Chemical vapor deposition merges MoS₂ grains into high-quality and centimeter-scale films on Si/SiO₂

Abstract

Two-dimensional molybdenum disulfide (MoS₂) has attracted increasing attention due to its promise for next-generation electronics. To realize MoS₂-based electronics, however, a synthesis method is required that produces a uniform single-layer material and that is compatible with existing semiconductor fabrication techniques. Here, we demonstrate that uniform films of single-layer MoS₂ can be directly produced on Si/SiO₂ at wafer-scale without the use of catalysts or promoters. Control of the precursor transport through oxygen dosing yielded complete coverage and increased connectivity between crystalline MoS₂ domains. Spectroscopic characterization and carrier transport measurements furthermore revealed a reduced density of defects compared to conventional chemical vapor deposition growth that increased the quantum yield over ten-fold. To demonstrate the impact of enhanced scale and optoelectronic performance, centimeter-scale arrays of MoS₂ photosensors were produced that demonstrate unprecedentedly high and uniform responsivity. Our approach improves the prospect of MoS₂ for future applications.

---

Introduction

Molybdenum disulfide (MoS₂) is a 2D material that has attracted the attention of researchers due to its unique electrical and optical properties.¹ Based on its high carrier mobility,^2,3 electrostatic controllability,⁴ environment stability,⁵ and thickness-induced bandgap tunability,^6,7 MoS₂ is considered an enabling material for future electronics.^8,9 This promise is corroborated by the reported impressive performance of MoS₂ in optoelectronic devices such as photodetectors,^10–15 light emitters,^16,17 and haptic sensors.¹⁸

However, to advance MoS₂-based electronics from research into commercially viable applications, MoS₂ production has to be compatible with existing, mature semiconductor fabrication technology. This requirement disqualifies mechanical exfoliation due to small crystallite size and unreliable placement. Moreover, synthesis on single-crystalline substrates such as sapphire,¹⁹ h-BN, mica^20–23 and metal^24,25 cannot be employed, despite their impressive performance, as the required large-scale and high-precision transfer step are beyond current capabilities. The most promising approach is the direct growth of MoS₂ on silicon dioxide (SiO₂) due to the experience of semiconductor industry with the material system^8,26 and the existing commercial infrastructure. Unfortunately the growth on SiO₂, to date, has only yielded discontinuous flakes compared to the uniform and continuous films grown on other substrates.^27–29 Several approaches have been explored to overcome this challenge but they introduce other issues. First, the use of organic seeding materials like perylene-3,4,9,10-tetracarboxylic acid tetrapotassium salt (PTAS) can enhance the continuity of MoS₂ but introduces impurities into the reactor and into the produced MoS₂.^30,31 Second, large grain sizes and continuous films can be achieved by liquid-substrate chemical vapour deposition (CVD) but salt-residue remains on the SiO₂.³²

In this study, we have introduced an approach to produce continuous single layer MoS₂ at large-scale on Si/SiO₂. This advance was accomplished by a modified CVD process that utilizes a small amount of oxygen. Using this method, high quality material with unprecedented continuity and uniformity could be synthesized, as evidenced by spectroscopic characterization and electron microscopy. The resulting material was applied to MoS₂-based optoelectronic devices that demonstrate record-breaking performance and uniform properties throughout wafer-scale device array.

Results and discussion

Chemical vapor deposition is an established process to synthesize MoS₂ from gaseous precursors.^33–35 We employ silicon oxide wafers as substrates and observe discontinuous triangular flakes with average dimensions of ∼10 μm size (Fig. 1(a)) when using established CVD parameters.^33,36 Even long growth durations cannot connect these individual flakes, raising the question as to the limiting mechanism. The MoS₂ growth by CVD is a multistep conversion process from gaseous MoO₃ precursor into adsorbed species that are subsequently integrated into the outgrowing MoS₂ crystal.³⁷ The self-limiting grain size indicates the kinetic hindrance of precursor supply at long growth durations: at low MoS₂ coverage, adsorption of MoO₃ precursor proceeds quickly and flake growth occurs. At high MoS₂ coverage, however, the adsorption efficiency decreases as less uncovered surface sites are available. Consequently, the MoS₂ growth rate decreases until precursor adsorption and desorption are in equilibrium, and no further growth occurs (Fig. 1(b)).

Fig. 1 (a) Optical microscopy image of MoS₂ films as deposited on Si/SiO₂ substrate without O₂. (b) Schematic of adsorption and desorption of gaseous phase MoO₃ on Si/SiO₂. (c) Optical microscopy image (taken after scratch the film gently by tweezer) of in situ O₂ processed samples. (d) Camera image of wafer scale continuous MoS₂ on Si/SiO₂ (in right) and a blank Si/SiO₂ substrate (in left) for the reference.

We overcome this limitation by introducing 0.5 sccm of oxygen. Previous work has demonstrated that increased chemisorption of oxygen on silicon oxide surfaces stabilizes the MoO₃ precursor³⁸ and thus reduces the impact of precursor desorption on growth rate (Fig. 1(b)). Indeed, optical microscopy images demonstrate that the addition of oxygen during the CVD growth process produces continuous films (Fig. 1(c)). The modified oxygen-assisted CVD method can also grow uniform and continuous wafer-scale MoS₂ (Fig. 1(d)). Detailed characterization of the oxygen-grown MoS₂ by atomic force microscopy (AFM) corroborates the high uniformity of the film (Fig. 2a). The thickness of ∼1 nm presents a clear evidence of uniform single-layer MoS₂ films^39,40 (more details about the uniformity of the films can be found in the ESI S1†). This evaluation is confirmed by Raman spectra which show characteristic peaks at 382 ± 0.5 cm⁻¹ and 404 ± 0.5 cm⁻¹ corresponding to in-plane E¹_2g mode and out-of-plane A_1g mode, respectively (Fig. 2b). These two peaks are well separated by 22 ± 0.5 cm⁻¹, which indicates the single/bi layer nature MoS₂ film.^39,41,42 Raman mapping was employed to characterize the uniformity of the MoS₂. Fig. 2c(i and ii) shows the spatial distribution of the E¹_2g and A_1g peaks postion over a 100 × 100 μm² area with a step size of 1 μm, demonstrating the absence of vacancies and the uniform peak separation of ∼22 ± 0.5 cm⁻¹ throughout the film. TEM images (Fig. 3a(i) and b(i)) further corroborate the uniformity for both type of the films. The selected area electron diffraction (SAED) (Fig. 3a(ii) and b(ii)) and high-resolution transmission electron microscopy (HRTEM) (Fig. 3a(iii) and b(iii)) indicate the polycrystalline nature of O₂-assisted MoS₂ in comparison to conventional MoS₂, where single crystal characteristic was observed. Apart from the optical characterizations, films were further characterized by electrical transport measurement in co-planar geometry for the centimetre scale area of MoS₂ continuity. Fig. 3c and d represent the electrical sheet conductance and electrical current respectively, in two terminal co-planar geometry, for the device with different channel length of 50, 100, 150, 200 and 250 μm and at a fixed channel width of 580 μm. Each channel length belonging to one particular row and there are five row in the device. So, the data shown here is for 25 devices. The uniformity of current in MoS₂ shows that films are continuous in centimetre scale. The current variation in each row indicates the polycrystalline nature of MoS₂ films.

Fig. 2 Characterization of in situ O₂ processed continuous MoS₂ films on Si/SiO₂ (a) AFM image (scale bar: 2 μm) (height profile given in inset). (b) Raman spectra of the as deposited films measured at several randomly selected locations. (c) Raman mapping of (i) E¹_2g and (ii) A_1g peak positions in 100 μm × 100 μm dimensions with 1 μm step size (histogram shown in inset).

Fig. 3 (a and b) TEM characterization of conventional (a) and oxygen-assisted (b) MoS₂:(i) low resolution TEM image after transfer onto carbon-coated grid, (ii) SAED pattern (scale bar: 5 nm⁻¹) (iii) HRTEM image (scale bar: 2 nm). Electrical characterization of uniformity: histogram of sheet conductance for 18 devices (c) in device array (see inset, (c)). Demonstration of channel continuity through current measurements within large device array of different channel lengths measured at fixed bias of 20 V in two terminal co-planar geometry.

The effect of adding oxygen was further investigated by photoelectron spectroscopy. X-ray photoelectron spectroscopy (XPS) demonstrate peaks corresponding to S 2s, Mo⁴⁺ 3d_5/2 and Mo⁴⁺ 3d_3/2 (indicating MoS₂) at 226.7 eV, 229.6 eV, 232.8 eV, respectively (Fig. 4(a) and (b)). Along with the Mo⁴⁺ oxidation states, a small fraction of higher oxidation states was also observed, such as peaks at 233.6 eV and 235.9 eV corresponding to Mo⁶⁺ 3d_5/2 and Mo⁶⁺ 3d_3/2 respectively, for the in situ oxygen processed samples (Fig. 4(b)).^43,44 These higher oxidation states could indicate molybdenum oxide formation, but their concentration is less than 7% (estimated from the fitted area of MoS₂ and MoO₃ oxidation), which could indicate their origin as surface contaminants.

Fig. 4 XPS spectra of Mo 3d binding energy for the films deposited (a) without O₂ and (b) with in situ O₂ samples. (c) PL spectra of single layer MoS₂ for both the films; without O₂ and with in situ O₂. (d and e) Deconvoluted PL spectra of MoS₂ without O₂ and with in situ O₂ processed samples respectively. (f) Schematic energy level diagram of conventional MoS₂ (i.e. without O₂) and oxygen-assisted MoS₂ films.

The high crystallinity of oxygen-assisted MoS₂ growth can be inferred from photoluminescence (PL) spectroscopy. Comparable to conventional MoS₂, oxygen-assisted MoS₂ shows a prominent peak near 1.8 eV which corresponds to the direct bandgap absorption of single-layer MoS₂ (Fig. 4(c)).⁴⁵ Surprisingly, an enhancement of PL peak intensity for oxygen-assisted CVD over conventional MoS₂ by 15 times can be seen. This significant increase in quantum yield provides an exciting route towards improving MoS₂ optoelectronic performance without the need for post-growth treatment.⁴⁶ Detailed characterization of the PL spectra (Fig. 4(d) and (e)) demonstrates a decrease in the concentration of charged excitons (i.e. trions) and a subsequent increase of the neutral exciton emission for O₂ processed samples. The schematic energy level diagram of conventional and oxygen-assisted MoS₂ is shown in Fig. 4(f), where transitions of excitons and trions are shown.

To identify the reason for this difference, we conduct Kelvin probe measurements. The work function for individual triangle MoS₂ samples (O₂ free) was 4.34 eV, whereas the work-function of in situ oxygen processed samples was increased to 4.73 eV. The increase of work function in in situ O₂ processed samples indicates that the usually n-type material is rendered more neutral.⁴⁷ This effect could be due to the decrease in sulfur vacancies that cause n-type doping.⁴⁸

These results indicate that the enhanced quantum yield's origin is a decrease in nonradiative recombination through trion states brought about by a lowered defect-induced doping.⁴⁹

We illustrate the potential of the increased scale and quality of MoS₂ growth by producing electronic device arrays. As a first example, transistor devices were fabricated using photo-lithography (Fig. 5(a)) (more details on the device fabrication can be found in the ESI S2†). The transfer curves of oxygen-processed MoS₂ show a higher threshold voltage than conventionally grown MoS₂, which corroborates the observed decrease in n-type character (Fig. 5(b)). Notably, the transistor channel for O₂ grown MoS₂ includes many polycrystalline regions, whereas the conventional grown MoS₂ transistor was designed around a single-crystalline flake (Fig. 5(a)), necessitating complex alignment schemes. Consequently, the observed transfer characteristics of the O₂-grown MoS₂ represent an averaging of the conductivity from the dozens of individual flakes and their boundaries. Despite this disadvantage, the transfer characteristics of large scale O₂ grown MoS₂ exhibits a similar behaviour as microscopic MoS₂ crystals, indicating the good connectivity between polycrystalline regions. The measured field-effect mobility for both device types is within the range 0.1–10 cm² V⁻¹ s⁻¹ of previous results in the literature for the similar structure of back gated MoS₂ field-effect transistor devices.^50,51

Fig. 5 (a) Optical microscopy image of MoS₂ back gate field effect transistor with channel length 6 μm (scale bar 20 μm). (b) Transfer characteristic of individual triangular MoS₂ domains with fixed bias voltage (V_ds) 10 V (for continuous MoS₂ films, transfer characteristic shown in inset). Photosensor characteristics of in situ O₂ processed continuous MoS₂ films; (c) camera image of large area MoS₂ photosensor array in centimetre dimension (channel length = 125 μm), (d) photocurrent variation with different laser power, (e) responsivity of the device with different laser power, (f) average responsivity of the MoS₂ films shown at fixed power of 1.4 × 10⁻⁴ W cm⁻².

As a second application of MoS₂ high optoelectronic performance, we produce centimeter-scale photosensor arrays (Fig. 5(c)). A variable photocurrent (I_ph) was observed when toggling illumination with different laser power (Fig. 5(d)). The responsivity (R) of the device, defined as the ratio of photocurrent to the incident power (P_in), increases with decreasing laser power (Fig. 5(e)) and reaches values in excess of ∼420 A W⁻¹ in two-terminal co-planar geometry. This value represents the quite higher responsivity of MoS₂ as compared to other reports^11,41,52,53 (for a comparison to literature, see the ESI S3†) and indicates the benefits of producing MoS₂ with low trap density and efficient carrier conduction. More importantly, uniformly high responsivity could be achieved for all devices throughout the centimetre-sized samples (Fig. 5(f)). The stability of photosensor also performed using I–T measurement for 10 cycles and details are given in the ESI S4.† In the support of large scale photodetector array, a uniform photosensitivity of MoS₂ on Si/SiO₂ is shown in ESI S5,† where, wafer scale photosensitivity is demonstrated (Table 1).

Table 1 Responsivity data comparison for MoS₂-based photosensors

Responsivity (A W^−1)	Wavelength and laser power	Area of devices (L/W) (in μm)	Materials	References
1 at Vds = 1.5 V	532 nm, 200 μW	A few micron/tens to 100 micron	CVD MoS2	1
1 × 10^−3 at Vds = 1 V	405 nm, 100 μW	2/20	CVD MoS2	2
59 at Vds = 1.2 V	532 nm, 1.69 × 10^−3 W cm^−2	5/∼30	Exfoliated MoS2	3
780 at Vds = 1 V	532 nm, 1.3 × 10^−4 W cm^−2	—	CVD MoS2	4
1.1 × 10^6 at Vds = 0.15 V	460 nm, 0.33 pW	—	Exfoliated MoS2	5
1.1 × 10^−3 at Vds = 1.5 V	514.5 nm, 1 μW	0.8/5	CVD MoS2	6
0.42 × 10^−3 at Vds = 1 V	550 nm, 80 W cm^−2	2.1/2.6	Exfoliated MoS2	7
7 at Vds = 1 V	488 nm, 1 μW	—	CVD MoS2	8
420 at Vds = 15 V	532 nm, 10^−5 W cm^−2	125/480	CVD MoS2	This work

---

Conclusion

Continuous single-layer MoS₂ films on Si/SiO₂ substrates were successfully fabricated through an oxygen-assisted CVD growth process. Modification of the precursor transport resulted in highly crystalline material with well-interconnected domains. The absence of defects was shown to decrease the n-type doping of MoS₂ and enhance its optoelectronic performance. Large array of highly sensitive photosensors were produced that exhibit uniform and 200-fold increased responsivities over the entire substrate as compared to other reports. The presented combination of increased MoS₂ quality, scale, and commercial appeal open up new routes towards the future of 2D materials.

Experimental section

MoS₂ synthesis

MoS₂ thin films on Si/SiO₂ were prepared by a CVD process, where 3 inch quartz tube furnace, having three different heating zone, was used. Si/SiO₂ substrates were cleaned by sonication in acetone for 15 min, and then isopropyl alcohol followed by water wash and then dry with N₂ gun. MoO₃ powder (20 mg) was placed into two alumina boat (10 mg in each) at the lower heating zone furnace, which was upstream relative to center zone furnace. H₂S gas (99% diluted with Ar) 400 sccm was used as a source of sulfur. It also acts as a carrier gas for MoO₃ vapor to react with H₂S and get deposited on the substrate. Separate Ar gas of 100 sccm was also used in the process. Substrate was placed at the center zone furnace facing polished surface upside and kept at a distance of 20 cm downstream relative to MoO₃ powder. MoS₂ films were deposited for 40 min at a low process pressure of 4 torr. Substrates were placed at 900 °C in the centre zone furnace, while MoO₃ powder were placed at lower temperature of 750 °C. After purging the furnace with pure Ar for 1 h, furnace was heated to 600 °C at a heating rate of 30 °C min⁻¹ and then maintained at the same temperature for 5 min. Further, temperature was increased to 900 °C at a rate of 30 °C min⁻¹. A small amount of oxygen (0.5 sccm) was also added during the deposition. After completion, furnace was cooled down naturally. Here, we have only reported for the films deposited with O₂ 0.5 sccm, because with higher oxygen rate the pressure inside the furnace was not stable and started to keep on increasing with increasing process time.

MoS₂ characterization

The surface morphology of the samples was characterized by AFM (model: Veeco, Dimension Icon) and TEM (model: JEOL, JEM-2100F) measurements. Raman spectra and PL measurements (model: HORIBA, Jobin Yvon Technology iHR 550) were recorded under ambient using a laser excitation wavelength of 532 nm. Laser spot size was ∼1 μm² and laser power at the sample surface was 3 mW. Chemical configuration was characterized by XPS (model: ULVAC-PHI, PHI Quanterall), where XPS peak was calibrated with C 1s (284.6 eV). Work function of the sample surface was measured by Kelvin probe force microscopy (model: KP Technology Ltd., DCU series-10) using Pt coated tip in non-contact mode. The cantilever tip was calibrated using a standard Au plate electrode.

Author contributions

M. S. synthesis of MoS₂ films via CVD and also fabricated the devices and performed the electrical and optical measurements. M. S., Y. S.-C., and Z. L.-Y. analyzed Raman and AFM measurements. R. G. and M. S. performed the photosensor measurements and analyzed the responsivity data. Y. P.-H., Y. F.-C., and M. H. supervised the project. M. S., Y. P.-H., Y. F.-C., and M. H. have contributed to writing the paper through the input of all the authors.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is financially supported by the Ministry of Science and Technology, Taiwan (Grant No. MOST107-2112-M-002-004-MY3, MOST107-2119-M-001-032-MY3 and MOST108-2112-M-001-040-MY3). Also, authors would like to acknowledge Centre for Emerging Material and Advanced Devices, National Taiwan University, Taiwan, for providing characterization facilities.

References

1. Y. L. Huang, Y. Chen, W. Zhang, S. Y. Quek, C. Chen, L. Li, W. Hsu, W. Chang, Y. J. Zheng, W. Chen and A. T. S. Wee, Nat. Commun., 2015, 6, 6298.
2. J. Jeon, S. K. Jang, S. M. Jeon, G. Yoo, Y. H. Jang, J. H. Park and S. Lee, Nanoscale, 2015, 7, 1688–1695.
3. B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti and A. Kis, Nat. Nanotechnol., 2011, 6, 147–150.
4. A. Azcatl, X. Qin, A. Prakash, C. Zhang, L. Cheng, Q. Wang, N. Lu, M. J. Kim, J. Kim, K. Cho, R. Addou, C. L. Hinkle, J. Appenzeller and R. M. Wallace, Nano Lett., 2016, 16, 5437–5443.
5. P. Budania, P. Baine, J. Montgomery, C. McGeough, T. Cafolla, M. Modreanu, D. McNeill, N. Mitchell, G. Hughes and P. Hurley, MRS Commun., 2017, 7, 813–818.
6. K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805.
7. J. K. Ellis, M. J. Lucero and G. E. Scuseria, Appl. Phys. Lett., 2011, 99, 261908.
8. H. Wang, L. Yu, Y. H. Lee, Y. Shi, A. Hsu, M. L. Chin, L. J. Li, M. Dubey, J. Kong and T. Palacios, Nano Lett., 2012, 12, 4674–4680.
9. P. V. K. Yadav, B. Ajitha, Y. A. Kumar Reddy and A. Sreedhar, Chemosphere, 2021, 279, 130473.
10. Z. Yin, H. Li, H. Li, L. Jiang, Y. Shi, Y. Sun, G. Lu, Q. Zhang, X. Chen and H. Zhang, ACS Nano, 2012, 6, 74–80.
11. N. Perea-López, Z. Lin, N. R. Pradhan, A. Iñiguez-Rábago, A. L. Elías, A. McCreary, J. Lou, P. M. Ajayan, H. Terrones, L. Balicas and M. Terrones, 2D Mater., 2014, 1, 011004.
12. P. V. K. Yadav, B. Ajitha, Y. A. Kumar Reddy and V. R. Minnam Reddy, ACS Appl. Electron. Mater., 2021, 3, 2056–2066.
13. P. V. K. Yadav, B. Ajitha, V. Annapureddy, Y. A. K. Reddy and A. Sreedhar, Mater. Lett., 2021, 301, 130296.
14. P. V. K. Yadav, B. Ajitha, Y. A. K. Reddy, V. R. M. Reddy, M. Reddeppa and M. D. Kim, Appl. Surf. Sci., 2021, 536, 147947.
15. P. V. K. Yadav, Y. A. K. Reddy, B. Ajitha and V. R. M. Reddy, J. Alloys Compd., 2020, 816, 152565.
16. E. Singh, P. Singh, K. S. Kim, G. Y. Yeom and H. S. Nalwa, ACS Appl. Mater. Interfaces, 2019, 11, 11061–11105.
17. W. Zhang, P. Zhang, Z. Su and G. Wei, Nanoscale, 2015, 7, 18364–18378.
18. M. Park, Y. J. Park, X. Chen, Y. K. Park, M. S. Kim and J. H. Ahn, Adv. Mater., 2016, 28, 2556–2562.
19. H. Yu, M. Liao, W. Zhao, G. Liu, X. J. Zhou, Z. Wei, X. Xu, K. Liu, Z. Hu, K. Deng, S. Zhou, J. A. Shi, L. Gu, C. Shen, T. Zhang, L. Du, L. Xie, J. Zhu, W. Chen, R. Yang, D. Shi and G. Zhang, ACS Nano, 2017, 11, 12001–12007.
20. D. Dumcenco, D. Ovchinnikov, K. Marinov, P. Lazic, M. Gibertini, N. Marzari, O. L. Sanchez, Y. Kung, D. Krasnozhon, M. Chen, S. Bertolazzi, P. Gillet, A. Fontcuberta, A. Radenovic and A. Kis, ACS Nano, 2015, 9, 4611–4620.
21. F. Zhang, Y. Wang, C. Erb, K. Wang, P. Moradifar, V. H. Crespi and N. Alem, Phys. Rev. B, 2019, 99, 155430.
22. Z. Ma, S. Wang, Q. Deng, Z. Hou, X. Zhou, X. Li, F. Cui, H. Si, T. Zhai and H. Xu, Small, 2020, 16, 2000596.
23. Q. Ji, Y. Zhang, T. Gao, Y. Zhang, D. Ma, M. Liu, Y. Chen, X. Qiao, P. H. Tan, M. Kan, J. Feng, Q. Sun and Z. Liu, Nano Lett., 2013, 13, 3870–3877.
24. Y. Zhan, Z. Liu, S. Najmaei, P. M. Ajayan and J. Lou, Small, 2012, 8, 966–971.
25. D. H. Cho, W. J. Lee, J. H. Wi, W. S. Han, S. J. Yun, B. Shin and Y. D. Chung, Phys. Chem. Chem. Phys., 2018, 20, 16193–16201.
26. J. L. Benton, D. C. Jacobson, B. Jackson, J. A. Johnson, T. Boone, D. J. Eaglesham, F. A. Stevie and J. Becerro, J. Electrochem. Soc., 1999, 146, 1929–1933.
27. Z. Lin, Y. Zhao, C. Zhou, R. Zhong, X. Wang, Y. H. Tsang and Y. Chai, Sci. Rep., 2016, 5, 18596.
28. L. Tao, K. Chen, Z. Chen, W. Chen, X. Gui, H. Chen, X. Li and J. B. Xu, ACS Appl. Mater. Interfaces, 2017, 9, 12073–12081.
29. J. Zhang, H. Yu, W. Chen, X. Tian, D. Liu, M. Cheng, G. Xie, W. Yang, R. Yang, X. Bai, D. Shi and G. Zhang, ACS Nano, 2014, 8, 6024–6030.
30. X. Ling, Y. H. Lee, Y. Lin, W. Fang, L. Yu, M. S. Dresselhaus and J. Kong, Nano Lett., 2014, 14, 464–472.
31. P. Yang, A. G. Yang, L. Chen, J. Chen, Y. Zhang, H. Wang, L. Hu, R. J. Zhang, R. Liu, X. P. Qu, Z. J. Qiu and C. Cong, Nano Res., 2019, 12, 823–827.
32. S. Li, Y. C. Lin, X. Y. Liu, Z. Hu, J. Wu, H. Nakajima, S. Liu, T. Okazaki, W. Chen, T. Minari, Y. Sakuma, K. Tsukagoshi, K. Suenaga, T. Taniguchi and M. Osada, Nanoscale, 2019, 11, 16122–16129.
33. Y. Xie, Z. Wang, Y. Zhan, P. Zhang, R. Wu, T. Jiang, S. Wu, H. Wang, Y. Zhao, T. Nan and X. Ma, Nanotechnology, 2017, 28, 084001.
34. Y. H. Lee, X. Q. Zhang, W. Zhang, M. T. Chang, C. Te Lin, K. Di Chang, Y. C. Yu, J. T. W. Wang, C. S. Chang, L. J. Li and T. W. Lin, Adv. Mater., 2012, 24, 2320–2325.
35. J. Jeon, S. K. Jang, S. M. Jeon, G. Yoo, Y. H. Jang, J. H. Park and S. Lee, Nanoscale, 2015, 7, 1688–1695.
36. H. Liu, Y. Zhu, Q. Meng, X. Lu, S. Kong, Z. Huang, P. Jiang and X. Bao, Nano Res., 2017, 10, 643–651.
37. Y. C. Lin, W. Zhang, J. K. Huang, K. K. Liu, Y. H. Lee, C. Te Liang, C. W. Chu and L. J. Li, Nanoscale, 2012, 4, 6637–6641.
38. A. N. Desikan, L. Huang and S. T. Oyama, J. Phys. Chem., 1991, 95, 10050–10056.
39. S. H. Baek, Y. Choi and W. Choi, Nanoscale Res. Lett., 2015, 10, 388.
40. N. A. Lanzillo, A. Glen Birdwell, M. Amani, F. J. Crowne, P. B. Shah, S. Najmaei, Z. Liu, P. M. Ajayan, J. Lou, M. Dubey, S. K. Nayak and T. P. O'Regan, Appl. Phys. Lett., 2013, 103, 093102.
41. S. Khadka, T. E. Wickramasinghe, M. Lindquist, R. Thorat, S. H. Aleithan, M. E. Kordesch and E. Stinaff, Appl. Phys. Lett., 2017, 110, 261109.
42. H. Li, Q. Zhang, C. C. R. Yap, B. K. Tay, T. H. T. Edwin, A. Olivier and D. Baillargeat, Adv. Funct. Mater., 2012, 22, 1385–1390.
43. S. Hussain, J. Singh, D. Vikraman, A. K. Singh, M. Z. Iqbal, M. F. Khan, P. Kumar, D. C. Choi, W. Song, K. S. An, J. Eom, W. G. Lee and J. Jung, Sci. Rep., 2016, 6, 30791.
44. P. Kumar, M. Singh, R. K. Sharma and G. B. Reddy, J. Alloys Compd., 2016, 671, 440–445.
45. X. Wang, H. Feng, Y. Wu and L. Jiao, J. Am. Chem. Soc., 2013, 135, 5304–5307.
46. J. Jadwiszczak, G. Li, C. P. Cullen, J. J. Wang, P. Maguire, G. S. Duesberg, J. G. Lunney and H. Zhang, Appl. Phys. Lett., 2019, 114, 091103.
47. S. Y. Lee, U. J. Kim, J. Chung, H. Nam, H. Y. Jeong, G. H. Han, H. Kim, H. M. Oh, H. Lee, H. Kim, Y. G. Roh, J. Kim, S. W. Hwang, Y. Park and Y. H. Lee, ACS Nano, 2016, 10, 6100–6107.
48. D. Liu, Y. Guo, L. Fang and J. Robertson, Appl. Phys. Lett., 2013, 103, 183113.
49. D. H. Lien, S. Z. Uddin, M. Yeh, M. Amani, H. Kim, J. W. Ager, E. Yablonovitch and A. Javey, Science, 2019, 364, 468–471.
50. M. Amani, M. L. Chin, A. G. Birdwell, T. P. O'Regan, S. Najmaei, Z. Liu, P. M. Ajayan, J. Lou and M. Dubey, Appl. Phys. Lett., 2013, 102, 193107.
51. N. Huo, Y. Yang, Y. N. Wu, X. G. Zhang, S. T. Pantelides and G. Konstantatos, Nanoscale, 2018, 10, 15071–15077.
52. A. E. Yore, K. K. H. Smithe, S. Jha, K. Ray, E. Pop and A. K. M. Newaz, Appl. Phys. Lett., 2017, 111, 043110.
53. R. Nur, T. Tsuchiya, K. Toprasertpong, K. Terabe, S. Takagi and M. Takenaka, Commun. Mater., 2020, 1, 103.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ra06933k

---