Metal−insulator transition in an isoelectronically doped transition metal dichalcogenide and its heterostructures

Abstract

Due to the limitations of traditional silicon-based semiconductors at the nanoscale, such as short-channel effects and quantum effects, two-dimensional (2D) transition metal dichalcogenides (TMDs) like MoS₂ and MoTe₂ are increasingly recognized for their remarkable characteristics. These materials exhibit unique properties, including tunable bandgaps and the ability to mitigate electron scattering. The metal−insulator transition (MIT), a special electrical property found in some 2D materials, holds great potential for various applications. The MIT in TMDs can be induced through external parameters, but challenges like charge inhomogeneity and the detrimental effects of ionic liquid gating complicate device fabrication and measurement. In this work, we report the MIT behavior in an isoelectronic doped transition metal dichalcogenide MoS_2(1−x)Se_2x. By studying the dependence of conductivity on temperature in MoS_2(1−x)Se_2x field-effect transistors employing a single back-gate device structure, we observe clear evidence of the metal−insulator transition in the electron carriers. More importantly, we demonstrate that this MIT behavior can be replicated in other 2D material systems that lack such properties by heterostructure engineering. Our research lays the foundation for further enhancing the performance of 2D materials and may lead to broader applications in functional electronic devices.

Graphical abstract

Keywords

2D semiconductors / metal−insulator transition / 2D transistors / van der Waals heterostructures / isoelectronic doping

1 Introduction

Due to the performance limitations encountered by traditional silicon-based semiconductor materials at the nanoscale, such as short-channel effects, increased power consumption, and quantum effects, it has become increasingly difficult to sustain Moore’s Law. Transition metal dichalcogenides (TMDs) like MoS₂ and MoTe₂ have many exceptional electrical and physical properties because of their atomic-scale thickness and the absence of surface dangling bonds, which has led to significant interest in recent years [1−11]. Due to their unique layered structure, where the layers are bound by van der Waals interactions, TMDs can also be reduced to stable monolayer structures through methods such as mechanical exfoliation. This process allows their semiconductor properties to be tuned from an indirect bandgap to a direct bandgap [12]. This characteristic makes certain TMDs complementary to graphene in nanoelectronic applications, as the presence of a bandgap enables efficient charge exchange, facilitating logic and optoelectronic operations [13−18]. TMDs are poised to become a powerful alternative for future electronic materials, as they mitigate the significant issue of electron surface scattering in channel materials, which becomes more pronounced as device sizes shrink in silicon-based devices.

The metal−insulator transition (MIT) refers to the transformation of a material’s electronic state from a conductive metallic phase to a non-conductive insulating phase, or vice versa. This phenomenon is particularly significant in the study of two-dimensional (2D) materials, as it is much more challenging to achieve MIT in traditional 3D materials due to strong inter-electron interactions and the rigid constraints of their crystal structures. In contrast, in TMDs, the atomic thickness and more flexible electronic structure enable MIT to be induced through external control parameters such as electric fields, temperature, or carrier concentration [19−24]. To achieve the MIT, it is usually necessary to tune the chemical potential near the valence or conduction bands in experiments. Monolayer MoS₂ has an electronic bandgap of over 2.0 eV when placed on weakly interacting substrates, making it difficult for back-gated devices to induce the MIT transition before breakdown occurs [25, 26]. Previous studies have reported the induction of a shift from an insulating phase to a conductive phase in MoS₂ using a dual-gate configuration in a field-effect transistor [27], but these works often require covering the surface of the 2D materials with high-κ dielectric layers to enhance carrier concentration. Furthermore, according to previous reports, the use of ionic liquid gating can also induce the MIT. However, ionic liquid gates often suffer from issues related to charge inhomogeneity, and the freezing of the liquid at low temperatures exacerbates this problem [28]. The ionic liquid gate generates substantial stress on the channel material during temperature fluctuations, which can compromise its structural integrity and damage the device. Furthermore, the potential side reactions between the ionic liquid and the channel material, coupled with the device’s sensitivity to moisture, further complicate both the fabrication and measurement processes of the device. Additionally, the liquid can interfere with technical measurements of the sample surface.

Here, we report a method to induce the metal-insulator transition in an isoelectronic doped transition metal dichalcogenide MoS_2(1−x)Se_2x, in which the sulfur (S) atoms were substituted with isoelectronic selenium (Se) atoms. By studying the dependence of conductivity on temperature in MoS_2(1−x)Se_2x field-effect transistors with a single back-gate device structure, we observe clear evidence of the metal−insulator transition in the electron carriers. More importantly, we demonstrate that this MIT behavior can be replicated in other 2D material systems that lack such properties by heterostructure engineering. This also reduces the Schottky barrier of MoTe₂ while increasing its carrier concentration and carrier mobility, opening a new path for further enhancing the performance of 2D semiconductor materials.

2 Experimental methods

2.1 Material characterizations

Few-layer MoS_2(1−x)Se_2x flakes were prepared by mechanically peeling off the bulk crystal (2D semiconductor). The morphology of the few-layer MoS_2(1−x)Se_2x flakes were analyzed using optical microscopy (OM, Olympus BX51M) and atomic force microscopy from Dimension Icon. Atomic-resolution high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images were acquired using JEOL ARM200F microscope, with an acceleration voltage of around 80 kV. Raman spectroscopy was conducted with a LabRAM HR Evolution instrument.

2.2 FET fabrication and electrical measurements

First, few-layer MoS_2(1−x)Se_2x flakes were exfoliated onto 300 nm SiO₂/Si substrates using the conventional typical scotch method. Then, We exposed and developed the source and drain regions using standard electron beam lithography (EBL), followed by depositing 5/50 nm Cr/Au through metal thermal evaporation. The electrical performance of the fabricated FET was measured using a probe station with a vacuum pump, in conjunction with a Keithley 4200 semiconductor analyzer. As for the heterostructure devices, MoTe₂ flakes were first exfoliated onto the 300 nm SiO₂/Si substrates. MoS_2(1−x)Se_2x flake was mechanically peeled onto the PPC-coated substrates, followed by its precise transfer onto the underlying MoTe₂. The sample was heated at 115 °C for 12 minutes, followed by dissolving the PPC layer with acetone. Same methods as above were adopted for the device fabrication and electrical measurements.

3 Results and discussion

A Ball-and-stick model representation of the MoS_2(1−x)Se_2x, where Se atoms randomly substitute the positions of S atoms, is shown in Fig.1(a). Raman spectroscopy was performed on the sample to confirm this isoelectronic substitution. As shown in Fig.1(b), the Raman peaks at 271 cm⁻¹, 402 cm⁻¹, and 376 cm⁻¹ correspond to the MoSe₂-like A_1g mode, MoS₂-like A_1g mode, and MoS₂-like E_2g mode, respectively, which are consistent with the previous reports [29]. Fig.1(c) gives the corresponding PL spectrum, and the bandgap emission energy of MoS_2(1−x)Se_2x is approximately 1.7 eV (i.e., 714 nm), which is smaller than that of MoS₂ (~1.86 eV) [30]. Fig.1(d) displays the HAADF-STEM image of MoS_2(1−x)Se_2x, where the yellow arrows indicate the positions of Se atoms. The clearly defined hexagonal atomic arrangement shows distinct contrast, further revealing the substitution of S atoms by Se atoms. Fig.1(e) shows the selected area electron diffraction (SAED) pattern with hexagonal symmetry, revealing the high crystallinity and robust atomic configuration of MoS_2(1−x)Se_2x [31]. Additionally, the atomic composition of MoS_2(1−x)Se_2x is displayed in the energy-dispersive X-ray spectroscopy (EDS) in Fig.1(f), where the elemental percentages of Mo, S, and Se are 22.68%, 28.43%, and 48.88%, respectively. More EDS measurement information is provided in Fig. S1.

FETs were fabricated to study the electrical properties of MoS_2(1−x)Se_2x flakes, as shown schematically in Fig.2(a). The optical and AFM images of the device are given in Fig. S2. The transport characteristics of MoS_2(1−x)Se_2x device were first measured at 300 K under different V_DS, as shown in Fig.2(b), exhibiting outstanding n-type characteristics and excellent gate tunability with On/Off current ratio ranging from 10⁸ to 10⁹. The output characteristics were also measured [Fig.2(c)], where the typical linear characteristic curve reveals good ohmic contact between the electrodes and MoS_2(1−x)Se_2x [32].

Fig.3(a) shows the transfer characteristics of MoS_2(1−x)Se_2x back-gate field-effect transistor at different temperatures. I_DS rises as the temperature increases at low V_GS, indicating the insulating behavior of MoS_2(1−x)Se_2x. When V_GS reaches approximately 45 V, MoS_2(1−x)Se_2x transitions into a metallic state, and I_DS increases as the temperature decreases. This MIT transition is manifested by the crossing of I_DS−V_GS curves obtained at different temperatures, indicating the existence of two distinct states [33]. In addition, the threshold voltage (V_TH) of the MoS_2(1−x)Se_2x FET also increases as the temperature decreases, and the carrier concentration increases as indicated by

(1)n_{2\mathrm{D}}={\displaystyle \frac{C_{\mathrm{O}\mathrm{X}}\left(V_{\mathrm{G}\mathrm{S}}-V_{\mathrm{T}\mathrm{H}}\right)}{e}} .

This further reveals the transition of MoS_2(1−x)Se_2x to the metallic state [Fig.3(b)]. Figure S3 shows the I_DS−V_DS curves for the MoS_2(1−x)Se_2x FET at different temperatures, with the typical linear output characteristic revealing good ohmic contact between MoS_2(1−x)Se_2x and the electrodes. Additionally, at V_GS = 80 V, the current decreases with increasing temperature, which also reveals the MIT in MoS_2(1−x)Se_2x.

To quantitatively obtain the Schottky barrier, we performed temperature-dependent electrical measurements between 200 K and 320 K. According to the thermionic emission theory, I_DS can be expressed as [34, 35]

(2)I_{2\mathrm{D}}=A_{2\mathrm{D}}^{\ast }{T}^{1.5}{\mathrm{e}}^{\left(-\frac{{\mathrm{\Phi }}_{\mathrm{B}}}{k_{\mathrm{B}}T}\right)}\left[1-{\mathrm{e}}^{\left(-\frac{V_{\mathrm{D}\mathrm{S}}}{k_{\mathrm{B}}T}\right)}\right],

where A^*_2D = q(8πk_B³m^*)^0.5/h² is the Richardson constant (where m^* is the effective electron mass), q represents the elementary charge, Φ_B is the effective contact barrier height, k_B denotes the Boltzmann constant, and T is the temperature. To determine the Schottky barrier height (SBH), the values of ln(I_DS/T^1.5) at various V_GS under V_DS = 1 V were plotted on an Arrhenius plot [(Fig.3(c)]. The barrier height was obtained by extracting the slope of the Arrhenius curve [Fig.3(d)]. Φ_B varies linearly with V_GS when V_GS is less than approximately −31 V. With further increase in V_GS, the plot line satisfies the flat-band voltage condition at the deviation from the linear relationship, corresponding to an extracted Schottky barrier height (SBH) of 118 meV.

Fig.3(e) shows the relationship between the conductivity σ (in units of e²/h) and temperature under different V_GS in the range of 10 V to 80 V. At lower V_GS, the σ increases as the temperature decreases, corresponding to the insulating behavior. As the V_GS increases, the σ gradually exhibits a phenomenon where it first increases and then decreases with the decreasing temperature. When V_GS reaches 80 V, the conductivity increases with the decreasing temperature, corresponding to metallic behavior. This trend is also consistent with Fig.2(a). Furthermore, as V_GS increases, the sample conductivity approaches the quantum conductivity e²/h, which is the minimum value for metallic conductivity. This indicates that with the increase of V_GS, not only does the sample’s conductivity approach metallic behavior, but its temperature-dependent conductivity behavior also gradually shifts from insulating to metallic behavior [36, 37].

The conductivity is defined as G = I_DS⋅L/(V_DS⋅W), where L and W represent the channel length and width, respectively. We measured the change in the sample conductivity G with temperature under a gate voltage range of −20 V to 20 V, and found that within the temperature range of 80 K to 320 K, the sample conductivity G follows the thermally activated transport model:

(3)G=G_0{\mathrm{e}}^{-\frac{E_{\mathrm{a}}}{k_{\mathrm{B}}\mathrm{T}}},

where E_a is the energy barrier that electrons need to overcome in order to transition from the Fermi level to the conduction band, and G₀ is the relevant parameter extracted from the fitting curve [38]. By plotting the logarithm of G as the vertical axis and 1/T as the horizontal axis, a fit of G was performed. The fitting line shows a good agreement with the experimental data of G [Fig.3(f)], indicating that the experimental data follow the thermally activated transport model well, which suggests that the charge transport in the sample is thermally activated [39]. Studies on the mobility also show that the mobility follows a μ ∝ T^−γ dependence above 200 K, consistent with the electron−phonon scattering transport model (Fig. S4) [40]. The thermal activation energy E_a at various V_GS values was obtained by analyzing the slope of the fitted line according to the thermal activation model equation. As shown in Fig. S5, E_a decreases as the V_GS increases, making it easier for carriers to be excited, which in turn increases the conductivity of the sample.

In addition to observing the MIT transition in MoS_2(1−x)Se_2x flakes within a single back-gate device structure, we also discovered that this MIT behavior can be replicated in MoTe₂, a material that does not inherently exhibit such properties, through heterostructure engineering. Fig.4(a) shows the optical image and schematic diagram of the bare MoTe₂ device and the MoTe₂ device overlapped with MoS_2(1−x)Se_2x flake (abbreviated as MoTe₂−MoS_2(1−x)Se_2x device). Fig.4(b) and (d) show the relationship between conductivity and back-gate voltage for both the MoTe₂ device and MoTe₂−MoS_2(1−x)Se_2x device, measured at temperatures ranging from 80 K to 320 K with a fixed bias voltage V_DS = 1 V. The transfer curves of MoTe₂−MoS_2(1−x)Se_2x device cross at different temperatures, a typical hallmark of MIT, while the bare MoTe₂ device does not exhibit this behavior. This indicates that the presence of MoS_2(1−x)Se_2x flake could duplicate its MIT behavior to other 2D material systems that have no such properties themselves by heterostructure engineering. As a control, we also measured the electrical properties of the independent MoTe₂ device to rule out the lateral effect of MoS_2(1−x)Se_2x on MoTe₂ (Fig. S6). Furthermore, we measured the electrical properties of MoS_2(1−x)Se_2x in the MoTe₂−MoS_2(1−x)Se_2x device, and the results indicated that MoS_2(1−x)Se_2x still retains this metal−insulator transition behavior (Fig. S7).

Fig.4(c) and (e) are the Arrhenius plots for the MoTe₂ and MoTe₂−MoS_2(1−x)Se_2x device, respectively. The corresponding Schottky barriers were extracted in Fig.4(f) and (g). Compared to the bare MoTe₂, the MoTe₂ device overlapped with MoS_2(1−x)Se_2x flake shows a significantly reduced Schottky barrier. Fig.4(h) shows the transfer curves for the MoTe₂ device and MoTe₂−MoS_2(1−x)Se_2x device at room temperature (300 K) with a fixed bias voltage V_DS = 1 V. The linear portion of the curve was fitted to obtain the threshold voltage V_TH for both devices. It was found that the additional MoS_2(1−x)Se_2x results in a lower threshold voltage. The electron concentration can be calculated using the parallel plate capacitor model:

(4)n_{2\mathrm{D}}={\displaystyle \frac{C_{\mathrm{O}\mathrm{X}}\mathrm{\Delta }V}{q}},

where n_2D is the electron concentration, C_OX = {\epsilon }_0{\epsilon }_{\mathrm{r}}/d, {\epsilon }_0 is the vacuum permittivity, {\epsilon }_{\mathrm{r}} represents the relative dielectric constant of the silicon dioxide layer on the back gate, d denotes the thickness of the SiO₂, while ΔV = V_GS − V_TH. By measuring the threshold voltage V_TH, the electron concentration can be obtained. The MoTe₂−MoS_2(1−x)Se_2x device exhibits a lower threshold voltage V_TH, signifying a higher electron concentration. Moreover, the MoTe₂−MoS_2(1−x)Se_2x device also demonstrates a considerably lower gate voltage at the minimum current, which further corroborates its higher carrier concentration (Fig. S8). In addition, we fabricated more MoTe₂−MoS_2(1−x)Se_2x devices to verify the universality of the metal−insulator transition behavior (Fig. S9).

To explain this phenomenon, we evaluated the conduction band position of the MoTe₂−MoS_2(1−x)Se_2x device. At a gate voltage of V_GS = 45 V and room temperature T = 300 K, we calculated the 2D electron concentration n_2D = 3.17 × 10¹² cm⁻². The thickness of the 2D electron gas layer is less than 2 nm, so the 3D electron concentration at the heterojunction interface is n_3D>1.58 × 10¹⁹ cm⁻³.

The electronic density of states at the lower edge of the conduction band for MoS_2(1−x)Se_2x is given by the equation:

(5)N_{\mathrm{C}}= 2{\left({\displaystyle \frac{2\pi m_{\mathrm{e}}^{\ast }{k}_{\mathrm{B}}T}{h^2}}\right)}^{\frac{3}{2}}=9\times {10}^{18}{\mathrm{c}\mathrm{m}}^{-3},

where the effective electron mass for MoS_2(1−x)Se_2x m_{\mathrm{e}}^{\ast } = 0.521m₀ [41], the Planck constant h is 6.626 × 10⁻³⁴ Js.

The relationship between the electron concentration n_3D and the density of states N_C is expressed as

(6)n_{3\mathrm{D}}={N_{\mathrm{C}}\mathrm{e}}^{-\frac{E_{\mathrm{C}}-E_{\mathrm{F}}}{k_{\mathrm{B}}T}}.

When n_3D > N_C, it indicates that E_F > E_C. At the MoTe₂−MoS_2(1−x)Se_2x heterojunction interface, due to band bending, the minimum of the conduction band of MoS_2(1−x)Se_2x is positioned below the Fermi level, enabling electrons to populate the conduction band and consequently increasing the electron concentration at the interface. The 2D electron gas at the interface caused by band bending and electron filling should be the key to the property change brought about by MoS_2(1−x)Se_2x.

In the context of the MIT in MoS_2(1−x)Se_2x, an increase in gate voltage or temperature results in a rise in electron concentration, prompting the system to shift progressively from the insulating state to the metallic state. The heterojunction of MoTe₂−MoS_2(1−x)Se_2x also gives rise to a higher electron concentration, the increase in electron concentration is always closely related to the MIT phenomenon, indicating that the enhancement of electron concentration has a pivotal impact on the occurrence of MIT, which might be the underlying cause of the MIT phenomenon in the MoTe₂−MoS_2(1−x)Se_2x device [42].

To characterize the Coulomb interactions among electrons in the material, the ratio [43] of the electron potential energy to kinetic energy is defined as follows:

(7)r_{\mathrm{s}}= {\displaystyle \frac{E_{\mathrm{C}}}{E_{\mathrm{F}}}}={\displaystyle \frac{n_{\mathrm{V}}}{a_{\mathrm{B}}^{\ast }\sqrt{\pi n_{2\mathrm{D}}}}}= {\displaystyle \frac{n_{\mathrm{V}}{m}_{\mathrm{e}}^{\ast }{e}^2}{4\pi \epsilon {\hslash }^2\sqrt{\pi n_{2\mathrm{D}}}}},

where n_V represents the number of degenerate valleys, a_{\mathrm{B}}^{\ast }=\left(4\pi \epsilon {\hslash }^2\right)/\left(m_{\mathrm{e}}^{\ast }{e}^2\right) is the effective Bohr radius, m_{\mathrm{e}}^{\ast } is the effective electron mass, and ε is the system’s dielectric constant. When ε lies within 1 to 3 times that of MoS₂, the value of r_s varies from 8.55 to 2.85, signifying strong Coulomb interactions among the electrons. Typically, electrons in metals are in delocalized states and the interactions among them are mutually screened. In contrast, electrons in insulators tend to be in localized states. Nevertheless, for the MoTe₂−MoS_2(1−x)Se_2x device, the electrons can possess rather potent interactions and concomitantly experience a transition from an insulating state to a metallic state, which implies that the electron interactions in this system are fundamentally distinct from those in common metals. Such strong interactions can prompt the system to shift from a strongly localized insulating state to a weakly localized metallic state [44].

Fig.4(i) shows the mobility μ of the MoTe₂ device and the MoTe₂−MoS_2(1−x)Se_2x device at different temperatures under V_DS = 1 V. The mobility μ is calculated using the formula:

(8)\mu ={\displaystyle \frac{L\left({\displaystyle \frac{\mathrm{d}{I}_{\mathrm{D}\mathrm{S}}}{\mathrm{d}{V}_{\mathrm{G}\mathrm{S}}}}\right)}{W{C}_{\mathrm{O}\mathrm{X}}{V}_{\mathrm{D}\mathrm{S}}}}.

The mobility can be determined from the output curve. The incorporation of MoS_2(1−x)Se_2x remarkably enhances the mobility. As the temperature rises, the mobility of the MoTe₂−MoS_2(1−x)Se_2x device first increases and then decreases. At elevated temperatures, the device exhibits metallic behavior, while the MoTe₂ device solely demonstrates insulating properties. The change in the mobility of the MoTe₂−MoS_2(1−x)Se_2x device at higher temperatures conforms to the electron−phonon scattering model, indicating that electron−phonon scattering becomes the dominant scattering mechanism at higher temperatures. Under the mode of electron−phonon scattering, the variation of the mobility with temperature follows the relationship μ~T^−γ, where the exponent γ depends on the dominant phonon scattering mechanism. Therefore, we used the general formula of the temperature dependence of the mobility μ~T^−γ to fit the part of the curve at higher temperatures. Through fitting the curve, we found that γ ≈ 1.01. This stark contrast in temperature-dependent behavior between the MoTe₂−MoS_2(1−x)Se_2x device and MoTe₂ devices correlates with the occurrence or absence of the MIT phenomenon, thereby confirming that MoS_2(1−x)Se_2x imparts MIT characteristics to MoTe₂. In addition, this also reflects that MoS_2(1−x)Se_2x has brought about performance enhancements in various aspects to the MoTe₂ device, providing insights for other research directions beyond the study of the MIT phenomenon, such as device performance.

4 Conclusion

To summarize, we propose a strategy for inducing the MIT in isoelectronic doped transition metal dichalcogenide MoS_2(1−x)Se_2x by substituting sulfur atoms with isoelectronic selenium atoms. The dependence of MIT on gate voltage and temperature in MoS_2(1−x)Se_2x was also investigated. In the study, the observed trend of increasing electron concentration coincided with the occurrence of the MIT, indicating that electron interactions play a crucial role in this transition. More importantly, we demonstrate that this MIT behavior can be replicated in other 2D material systems, which do not inherently exhibit such properties, through heterostructure engineering. The difference in electron concentration between MoTe₂ and MoS_2(1−x)Se_2x devices suggests that the inclusion of MoS_2(1−x)Se_2x results in a higher electron concentration. Strong electron interactions within the system account for the observed MIT. Furthermore, the integration of MoS_2(1−x)Se_2x reduces the Schottky barrier, enhancing the device’s mobility and carrier concentration, thereby improving its overall performance. This work contributes to a deeper understanding of the MIT phenomenon in 2D transition metal dichalcogenide systems and provides valuable strategies for improving the efficiency of 2D material-based transistors.

References

[1] L. Yin, R. Cheng, J. Ding, J. Jiang, Y. Hou, X. Feng, Y. Wen, and J. He, Two-dimensional semiconductors and transistors for future integrated circuits, ACS Nano 18(11), 7739 (2024)

[2] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, and M. S. Strano, Electronics and optoelectronics of two-dimensional transition metal dichalcogenides, Nat. Nanotechnol. 7(11), 699 (2012)

[3] M. Chhowalla, D. Jena, and H. Zhang, Two-dimensional semiconductors for transistors, Nat. Rev. Mater. 1(11), 16052 (2016)

[4] M. Y. Li, S. K. Su, H. S. P. Wong, and L. J. Li, How 2D semiconductors could extend Moore’s law, Nature 567(7747), 169 (2019)

[5] S. Manzeli,D. Ovchinnikov,D. Pasquier,O. V. Yazyev,A. Kis, 2D transition metal dichalcogenides, Nat. Rev. Mater. 2(8), 17033 (2017)

[6] K. S. Novoselov,A. Mishchenko,A. Carvalho,A. H. Castro Neto, 2D materials and van der Waals heterostructures, Science 353(6298), aac9439 (2016)

[7] T. Song, X. Cai, M. W. Y. Tu, X. Zhang, B. Huang, N. P. Wilson, K. L. Seyler, L. Zhu, T. Taniguchi, K. Watanabe, M. A. McGuire, D. H. Cobden, D. Xiao, W. Yao, and X. Xu, Giant tunneling magnetoresistance in spin-filter van der Waals heterostructures, Science 360(6394), 1214 (2018)

[8] D. Akinwande, C. Huyghebaert, C. H. Wang, M. I. Serna, S. Goossens, L. J. Li, H. S. P. Wong, and F. H. L. Koppens, Graphene and two-dimensional materials for silicon technology, Nature 573(7775), 507 (2019)

[9] R. Cheng, L. Yin, R. Hu, H. Liu, Y. Wen, C. Liu, and J. He, Modulation of negative differential resistance in black phosphorus transistors, Adv. Mater. 33(25), 2008329 (2021)

[10] L. Yin, P. He, R. Cheng, F. Wang, F. Wang, Z. Wang, Y. Wen, and J. He, Robust trap effect in transition metal dichalcogenides for advanced multifunctional devices, Nat. Commun. 10(1), 4133 (2019)

[11] R. Yang, J. Fan, and M. Sun, Transition metal dichalcogenides (TMDCs) heterostructures: Optoelectric properties, Front. Phys. (Beijing) 17(4), 43202 (2022)

[12] C. H. Lui, K. F. Mak, J. Shan, and T. F. Heinz, Ultrafast photoluminescence from graphene, Phys. Rev. Lett. 105(12), 127404 (2010)

[13] L. Yin, R. Cheng, X. Wan, J. Ding, J. Jia, Y. Wen, X. Liu, Y. Guo, and J. He, High-κ monocrystalline dielectrics for low-power two-dimensional electronics, Nat. Mater. 24(2), 197 (2025)

[14] R. Cheng, F. Wang, L. Yin, Z. Wang, Y. Wen, T. A. Shifa, and J. He, High-performance, multifunctional devices based on asymmetric van der Waals heterostructures, Nat. Electron. 1(6), 356 (2018)

[15] H. Liu, A. T. Neal, and P. D. D. Ye, Channel Length Scaling of MoS₂ MOSFETs, ACS Nano 6(10), 8563 (2012)

[16] Z. Y. Yin, H. Li, H. Li, L. Jiang, Y. M. Shi, Y. H. Sun, G. Lu, Q. Zhang, X. D. Chen, and H. Zhang, Single-layer MoS₂ phototransistors, ACS Nano 6(1), 74 (2012)

[17] G. Eda and S. A. Maier, Two-dimensional crystals: Managing light for optoelectronics, ACS Nano 7(7), 5660 (2013)

[18] Z. Wang, Y. Yang, B. Hua, and Q. Ji, Synthetic two-dimensional electronics for transistor scaling, Front. Phys. (Beijing) 18(6), 63601 (2023)

[19] J. T. Ye, Y. J. Zhang, R. Akashi, M. S. Bahramy, R. Arita, and Y. Iwasa, Superconducting dome in a gate-tuned band insulator, Science 338(6111), 1193 (2012)

[20] B. H. Moon, J. J. Bae, G. H. Han, H. Kim, H. Choi, and Y. H. Lee, Anomalous conductance near percolative metal−insulator transition in monolayer MoS₂ at low voltage regime, ACS Nano 13(6), 6631 (2019)

[21] X. Chen, Z. Wu, S. Xu, L. Wang, R. Huang, Y. Han, W. Ye, W. Xiong, T. Han, G. Long, Y. Wang, Y. He, Y. Cai, P. Sheng, and N. Wang, Probing the electron states and metal−insulator transition mechanisms in molybdenum disulphide vertical heterostructures, Nat. Commun. 6(1), 6088 (2015)

[22] J. H. Lee, H. Z. Gul, H. Kim, B. H. Moon, S. Adhikari, J. H. Kim, H. Choi, Y. H. Lee, and S. C. Lim, Photocurrent switching of monolayer MoS₂ using a metal−insulator transition, Nano Lett. 17(2), 673 (2017)

[23] C. L. Wu, H. Yuan, Y. Li, Y. Gong, H. Y. Hwang, and Y. Cui, Gate-induced metal−insulator transition in MoS₂ by solid superionic conductor LaF₃, Nano Lett. 18(4), 2387 (2018)

[24] Y. Li, M. Yang, Y. Lu, D. Cao, X. Chen, and H. Shu, Reversible doping polarity and ultrahigh carrier density in two-dimensional van der Waals ferroelectric heterostructures, Front. Phys. (Beijing) 18(3), 33307 (2023)

[25] P. V. Nguyen, N. C. Teutsch, N. P. Wilson, J. Kahn, X. Xia, A. J. Graham, V. Kandyba, A. Giampietri, A. Barinov, G. C. Constantinescu, N. Yeung, N. D. M. Hine, X. Xu, D. H. Cobden, and N. R. Wilson, Visualizing electrostatic gating effects in two-dimensional heterostructures, Nature 572(7768), 220 (2019)

[26] Z. Qiu, M. Trushin, H. Fang, I. Verzhbitskiy, S. Gao, E. Laksono, M. Yang, P. Lyu, J. Li, J. Su, M. Telychko, K. Watanabe, T. Taniguchi, J. Wu, A. H. C. Neto, L. Yang, G. Eda, S. Adam, and J. Lu, Giant gate-tunable bandgap renormalization and excitonic effects in a 2D semiconductor, Sci. Adv. 5(7), eaaw2347 (2019)

[27] B. Radisavljevic and A. Kis, Mobility engineering and a metal−insulator transition in monolayer MoS₂, Nat. Mater. 12(9), 815 (2013)

[28] Y. Ren, H. Yuan, X. Wu, Z. Chen, Y. Iwasa, Y. Cui, H. Y. Hwang, and K. Lai, Direct imaging of nanoscale conductance evolution in ion-gel-gated oxide transistors, Nano Lett. 15(7), 4730 (2015)

[29] Q. Feng, Y. Zhu, J. Hong, M. Zhang, W. Duan, N. Mao, J. Wu, H. Xu, F. Dong, F. Lin, C. Jin, C. Wang, J. Zhang, and L. Xie, Growth of large-area 2D MoS_2(l−x)Se_2x semiconductor alloys, Adv. Mater. 26(17), 2648 (2014)

[30] K. F. Mak, C. Lee, J. Hone, J. Shan, and T. F. Heinz, Atomically thin MoS₂: A new direct-gap semiconductor, Phys. Rev. Lett. 105(13), 136805 (2010)

[31] M. Li,J. Yao,X. Wu,S. Zhang,B. Xing, X. Niu,X. Yan,Y. Yu,Y. Liu,Y. Wang, P-type doping in large-area monolayer MoS₂ by chemical vapor deposition, ACS Appl. Mater. Interfaces 12(5), 6276 (2020)

[32] J. Gao, Y. D. Kim, L. Liang, J. C. Idrobo, P. Chow, J. Tan, B. Li, L. Li, B. G. Sumpter, T. M. Lu, V. Meunier, J. Hone, and N. Koratkar, Transition-metal substitution doping in synthetic atomically thin semiconductors, Adv. Mater. 28(44), 9735 (2016)

[33] I. Castillo, T. Sohier, M. Paillet, D. Cakiroglu, C. Consejo, C. Wen, F. W. Klein, M. Q. Zhao, A. Ouerghi, S. Contreras, A. T. C. Johnson, M. J. Verstraete, B. Jouault, and S. Nanot, Metal−insulator crossover in monolayer MoS₂, Nanotechnology 34(33), 335202 (2023)

[34] X. Cui, G. H. Lee, Y. D. Kim, G. Arefe, P. Y. Huang, C. H. Lee, D. A. Chenet, X. Zhang, L. Wang, F. Ye, F. Pizzocchero, B. S. Jessen, K. Watanabe, T. Taniguchi, D. A. Muller, T. Low, P. Kim, and J. Hone, Multi-terminal transport measurements of MoS₂ using a van der Waals heterostructure device platform, Nat. Nanotechnol. 10(6), 534 (2015)

[35] L. Yu, Y. H. Lee, X. Ling, E. J. G. Santos, Y. C. Shin, Y. Lin, M. Dubey, E. Kaxiras, J. Kong, H. Wang, and T. Palacios, Graphene/MoS₂ hybrid technology for large-scale two-dimensional electronics, Nano Lett. 14(6), 3055 (2014)

[36] S. Das Sarma, M. P. Lilly, E. H. Hwang, L. N. Pfeiffer, K. W. West, and J. L. Reno, Two-dimensional metal−insulator transition as a percolation transition in a high-mobility electron system, Phys. Rev. Lett. 94(13), 136401 (2005)

[37] H. J. Chuang, X. Tan, N. J. Ghimire, M. M. Perera, B. Chamlagain, M. M. C. Cheng, J. Yan, D. Mandrus, D. Tomanek, and Z. Zhou, High mobility WSe₂ p- and n-type field-effect transistors contacted by highly doped graphene for low-resistance contacts, Nano Lett. 14(6), 3594 (2014)

[38] B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti, and A. Kis, Single-layer MoS₂ transistors, Nat. Nanotechnol. 6(3), 147 (2011)

[39] S. Ghatak, A. N. Pal, and A. Ghosh, Nature of electronic states in atomically thin MoS₂ field-effect transistors, ACS Nano 5(10), 7707 (2011)

[40] K. Kaasbjerg, K. S. Thygesen, and K. W. Jacobsen, Phonon-limited mobility in n-type single-layer MoS₂ from first principles, Phys. Rev. B 85(11), 115317 (2012)

[41] M. Farkous, M. El-Yadri, H. Erguig, L. M. Perez, D. Laroze, C. V. Nguyen, N. T. T. Binh, N. N. Hieu, H. V. Phuc, M. Sadoqi, G. Long, and E. Feddi, Anisotropy of effective masses induced by strain in Janus MoSSe and WSSe monolayers, Physica E 134, 114826 (2021)

[42] A. Punnoose and A. M. Finkel’stein, Metal−insulator transition in disordered two-dimensional electron systems, Science 310(5746), 289 (2005)

[43] L. H. Ho, A. P. Micolich, A. R. Hamilton, and O. P. Sushkov, Ground-plane screening of Coulomb interactions in two-dimensional systems: How effectively can one two-dimensional system screen interactions in another, Phys. Rev. B 80(15), 155412 (2009)

[44] E. Abrahams, P. W. Anderson, D. C. Licciardello, and T. V. Ramakrishnan, Scaling theory of localization: Absence of quantum diffusion in two dimensions, Phys. Rev. Lett. 42(10), 673 (1979)

RIGHTS & PERMISSIONS

Higher Education Press