Defect‐Engineered n‐Doping of WSe₂ via Argon Plasma Treatment and Its Application in Field‐Effect Transistors

Abstract

limits the further downscaling of Si-based electronic devices. 2D layered materials have attracted significant attention as post-Si nanoelectronic materials owing to their intriguing properties such as atomic thickness, having no dangling bond in a surface, high surface-to-volume ratio, and immunity to the short-channel effect. [1] Among the various 2D materials, 2H transition metal dichalcogenides (TMDCs), represented by MoS 2 and WSe 2 , are promising candidates for next-generation semiconductor materials in a post-Si era. They have suitable energy bandgaps and excellent electrical and optoelectronic properties that can overcome the fundamental limitations of graphene, which has zero energy bandgap. [2] TMDCs have the chemical structure of MX 2 , where M is a transition metal and X is a chalcogen atom, and they present a versatile platform with unique characteristics that depend on the combination of atoms. [3] In particular, WSe 2 exhibits ambipolar characteristics with high carrier mobilities (≈250 cm 2 V −1 •s for electron and 270 cm 2 V −1 •s for hole) and a bandgap that can be tuned according to the thickness (1.2 eV for monolayer and 1.7 eV for bulk). [4] It has been utilized in various electronic and optoelectronic applications, such as volatile and non-volatile memory, transistors, and photodetectors. [5][6][7][8][9] Despite the desirable characteristics of 2D TMDCs, the performance of intrinsic 2D TMDC-based electronic devices is still limited by low carrier concentrations and high contact resistance. [10][11][12] Impurity doping is a useful strategy to address these limitations. In conventional Si-based complementary metal-oxide-semiconductor (CMOS) technology, substitutional doping methods using ion implantation have been used to obtain electron-dominant and hole-dominant transport. [13,14] However, the ion implantation technique is inadequate for atomically thin semiconductor materials because it is extremely challenging to control the stopping ranges within a few nanometers. Ambipolar semiconductors which can exploit electrons and holes simultaneously in one channel are expected to simplify the complex ion implantation and subsequent rapid thermal processes. [15] Moreover, the use of ambipolar 2D semiconductor materials facilitates a reduction in footprint, providing opportunities for further miniaturization. If the majority carrier of an ambipolar 2D semiconductor can be effectively controlled, a high-performance multi-mode homojunction nanodevice Doping of van der Waals layered semiconductor materials is an essential technique to realize their full potential for implementation in nanoelectronics.Herein, defect-engineered and area-selective n-doping of ambipolar multi-layer WSe 2 are demonstrated via Ar plasma treatment. The contact regions of the WSe 2 are exposed to a mild Ar plasma treatment to induce Se vacancy, while the channel region is protected by a hexagonal boron nitride. The results are systematically analyzed using structural and optical characterization methods, and the origin of the...

Selective Electron Beam Patterning of Oxygen‐Doped WSe2 for Seamless Lateral Junction Transistors

“…Defects, such as Se vacancies, can induce n‐type doping. [ 33 , 34 ] However, the formation of Se vacancies requires a high‐energy process, for example, Ar + plasma treatment or strong e‐beam irradiation with an ultrahigh dose (>4000 µC cm –2 ) typically used in transmission electron microscopy (TEM). Our e‐beam irradiation used the standard EBL process with a low area dose (≈600 µC cm –2 ).…”

“…Our e‐beam irradiation used the standard EBL process with a low area dose (≈600 µC cm –2 ). [ 33 , 34 , 35 ] Matsunaga et al. found that a low‐energy e‐beam created an S vacancy‐induced strain in the MoS 2 crystal, which modified the bandgap of MoS 2 .…”

Normally Off WSe2 Nanosheet-Based Field-Effect Transistors with Self-Aligned Contact Doping

“…The n-doping of intrinsically ambipolar multilayer-WSe 2 in the contact region was achieved by intentionally creating Se vacancies that can act as a donor site near the edge of the conduction band (shallow defect bands) by using a mild Ar-ion bombardment. , Tosun et al reported that the Fermi level of n-doped WSe 2 by the mild H 2 plasma treatment was estimated to be 10 meV below the conduction band edge . Kim et al reported that 0.33 and 0.46 eV defect levels of Se vacancy were generated below the conduction band under the mild Ar-ion plasma treatment. The defect engineering is advantageous for doping atomically thin 2D materials, in which case, the conventional ion implantation method is inappropriate.…”

“…32,34 Tosun et al reported that the Fermi level of n-doped WSe 2 by the mild H 2 plasma treatment was estimated to be 10 meV below the conduction band edge. 34 Kim et al 35 reported that 0.33 and 0.46 eV defect levels of Se vacancy were generated below the conduction band under the mild Ar-ion plasma treatment. The defect engineering is advantageous for doping atomically thin 2D materials, in which case, the conventional ion implantation method is inappropriate.…”

Interlayer Hopping Kinetics of Vacancies in CrI3 Layers Leading to Monolayer/Bilayer Heterostructures

“…[8] quite prominent, indicating its possible application in nonlinear and nonreciprocal optical devices. [17] Point defects have significant influence on physical and chemical properties of 2D materials [18][19][20][21][22][23][24] and are important with respect to the performance of devices. Various types of point defects in CrI 3 monolayer, including Cr vacancy and I vacancy, have been investigated theoretically.…”