Large-area synthesis of monolayer MoSe₂ films on SiO₂/Si substrates by atmospheric pressure chemical vapor deposition

Abstract

We report the synthesis of large-scale continuous MoSe₂ films on SiO₂/Si substrates by atmospheric pressure chemical vapor deposition (CVD). As-grown thin films were composed of a continuous monolayer of MoSe₂ and extended up to a millimeter scale. The CVD-grown monolayer MoSe₂ films were uniform in thickness and highly crystalline with hexagonal crystal structures. Raman and photoluminescence spectra showed that CVD-grown monolayer MoSe₂ films have similar vibrational and optical properties to those of mechanically exfoliated monolayer MoSe₂. These results demonstrate that the CVD-grown monolayer MoSe₂ films have reasonably high quality comparable to that of mechanically exfoliated monolayer MoSe₂ flakes.

---

Introduction

In recent years, transition metal dichalcogenides (TMDs) with an atomically thin two-dimensional layer structure have attracted enormous attention because of their excellent electronic and optical properties.^1–8 The band gap of TMDs semiconductors (1–2 eV) makes them promising candidates for the channel materials of field-effect transistors (FETs).^9–11 Moreover, TMDs materials such as MoS₂ or MoSe₂ show indirect-to-direct bandgap transition when their thickness decreases down to monolayer.^12,13 This suggests that monolayer TMDs semiconductors are very promising for potential applications in optoelectronic devices such as photo detectors.^14,15

So far, various methods have been tried to obtain large-area TMDs films as the synthesis of large-area TMDs is one of the critical challenges for their real applications. Most emphasis has been put on MoS₂. Mechanical exfoliation provides high-quality crystalline flakes. But, their size is limited, the number of layer is uncontrollable, and the product yield is very low.^4,16,17 Chemical exfoliation and electrochemical exfoliation provide a simple and easy way but obtained flakes are small in the size and uncontrollable in thickness.^18,19 Sulfurization of transition metals or transition metal oxides provides large-area films, but obtained films show uncontrollable thickness uniformity.²⁰ In comparison, chemical vapor deposition (CVD) with solid-phase precursors (such as S and MoO₃ powder) can provide an effective method to synthesize monolayer MoS₂ with reasonable quality on different substrates including SiO₂/Si wafers, metal foils, or sapphire wafers.^21–24 Moreover, CVD with metalorganic (MO) precursors can provide wafer-scale continuous MoS₂ films.²⁵

The use of MoSe₂ could be more suitable than MoS₂ for the application of tunnel FETs, optoelectronic devices and spintronic devices with its narrower bandgap,¹³ higher optical absorbance²⁶ and larger spin-splitting energy²⁷ than MoS₂. However, compared to CVD-grown monolayer MoS₂, there have been much less reports on CVD-grown monolayer MoSe₂.^28–37 Thin films of MoSe₂ has not yet been obtained by CVD with MO precursors. Furthermore, it is much more difficult to synthesize monolayer MoSe₂ films by CVD with solid-phase precursors than monolayer MoS₂ films because of the low chemical reactivity of Se. CVD with solid-phase precursors commonly results in triangular-shaped discontinuous domains of either single-layer MoSe₂ or mixtures of single- and few-layer MoSe₂.^{28–30,32–35,37} While continuous thin films of monolayer MoSe₂ can be obtained on substrates up to about 1 cm² in size by meticulously tuning and optimizing experimental parameters of low pressure CVD processes,³⁶ continuous MoSe₂ films with such size have not yet been reported by atmospheric pressure CVD. As atmospheric pressure CVD can allow higher deposition rate and lower substrate temperature than low pressure CVD in general, it is still of interest to synthesize large-area MoSe₂ films with a uniform thickness by atmospheric pressure CVD. Here, we demonstrate the synthesis of large-area monolayer MoSe₂ films extended up to a millimeter scale on SiO₂/Si substrates by atmospheric pressure CVD. The obtained monolayer MoSe₂ shows very uniform film thickness, good continuity, highly crystalline hexagonal crystal structures. Furthermore, we compare for the first time the vibrational and optical properties of monolayer MoSe₂ thin films to those of mechanically exfoliated monolayer MoSe₂ single crystals.

Experimental

The schematic illustration of the synthesis of MoSe₂ films is shown in Fig. 1. The atmospheric pressure CVD reaction was carried out using a two-zone horizontal tube furnace with a one inch diameter quartz tube. Se powder was loaded in an alumina boat at the center of zone 1, and MoO₃ powder was loaded in an alumina boat at the center of zone 2. A 300 nm-SiO₂/Si substrate was located at the center of zone 2 with its face down on another alumina boat. During the synthesis of MoSe₂ film, zone 1 was heated to 300 °C with flowing Ar/H₂ carrier gases under atmospheric pressure while zone 2 was heated to 750 °C. The Ar and H₂ flow rates were 1500 sccm and 150 sccm, respectively. Se powder was evaporated at the center of zone 1 and transported to zone 2 downstream by Ar carrier gas, and MoO₃ powder was reduced by H₂ gas at the same time. The selenization of vapor-phase Mo produced nucleation of MoSe₂ species on the SiO₂/Si substrate. Finally, the nucleated MoSe₂ species grew into large-area MoSe₂ films. It is well understood that hydrogen plays a critical role in the growth of MoSe₂ films as an additional reducing agent with Se and MoO₃ powder precursors.^30,38 MoSe₂ films were not observed on the substrates without H₂ gas.

Fig. 1 Schematic illustration for the growth of MoSe₂ films on a SiO₂/Si substrate by CVD.

Results and discussion

Fig. 2(a)–(d) show the optical images of as-grown MoSe₂. The area covered with monolayer MoSe₂ looks dark while uncovered area looks bright. The monolayer MoSe₂ films with several hundred micrometers were found on SiO₂/Si substrates as shown in Fig. 2(a) and (b) (we intentionally scratched the surface to leave the pink colored line). Fig. 2(c) shows some triangular domains of MoSe₂ with several micrometers on SiO₂/Si substrates. Those domains merge together and become a continuous and large-area film up to several hundred micrometers. Partly some triangular domains are folded, resulting in bilayers or triple layers as shown in Fig. 2(d). The homogeneous color of MoSe₂ films in optical images suggests that the as-grown MoSe₂ films have good continuity and uniformity. Atomic force microscopy (AFM) was used to measure the morphology and the thickness of the MoSe₂ films. Fig. 2(e) and (f) show the AFM image and the height of as-grown MoSe₂ film on the SiO₂/Si substrate, respectively. The measured height is about 0.7 nm, which is consistent with the thickness of mechanically exfoliated monolayer MoSe₂ flakes.

Fig. 2 (a–d) Optical images of the as-grown monolayer MoSe₂ film. (e) AFM image of the as-grown MoSe₂ film. (f) Height profile of the MoSe₂ film obtained from the corresponding white line shown in (e).

Raman spectrum is an effective method to characterize layer numbers of TMDs materials. Fig. 3(a) shows the Raman spectra of as-grown mono-, bi-, and multilayer MoSe₂ films with a 514 nm excitation laser. There are two main peaks in the Raman spectra, one is a sharp peak at low wavenumber corresponding to A_1g mode (out of plane vibration), and another is a broad peak at high wavenumber corresponding to E_2g mode (in plane vibration). In general, the location of Raman modes can be used to determine the thickness of TMDs materials. In this work, the A_1g and E_2g modes of monolayer MoSe₂ films are located at 239.2 and 290.3 cm⁻¹, respectively, which is well agreed with the previous results of 2H-phase exfoliated monolayer MoSe₂ flakes and CVD-grown monolayer MoSe₂ films.^12,33 As the thickness of MoSe₂ film increases from monolayer to multilayer, the A_1g mode is blue-shifted to 240.2 cm⁻¹, and the E_2g mode is red-shifted to 285.9 cm⁻¹. Similar results were also previously reported by other groups.^15,30,32,33

Fig. 3 (a) Raman spectra of MoSe₂ films with different number of layers. (b) Photoluminescence (PL) spectra of MoSe₂ with different number of layers. (c) Raman mapping of the monolayer MoSe₂ film (A_1g mode). (d) Raman mapping of the monolayer MoSe₂ film (E_2g mode).

In order to confirm the bandgap of CVD-grown MoSe₂ films, photoluminescence (PL) spectra were measured. Fig. 3(b) shows the PL spectra of mono-, bi-, and multilayer MoSe₂ films. A monolayer MoSe₂ film shows a high intensity peak at 790 nm (1.57 eV), which can be ascribed to the direct bandgap at K point of Brillouin zone.¹³ The observed PL emission at 1.57 eV is in good agreement with the reported values in literature.^32,33 The bandgap of MoSe₂ changes from direct to indirect as the thickness increases from monolayer to multilayer. This significantly decreases PL peak intensity.¹³ The PL peak of bilayer MoSe₂ films is red-shifted to 810 nm (1.53 eV), and the peak intensity is approximately 15 times weaker than that of monolayer MoSe₂ films. This is also in good agreement with the results on MoSe₂ bilayer in literature.³³ Moreover, it is difficult to find noticeable peak intensity for multilayer MoSe₂ films.

It is well known that Raman mapping can further confirm uniformity and continuity of the CVD-grown MoSe₂ films. Fig. 3(c) and (d) show Raman mapping of monolayer MoSe₂ films in the area of 40 × 40 μm² where the distribution in A_1g and E_2g mode is presented, respectively. The uniform color of the two mapping images indicates that the CVD-grown MoSe₂ films have considerably good uniformity and continuity. Raman and PL analysis suggests that our CVD-grown monolayer MoSe₂ films are uniform in thickness.

In order to investigate elemental composition and binding energy of the as-grown monolayer MoSe₂ films, X-ray photoelectron spectroscopy (XPS) was used. Carbon binding energy was set as a reference to remove any effects of charge accumulation on these samples, and the backgrounds were estimated as Shirley-type.²⁸ Fig. 4(a) shows that MoSe₂ films have the binding energy of 229.1 and 232.2 eV at Mo 3d_5/2 and 3d_3/2 peaks, respectively. In this work, the binding energies of Mo are significantly shifted from those of hexavalent Mo (∼232.5 and 235.9 eV), suggesting the reduction of Mo from Mo⁶⁺ (MoO₃) to Mo⁴⁺.³⁹ Fig. 4(b) shows the binding energies of 54.5 and 55.3 eV corresponding to divalent Se ions (Se 3d_5/2 and 3d_3/2, respectively). It is worth noting that XPS results are consistent with previous works in literature confirming the chemical valence states of the CVD-grown monolayer MoSe₂ films.^28,31

Fig. 4 XPS characterization of monolayer MoSe₂ film. (a) XPS spectra of Mo 3d binding energy and (b) XPS spectra of Se 3d binding energy.

The crystal structure of CVD-grown monolayer MoSe₂ films was characterized using transmission electron microscopy (TEM). The MoSe₂ films were transferred to a carbon-coated Cu TEM grid. Fig. 5(a) shows that the transferred MoSe₂ film has good continuity but some folds and small particles exist due to residual organics or unskillful transfer technique. Fig. 5(b) shows a triangular grain with several micrometers in size inside a continuous MoSe₂ film, indicating the polycrystalline nature of CVD-grown monolayer MoSe₂ films. Fig. 5(c) shows a TEM image of a monolayer MoSe₂ film with a corresponding fast Fourier transformation (FFT) pattern in the inset. The FFT pattern exhibits clear six-fold symmetry of diffraction spots, demonstrating single layer MoSe₂ film with a hexagonal structure. Fig. 5(d) shows a TEM image of a bilayer MoSe₂ film with a corresponding FFT pattern in the inset. The FFT pattern indicates a twist between two layers with additional six diffraction spots. Fig. 5(e) clearly demonstrates the expected hexagonal crystal structure of monolayer MoSe₂ in the high resolution TEM (HRTEM) image. The lattice constant is ∼0.3 nm, which is well agreed with the previous results of monolayer MoSe₂.^28,30,31,33 HRTEM images and FFT patterns confirm the high crystallinity of the CVD-grown monolayer MoSe₂ films.

Fig. 5 TEM characterization of the sample. (a) TEM image of a CVD-grown monolayer MoSe₂ film transferred onto a TEM grid. The folds were produced during the transfer onto the grid. (b) Magnified TEM image of a monolayer MoSe₂ film. The dot line indicates a triangular grain. (c) TEM image of a monolayer MoSe₂ film with its corresponding FTT (inset). (d) TEM image of a bilayer MoSe₂ film with its corresponding FTT (inset). (e) HRTEM image of a monolayer MoSe₂ film. (f) Schematics illustrating the atomic structure of MoSe₂.

It needs to be mentioned that, while perfect 2H-phase MoS₂ is known to be inert, defects such as vacancies and grain boundaries in real MoS₂ can lead to poor long-term stability in air.^40,41 As a theoretical calculation predicts comparable oxidation behavior between single layer MoS₂ and MoSe₂,⁴⁰ we expect our CVD-grown MoSe₂ thin films could be more resistant to oxidation with improved crystallinity in the future.

Finally, the quality of the CVD-grown monolayer MoSe₂ film was compared to a mechanically exfoliated monolayer MoSe₂ flake through Raman and PL spectra. Fig. 6(a) shows that Raman spectrum of CVD-grown monolayer MoSe₂ film is consistent with that of an exfoliated monolayer MoSe₂ flake, which indicates the high quality of the CVD-grown monolayer MoSe₂ film. Even though the PL spectrum in Fig. 6(b) shows that the emission peak of CVD-grown monolayer MoSe₂ film exhibits very similar intensity and energy to those of the mechanically exfoliated monolayer MoSe₂ flake, the CVD-grown monolayer MoSe₂ film shows a slightly wider full-width at half-maximum of PL peak than that of the mechanically exfoliated monolayer MoSe₂ flake. It may imply that the CVD-grown monolayer MoSe₂ film has a little higher defect density than that of the mechanically exfoliated monolayer MoSe₂ flake. In fact, the interpretation of the crystal quality by optical analysis like Raman or PL is complicated to assess the crystal quality of two-dimensional materials.⁴² Nevertheless, these results suggest that the CVD-grown monolayer MoSe₂ film has reasonably high quality comparable to that of the mechanically exfoliated monolayer MoSe₂ flake.

Fig. 6 (a) Raman spectra of the mechanically exfoliated monolayer MoSe₂ flake and the CVD-grown monolayer MoSe₂ film. (b) Photoluminescence (PL) spectra of the mechanically exfoliated monolayer MoSe₂ flake and CVD-grown monolayer MoSe₂ film.

Conclusions

We investigated the synthesis of large-area monolayer MoSe₂ films on SiO₂/Si substrates by atmospheric pressure CVD. The CVD-grown monolayer MoSe₂ exhibited continuous and large- scale films with uniform thickness. The HRTEM images with corresponding FFT patterns revealed that CVD-grown monolayer MoSe₂ films have hexagonal crystal structure with high crystallinity. Raman and PL spectra analysis showed that the CVD-grown monolayer MoSe₂ films have comparable vibrational and optical properties with those of mechanically-exfoliated monolayer MoSe₂ flakes. These results demonstrate that our large-area CVD-grown monolayer MoSe₂ films have reasonably high quality comparable to that of the mechanically exfoliated monolayer MoSe₂ flake. We suggest that our large-area and high quality CVD-grown monolayer MoSe₂ films can expand understanding on the synthesis of TMDs materials, providing potentially important implications on their applications in various electronic and photonic devices in the future.

Acknowledgements

This work was supported by Brain Korea 21 Plus Project (T1300304) and Carbon-based Technology for the Internet-of-Things Era Project (K1518861).

References

1. A. Splendiani, L. Sun, Y. Zhang, T. Li, J. Kim, C. Y. Chim, G. Galli and F. Wang, Nano Lett., 2010, 10, 1271–1275.
2. B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti and A. Kis, Nat. Nanotechnol., 2011, 6, 147–150.
3. G. Eda, H. Yamaguchi, D. Voiry, T. Fujita, M. Chen and M. Chhowalla, Nano Lett., 2011, 11, 5111–5116.
4. Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman and M. S. Strano, Nat. Nanotechnol., 2012, 7, 699–712.
5. W. Choi, M. Y. Cho, A. Konar, J. H. Lee, G. B. Cha, S. C. Hong, S. Kim, J. Kim, D. Jena, J. Joo and S. Kim, Adv. Mater., 2012, 24, 5832–5836.
6. 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 and K. Kim, Nat. Commun., 2012, 3, 1011.
7. H. Kwon, W. Choi, D. Lee, Y. Lee, J. Kwon, B. Yoo, C. P. Grigoropoulos and S. Kim, Nano Res., 2014, 7, 1137–1145.
8. H. T. Mohammad and J. S. Volker, Nanotechnology, 2015, 26, 344005.
9. O. Lopez-Sanchez, D. Lembke, M. Kayci, A. Radenovic and A. Kis, Nat. Nanotechnol., 2013, 8, 497–501.
10. B. Chamlagain, Q. Li, N. J. Ghimire, H.-J. Chuang, M. M. Perera, H. Tu, Y. Xu, M. Pan, D. Xaio, J. Yan, D. Mandrus and Z. Zhou, ACS Nano, 2014, 8, 5079–5088.
11. D. Ovchinnikov, A. Allain, Y.-S. Huang, D. Dumcenco and A. Kis, ACS Nano, 2014, 8, 8174–8181.
12. K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805.
13. S. Tongay, J. Zhou, C. Ataca, K. Lo, T. S. Matthews, J. Li, J. C. Grossman and J. Wu, Nano Lett., 2012, 12, 5576–5580.
14. H. R. Gutiérrez, N. Perea-López, A. L. Elías, A. Berkdemir, B. Wang, R. Lv, F. López-Urías, V. H. Crespi, H. Terrones and M. Terrones, Nano Lett., 2013, 13, 3447–3454.
15. P. Tonndorf, R. Schmidt, P. Bottger, X. Zhang, J. Borner, A. Liebig, M. Albrecht, C. Kloc, O. Gordan, D. R. T. Zahn, S. M. de Vasconcellos and R. Bratschitsch, Opt. Express, 2013, 21, 4908–4916.
16. H. Li, J. Wu, Z. Yin and H. Zhang, Acc. Chem. Res., 2014, 47, 1067–1075.
17. M. Chhowalla, H. S. Shin, G. Eda, L.-J. Li, K. P. Loh and H. Zhang, Nat. Chem., 2013, 5, 263–275.
18. J. N. Coleman, M. Lotya, A. O'Neill, S. D. Bergin, P. J. King, U. Khan, K. Young, A. Gaucher, S. De, R. J. Smith, I. V. Shvets, S. K. Arora, G. Stanton, H. Y. Kim, K. Lee, G. T. Kim, G. S Duesberg, T. Hallam, J. J. Boland, J. J. Wang, J. F. Donegan, J. C. Grunlan, G. Moriarty, A. Shmeliov, R. J. Nicholls, J. M. Perkins, E. M. Grieveson, K. Theuwissen, D. W. McComb, P. D. Nellist and V. Nicolosi, Science, 2011, 331, 568–571.
19. Z. Zeng, Z. Yin, X. Huang, H. Li, Q. He, G. Lu, F. Boey and H. Zhang, Angew. Chem., Int. Ed., 2011, 50, 11093–11097.
20. Y.-C. Lin, W. Zhang, J.-K. Huang, K.-K. Liu, Y.-H. Lee, C.-T. Liang, C.-W. Chu and L.-J. Li, Nanoscale, 2012, 4, 6637–6641.
21. Y.-H. Lee, L. Yu, H. Wang, W. Fang, X. Ling, Y. Shi, C.-T. Lin, J.-K. Huang, M.-T. Chang, C.-S. Chang, M. Dresselhaus, T. Palacios, L.-J. Li and J. Kong, Nano Lett., 2013, 13, 1852–1857.
22. J. Shi, D. Ma, G.-F. Han, Y. Zhang, Q. Ji, T. Gao, J. Sun, X. Song, C. Li, Y. Zhang, X.-Y. Lang, Y. Zhang and Z. Liu, ACS Nano, 2014, 8, 10196–10204.
23. L. Ma, D. N. Nath, E. W. Lee, C. H. Lee, M. Yu, A. Arehart, S. Rajan and Y. Wu, Appl. Phys. Lett., 2014, 105, 072105.
24. W. Shanshan, P. Merce, B. Harish and H. W. Jamie, Nanotechnology, 2016, 27, 085604.
25. K. Kang, S. Xie, L. Huang, Y. Han, P. Y. Huang, K. F. Mak, C.-J. Kim, D. Muller and J. Par, Nature, 2015, 520, 656–660.
26. M. Bernardi, M. Palummo and J. C. Grossman, Nano Lett., 2013, 13, 3664.
27. Y. Zhang, T.-R. Chang, B. Zhou, Y.-T. Cui, H. Yan, Z. Liu, F. Schmitt, J. Lee, R. Moore, Y. Chen, H. Lin, H.-T. Jeng, S.-K. Mo, Z. Hussain, A. Bansil and Z.-X. Shen, Nat. Nanotechnol., 2014, 9, 111.
28. Y. H. Chang, W. Zhang, Y. Zhu, Y. Han, J. Pu, J. K. Chang, W. T. Hsu, J. K. Huang, C. L. Hsu, M. H. Chiu, T. Takenobu, H. Li, C. I. Wu, W. H. Chang, A. T. S. Wee and L. J. Li, ACS Nano, 2014, 8, 8582–8590.
29. A. Bachmatiuk, R. F. Abelin, H. T. Quang, B. Trzebicka, J. Eckert and M. H. Rummeli, Nanotechnology, 2014, 25, 365603.
30. J. Shaw, H. Zhou, Y. Chen, N. Weiss, Y. Liu, Y. Huang and X. Duan, Nano Res., 2014, 7, 511–517.
31. X. Wang, Y. Gong, G. Shi, W. L. Chow, K. Keyshar, G. Ye, R. Vajtai, J. Lou, Z. Liu, E. Ringe, B. K. Tay and P. M. Ajayan, ACS Nano, 2014, 8, 5125–5131.
32. X. Lu, M. I. B. Utama, J. H. Lin, X. Gong, J. Zhang, Y. Y. Zhao, S. T. Pantelides, J. X. Wang, Z. L. Dong, Z. Liu, W. Zhou and Q. H. Xiong, Nano Lett., 2014, 14, 2419–2425.
33. J. Xia, X. Huang, L.-Z. Liu, M. Wang, L. Wang, B. Huang, D.-D. Zhu, J.-J. Li, C.-Z. Gu and X.-M. Meng, Nanoscale, 2014, 6, 8949–8955.
34. G. W. Shim, K. Yoo, S.-B. Seo, J. Shin, D. Y. Jung, I.-S. Kang, C. W. Ahn, B. J. Cho and S.-Y. Choi, ACS Nano, 2014, 8, 6655–6662.
35. C. Jung, S. M. Kim, H. Moon, G. Han, J. Kwon, Y. K. Hong, I. Omkaram, Y. Yoon, S. Kim and J. Park, Sci. Rep., 2015, 5, 15313.
36. Y. Gong, G. Ye, S. Lei, G. Shi, Y. He, J. Lin, X. Zhang, R. Vajtai, S. T. Pantelides, W. Zhou, B. Li and P. M. Ajayan, Adv. Funct. Mater., 2016, 26, 2009–2015.
37. H. Liu, Z. Chen, X. Chen, S. Chu, J. Huang and R. Peng, J. Mater. Chem. C, 2016, 4, 9399–9404.
38. J.-K. Huang, J. Pu, C.-L. Hsu, M.-H. Chiu, Z.-Y. Juang, Y. H. Chang, W.-H. Chang, Y. Iwasa, T. Takenobu and L.-J. Li, ACS Nano, 2014, 8, 923–930.
39. M. Bougouma, A. Batan, B. Guel, T. Segato, J. B. Legma, F. Reniers, M.-P. Delplancke-Ogletree, C. Buess-Herman and T. Doneux, J. Cryst. Growth, 2013, 363, 122–127.
40. H. Liu, N. Han and J. Zhao, RSC Adv., 2015, 5, 17572–17581.
41. J. Gao, B. Li, J. Tan, P. Chow, T.-M. Lu and N. Koratkar, ACS Nano, 2016, 10, 2628–2635.
42. S. Tongay, J. Suh, C. Ataca, W. Fan, A. Luce, J. S. Kang, J. Liu, C. Ko, R. Raghunathanan, J. Zhou, F. Ogletree, J. B. Li, J. C. Grossman and J. Wu, Sci. Rep., 2013, 3, 2657.

---