Growth of large area few-layer or monolayer MoS₂ from controllable MoO₃ nanowire nuclei

Abstract

Here, we report a novel intermediate state (core–shell MoO₃–MoS₂ nanowires) in the synthesis of large area few-layer or monolayer MoS₂ films on Si/SiO₂ substrates coated with 3,4,9,10-perylene-tetracarboxylic-dianhydride (PTCDA) using a chemical vapour deposition (CVD) method. In our experiments, water vapor present in the carrier gas (N₂) should help enhance the interaction between PTCDA and MoO₃. In the intermediate state, the morphology of the nuclei is controlled to be nanowires with 20–180 nm diameters and 30–70 μm lengths. We investigate the formation mechanism of the nucleation-controlled intermediate state and the formation process of monolayer MoS₂ using scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), photoluminescence (PL), Raman spectroscopy and transmission electron microscopy (TEM) measurements. We also describe a method to control the diameter of the nanowires and the stacking of the nanowires into MoS₂ nanosheets on the substrate, which is meaningful for producing large area and highly crystalline MoS₂ monolayers.

---

1. Introduction

The great success of graphene has shown that it is possible to produce stable, single-layered two-dimensional (2D) materials, exhibiting fascinating and technologically useful properties.^1–3 Transition metal dichalcogenides (TMDs), MX₂ (M = Mo, W, Nb; X = S, Se, Te), as a series of attractive 2D materials, have been investigated for their potential in friction,^4–7 catalysis,^8–10 electrochemistry,^11–13 biology,¹⁴ optoelectronics,¹⁵ etc. As a typical layered TMD, MoS₂ has attracted great interest because of its distinctive electronic, optical, and catalytic properties.^16–21 Many methods of obtaining TMD thin films, especially MoS₂ thin films, have been investigated, such as the intercalation-assisted thermal cleavage method,²² micromechanical exfoliation,¹⁸ liquid exfoliation,²³ and vapor deposition.^24,25 The traditional micromechanical exfoliation method can produce high quality MoS₂ monolayers, however low yield and uncontrollable layers limit its application in commercially viable devices. Liquid exfoliation is insensitive to ambient conditions, but it yields a relatively low concentration of monolayer flakes. Therefore, synthesis of monolayer or few-layer MoS₂ is in accordance with current micro- or nano-fabrication processes, which largely promote the integration of this fascinating material into future devices. Recently, the sulfurization of molybdenum trioxide (MoO₃) by CVD has been one of the most attractive methods for synthesizing large area and highly crystalline MoS₂.^26,27 There are a few studies about the formation mechanism of MoS₂ thin films, for example, Lee et al. reported the nucleation and growth process of CVD-grown highly crystalline MoS₂ atomic layers facilitated by seeding the substrate with graphene-like species.²⁵ Najmaei et al. investigated a method to control the nucleation by strategically creating step edges on substrates using conventional lithography processes to obtain large area monolayer MoS₂.²⁶ Li et al. reported the synthesis of MoS₂ nano- and microribbons by vulcanizing MoO₂ nanowires, which served as seeds, providing a new method to produce MoS₂ thin films.²⁸ Herein, we introduce an easy and natural method to control the nucleation process when producing monolayer MoS₂, which provides the possibility of producing large area, highly crystalline monolayer MoS₂.

2. Experimental

A ceramic boat containing MoO₃ powder was placed 1 cm away from the center of a tube furnace, and Si/SiO₂ substrates spun with a drop of PTCDA/NaOH solution (8 mg in 2 mL 0.2 mol L⁻¹ NaOH solution) were placed on top of the MoO₃ powder. Another ceramic boat containing pure sulfur (S) (0.5–1 g) was placed 1 cm away from the upwind low temperature zone in the quartz tube. During the reaction, the temperature of the low temperature zone was controlled to be slightly above the melting point of S (113 °C). The quartz tube was kept in a flowing protective atmosphere of high purity N₂ (99.9999%) and water vapor by passing the N₂ through a water bubbler, with a flow rate of 150–200 sccm. After 15 minutes of N₂ purging, the furnace temperature was gradually increased from room temperature to 700 °C over 20 minutes. Then, the temperature was kept at 700 °C for 6 minutes. After the first X (X = 40, 60, 80, 100, 120) seconds, the tube was moved approximately 1 cm along the direction of N₂ flow (this is a very important step; different values of X result in different diameters of the nanowires). Fig. 1 shows a schematic illustration of the reaction conditions in this CVD process.

Fig. 1 The configuration used for monolayer MoS₂ preparation in our experiments.

3. Results and discussion

3.1 Reaction mechanism

It has been confirmed that the reaction between MoO₃ and H₂S under a H₂ atmosphere includes a stepwise reduction of Mo^VI in MoO₃ to Mo^IV in MoS₂,^29–33 and that MoO_3−x species are present. The reaction also involves the subsequent formation of oxysulfide (MoOS₂), which is a composite of MoS₂ and MoO_3−x.²⁹ The reaction between MoO₃ and S in our experiment is given in eqn (1):

2MoO₃ + 7S → 2MoS₂ + 3SO₂ (1)

It is postulated that the transition process involves reduction and sulfurization. A possible stepwise process for the reaction of MoO₃ with S is given in eqn (2) and (3) (ref. 34):

MoO₃ + x/2S → MoO_3−x + x/2SO₂ (2)

MoO_3−x + (7 − x)/2S → MoS₂ + (3 − x)/2SO₂ (3)

3.2 Growth process

The growth process consists of the following steps: (1) MoO₃ deposited on the substrate (Si/SiO₂) first forms dendritic MoO₃ nanowires, which is induced by dendritic PTCDA coating on the substrate (Fig. 3c). We also carried out a control experiment, in which the substrate is not coated with PTCDA (Fig. 3a), and it was found that only MoS₂ particles are obtained. (2) Owing to the presence of S, the surface of the MoO₃ nanowires quickly transforms into MoO_3−x, and dangling bonds are generated at the same time because fully stoichiometric MoO₃ is easily reduced (Fig. 4a).³⁵ S atoms are bound to these dangling bond sites due to the presence of large numbers of dangling bonds on the surface of the nanowires; thin nanosheets are formed on the nanowires because growth in the lateral dimension is much faster than growth in the vertical thickness dimension (Fig. 2a). (3) During the sulfurization process, more and more MoS₂ nanosheets gradually spread out from the nanowires to the substrate, and the initial MoO₃ nanowires are covered (Fig. 2b). As more and more MoO_3−x becomes MoS₂, which tends to form films, the nanowires are slowly “corroded” (Fig. 2c and d), and become flatter and flatter to form large area MoS₂ ultrathin films with a series of terraces between the center of the nanowire and the edge of the thin film (Fig. 7b); monolayer MoS₂ is present at the edge of the thin film.

Fig. 2 SEM images of the core–shell monolayer MoS₂ structure during the growth process. (a) The initial phase of sulfurization. (b)–(d) With further sulfurization, more MoS₂ ultrathin films form. Inset: close-up of the white rectangle.

3.3 Effect of water vapor on the growth process

It has been reported that PTCDA can promote the formation of MoO₃ seeds,²⁵ but the reason has not yet been explained in detail. Here, through comparing two experimental conditions: (a) with water vapor in the carrier gas, and (b) without water vapor in the carrier gas, it is found that the nanowires are discontinuous without water vapor (Fig. 3b), and that water vapor can enhance the interaction between MoO₃ and PTCDA. A possible mechanism can be described as follows. Firstly, MoO₃ reacts with water vapor according to the following equilibrium:³⁶

MoO₃(s) + H₂O(g) ⇌ MoO₂(OH)₂(g) (4)

Fig. 3 (a) Optical image for the experiment without PTCDA coating on the substrate. (b) Optical image of the core–shell structure obtained without water vapor in the carrier gas. (c) Optical image of PTCDA spin-coated on the substrate at 3000 rpm; inset: the molecular structure of PTCDA. The scale bars are 10 μm.

Based on the solid/gas equilibrium in eqn (4), it is reasonable to assume that the MoO₂(OH)₂ species reacts with hydroxyl groups on the surface of the substrate (supported by NaOH):

MoO₂(OH)₂(g) + 2OH⁻(surface) → MoO₄²⁻(surface) + 2H₂O(g) (5)

As a result, the MoO₃ forms MoO₄²⁻, which enhances the interaction between MoO₃ and PTCDA due to the presence of dangling bonds in MoO₄²⁻ (ref. 37–40).

3.4 Detailed analysis

XPS is used to investigate the chemical states of Mo and S in the surface regions of the core–shell nanowires, as shown in Fig. 4. The Mo 3d_5/2 electron peak at 233.2 eV corresponds to the 6+ oxidation state consistent with the core MoO₃. The Mo 3d_5/2 electron peak at 229.8 eV corresponds to the 4+ oxidation state (Fig. 4a), and the S 2p_3/2 peak at 162.3 eV shows that the S is entirely in the 2− oxidation state (Fig. 4b); quantification of the peak areas confirms a Mo : S ratio of 1 : 2, indicating the presence of MoS₂. In general, XPS indicates that MoO₃ and MoS₂ are present in the surface of the nanowires, and that MoO_3−x is not present. The MoO_3−x content is much less than the MoO₃ and MoS₂ content because MoO_3−x is a transitional product.

Fig. 4 XPS spectra of the core–shell nanowires showing (a) the Mo 3d and S 2s peaks, and (b) the S 2p peaks. Fitted peaks are shown as colored lines.

Raman spectra are used to confirm the formation of MoO₃–MoS₂ nanowires with core–shell structure (Fig. 5), and it is notable that the frequency difference (Δ) between the out-of-plane phonon mode (A_1g) and the in-plane phonon mode (E¹_2g) can be used to identify the number of MoS₂ layers. It is observed that, for the surface of the nanowires, other peaks besides the A_1g and E¹_2g Raman peaks are also present (Fig. 5a), which is in accordance with others' work, indicating that this region contains oxysulfide (MoOS₂).^26,29 On comparing the values of Δ shown in Fig. 5b and c, it is evident that the MoS₂ thin film on the edge of the nanowires is thinner than that on the regions away from the nanowires because the thin film at the edge of the nanowires is nearly suspended in the air (Fig. 2b), and is formed from the nanowires by the disappearance of dangling bonds. The value of Δ shown in Fig. 5d indicates that the MoS₂ thin film on the furthest edge of the core–shell structure is a monolayer.

Fig. 5 Raman spectra acquired from different regions highlighted in the inset of a: (a) (black), (b) (red), (c) (green) and (d) (purple) stand for the surface of the MoO₃–MoS₂ nanowires, the edge of the nanowires, the MoS₂ ultrathin film away from the nanowires and the furthest edge of the MoS₂ ultrathin film, respectively.

The photoluminescence (PL) spectra are in accordance with the Raman spectra. The PL spectrum of monolayer MoS₂ exhibits two peaks at ∼625 and 672 nm, corresponding to the A₁ and B₁ direct excitonic transitions, respectively. As the number of layers decreases, the PL intensities apparently become stronger. The PL intensity for the surface of core–shell MoO₃–MoS₂ is relatively weak because partially reduced MoO₃ shows high conductivity,^35,41 which depresses the PL intensity (Fig. 6).

Fig. 6 PL spectra acquired from different regions highlighted in Fig. 5. Inset: close-up of the black rectangle.

We further investigate the core–shell structure by AFM measurements (Fig. 7), and clearly observe a series of terraces between the centers of the nanowires and the edge of the thin film. It is noted that the terraces are not very flat, which is attributed to the curvature and branching of the MoS₂ nanosheets, especially on the edge of the terraces, and the terraces are not induced by the roughness of the substrate, as the surface of the bare substrate is very smooth (Fig. 7b, left inset). The distance between the terraces is approximately 7 Å, which is consistent with the distance between the MoS₂ layers, and the thickness is 9.5 Å at the edge of the thin film, indicating that the furthest edge of the thin film is a monolayer.

Fig. 7 (a) AFM image of core–shell MoO₃–MoS₂ nanowires. (b) A selected core–shell MoO₃–MoS₂ nanowire height profile along the black line indicated in (a). Insets: close-ups of the black rectangles.

The high resolution TEM image (Fig. 8a) and the corresponding selected area electron diffraction (SAED) pattern with [001] zone axis (Fig. 8b) reveal the hexagonal lattice structure with lattice spacings of 0.27 and 0.16 nm assigned to the (100) and (110) planes, respectively.

Fig. 8 (a) High resolution TEM image of few-layer or monolayer MoS₂. (b) The corresponding selected area electron diffraction (SAED) pattern.

3.5 Effect of the time (X) and the sulfur steam concentration during the growth process

Li et al. have studied the growth mechanism of ZnO nanostructures, from nanopapers to nanowires, via an indium-assisted vapor-phase transport (VPT) method;^43–45 the morphologies of the ZnO nanostructures were controlled by the substrate temperature and the vapor phase transportation distance between the source and the substrates.⁴² In our experiment, we find that the morphologies of the core–shell structures depend on the sulfur steam concentration and the time (X) during the growth process. Firstly, the time X (as mentioned in the experimental part) can influence the diameter of the nanowires. The results also verify that the MoO₃ nanowires are formed in the first X seconds; after X seconds, much sulfur stream begins to enter the quartz tube and the sulfurization process starts, which plays a dominant role in the growth of the large area MoS₂ ultrathin film. During the sulfurization process, the formation of MoO₃ nanowires is inhibited, owing to sulfurization on the surface of the MoO₃ powder (Mo source) in the ceramic boat. More and more MoS₂ powder, which is more difficult to evaporate than MoO₃ power for a given temperature, is produced in the ceramic boat, so that evaporation of the MoO₃ power is hindered to some extent (Fig. 9).

Fig. 9 AFM images of nanowires with different diameters, which correspond to different values of X. (a), (b), (c), (d) and (e) represent values of X of 40 s, 60 s, 80 s, 100 s and 120 s, and the diameters are 30 nm, 50 nm, 80 nm, 150 nm and 200 nm, respectively.

We have also found that the amount of sulfur is a critical factor in the process of synthesizing MoS₂ ultrathin films. At a given time, the sulfur steam concentration in the experiment depends on the position of the ceramic boat filled with sulfur relative to the center of the furnace which affects the temperature gradient, the outgoing flow of carrier gas from the chamber, the initial sulfur loading, and the partial pressure of the sulfur steam. Therefore, it is difficult to precisely analyze how different sulfur steam concentrations affect the morphology of MoS₂ due to the various influencing factors, however the experiments demonstrate that, for small amounts of sulfur, almost no MoS₂ ultrathin films are formed from the MoO₃ nanowires because of the extremely inadequate sulfurization (Fig. 10a). On increasing the level of sulfur loading, the sulfurization becomes sufficient, and more and more MoO_3−x is converted into MoS₂. As MoS₂ tends to form films, the nanowires become wider and wider, and core–shell MoO₃–MoS₂ nanowires with monolayer MoS₂ form (Fig. 10b and c). A large area MoS₂ ultrathin film is obtained at 0.8 grams, and the nanowires lose their shape, which is attributed to extensive sulfurization (Fig. 10d).

Fig. 10 Effect of sulfur concentration on the growth of MoS₂. For (a), (b), (c) and (d) the amounts of sulfur are: <0.3 grams, 0.4–0.5 grams, 0.6–0.7 grams and ≥0.8 grams, respectively. The scale bars are 10 μm.

4. Conclusion

We have employed a novel nucleation-controlled intermediate state in the synthesis of monolayer MoS₂ films, and have found that water vapor in the carrier gas was important for the experiment. Thinner and larger MoS₂ thin films were produced by controlling the amount of sulfur. We also investigated the terraces between the centers of the nanowires and the edge of the thin film, and found that the furthest edge of the thin film was a monolayer, which is meaningful for producing large area highly crystalline MoS₂ monolayers. Our next work will explore using the concentration of PTCDA and the spin-coating rotation speed to control the density of PTCDA coated on the substrate, because a suitable PTCDA density can induce the formation of thinner large area MoS₂ ultrathin films with fewer grain boundaries. The sulfur concentration and the pressure during the growth process are also important, as adequate sulfurization of the nanowires results in the production of large area highly crystalline MoS₂ monolayers.

Acknowledgements

This work was supported by the China Postdoctoral Science Foundation (no. 2013M540127), the National Natural Science Foundation of China (grant no. 91233120) and the National Basic Research Program of China (2011CB921901).

References

1. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
2. K. S. Novoselov, A. K. Geim, S. Morozov, D. Jiang, Y. Zhang, S. Dubonos, I. Grigorieva and A. Firsov, Science, 2004, 306, 666–669.
3. K. Novoselov, D. Jiang, F. Schedin, T. Booth, V. Khotkevich, S. Morozov and A. Geim, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 10451–10453.
4. M. Chhowalla and G. A. Amaratunga, Nature, 2000, 407, 164–167.
5. Y. Golan, C. Drummond, M. Homyonfer, Y. Feldman, R. Tenne and J. Israelachvili, Adv. Mater., 1999, 11, 934–937.
6. T. Kubart, T. Polcar, L. Kopeckỳ, R. Novak and D. Novakova, Surf. Coat. Technol., 2005, 193, 230–233.
7. D. Shtansky, T. Lobova, V. Y. Fominski, S. Kulinich, I. Lyasotsky, M. Petrzhik, E. Levashov and J. Moore, Surf. Coat. Technol., 2004, 183, 328–336.
8. B. Hinnemann, P. G. Moses, J. Bonde, K. P. Jørgensen, J. H. Nielsen, S. Horch, I. Chorkendorff and J. K. Nørskov, J. Am. Chem. Soc., 2005, 127, 5308–5309.
9. Y. Li, H. Wang, L. Xie, Y. Liang, G. Hong and H. Dai, J. Am. Chem. Soc., 2011, 133, 7296–7299.
10. D. Kong, H. Wang, J. J. Cha, M. Pasta, K. J. Koski, J. Yao and Y. Cui, Nano Lett., 2013, 13, 1341–1347.
11. M. S. Whittingham, Chem. Rev., 2004, 104, 4271–4302.
12. K. Brandt, Solid State Ionics, 1994, 69, 173–183.
13. J. Gobrecht, H. Tributsch and H. Gerischer, J. Electrochem. Soc., 1978, 125, 2085–2086.
14. R. F. Hanson, P. Szczepanik-VanLeeuwen, G. C. Williams, G. Grabowski and H. L. Sharp, Science, 1979, 203, 1107–1108.
15. A. Matthäus, A. Ennaoui, S. Fiechter, S. Tiefenbacher, T. Kiesewetter, K. Diesner, I. Sieber, W. Jaegermann, T. Tsirlina and R. Tenne, J. Electrochem. Soc., 1997, 144, 1013–1019.
16. K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805.
17. A. Splendiani, L. Sun, Y. Zhang, T. Li, J. Kim, C.-Y. Chim, G. Galli and F. Wang, Nano Lett., 2010, 10, 1271–1275.
18. R. Sundaram, M. Engel, A. Lombardo, R. Krupke, A. Ferrari, P. Avouris and M. Steiner, Nano Lett., 2013, 13, 1416–1421.
19. H. Zeng, J. Dai, W. Yao, D. Xiao and X. Cui, Nat. Nanotechnol., 2012, 7, 490–493.
20. B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti and A. Kis, Nat. Nanotechnol., 2011, 6, 147–150.
21. S. Tongay, J. Zhou, C. Ataca, J. Liu, J. Kang, T. Matthews, L. You, J. Li, J. C. Grossman and J. Wu, Nano Lett., 2013, 13, 2831–2836.
22. J.-M. Li, Appl. Phys. A: Mater. Sci. Process., 2010, 99, 229–235.
23. J. N. Coleman, M. Lotya, A. ONeill, S. D. Bergin, P. J. King, U. Khan, K. Young, A. Gaucher, S. De and R. J. Smith, Science, 2011, 331, 568–571.
24. J. V. Lauritsen, J. Kibsgaard, S. Helveg, H. Topsøe, B. S. Clausen, E. Lægsgaard and F. Besenbacher, Nat. Nanotechnol., 2007, 2, 53–58.
25. Y.-H. Lee, X.-Q. Zhang, W. Zhang, M.-T. Chang, C.-T. Lin, K.-D. Chang, Y.-C. Yu, J. T.-W. Wang, C.-S. Chang and L.-J. Li, Adv. Mater., 2012, 24, 2320–2325.
26. J.-C. Idrobo, P. M. Ajayan and J. Lou, Nat. Mater., 2013, 12, 754–759.
27. X. Wang, H. Feng, Y. Wu and L. Jiao, J. Am. Ceram. Soc., 2013, 135, 5304–5307.
28. Q. Li, J. Newberg, E. Walter, J. Hemminger and R. Penner, Nano Lett., 2004, 4, 277–281.
29. T. Weber, J. Muijsers, J. Van Wolput, C. Verhagen and J. Niemantsverdriet, J. Phys. Chem., 1996, 100, 14144–14150.
30. R. Tenne, L. Margulis, M. Genut and G. Hodes, Nature, 1992, 360, 444–446.
31. Y. Feldman, E. Wasserman, D. Srolovitz and R. Tenne, Science, 1995, 267, 222–225.
32. Y. Feldman, G. Frey, M. Homyonfer, V. Lyakhovitskaya, L. Margulis, H. Cohen, G. Hodes, J. Hutchison and R. Tenne, J. Am. Chem. Soc., 1996, 118, 5362–5367.
33. A. Zak, Y. Feldman, V. Alperovich, R. Rosentsveig and R. Tenne, J. Am. Chem. Soc., 2000, 122, 11108–11116.
34. X. L. Li and Y. D. Li, Chem. - Eur. J., 2003, 9, 2726–2731.
35. D. O. Scanlon, G. W. Watson, D. Payne, G. Atkinson, R. Egdell and D. Law, J. Phys. Chem. C, 2010, 114, 4636–4645.
36. J. Leyrer, M. Zaki and H. Knözinger, J. Phys. Chem., 1986, 90, 4775–4780.
37. A. Morozov, T. Kampen and D. Zahn, Surf. Sci., 2000, 446, 193–198.
38. G. Salvan, D. Tenne, A. Das, T. Kampen and D. Zahn, Org. Electron., 2000, 1, 49–56.
39. T. Soubiron, F. Vaurette, J. Nys, B. Grandidier, X. Wallart and D. Stiévenard, Surf. Sci., 2005, 581, 178–188.
40. J. Gustafsson, E. Moons, S. Widstrand and L. Johansson, Surf. Sci., 2004, 572, 23–31.
41. Z. Chen, D. Cummins, B. N. Reinecke, E. Clark, M. K. Sunkara and T. F. Jaramillo, Nano Lett., 2011, 11, 4168–4175.
42. J.-M. Li, X.-P. Wang, L.-G. Dai and Z.-A. Xu, CrystEngComm, 2013, 15, 1179–1184.
43. J.-M. Li, L.-G. Dai, X.-P. Wan and X.-L. Zeng, Appl. Phys. Lett., 2012, 101, 173105.
44. J.-M. Li, Nanotechnology, 2010, 21, 175603.
45. J.-M. Li, X.-L. Zeng, Q. Huang and Z.-A. Xu, CrystEngComm, 2012, 14, 7800–7806.

---