ultrasensitive multilayer mos 2 ‐based photodetector with permanently grounded...

1901256 (1 of 8) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advelectronicmat.de

CommuniCation

Ultrasensitive Multilayer MoS2-Based Photodetector with Permanently Grounded Gate Effect

Muhammad Naqi, Manasa Kaniselvan, Sooho Choo, Gyuchull Han, Sangjin Kang, Jeonghun Kim, Youngki Yoon,* and Sunkook Kim*

DOI: 10.1002/aelm.201901256

chemical doping, elemental doping, and electrostatic gating techniques. The pho-toresponsive performances of PN-junction photodiodes have thus been effectively increased compared to intrinsic 2D mate-rial-based photodiodes.[18,19] Although PN-homojunction photodiodes exhibit better optical properties than intrinsic 2D materials photodiodes, there are many challenges present in fabricating the junctions. Chemical doping techniques are highly dependent on environmental conditions,[12,13] the elemental doping method causes limited vertical junction formation which leads to lower charge mobility,[5] and the electrostatic doping method has led to devices with slow photoresponses.[14] Heterojunction pho-todiodes with dissimilar materials and unequal bandgap have also been studied, using MoS2/Si,[20,21] MoS2/GaN,[22] MoS2/graphene,[23,24] MoS2/Black Phosphorous (BP),[6,25] and MoS2/pentacene,[26,27] and

show improved electrical and optical properties compared to homojunction photodiodes. However, despite notable perfor-mance improvements in terms of photoresponsivity and sensi-tivity, homo- and heterojunction photodiodes are limited by low carrier mobility, difficult processing techniques, and unstable photoresponsive behavior.[18,19] In these designs, the chosen fabrication technique also has a significant influence on the device performance.

Many studies have also been reported on Schottky bar-rier phototransistors, which incorporate a gate terminal to a metal–semiconductor–metal (M-S-M) design to obtain reliable electrical and optical properties along with a stable photore-sponse and easier processing methods.[28,29] The gating effect is a key device function in these devices as it tunes the Schottky barrier, which controls charge transfer in the device.[30,31] The forward and reverse saturation currents can be determined by modulating the Schottky barrier by applying positive, nega-tive, or zero bias.[28] However, electrical characteristics are often limited by the interface properties of the materials used, and more generally by fabrication techniques. The investigation of the Schottky contact between metals and MoS2 is now gaining interest due to possible applications in photodetectors, biosen-sors, photovoltaics, as well as field-effect transistors (FETs).

In this work, we introduce a unique design for MoS2 photo-diodes. This design uses a multilayer MoS2 semiconductor layer

2D materials, specifically MoS2 semiconductors, have received tremendous attention for photo-sensing applications due to their tunable bandgap and low noise levels. A unique photodetector using multilayer MoS2 as the semiconductor channel, in which the gate electrode of the device is perma-nently connected to the grounded source electrode to introduce rectification, is reported. The proposed grounded-gate photodiode exhibits high photo-responsivity of 1.031 A W−1, excellent photodetectivity (>6 × 1010 jones), and highly stable rise/fall time response (100–200 ms) under illumination of visible light (at the wavelengths of 405, 532, and 638 nm). Numerical device simulations using quantum transport methods and photoconductive effects are used to explain the device operation. It is also suggested that the gate metal work function can be carefully chosen to increase the sensitivity of the grounded-gate photodetector by suppressing the dark current. The grounded-gate device proposed, owing to the properties of rectifying behavior, low contact resistance, consistent photoresponsivity, and linear sensitivity, provides a new platform for next-generation applications in the field of electronics and optoelectronics.

M. Naqi, S. Choo, S. Kang, J. Kim, Prof. S. KimMultifunctional Nano Bio-Electronics LabDepartment of Advanced Materials and Science EngineeringSungkyunkwan UniversityGyeonggi-do, Suwon 16419, South KoreaE-mail: [email protected]. Kaniselvan, G. Han, Prof. Y. YoonWaterloo Institute for Nanotechnology (WIN) & Department of Electrical and Computer EngineeringUniversity of WaterlooWaterloo, Ontario N2L 3G1, CanadaE-mail: [email protected]

The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/aelm.201901256.

Photodetectors based on transition metal dichalcogenides, specifically molybdenum disulfide (MoS2), have gained interest in the fields of electronics and optoelectronics.[1–4] Some have attained good photodetection based on homo- and heterojunction designs,[5–7] and others exhibit modest pho-toresponsivity and photodetectivity by efficiently separating photogenerated electron-hole pairs.[8,9] In terms of homo-junction photodiodes, 2D materials such as graphene,[10,11] MoS2,[12,13] WSe2,[14] MoSe2,[15,16] and black phosphorus[17] have been extensively studied to form PN-junctions through

Adv. Electron. Mater. 2020, 1901256

www.advancedsciencenews.com

© 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim1901256 (2 of 8)


between metal source and drain electrodes made of titanium and gold (Ti/Au), as well as a global bottom gate terminal which is permanently shorted to the ground contact of the source (VG = 0 V) to preset Schottky barriers. The multilayer MoS2 was transferred onto the global bottom-gate/dielectric (Al2O3) layers through mechanical exfoliation, and material characterization has been done by Raman spectroscopy. The electrical current–voltage (ID–VD) curve shows clear rectification behavior, and the proposed device operation is explained through numerical simulations. The device has been investigated under different incident visible light, that is, blue (405 nm), green (532 nm), and red (638 nm). The optical results show linear sensitivity, high photoresponsivity of 1.031 A W−1, excellent photodetec-tivity (>6 × 1010 jones) and highly stable rise/fall time response (100–200 ms) under different light illumination intensities. Although the photodiode presented here was initially intended to measure photoresponse at different wavelengths, it may be further enhanced to be used in high-performance touch panels, large-scale photodetectors, and rectifier devices, thus providing an innovative platform for next-generation electronic and opto-electronic applications.

Photodiodes based on PN-junctions have been previously studied to obtain good photoresponse by optimizing doping methods, but it remains challenging to obtain linear photosen-sitivity over a wide spectral wavelength range. [18,19] Similarly,

the presence of a gate terminal in FET presents another way of optimizing the diode characteristics, but this has not expan-sively studied in the field of optoelectronic devices.[28,30,31] Our design attempts to overcome the existing limitations by using a permanently grounded-gate photodiode, which has a preset Schottky barrier and rectification behavior. Figure 1a presents a schematic of the structural design of the photodiode, in which the gate is permanently shorted to the ground (source) contact. To fabricate the proposed photodiode, the gate is first patterned on a rigid glass substrate using photolithography, and then a dielectric layer of Al2O3 (80 nm) is deposited via atomic layer deposition (ALD). Then, multilayer MoS2 is transferred through mechanical exfoliation onto a pre-patterned gate/dielectric layer substrate. After making a via for the gate electrode by an etching process, the source and drain electrodes have been patterned such that the source electrode is shorted to gate elec-trode by the via. The sequential fabrication process of proposed grounded-gate photodiode can be seen in the schematic layout of Figure 1b. The real image of the proposed photodiode is pre-sented in Figure 1c along with an inset of the integrated layout. The photodiode has a channel length of 12 μm and width of 7.74 μm, both measured in ambient conditions. Recent studies on monolayer and multilayer MoS2-based photodetectors show that both are capable of impressive photoresponsivity and high detectivity,[32,33] but multilayer MoS2 in particular offers a wider


Figure 1. Multilayer MoS2 photodiode with a permanently grounded gate. a) A schematic layout of the presented photodiode under various light illuminations of different wavelengths (blue: 405 nm, green: 532 nm, and red: 638 nm). b) A sequential fabrication process of the proposed photodiode. c) An optical image of the photodiode. d) Raman spectrum of the multilayer MoS2.




spectral range due to its narrower bandgap and higher optical absorption.[34] Multilayer MoS2 can also produce substantially higher currents compared to the single-layer MoS2 due to its higher density of states (DOS).[35,36] The exfoliated multilayer MoS2 was characterized by Raman spectroscopy, and the measured intensity spectrum exhibits A1g and E12g peaks with a difference of 26.66 cm−1, which indicates the presence of multilayer MoS2 (Figure 1d).

After fabrication of the presented device, its diode charac-teristics were measured and the ID–VD curve exhibited clear rectifying behavior from reverse bias (VD < 0 V) to forward bias (VD > 0 V) in ambient conditions (Figure S1, Supporting Information). A diode curve in linear scale can also be seen in the inset of Figure S1, Supporting Information. In addition, the rectification effect of our proposed grounded-gate photo-diode was further verified by measuring ten fabricated devices; the consistent on/off ratio is shown in Figure S2, Supporting Information. The difference between the ID–VD curves under dark and illuminated conditions can be seen in Figure 2a. The ID–VD curve has an increase in current (ID) by three orders of magnitude under forward bias when illuminated with blue light (405 nm) with a power intensity of 430 W m−2. On the contrary, the change in current at the reverse bias is relatively insignificant under illumination. To better understand the

device performance, quantum transport simulations using the non-equilibrium Green's function formalism were performed for this device architecture. Model parameters (e.g., Schottky barrier height, gate metal work function) were found such that the simulated ID–VD characteristics (which are presented in Figure 2b) resemble the corresponding experimental data (Figure 2a). These parameters result in the energy band dia-grams shown in Figures 2c (reverse bias) and 2d (forward bias), which can be used to investigate current flow in the device in each state of operation. Under reverse bias, due to the gating effect, the energy barrier created at the gate-drain junction is sharp enough to allow for tunneling current in the same direc-tion as the photocurrent. However, in forward bias, the energy barrier at the source-gate junction is too wide to permit tun-neling current and high enough to prevent significant thermi-onic current, while photocurrent is unobstructed.

The rectification effect caused in the presence of a grounded-gate terminal serves to increase the photosensitivity of the device by decreasing the dark current under the for-ward bias condition. This can be seen clearly by comparing it to a regular two-terminal M-S-M photodetector, in which the energy bands are symmetric under forward and reverse bias (see Figure S3, Supporting Information for experimental data). To fabricate the two-terminal M-S-M photodetector,


Figure 2. Electrical properties of the grounded-gate photodiode with multilayer MoS2. a) Electrical measurement of the photodiode under dark and light illuminations [blue light (405 nm), incident power density (Pinc) = 430 W m−2]. b) Simulated electrical measurement of the photodiode design under dark and illuminated conditions [blue light (405 nm), Pinc = 430 W m−2]. Simulated band structure of device under c) reverse (VD = −1 V) and d) forward bias (VD = +1 V), where blue circles represent injected carriers and red circles represent photo-generated carriers.




the multilayer MoS2 was exfoliated and transferred onto the Si wafer substrate and then source/drain electrodes were patterned by using the photolithography process. Figure 3a presents the simulated ID–VD characteristics, where at posi-tive drain voltages, the dark current for the grounded-gate photodiode is negligible but that of the two-terminal device is steadily increasing. As illustrated in Figure 3b, without a gating effect under forward bias to obstruct dark current flow, tunneling current is significant as compared to that with the grounded-gate photodiode (see Figure 2d). The addition of the gate terminal adds the capability to have a preset modu-lation of the potential along the channel without needing to operate it as a three-terminal device. Therefore, the grounded-gate photodiode architecture forgoes part of the complexity of a three-terminal device while maintaining the performance advantages of the gating effect.

The optoelectronic behavior of the proposed grounded-gate photodiode in the dark and under different light illumination intensities was then examined. A gradual increase in photo-generated current was observed under forward bias when exposed to red (638 nm), green (532 nm), and blue (405 nm) light at the same power (500 μW), described in Figure 4a. Due to its higher photon energy, blue light generates more photoinduced carriers than red and green at the same power.[37] To evaluate photoresponsivity as a function of incident light power, the photodiode was examined under blue light at dif-ferent intensities ranging from 200 μW to 3 mW at an applied drain voltage range of − 5 to 5 V, as shown in Figure 4b. The same measurements were taken under green and red light, which are shown in Figure S4a,b, Supporting Information. The measurement results show excellent photodetection capabilities and a clear distinction between the photocurrents under each condition. The device was also evaluated according to several key figures of merit for photodetection, such as responsivity (R), sensitivity (S), and specific detectivity (D*). Photorespon-sivity is a measure of photodetector gain, and is calculated as R = Iph/Pinc (A W−1), where Iph represents the photocurrent and Pinc represents the incident light power. Photosensitivity is defined as S = Iph/Idark, where Idark represents the dark cur-rent (i.e., current under dark conditions). Photocurrent can be obtained by Iph = Itotal − Idark, where Itotal is the total current measured at a specific power of light illumination. Due to the dark current being the dominant source of noise in this photo-detector, the sensitivity can be thought of as a signal-to-noise ratio. The specific detectivity is defined as D* = RA1/2/(2eIdark)1/2, where A is the illuminated area and e is the elementary charge. The responsivity and the sensitivity of the photodiode were measured under blue light at various power intensities from 200 μW to 3 mW and the functional relationship of photore-sponsivity and light power is fitted by a power law behavior (R ≈ Pincβ-1) with the fitting parameter β = 0.37 (Figure 4c). Results show responsivities in the range of 0.2–1.1 A W−1, exhibiting high reliability in photodetection. The sensitivity was in the range of 20–180 and also distinctly linear. The same measurements were performed for green and red light, and are presented in Figure S5a,b, Supporting Information. The spe-cific detectivity of the photodiode was measured under different wavelengths of light at an incident power of 500 μW, which is shown in Figure 4d. As the wavelength increases from blue to red, there are gradual decreases in detectivity and responsivity due to the reduced photon energy.[37] The results show linear trends in the range of 6.5–3.2 × 1010 jones for detectivity and 1.1–0.5 A W−1 for responsivity, exhibiting highly sensitive and responsive performance.

Next, the time-domain behavior of the device was charac-terized by measuring its photoresponse under pulses of light with different wavelengths (638, 532, and 405 nm) at a con-stant VD = 1 V. Under a pulse of red light (638 nm) of inten-sity 200 μW (Figure 5a), there was an increase in current by one order of magnitude with a rise time of 194.19 ms and fall-time of 48.53 ms. Green (blue) light pulses of the same intensity generated a larger increase in current (1.5 orders of magnitude) with a rise time of 234.31 ms (127.14 ms) and fall time of 24.29 ms (127.12 ms), which can be seen in Figure 5b (Figure 5c). The reproducibility of the photoresponse was


Figure 3. Introduction of rectification in proposed grounded-gate photo-diode compared to the two-terminal photodiode. a) Electrical character-istics (dark current) of both the proposed grounded-gate photodiode and a two-terminal device. b) Band structure of the simulated two-terminal device under forward bias (VD = +1 V), indicating lack of tunneling cur-rent suppression.




examined by exposing the device to each wavelength of light at 500 μW in seven distinct pulses (repeats of 10 s on and 10 s off with 150 s in total), presented in Figure 5d. Both the photoresponse amplitude and switching speed were stable throughout the measurement, demonstrating robustness. We compare this grounded-gate photodiode with recently reported PN homo/heterojunctions using the benchmark parameters of device configuration, deposition method, responsivity, detec-tivity, and rise and fall times; a full summary is presented in Table S1, Supporting Information. In comparison to the other reported devices, the photodiode presented in this work dem-onstrates reliable photoresponsivity (>1 A W−1), high photo-detectivity (>6 × 1010 jones), and highly stable response time (100–200 ms), which can be achieved through simple pro-cessing methods.

To further improve the photodetection capabilities of this design, two areas could be targeted; increasing the photo-current by making the generation of excess carriers more effi-cient, or suppressing dark current. Since the former was inves-tigated by varying wavelength and incident power density in Figure 4, here we focus on the suppression of the dark current to increase the sensitivity of the photodetector while leaving the photocurrent unchanged. In principle, for three-terminal devices, dark current can be suppressed by increasing the channel potential barrier through gate voltage modulation. A similar effect can be achieved in the presented grounded-gate

device by choosing a gate metal with a different work function, defined as Φm = − eΦV − EF, where ΦV is the electrostatic poten-tial of the vacuum and EF is the Fermi level of the metal. For the device simulations in Figure 2b, a nominal gate metal work function m

0Φ was used for the photodiode design. Figure 6a pre-sents the effect of changing the gate metal work function with respect to m

0Φ on the dark current of the photodiode. When increased (i.e., 0m m

0Φ − Φ > eV), the energy barrier for elec-tron injection from the source to the channel under forward bias is increased, which further suppresses the thermionic current. This reduction in dark current serves to improve the sensitivity of the design. Figure 6b presents the change in sen-sitivity based on the simulated range of the variation of gate metal work function. Notably, an increase of 0.20 eV causes over three orders of magnitude of increase in the sensitivity. Therefore, the device design can be optimized to detect very low levels of light by using a gate metal with a larger metal work function.

We have introduced a unique photodiode design which uses a multilayer MoS2 semiconductor material and a grounded-gate terminal. Using energy band diagrams from quantum transport simulations, we explained how the existence of the grounded-gate terminal adds a rectifi-cation effect and a preset Schottky barrier. The device was fabricated using simple processing techniques, and its optoelectronic performance was characterized by exposure


Figure 4. Optical measurements of the proposed photodiode. a) Diode curves of different light illumination (blue, green, and red) at the same power of 500 μW along with the dark condition. b) The ID–VD characteristics of the device under blue (405 nm) light illumination at different power ranges from 200 μW to 3 mW along with the dark condition. c) The photoresponsivity and sensitivity measurements of the proposed photodiode under blue (405 nm) light illumination at different power ranges from 200 μW to 3 mW. The photoresponsivity versus power intensity curve is fitted to the power law (R ≈ Pincβ-1). d) The photoresponsivity and photodetectivity representations of various light illuminations with different wavelength (405, 532, and 638 nm) at same power of 500 μW.




to three different wavelengths of visible light (405, 532, and 638 nm) under different incident power intensities ranging from 200 μW to 3 mW. The measurement results show that this design is capable of high photoresponsivity (1.031 A W−1), reliable photodetectivity (≈6 × 1010 per jones), and highly stable response time (100–200 ms) and wide spectral range of detection when compared to other reported homo/heterojunction and M-S-M photo diodes. The demon-strated grounded-gate MoS2 photodiode shows promising potential for next-generation applications in the field of optoelectronics.

Experimental SectionFabrication of the Proposed Photodiode: In fabricating the proposed

photodiode, the gate electrode was patterned as a first step, in which Titanium and Gold (Ti/Au) with the thickness of 10/50 nm were deposited by e-beam evaporation and then patterned as a global bottom gate electrode through the photolithographic process. In patterning the gate electrode, the photoresist (PR) was spin-coated with 3000 rpm for 30 s, then exposed to a UV light in the presence of the designed mask and developed in the developer for 30 s. After developing, the presented device proceeded to the etching process, in which it was etched in Au and diluted buffer oxide etchant for 5 and 20 s, respectively. After patterning the gate electrode, the dielectric layer of Al2O3 with a


Figure 5. Time response measurements of the grounded-gate photodiode at an applied voltage of 1 V under various light illuminations. a) Red light (638 nm), b) green light (532 nm), and c) blue light (405 nm). d) The photo-switching characteristics of the device under blue, green and red light illumination in 10 s pulses, up to 150 s at the same power of 200 μW.

Figure 6. Suppressing dark current by changing the gate metal work function. a) Effect of changing the gate metal work function on the electrical characteristics (dark current) of the device. b) Effect of changing the gate metal work function on the sensitivity of the grounded-gate photodiode.




thickness of 80 nm was deposited through the atomic layer deposition (ALD) technique and then the via of the gate electrode was patterned through the above-mentioned process of photolithography. In the next step, the single crystalline multilayer MoS2 semiconductor was mechanically exfoliated by the conventional scotch tape method onto the pre-deposited Al2O3 and global bottom gate substrate. As a final step, the anode (drain) and cathode (source) were patterned with a selective contact of cathode (source)—gate through the lift-off process, in which negative photoresist, LOR3B, was spin-coated with 1000 rpm for 40 s annealed at 180 °C for 4 min and followed by the positive photoresist (PR), spin coating of 3000 rpm for 30 s annealed at 90 °C for 3 min. Then, UV light was exposed for 1 s onto the pre-deposited substrate in the presence of the designed mask and developed in the developer for 20 s. After that, the Ti/Au with a thickness of 20/100 nm was deposited by e-beam evaporation process onto the pre-patterned anode (drain) and cathode (source) electrodes and the Ti/Au was removed from the unselected area by using the PG-remover. Negative photoresist LOR3B (Product no. G3167070500L1GL, MicroChem), Positive photoresist (AZGXR-601, MERCK), single-crystalline MoS2 (SPI supplies, USA), and PG-remover (mr-Rem 700, Micro-Resist Technology) were purchased and used as received.

Material Characterizations: To obtain the material characterization, the Raman spectra was measured through the Raman spectroscopy system (WITec Co., ALPHA300) with an excitation laser of 532 nm at the power of 2 mW.

Electrical Characterizations: The electrical and photoresponse characteristics of the multilayer MoS2 photodiode were measured using a semiconductor characterization system (Keithley, 4200 SCS) under different light illumination wavelengths of 405 and 638 nm (Thorlabs, S405-HP for 405 nm and SM600 for 638 nm), and 532 nm (Changchun New Industries Optoelectronics Technology Co., MGL-FN-532) in ambient conditions.

Device Simulation: Dark current in the device was simulated using the non-equilibrium Green's function (NEGF) formalism with electrostatics described by the Poisson equation.[38] The simulated device structure was a Schottky barrier FET with a Schottky barrier height of ΦBn = 0.1 eV, and a 40 nm-long monolayer MoS2 channel. A 20-nm gate was used with 10-nm gate underlap on each side. The gate oxide was 2.5 nm-thick SiO2. Although the simulation considered different channel dimensions than those of the fabricated device, the underlying physics explaining the trends in electrical and optical characteristics would not be changed. The Hamiltonian for the NEGF transport solver was constructed using an effective mass approximation (m* = 0.45m0). Due to the relatively short channel length, ballistic transport was assumed through the channel' is providing clear message of our work. The photoconductive effect was considered to calculate photocurrent.[39]

Supporting InformationSupporting Information is available from the Wiley Online Library or from the author.

AcknowledgementsM.N., M.K., and S.C. equally contributed to this work. This research was supported in part by the National Research Foundation of Korea (2018R1A2B2003558), NSERC Discovery Grant (RGPIN-05920-2014), and Ontario MRIS Early Researcher Award (ER17-13-205). Computing resources were provided by Compute Canada.

Conflict of InterestThe authors declare no conflict of interest.

Keywordsgrounded-gate effect, molybdenum disulfide, photodetectors, photodiodes, transition metal dichalcogenides

Received: November 13, 2019Revised: December 25, 2019

Published online:

[1] O. Lopez-Sanchez, D. Lembke, M. Kayci, A. Radenovic, A. Kis, Nat. Nanotechnol. 2013, 8, 497.

[2] N. Huo, G. Konstantatos, Nat. Commun. 2017, 8, 1.[3] S. Kim, A. Konar, W. S. Hwang, J. H. Lee, J. Lee, J. Yang, C. Jung,

H. Kim, J. B. Yoo, J. Y. Choi, Y. W. Jin, S. Y. Lee, D. Jena, W. Choi, K. Kim, Nat. Commun. 2012, 3, 1011.

[4] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, M. S. Strano, Nat. Nanotechnol. 2012, 7, 699.

[5] Y. Jin, D. H. Keum, S. J. An, J. Kim, H. S. Lee, Y. H. Lee, Adv. Mater. 2015, 27, 5534.

[6] L. Ye, H. Li, Z. Chen, J. Xu, ACS Photonics 2016, 3, 692.[7] P. Vabbina, N. Choudhary, A. A. Chowdhury, R. Sinha, M. Karabiyik,

S. Das, W. Choi, N. Pala, ACS Appl. Mater. Interfaces 2015, 7, 15206.

[8] F. H. L. Koppens, T. Mueller, P. Avouris, A. C. Ferrari, M. S. Vitiello, M. Polini, Nat. Nanotechnol. 2014, 9, 780.

[9] W. Zhang, Q. Wang, Y. Chen, Z. Wang, A. T. S. Wee, 2D Mater. 2016, 3, 022001.

[10] E. C. Peters, E. J. H. Lee, M. Burghard, K. Kern, Appl. Phys. Lett. 2010, 97, 193102.

[11] M. C. Lemme, F. H. L. Koppens, A. L. Falk, M. S. Rudner, H. Park, L. S. Levitov, C. M. Marcus, Nano Lett. 2011, 11, 4134.

[12] M. S. Choi, D. Qu, D. Lee, X. Liu, K. Watanabe, T. Taniguchi, W. J. Yoo, ACS Nano 2014, 8, 9332.

[13] S. Wi, H. Kim, M. Chen, H. Nam, L. J. Guo, E. Meyhofer, X. Liang, ACS Nano 2014, 8, 5270.

[14] D. J. Groenendijk, M. Buscema, G. A. Steele, S. Michaelis De Vas-concellos, R. Bratschitsch, H. S. J. Van Der Zant, A. Castellanos-Gomez, Nano Lett. 2014, 14, 5846.

[15] Y. Yang, N. Huo, J. Li, J. Mater. Chem. C 2017, 5, 7051.[16] S. Fan, W. Shen, C. An, Z. Sun, S. Wu, L. Xu, D. Sun, X. Hu,

D. Zhang, J. Liu, ACS Appl. Mater. Interfaces 2018, 10, 26533.[17] M. Buscema, D. J. Groenendijk, G. A. Steele, H. S. J. Van Der Zant,

A. Castellanos-Gomez, Nat. Commun. 2014, 5, 4651.[18] M. Long, P. Wang, H. Fang, W. Hu, Adv. Funct. Mater. 2018, 29,

1803807.[19] N. Huo, G. Konstantatos, Adv. Mater. 2018, 30, 1801164.[20] L. Wang, J. Jie, Z. Shao, Q. Zhang, X. Zhang, Y. Wang, Z. Sun,

S. T. Lee, Adv. Funct. Mater. 2015, 25, 2910.[21] O. Lopez-Sanchez, E. Alarcon Llado, V. Koman, A. Fontcuberta I

Morral, A. Radenovic, A. Kis, ACS Nano 2014, 8, 3042.[22] Z. Xu, S. Lin, X. Li, S. Zhang, Z. Wu, W. Xu, Y. Lu, S. Xu, Nano

Energy 2016, 23, 89.[23] J. Y. Kwak, J. Hwang, B. Calderon, H. Alsalman, N. Munoz,

B. Schutter, M. G. Spencer, Nano Lett. 2014, 14, 4511.[24] Z. Huang, W. Han, H. Tang, L. Ren, D. S. Chander, X. Qi, H. Zhang,

2D Mater. 2015, 2, 035011.[25] Y. Deng, Z. Luo, N. J. Conrad, H. Liu, Y. Gong, S. Najmaei,

P. M. Ajayan, J. Lou, X. Xu, P. D. Ye, ACS Nano 2014, 8, 8292.[26] Q. Ren, Q. Xu, H. Xia, X. Luo, F. Zhao, L. Sun, Y. Li, W. Lv,

L. Du, Y. Peng, Z. Zhao, Org. Electron. Phys. Mater. Appl. 2017, 51, 142.

[27] Y. Peng, R. Ding, Q. Ren, S. Xu, L. Sun, Y. Wang, F. Lu, Appl. Surf. Sci. 2018, 459, 179.





[28] A. Di Bartolomeo, A. Grillo, F. Urban, L. Iemmo, F. Giubileo, G. Luongo, G. Amato, L. Croin, L. Sun, S. J. Liang, L. K. Ang, Adv. Funct. Mater. 2018, 28, 1800657.

[29] A. Di Bartolomeo, F. Giubileo, G. Luongo, A. Di Bartolomeo, G. Luongo, F. Giubileo, N. Funicello, G. Niu, 2D Mater. 2017, 4, 025075.

[30] S. Barrier, H. Yang, J. Heo, S. Park, H. J. Song, D. H. Seo, K. Byun, P. Kim, I. Yoo, H. Chung, K. Kim, Science 2012, 336, 1140.

[31] C. P. Lu, G. Li, J. Mao, L. M. Wang, E. Y. Andrei, Nano Lett. 2014, 14, 4628.

[32] N. Liu, J. Baek, S. M. Kim, S. Hong, Y. K. Hong, Y. S. Kim, H. S. Kim, S. Kim, J. Park, ACS Appl. Mater. Interfaces 2017, 9, 42943.

[33] D. J. Late, B. Liu, H. S. S. R. Matte, C. N. R. Rao, V. P. Dravid, Adv. Funct. Mater. 2012, 22, 1894.

[34] W. Choi, M. Y. Cho, A. Konar, J. H. Lee, G. B. Cha, S. C. Hong, S. Kim, J. Kim, D. Jena, J. Joo, S. Kim, Adv. Mater. 2012, 24, 5832.

[35] K. Natori, J. Appl. Phys. 1994, 76, 4879.[36] H. Wang, L. Yu, Y. H. Lee, Y. Shi, A. Hsu, M. L. Chin, L. J. Li,

M. Dubey, J. Kong, T. Palacios, Nano Lett. 2012, 12, 4674.[37] J. Yu, B. J. Kim, S. Park, I. K. Han, S. J. Kang, Jpn. J. Appl. Phys. 2018,

57, 044001.[38] S. Datta, Quantum Transport: Atom to Transistor, Cambridge Univer-

sity Press, Cambridge 2005.[39] M. M. Furchi, D. K. Polyushkin, A. Pospischil, T. Mueller, 2014, 14, 11.


ultrasensitive multilayer mos 2 ‐based photodetector with permanently grounded...

Documents