An in-plane WSe2 p–n homojunction two-dimensional diode by laser-induced doping

Sujeong Yang a, Geonyeop Lee a, Janghyuk Kim a, Seunghoon Yang b, Chul-Ho Lee *b and Jihyun Kim *a
aDepartment of Chemical and Biological Engineering, Korea University, Seoul 02841, Korea. E-mail: chlee80@korea.ac.kr; hyunhyun7@korea.ac.kr
bKU-KIST Graduate School of Converging Science and Technology, Korea University, Seoul 02841, Korea

Received 10th April 2020 , Accepted 26th May 2020

First published on 28th May 2020


Conventional doping schemes of Si microelectronics are inadequate for atomic-thickness two-dimensional (2D) semiconductors, which makes it challenging to construct 2D p–n homojunctions. Herein, a UV laser-assisted doping method with addressability is proposed for seamlessly building a 2D WSe2 p–n homojunction. WSe2 with ambipolar transport properties was exposed to a focused UV laser to form WOx in a self-limiting and area-selective process that induces hole doping in the underlying WSe2via electron transfer. Different electrical behaviors, ranging from p–p to p–n in-plane homojunctions, were observed between the as-exfoliated (ambipolar) region and the UV laser-treated (p-doped) region, under the electrostatic modulation of the back-gate bias (VBG), resulting in the multi-state rectification ratios of 895 (positive VBG) and ∼4 (negative VBG). The evolution of the depletion region in the WSe2 in-plane homojunction was analyzed at different VBG using the scanning photocurrent mapping approach, yielding a high photocurrent of 1.8 nA for positive VBG, owing to the development of the p–n junction. Finally, a WSe2-based 2D homogeneous complementary inverter is demonstrated with a voltage gain of 1.8, thereby paving the way for next-generation atomic-thickness circuitry.


Introduction

Two-dimensional (2D) materials, also called van der Waals (vdW) materials, bear the promise of offering solutions to overcome the physical limitations of silicon-based microelectronic devices, and they are likely to enable device downscaling and atomic-level circuit miniaturization.1,2 Owing to their unique electrical and optical properties, versatile 2D materials, such as graphene, transition metal dichalcogenides (TMDCs, e.g., MoS2, WSe2, and WS2), and black phosphorus, have opened the door to nanometer-thick electronic and optoelectronic devices with unprecedented properties.3,4 Numerous 2D material-based transistors,5,6 solar cells,7,8 memory devices,9 light-emitting devices,10 and sensors11 have been reported. Their chemical and mechanical robustness, combined with excellent mechanical flexibility, can be used for fabricating flexible electronic and optoelectronic devices.12 2D materials with high surface-to-volume ratios are ideal for high-performance chemical sensors and photodetectors.13–15

Tungsten diselenide (WSe2) is an intriguing 2D material because it exhibits ambipolar transport characteristics with thickness-dependent band-gaps of 1.7 eV (direct, monolayer) and ∼1.2 eV (indirect, multi-layer), along with theoretical electron and hole motilities of ∼250 and ∼270 cm2 V−1 s−1 respectively.16 The ambipolar characteristics of WSe2 permit selective amplification of electron or hole transport via doping.17–20 A chemical method based on the interfacial charge transfer interaction has been proposed by exposing a 2D material to an electron-donating agent (polyethyleneimine (PEI)) and an electron-withdrawing agent (fluoropolymer (CYTOP)).16,21 The carrier type of WSe2 was controlled by using iodine vapor and various molecules such as triphenylboron, tris(pentafluorophenyl)borane, and tritolyborane.22,23 Although the method is simple and effective, the involved chemical agent should be protected from the air. As an alternative, an electrostatic gating (or doping) method has been introduced, for controlling the type and concentration of the majority carrier, by applying a bias locally or globally to the WSe2 channel.24 Recently, it was reported that the under-stoichiometric oxidation of WSe2 into WOx (x ≤ 3) induces p-type doping in the neighboring (or underlying) WSe2via a charge transfer. Tungsten oxide itself is known to be an effective hole-injection layer in organic electronics, owing to its high electron affinity. Yamamoto et al. demonstrated a self-limiting doping process, in which exposure of WSe2 to ozone converted it into sub-stoichiometric WOx, thereby making the underlying WSe2 hole-doped.25 In addition, air exposure,26 oxygen plasma,27,28 UV–O3 treatment29 and thermal treatment30 were used for oxidizing WSe2, which converted ambipolar transport into hole-dominant transport in the WSe2 channel.

A p–n junction diode is one of the fundamental building blocks of semiconductor devices.31 However, p–n homojunctions based on 2D materials have been rarely reported, although homojunctions offer a clean and self-aligned interface when compared to heterojunctions, which suffer from unavoidable residues and a complicated flake alignment process.32 The atomic-scale thickness of 2D materials makes it inappropriate to use conventional doping methods, including ion implantation and diffusion. Moreover, both methods can unintentionally damage the 2D materials, which is generally undesirable for device performance.33 In our experiment, a seamless in-plane WSe2 2D p–n homojunction was fabricated using a laser-assisted doping technique with area selectivity. The combination of laser-induced p-doping and electrostatic doping enabled us to tune the device characteristics of the in-plane WSe2 homojunction from a p–p junction to a p–n junction. Micro-Raman spectroscopy and atomic force microscopy were used for analyzing the oxidation of the obtained exfoliated WSe2 flakes. The electrical and optoelectronic properties of the WSe2 p–p and p–n homojunctions were systemically characterized.

Experimental details

Fig. 1(a) shows optical microscope images depicting the fabrication process of the proposed seamless WSe2 homojunction device. WSe2 flakes were mechanically exfoliated from a bulk crystal (HQ graphene) using an adhesive tape, and they were then transferred to a transparent gel film (Gel-Pak). The WSe2 flakes were identified using an optical microscope (BX51M, Olympus) and subsequently dry-transferred onto a SiO2/Si substrate using a micromanipulator. Ti/Au electrodes (50/100 nm) were defined to contact the as-exfoliated region of WSe2 using standard electron-beam lithography, electron-beam evaporation, and a lift-off process. A He–Cd laser (KIMMON KOHA), with a wavelength of 325 nm, was used for selectively oxidizing the WSe2 flakes. The power density of the laser was 24.6 mW, and its resolution was ∼1 μm. The laser irradiation for each point was performed for 1 min under ambient conditions. Pt/Au (20/80 nm) electrodes were defined to contact laser-treated WSe2 flakes, using the same method as that used for the Ti/Au electrodes.

An atomic force microscope (AFM) (Innova, Bruker) was used to measure the thickness of the WSe2 flakes before and after the laser irradiation. The structural properties of the WSe2 before and after laser irradiation were measured by micro-Raman spectroscopy, using a solid-state laser (Omicron) with a wavelength of 532 nm in the back-scattering geometry. The electrical properties of the fabricated devices were characterized by using a semiconductor parameter analyzer (4155C, Agilent) connected to a vacuum probe station under a pressure of ∼10 mTorr. The scanning photocurrent was obtained using a probe station connected to a DC voltage source (GS 20, Yokogawa), a current amplifier (1211, DL Instrument), and a multimeter (34401A, Keysight), under irradiation by a 3 mW continuous-wave of a 532 nm-wavelength laser beam, focused using a 100× microscope objective lens.

Results and discussion

Fig. 1(b) shows the Raman spectra of the WSe2 flakes before and after the laser treatment, where the thickness of the as-exfoliated WSe2 flakes was in the 3–5 nm range. For thicker WSe2 flakes (>5 nm), the change in the Raman spectra induced by the laser treatment was indistinguishable owing to the strong intensity of the phonons scattered from the underlying WSe2. The E12g and A1g phonons with vibration frequencies of ∼249 cm−1 and ∼259 cm−1 diminished after the laser treatment, as shown in Fig. 1(b). This can be attributed to the oxidation of WSe2, which is consistent with the previous reports.27,34 The Raman spectroscopy results imply that the oxidized layer, which is assumed to be amorphous, is decoupled from the underlying WSe2. The optical microscope image (Fig. S1, ESI) shows that the laser-treated WSe2 flakes are transparent, although the as-exfoliated ones are still opaque, thereby indicating their conversion into transparent WOx.34Fig. 1(c) and (d) present the AFM analysis results for the as-exfoliated (red) and laser-treated (blue) flakes, revealing that the laser-assisted oxidation reduced the thickness of the flakes, which is in good agreement with the previous reports.35 Liu et al. reported that the thermal oxidation of 2D WSe2 resulted in the reduction of its thickness by forming WO3, which was initiated from the edges of 2D WSe2.30 These figures also show that the oxidation process was more effective at the edges of the flakes, where they are highly defective.36 A previous report shows that the oxidation process begins with the substitution of the Se vacancy with an O atom (VSe + Oad → OSe), which is a spontaneous reaction with an enthalpy of –4.7 eV.37 It should be noted that the roughness of the flakes was reduced after the laser treatment owing to the removal of surface residues.
image file: d0tc01790f-f1.tif
Fig. 1 (a) Optical microscope images showing the fabrication process of a seamless 2D WSe2 homojunction device. The white dashed line represents the laser-treated region of the WSe2 flake. The scale bar is 10 μm. (b) Micro-Raman spectra of the as-exfoliated WSe2 and laser-treated WSe2. Note that they are normalized to the Si peak. (c) AFM height profile and (d) AFM image of the as-exfoliated (red) and laser-treated (blue) WSe2.

Electrical properties were compared by fabricating the back-gated field-effect transistors (FETs) on both as-exfoliated and laser-treated areas. Fig. 2(a) schematically shows the fabricated FETs, where the electrodes (Ti/Au and Pt/Au) were chosen for improving the electrical contacts to each region. The work functions of the Ti (ΦTi = 4.3 eV) and Pt (ΦPt = 5.6 eV) facilitate the carrier transport for electrons and holes at the metal-semiconductor junction.38 The linear transfer curves at VDS = +1 V, for both as-exfoliated and laser-treated regions, are compared in Fig. 2(b) and (c) (insets are in the semi-logarithmic scale). The as-exfoliated WSe2 FET exhibited an ambipolar transfer characteristic, with the on/off ratios of 1.1 × 105 (hole conduction) and 3.4 × 104 (electron conduction). The laser-treated WSe2 FET displayed a p-type unipolar behavior, with an on/off ratio of 2.6 × 104, which is comparable to that of the as-exfoliated WSe2 FET in the hole-dominant region. Amorphous WOx is known to attract electrons from WSe2 owing to its higher electron affinity, thereby inducing the underlying hole-doped WSe2.25 The maximal hole current density greatly increased following the laser treatment, indicating area-selective p-doping of the WSe2 (see the ESI, Fig. S2). Liu et al. reported that partially oxidized WSe2 is more conductive than WSe2 with a sheet resistance of ∼105 Ω sq−1 by using microwave impedance microscopy.30 However, fully oxidized WO3 acted as a good electrical insulator with a sheet resistance of >109 Ω sq−1. Addou et al. reported that WSe2 exposed to ambient air exhibited metallic properties owing to the formation of WOx, which in turn induced hole conductivity in WSe2.39


image file: d0tc01790f-f2.tif
Fig. 2 (a) Schematic of the two 2D FETs fabricated on the as-exfoliated (Ti/Au electrodes, right) and laser-treated (Pt/Au electrodes, left) areas. DC transfer curves for both (b) as-exfoliated and (c) laser-treated WSe2, on the linear scale. Insets in (b and c) are on the semi-logarithmic scale.

The majority carrier in ambipolar WSe2 (as-exfoliated region) could be chosen by varying the VBG between the negative (hole-doped) and positive (electron-doped), while the majority carrier in the laser-treated region remained hole-dominant (p+ to p). Therefore, the seamless homojunction between the as-exfoliated and laser-treated regions could be set to either the p–p diode or the p–n diode, depending on the back-gate bias. The homojunction device is schematically depicted in Fig. 3(a). The output characteristics of the homojunction are summarized in Fig. 3(b) for different back-gate bias conditions, showing how the homojunction behavior changes from the p–p junction to the p–n junction. The VBG-dependent rectification ratio (|I+2V|/|I−2V|, (forward current at VDS = +2 V)/(reverse current at VDS = −2 V)) was calculated, as shown in Fig. 3(c). At VBG ≲ −50 V, the low rectification ratio can be explained by the formation of the p+–p junction, because of the electrostatic p-doping effect on both regions. The low rectification ratio (∼4) in the −50 V < VBG ≲ −20 V range can be attributed to the low carrier density, resulting in a high series resistance. However, in the −20 V < VBG ≲ +20 V range, I+2V more significantly increased than I−2V, leading to a high rectification ratio of 895 at VBG = +20 V. The decrease of rectification ratio from VBG > +20 V can be explained by the increased I−2V.


image file: d0tc01790f-f3.tif
Fig. 3 (a) Schematic of the homojunction WSe2 diode with back-gating configuration. (b) Output characteristics of the 2D homojunction for varying VBG. (c) Corresponding rectification ratio of the 2D homojunction device, defined as |I+2V|/|I−2V|.

Photocurrents at zero VDS were collected under exposure to a laser wavelength of 532 nm. Fig. 4(a) shows an optical microscope image of the WSe2 homojunction. The white dotted line in the WSe2 flake separates the laser-treated (left) and the as-exfoliated (right) regions. Fig. 4(b) shows the mapping images of the photocurrents at different VBG. Additional results for the photocurrent mapping can be found in the ESI (Fig. S3). The photocurrents analysis indicates that a depletion region is formed between the as-exfoliated and laser-treated regions in our measurement range (−60 V to +60 V). Fig. 4(c) summarizes the maximal photocurrents at different VBG. As VBG changes from negative to positive, the maximal photocurrent increases with increasing the collecting area. The wide depletion width of the as-exfoliated region can be attributed to its lower carrier concentrations compared with those of the laser-treated one. For VBG ≈ +40 V, a p–n junction forms between the laser-treated WSe2 and the as-exfoliated one, thereby increasing the photocurrent to as much as 1.8 nA. The energy-band diagrams for different VBG show that the photo-generated carriers can be effectively separated at high positive VBG, because the as-exfoliated WSe2 becomes n-type (Fig. 4(d)). These results suggest that the atomically thin in-plane logic circuits can be addressed via laser-induced doping, which allows us to control the carrier type and doping concentration in 2D material-based electronic and optoelectronic devices.


image file: d0tc01790f-f4.tif
Fig. 4 (a) Optical microscope image of the fabricated WSe2 homojunction. (b) Mapping images of the photocurrents for VBG at −60 V, 0 V, and +60 V. (c) Plot of maximal photocurrents as a function of VBG. (d) Energy-band diagrams of the fabricated 2D WSe2 homojunction under different VBG.

By integrating the laser-treated and as-exfoliated regions of the WSe2 layer, a homogeneous complementary 2D inverter, which is a basic building block of logic circuits, was monolithically implemented on a single WSe2 layer. Fig. 5(a) shows an optical microscope image (top) and schematic of the 2D inverter circuit (bottom) fabricated on the WSe2 layer, where the supply voltage, output voltage, and input voltage are labeled as VDD, VOUT, and VIN, respectively. The electrodes formed in the laser-treated region were utilized as VDD and VOUT terminals. The ground (GND) terminal was formed in the as-exfoliated region of the WSe2 layer. VIN was applied with biasing the back-gate. In other words, the p-doped region and p–n junction channel were used as the load and the driver of the inverter, respectively. Fig. 5(b) exhibits the voltage-transfer characteristics (VTC) of the fabricated WSe2 2D inverter, revealing a clear inversion function for VDD ranging from +3 V to +5 V. The gain curves were calculated as |dVOUT/dVIN| from the VTC and are shown in Fig. 5(c). The maximal voltage gain at VDD = 4 V was ∼1.8. These results suggest the feasibility of localized laser treatment of WSe2 flakes for modulating the electrical properties of WSe2 and for fabricating WSe2-based electrical logic applications.


image file: d0tc01790f-f5.tif
Fig. 5 (a) Optical microscope image (top) and schematic (bottom) of the fabricated WSe2 2D inverter. The scale bar is 5 μm. (b) Voltage transfer characteristics and (c) voltage gain of the WSe2 inverter for different VDD.

Conclusions

We demonstrated an in-plane WSe2 homojunction 2D diode via addressable one-step UV laser-assisted doping. The exposure of 2D WSe2 to a position-controlled UV laser selectively converted ambipolar behavior into a hole-dominant one owing to the electron transfer between the oxidized layer (top WOx) and the underlying WSe2, while the non-exposed WSe2 remain ambipolar. Electrostatic back-gating was applied for varying the electrical properties of WSe2 2D homojunctions, ranging from the p–p to p–n junctions with a rectification ratio as high as 895. Scanning photocurrent mapping revealed maximal photocurrents of 1.8 nA under the formation of lateral p–n homojunctions at positive VBG. The feasibility of WSe2 p–n homojunctions for use in atomic-thin circuitry was investigated via fabrication of a 2D inverter with a voltage gain of 1.8. The laser-assisted hole doping of 2D WSe2 provides a facile and effective method for controlling the electrical properties of low-dimensional materials, enabling versatile nano-electronic and nano-optoelectronic applications.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was supported by the National Research Foundation of Korea (2018R1D1A1A09083917) and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) (20172010104830). C.-H. L. acknowledges the support from the KU-KIST school project.

References

  1. R. Frisenda, E. Navarro-Moratalla, P. Gant, D. Perez De Lara, P. Jarillo-Herrero, R. V. Gorbachev and A. Castellanos-Gomez, Chem. Soc. Rev., 2018, 47, 53 RSC.
  2. D. Sarkar, X. Xie, W. Liu, W. Cao, J. Kang, Y. Gong, S. Kraemer, P. M. Ajayan and K. Banerjee, Nature, 2015, 526, 91 CrossRef CAS PubMed.
  3. H. Tian, M. L. Chin, S. Najmaei, Q. Guo, F. Xia, H. Wang and M. Dubey, Nano Res., 2016, 9, 1543 CrossRef CAS.
  4. J. Cheng, C. Wang, X. Zou and L. Liao, Adv. Opt. Mater., 2019, 7, 1800441 CrossRef.
  5. T. Roy, M. Tosun, J. S. Kang, A. B. Sachid, S. B. Desai, M. Hettick, C. C. Hu and A. Javey, ACS Nano, 2014, 8, 6259 CrossRef CAS.
  6. L. Li, Y. Yu, G. J. Ye, Q. Ge, X. Ou, H. Wu, D. Feng, X. H. Chen and Y. Zhang, Nat. Nanotechnol., 2014, 9, 372 CrossRef CAS PubMed.
  7. M. L. Tsai, S. H. Su, J. K. Chang, D. S. Tsai, C. H. Chen, C. I. Wu, L. J. Li, L. J. Chen and J. H. He, ACS Nano, 2014, 8, 8317 CrossRef CAS PubMed.
  8. A. Pospischil, M. M. Furchi and T. Mueller, Nat. Nanotechnol., 2014, 9, 257 CrossRef CAS PubMed.
  9. M. S. Choi, G. H. Lee, Y. J. Yu, D. Y. Lee, S. H. Lee, P. Kim, J. Hone and W. J. Yoo, Nat. Commun., 2013, 4, 1624 CrossRef PubMed.
  10. F. Withers, O. Del Pozo-Zamudio, A. Mishchenko, A. P. Rooney, A. Gholinia, K. Watanabe, T. Taniguchi, S. J. Haigh, A. K. Geim, A. I. Tartakovskii and K. S. Novoselov, Nat. Mater., 2015, 14, 301 CrossRef CAS PubMed.
  11. K. Lee, R. Gatensby, N. McEvoy, T. Hallam and G. S. Duesberg, Adv. Mater., 2013, 25, 6699 CrossRef CAS PubMed.
  12. Z. Zheng, T. Zhang, J. Yao, Y. Zhang, J. Xu and G. Yang, Nanotechnology, 2016, 27, 225501 CrossRef PubMed.
  13. C. Anichini, W. Czepa, D. Pakulski, A. Aliprandi, A. Ciesielski and P. Samori, Chem. Soc. Rev., 2018, 47, 4860 RSC.
  14. B. Cho, J. Yoon, S. K. Lim, A. R. Kim, D. H. Kim, S. G. Park, J. D. Kwon, Y. J. Lee, K. H. Lee, B. H. Lee, H. C. Ko and M. G. Hahm, ACS Appl. Mater. Interfaces, 2015, 7, 16775 CrossRef CAS PubMed.
  15. F. H. Koppens, T. Mueller, P. Avouris, A. C. Ferrari, M. S. Vitiello and M. Polini, Nat. Nanotechnol., 2014, 9, 780 CrossRef CAS PubMed.
  16. Z. H. Jin, X. D. Li, J. T. Mullen and K. W. Kim, Phys. Rev. B: Condens. Matter Mater. Phys., 2014, 90, 1 Search PubMed.
  17. K. Chen, D. Kiriya, M. Hettick, M. Tosun, T.-J. Ha, S. R. Madhvapathy, S. Desai, A. Sachid and A. Javey, APL Mater., 2014, 2, 092504 CrossRef.
  18. H. Fang, M. Tosun, G. Seol, T. C. Chang, K. Takei, J. Guo and A. Javey, Nano Lett., 2013, 13, 1991 CrossRef CAS PubMed.
  19. H. M. Oh, C. Park, S. Bang, S. J. Yun, N. T. Duong and M. S. Jeong, Adv. Electron. Mater., 2019, 5, 1800608 CrossRef.
  20. C.-H. Chen, C.-L. Wu, J. Pu, M.-H. Chiu, P. Kumar, T. Takenobu and L.-J. Li, 2D Mater., 2014, 1, 034001 CrossRef.
  21. Y. Tang, Z. Wang, P. Wang, F. Wu, Y. Wang, Y. Chen, H. Wang, M. Peng, C. Shan, Z. Zhu, S. Qin and W. Hu, Small, 2019, 15, 1 Search PubMed.
  22. S. Fan, M. Cao, J. Liu, J. Liu and J. Su, J. Mater. Chem. C, 2020, 8, 4365 RSC.
  23. S. Fan, X. Tang, D. Zhang, X. Hu, J. Liu, L. Yang and J. Su, Nanoscale, 2019, 11, 15359 RSC.
  24. B. W. Baugher, H. O. Churchill, Y. Yang and P. Jarillo-Herrero, Nat. Nanotechnol., 2014, 9, 262 CrossRef CAS PubMed.
  25. M. Yamamoto, S. Nakaharai, K. Ueno and K. Tsukagoshi, Nano Lett., 2016, 16, 2720 CrossRef CAS PubMed.
  26. R. Addou, C. M. Smyth, J.-Y. Noh, Y.-C. Lin, Y. Pan, S. M. Eichfeld, S. Fölsch, J. A. Robinson, K. Cho, R. M. Feenstra and R. M. Wallace, 2D Mater., 2018, 5, 025017 CrossRef.
  27. Z. Li, S. Yang, R. Dhall, E. Kosmowska, H. Shi, I. Chatzakis and S. B. Cronin, ACS Nano, 2016, 10, 6836 CrossRef CAS PubMed.
  28. W.-M. Kang, S. Lee, I.-T. Cho, T. H. Park, H. Shin, C. S. Hwang, C. Lee, B.-G. Park and J.-H. Lee, Solid-State Electron., 2018, 140, 2 CrossRef CAS.
  29. A. Azcatl, S. Kc, X. Peng, N. Lu, S. McDonnell, X. Qin, F. de Dios, R. Addou, J. Kim, M. J. Kim, K. Cho and R. M. Wallace, 2D Mater., 2015, 2, 014004 CrossRef.
  30. Y. Liu, C. Tan, H. Chou, A. Nayak, D. Wu, R. Ghosh, H. Y. Chang, Y. Hao, X. Wang, J. S. Kim, R. Piner, R. S. Ruoff, D. Akinwande and K. Lai, Nano Lett., 2015, 15, 4979 CrossRef CAS PubMed.
  31. D. Li, M. Chen, Z. Sun, P. Yu, Z. Liu, P. M. Ajayan and Z. Zhang, Nat. Nanotechnol., 2017, 12, 901 CrossRef CAS.
  32. I. Gasparutti, S. H. Song, M. Neumann, X. Wei, K. Watanabe, T. Taniguchi and Y. H. Lee, ACS Appl. Mater. Interfaces, 2020, 12, 7701 CrossRef CAS PubMed.
  33. S. Lei, X. Wang, B. Li, J. Kang, Y. He, A. George, L. Ge, Y. Gong, P. Dong, Z. Jin, G. Brunetto, W. Chen, Z. T. Lin, R. Baines, D. S. Galvao, J. Lou, E. Barrera, K. Banerjee, R. Vajtai and P. Ajayan, Nat. Nanotechnol., 2016, 11, 465 CrossRef CAS PubMed.
  34. M. Yamamoto, S. Dutta, S. Aikawa, S. Nakaharai, K. Wakabayashi, M. S. Fuhrer, K. Ueno and K. Tsukagoshi, Nano Lett., 2015, 15, 2067 CrossRef CAS PubMed.
  35. C. Tan, Y. Liu, H. Chou, J. S. Kim, D. Wu, D. Akinwande and K. Lai, Appl. Phys. Lett., 2016, 108, 083112 CrossRef.
  36. M. J. Shearer, W. Li, J. G. Foster, M. J. Stolt, R. J. Hamers and S. Jin, ACS Energy Lett., 2018, 4, 102 CrossRef.
  37. J. Lu, A. Carvalho, X. K. Chan, H. Liu, B. Liu, E. S. Tok, K. P. Loh, A. H. Castro Neto and C. H. Sow, Nano Lett., 2015, 15, 3524 CrossRef CAS PubMed.
  38. D. S. Schulman, A. J. Arnold and S. Das, Chem. Soc. Rev., 2018, 47, 3037 RSC.
  39. C. M. Smyth, R. Addou, S. McDonnell, C. L. Hinkle and R. M. Wallace, 2D Mater., 2017, 4, 025084 CrossRef.

Footnotes

Electronic supplementary information (ESI) available. See DOI: 10.1039/d0tc01790f
The authors contributed equally to this work.

This journal is © The Royal Society of Chemistry 2020