSe 2 demonstrated ferromagnetic domains by controlling the atomic ratio of V to W in a precursor solution from 0.1% to 40%, as illustrated in Figure 3a,b. Furthermore, in this growth, the doping concentration was controlled from 1% to a relatively high atom weight percent [57]. Lastly, the one-pot mixed-salt-intermediated CVD method was employed to synthesize single-crystal magnetic group VIII transition metal-doped MoSe 2 , which demonstrated excellent controllability and reproducibility (Figure 3c), and Fe-doped MoSe 2 monolayers were obtained by adjusting the FeCl 3 /Na 2 MoO 4 ratios to control Fe-doping concentrations ranging from 0.93% to 6.10% [58].

MOCVD Growth and Doping
The metal-organic chemical vapor deposition (MOCVD) uses a pulsed precursor gas vapor source. It is a highly complex process for growing wafer-scale 2D crystalline layers with an excellent uniformity of deposition rates, dopant concentrations, and layer thickness [59], which are crucial for practical nanoelectronics and optoelectronics applications [60]. For example, large-area Nb-doped monolayer MoS 2 was grown by employing molybdenum hexacarbonyl (Mo(CO) 6 ) and diethyl sulfide (DES) as Mo and S precursors. Figure 3d,e show the schematic of the MOCVD process and as-grown Nb-MoS 2 [60]. Multi-layer V-doped WSe 2 has been demonstrated on quasi-freestanding epitaxial graphene on 6H-SiC (Figure 3j-l) [61]. WSe 2 films with coalesced 2D surfaces were doped with 0.5-1.1% of Re atoms in situ in another systematic study: a precisely controlled doping concentration was achieved by tuning the precursor partial pressure, with decreased domain size and increased doping concentration. The Re dopant and decreased domain size are illustrated in Figure 3f-i [62].  As summarized in Table 1, solid source-based thermal CVD growth and doping are convenient means to grow various TMDs by vaporizing the metal-oxide precursors [52]. The liquid phase-assisted CVD growth and doping can achieve a more homogeneous mixing of dopants and transition metal sources at the molecular level than the solid sourcebased CVD growth [58]. However, the asynchronous evaporation of these dopants and precursors during CVD can lead to heterogeneous element distribution and aggregation, impeding material characterization and limiting the performance of resulting devices. MOCVD enables better control of the precursor supply in growing TMDs than thermal CVD. The metal-organic precursors can easily be turned to the vapor phase [63], permitting a lower growth temperature than thermal CVD growth. However, residual carbon contamination may occur with some precursors in the grown 2D film [64]. As summarized in Table 1, solid source-based thermal CVD growth and doping are convenient means to grow various TMDs by vaporizing the metal-oxide precursors [52]. The liquid phase-assisted CVD growth and doping can achieve a more homogeneous mixing of dopants and transition metal sources at the molecular level than the solid sourcebased CVD growth [58]. However, the asynchronous evaporation of these dopants and precursors during CVD can lead to heterogeneous element distribution and aggregation, impeding material characterization and limiting the performance of resulting devices. MOCVD enables better control of the precursor supply in growing TMDs than thermal CVD. The metal-organic precursors can easily be turned to the vapor phase [63], permitting a lower growth temperature than thermal CVD growth. However, residual carbon contamination may occur with some precursors in the grown 2D film [64].

Optical Properties
The electron-phonon coupling strengths are impacted by the modification of the A' 1 vibrational mode (out-of-plane vibration of chalcogen atoms), which is caused by the electron density shift resulting from the substitutional doping [67][68][69]. For example, in Figure 4a, Fe-doped MoS 2 showed the n-type doping effect, which is revealed by a blue shift of the A 1g peak compared to undoped MoS 2 [34]. Co-or Cr-doped MoS 2 showed a moderate blue shift of the A 1g peak, while unaffecting the E 1 2g peak (in-plane vibration mode of chalcogen atoms), showing the sensitivity of A 1g peak to dopant atoms [65]. The density and type of carriers can be modified after doping, affecting the PL spectra of the materials dramatically. For example, the Re doping of Wse 2 showed a PL redshift by 40 meV and quenching by a factor of two, caused by non-radiative trion recombination [62]. A similar redshift was observed in Nb-doped MoS 2 with a~150% intensity enhancement due to impurity state formation caused by the Nb atoms [60]. Moreover, Fe-doped WS 2 showed a PL blueshift by 13 ± 0.5 meV and PL quenching by 40%. Fe-doped MoS 2 exhibited a PL quenching of 35% and a redshift of 29 ± 0.5 meV [55]. The difference between Fe:MoS 2 and Fe:WS 2 is attributed to the interaction of neutral excitons and negative/positive trions [70]. The detrimental effect of dopant-induced defects on the optical properties of 2D materials can be mitigated through the growth of dual dopant monolayers, thereby tuning these properties. For example, incorporating Co and Cr atoms in MoS 2 monolayers suppresses nonradiative recombination, significantly improving the photoluminescence intensity [65].

Magnetic Properties
In TMD crystals, direct interband transitions at the K and K valleys are coupled exclusively with the left and right circularly polarized light [71]. Furthermore, the inversion symmetry enhanced by dopant atoms leads to the existence of a magnetic moment within the orbit of the electrons. Therefore, the valley magnetic moment can be controlled in monolayer TMDs by circularly polarized light and the out-of-plane magnetic field, permitting control of the valley degree of freedom [72]. Strong circular dichroism (CD) was observed in the Fe-related peak at 2.28 eV from the Fe-doped monolayer MoS 2 when opposite circularly polarized lights were applied. The magnetic circular dichroism (MCD), as a function of increasing and decreasing magnetic fields, also showed a distinct hysteresis loop in Figure 4i,j [47]. Figure 4d-f show the valley Zeeman shift of Fe-doped MoS 2 . At zero field, the PL spectra were completely overlapped when σ− and σ+ light were applied; however, they split at high magnetic fields with inverse shifting directions for opposite magnetic field directions [34].

Magnetic Properties
In TMD crystals, direct interband transitions at the K and K′ valleys are coupled exclusively with the left and right circularly polarized light [71]. Furthermore, the inversion symmetry enhanced by dopant atoms leads to the existence of a magnetic moment within the orbit of the electrons. Therefore, the valley magnetic moment can be controlled in monolayer TMDs by circularly polarized light and the out-of-plane magnetic field, permitting control of the valley degree of freedom [72]. Strong circular dichroism (CD) was observed in the Fe-related peak at 2.28eV from the Fe-doped monolayer MoS2 when opposite circularly polarized lights were applied. The magnetic circular dichroism (MCD), as a function of increasing and decreasing magnetic fields, also showed a distinct hysteresis loop in Figure 4i,j [47]. Figure 4d-f show the valley Zeeman shift of Fe-doped MoS2. At zero field, the PL spectra were completely overlapped when σ− and σ+ light were applied; however, they split at high magnetic fields with inverse shifting directions for opposite magnetic field directions [34].
The Goodenough-Kanamori-Anderson rules state that superexchange can lead to ferromagnetic short-range coupling between overlapping orbitals [73]. In Nb and Co codoped WSe2, the magnetic signal was strongly enhanced with the doping of Co at 4%. The saturation magnetization (Ms) reached 63.6 emu cm −3 at 10 K, as shown in Figure 4c [48]. The magnetization versus magnetic field strength (M−H) test for Cr-doped MoTe2 showed a significant enhancement in an out-of-plane magnetic field with a Curie temperature around 275 K. The maximum Ms value occurred at 4.78 emu g −1 for the bulk 4.3% Crdoped MoTe2 [66]. Semiconducting V-doped WSe2 monolayers showed temperature- The Goodenough-Kanamori-Anderson rules state that superexchange can lead to ferromagnetic short-range coupling between overlapping orbitals [73]. In Nb and Co co-doped WSe 2 , the magnetic signal was strongly enhanced with the doping of Co at 4%. The saturation magnetization (Ms) reached 63.6 emu cm −3 at 10 K, as shown in Figure 4c [48]. The magnetization versus magnetic field strength (M−H) test for Crdoped MoTe 2 showed a significant enhancement in an out-of-plane magnetic field with a Curie temperature around 275 K. The maximum Ms value occurred at 4.78 emu g −1 for the bulk 4.3% Cr-doped MoTe 2 [66]. Semiconducting V-doped WSe 2 monolayers showed temperature-dependent ferromagnetic domains with long-range order, as shown in Figure 4g,h. Magnetic force microscopy (MFM) was used to detect the ferromagnetic domain stripes with a solid contrast, where the splitting and merging phenomenon of domains was observed with the temperature changes from low to high temperature. The magnetic domain walls retained the vestige even at 420 K. Meanwhile, V-doped WSe 2 showed p-type semiconductor behavior after doping [57]. The reported magnetization and coercivity in doped TMDs corresponding to measurement temperatures are summarized in Table 2.

Other Properties
Recent research suggests that doping can modify the interlayer chemical bond in layer-structured materials, allowing access to layer-dependent transport properties [75]. In the case of MoS 2 , vanadium (V) doping has been shown to enhance in-plane conductivity and improve carrier concentration [76]. Additionally, dopants can activate out-of-plane interactions between adjacent layers, further tailoring the electrical carrier transport behavior of bilayers [77]. Doping of Mn in MoS 2 also results in improved electrical contact in field effect transistors due to the Fermi level shift towards the conduction band [78]. However, the mobility slightly decreases when doping Cr in monolayer MoS 2 , which is attributed to the change in intrinsic defect density [22,57]. When reacting with Ni atoms, the intrinsic defects and distorted lattice improve the carrier injection, modifying the electronic properties [79], as summarized in the previous review [80].
In addition, TMDs show excellent catalytic properties owing to their high surface area and edge sites, which are highly active for catalytic reactions. These properties make TMDs promising candidates for catalytic applications in energy conversion, environmental remediation, and chemical synthesis [81]. For example, hydrogen evolution reaction (HER) using TMDs has been studied due to their remarkable catalyst activation performance compared to bulk forms [82]. Using elemental doping of TMDs has emerged as an effective approach to tune the electronic structure of catalytic sites, optimizing the Gibbs free energy of hydrogen adsorption (∆GH) to promote HER kinetics [82,83]. Notably, various dopants have been found to activate the inactive basal plane of 2D TMDs into active ones by creating defects [84]. Previous studies have already summarized the discussion on the catalytic properties of dopant TMDs, indicating their potential for various applications [85]. For a more comprehensive review of the application of TMDs in different structures, such as heterostructures or nanoribbons, we refer the readers to [86][87][88].

Conclusions
Doping of TMDs is a popular topic since the external atoms can bring or enhance multiple properties of origin TMDs. In magnetic doping, the promoted magnetic properties due to the large atomic mass of doped transition metals can shed new light on their potential applications. This review has introduced different doping methods via CVD, including solid source-based doping, liquid-assisted doping, and MOCVD methods. Dopant-induced optical and magnetic properties of doped TMDs were also summarized.
The exploration of magnetically doped TMDs poses several challenges. First, the uniform and large-area TMD synthesis remains challenging due to the difficulty in controlling the flow of vaporized precursors and uncontrollable domain size [36]. Though room temperature ferromagnets were demonstrated, most of the magnetic TMDs show low Curie temperatures hampering the practical applications of these materials. Optimizing the exchange interaction and magnetic anisotropy would be key parameters to enhance the critical temperatures [1]. Another challenge includes randomly distributed dopants in the crystal lattice, causing localized magnetic fields. The long-range behavior of magnets is attributed to the finite lattice size, concentrated and ordered positions of magnetic ions, and different saturation magnetizations caused by different kinds of dopants. [9]. The co-doping in TMDs has been demonstrated as a promising approach for tuning the properties of 2D materials. Incorporating two dopant elements in the material can suppress nonradiative recombination, enhancing optical properties, and inducing ferromagnetic ordering. In addition, the electronic interactions between the dopants can further modify the band structure and energy levels of the materials, providing additional opportunities for tailoring their electrical and magnetic properties. The topological boundary states, magnetic proximity effects, and van der Waals heterostructures are emerging fields utilizing magnetic 2D TMDs. With groundbreaking research in the coming years, transition metal-doped TMDs are promising materials in optomagnetism, spintronic devices, and quantum information science.

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