Synthesis and Transfer of Single-Layer Transition Metal Disulfides on

Letter pubs.acs.org/NanoLett

Synthesis and Transfer of Single-Layer Transition Metal Disulfides on Diverse Surfaces Yi-Hsien Lee,†,‡ Lili Yu,† Han Wang,† Wenjing Fang,† Xi Ling,†,∥ Yumeng Shi,† Cheng-Te Lin,‡ Jing-Kai Huang,‡ Mu-Tung Chang,§ Chia-Seng Chang,§ Mildred Dresselhaus,†,∥ Tomas Palacios,† Lain-Jong Li,*,‡ and Jing Kong*,† †

Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge Massachusetts 02139, United States ‡ Institute of Atomic and Molecular Sciences, Academia Sinica, No. 1, Roosevelt Rd., Sec. 4, Taipei, 10617, Taiwan § Institute of Physics, Academia Sinica, 128 Sec. 2, Academia Rd., Nankang, Taipei 11529, Taiwan ∥ Department of Physics, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States S Supporting Information *

ABSTRACT: Recently, monolayers of layered transition metal dichalcogenides (LTMD), such as MX2 (M = Mo, W and X = S, Se), have been reported to exhibit significant spin-valley coupling and optoelectronic performances because of the unique structural symmetry and band structures. Monolayers in this class of materials offered a burgeoning field in fundamental physics, energy harvesting, electronics, and optoelectronics. However, most studies to date are hindered by great challenges on the synthesis and transfer of high-quality LTMD monolayers. Hence, a feasible synthetic process to overcome the challenges is essential. Here, we demonstrate the growth of high-quality MS2 (M = Mo, W) monolayers using ambient-pressure chemical vapor deposition (APCVD) with the seeding of perylene-3,4,9,10-tetracarboxylic acid tetrapotassium salt (PTAS). The growth of a MS2 monolayer is achieved on various surfaces with a significant flexibility to surface corrugation. Electronic transport and optical performances of the as-grown MS2 monolayers are comparable to those of exfoliated MS2 monolayers. We also demonstrate a robust technique in transferring the MS2 monolayer samples to diverse surfaces, which may stimulate the progress on the class of materials and open a new route toward the synthesis of various novel hybrid structures with LTMD monolayer and functional materials. KEYWORDS: Metal dichalcogenides, 2D materials, monolayer, transfer

L

hydrogen.16 Moreover, monolayer MoS2 and WS2 have been considered as an ideal material for exploring valleytronics and valley-based optoelectronic applications.17−20 The broken inversion symmetry of the monolayer and the strong spin− orbit coupling lead to a fascinating interplay between spin and valley physics, enable simultaneous control over the spin and valley degrees of freedom, and create an avenue toward the integration of spintronics and valleytronics applications. Considerable efforts have been devoted to synthesize an MoS 2 monolayer, including various kinds of exfoliations,10,14,15,22,23 physical vapor deposition,8,24 and chemical vapor deposition (CVD) approaches.25−27 Recently, a CVDMoS2 monolayer was presented with sulfurization of the thin Mo layer26 and the layer growth induced using fragments of reduced graphene oxide as seeds.25 However, the as-grown layers display obvious thickness variations, and their optoelec-

ayered transition metal dichalcogenides (LTMD), including MX2 (M = Mo, W; X = S, Se), have attracted extensive research efforts in the fields of nanotribology, catalysis, energy harvesting, and optoelectronics.1−11 Monolayers of two-dimensional crystals, such as graphene, have been highlighted regarding both scientific and industrial aspects due to novel physical phenomenon inherited from the reduced dimensionality.12 Similarly, the broken inversion symmetry and the indirect-to-direct bandgap transition of LTMD are observed when the dimension is reduced from multilayers to a monolayer.13−15 The LTMD monolayers, being considered as the thinnest semiconductor, exhibit great potential for advanced short-channel devices.11 The transistor fabricated with an exfoliated MoS2 monolayer displays a high on−off current ratio and good electrical performance, which are both necessary for electronic circuit requiring low stand-by power.10 Recent theoretical predictions suggest that the dissociation of H2O could be realized at defects of single layer MoS2, which is critical for developing clean and sustainable energy from © 2013 American Chemical Society

Received: February 22, 2013 Published: March 18, 2013 1852

dx.doi.org/10.1021/nl400687n | Nano Lett. 2013, 13, 1852−1857

Nano Letters

Letter

Figure 1. (a) Chemical structure of PTAS (right) and schematic picture for the growth process on diverse surfaces (left); (b) temperature dependence of the weight loss and differential weight loss of PTAS using thermogravimetry analysis (TGA). (c−d) SEM images of (c) MoS2 and (d) WS2 grown on the cleaved side-wall of Si substrates. (e−g) SEM images of monolayer MoS2 on (e) a 5 μm Si particle; (f) aggregates of TiO2 nanoparticles and (g) sapphire; (h) Optical microscope (OM) image of monolayer MoS2 on quartz.

tronic performance is a few orders of magnitude worse than the exfoliated ones. Further applications and scientific study have been hindered with a reduced mobility and a low on−off current ratio because of the high defect concentration and small grain size. Most studies still use exfoliated samples since the synthesis of high-quality LTMD monolayers remains a great challenge so far. However, the synthesis of an LTMD monolayer is possible to be achievable using various aromatic molecules as seeds. An aromatic molecule seed with high thermal stability and better control of the seeding treatment on surfaces is essential to overcome the challenges for the synthesis of a high quality LTMD monolayer, which may also provide more understanding on the novel seed-enhanced growth. A robust transfer technique to avoid degradation in quality and contamination is essential for fundamental physics and optoelectronic applications. In this study, we demonstrate that high-quality MS2 monolayers can be directly synthesized on various surfaces using a scalable APCVD process with the seeding of perylene-3,4,9,10-tetracarboxylic acid tetrapotassium salt (PTAS). Not only is the growth successful for surfaces of different materials, but it has been found that the deposition

method is also applicable for surfaces with various morphologies. The as-synthesized MS2 monolayer exhibits a single crystalline structure with a specific flake shape even on amorphous surfaces. Meanwhile, we also present a reliable transfer technique to enable MS2 monolayer growth on flexible substrates or surfaces of various functional materials while maintaining their high quality. Figure 1a shows the chemical formula of the PTAS and a schematic diagram for a possible growth mechanism. The high solubility of PTAS in water enables the seed solution to be uniformly distributed on hydrophilic substrate surfaces. Uniform tiny aggregations of PTAS appear on the substrate surfaces after the drying of water (Figure S1, Supporting Information). Compared to other aromatic molecules, PTAS survives better at a higher temperature. In Figure 1b, thermogravimetry analysis (TGA) of PTAS demonstrates good thermal stability and a slow decomposition rate when the growth temperature is below 820 °C. The existence of seeds is further confirmed by the atomic force microscopy (AFM) images of the as-grown WS2 and of the substrate after removal of the as-grown MoS2 (Figures S1c,d). With the seeding of 1853

dx.doi.org/10.1021/nl400687n | Nano Lett. 2013, 13, 1852−1857

Nano Letters

Letter

Figure 2. (a, b) OM images of MoS2 and WS2 monolayer near the edge region; (c, d) enlarged OM images in the marked area, with the inset showing the corresponding AFM images; (e, f) low-magnification and (g, h) high-resolution TEM images of as-grown MS2. Insets in (e, f) show the corresponding SAED patterns.

for the synthesis of an MS2 monolayer are summarized in Figure S2 and Table T1. At the growth temperature, MO3 powders were reduced by sulfur vapor to form volatile MO3−x.28 Substrates are facing down on the crucible, and the arriving MO3−x molecule would react with sulfur vapor to form MS2 on the substrates. Without the seeds, island growth of MoS 2 particles was commonly observed on bare SiO 2 surfaces.25 In contrast, the presence of PTAS on the surface significantly enables layer growth, possibly via assisting the adsorption of molecules and the initiation of heterogeneous nucleation. As shown in Figure 1c−h, as-grown MoS2 shows great flexibility to surface corrugations. In Figure 1c,d, MS2 flakes are uniformly grown on the cleaved side-wall of Si substrates. Most of the MoS2 flakes are single-layer, while WS2 flakes exhibit a variation on layer number. In Figure 1e, a micrometer-sized Si particle is covered with single-layer MoS2 flakes. Figure 1f shows that the growth of MoS2 flakes can even be achieved on aggregations of TiO2 nanoparticles. Nano-Auger electron spectroscopy is utilized to verify the existence of MS2 layers as shown in Figure S3. Furthermore, the growth of monolayer MoS2 is achievable on crystalline surfaces, including quartz and sapphire, as shown in Figure 1g,h. The triangle single-layer MoS2 flakes are commonly observed in the early stages of the growth. The ability to synthesize an LTMD monolayer with high tolerance to surface corrugation on diverse surfaces opens a route toward the synthesis of heteroand composite structures. In Figure 2a,b, uniform MoS2 and WS2 monolayers are grown on a SiO2/Si substrate, and these monolayers have centimeters and hundred micrometer sizes, respectively. The isolated MS2 flakes appear on the edge regions of the substrates (Figure 2c,d, the insets show AFM images of MoS2 and WS2 monolayers with thicknesses of 0.71 and 0.86 nm, respectively). In the inset of Figure 2c, there is an island in the center formed with the same edge orientation as the underneath MoS2 flake, which is consistent with its single crystal geometry. It is a crucial implication that the nucleation is the rate-controlling step in the seed-initiated growth process. The as-synthesized WS2 monolayer over a small amounts of seeds (the inset of

Figure 3. Mapping of (a) Raman peak intensity, (b) OM image, and (c) the PL peak intensity of a MoS2 monolayer; (d) mapping of Raman peak intensity, (e) OM image, and (f) PL peak intensity of WS2 flakes. Comparisons of the MS2 monolayer and bulk for the (g) Raman spectra and (h) PL spectra. Both Raman and PL experiments were performed in a confocal spectrometer using a 473 nm excitation laser.

PTAS, single-layer MS2 flakes can directly grow on diverse substrates. The experimental setup and optimized parameters 1854

dx.doi.org/10.1021/nl400687n | Nano Lett. 2013, 13, 1852−1857

Nano Letters

Letter

Figure 4. (a) Transport characteristics of the FETs fabricated on as-grown MoS2 on a linear scale (right y-axis) and a log scale (left y-axis). (b) Output characteristics of the MoS2 FET. The current is linear with the source drain voltage in the low electronic field region, indicating that the metal electrodes form ohmic contact with MoS2. (c) Transport characteristics of the FET fabricated on an as-grown WS2 monolayer on a linear scale (right y-axis) and a log scale (left y-axis). (d) Output characteristics of the WS2 FET. In the MS2 FETs, L = 1 μm.

modes of single-layer WS2 are located at 358 and 419 cm−1 with the fwhm values of 4.3 and 5.3 cm−1, while those of bulk WS2 are at 356 and 421 cm−1 with fwhm values of 3.6 and 3.5 cm−1, respectively. The Raman intensity of MS2 increases with thickness, whereas their PL intensity rapidly decreases with the increase of layer number (compare Figure 3g with h). The PL peaks of as-grown MoS2 and WS2 are approximately located at 670 and 633 nm, which is consistent with the published bandgap .13,21 Note that the PL peak of single layer MS2 is much stronger than the Raman signal, indicating good crystallinity and a low defect concentration of the as-grown MS2 monolayer.14,15 To evaluate the electrical performance of the as-grown MS2 monolayer, we fabricated bottom-gated transistors with the asgrown samples on SiO2/Si. Figure 4a−d shows the typical electrical performance of MS2 FETs. Both of them show n-type behavior, consistent with previous reports.10,11 We extract the field-effect electron mobility from the linear regime of the transfer properties using the equation μ= [dId/dVbg] × [L/ (WCgVd)], where L, W, and Cox are the channel length, width, and the gate capacitance per unit area, respectively. From the characteristics of the MoS2 FET shown in Figure 4a, the on−off current ratio exceeds 107, and the mobility is up to 1.2 cm2/V·s, which is comparable to an exfoliated MoS2 monolayer fabricated without high k-dielectrics.10 The excellent electrical performance demonstrates the low defect and high quality of our single-layer MoS2. To estimate the doping level of as-grown MoS2, the source/drain current at zero gate voltage was modeled as Id = qn2DWμ(Vd/L), where n2D is the 2-dimensional carrier concentration, q is the electron charge, μ is the calculated mobility, and Vd is the source/drain voltage,

Figure 2d) as well as the observation of an as-grown monolayer of large-area demonstrates that the growth of MS2 favors layer growth in the initial growth stage with the seeding of PTAS. The crystal structure and edge structure of the as-grown MS2 flakes are studied with TEM. In Figure 2, high-resolution TEM images and the corresponding SAED pattern with an [001] zone reveals the same hexagonal lattice structure and a similar lattice spacing for MoS2 and WS2.29,30 The spacing of (100) and (110) planes of both materials are 0.27 and 0.16 nm, respectively. Figure 2f shows that the domain facets clearly align along (100), (010), and (1−10) planes. In Figure S4, a fewlayer WS2 flake is shown, and the SAED patterns at different locations indicate that the flake is single crystal without any mis-orientations in the stacking of the layers. The single crystal structure and specific edge structures are essential to explore fundamental edge states in this class of materials. To study the spectroscopy and PL performance of the asgrown MS2, Raman and PL mapping in confocal measurements are shown in Figure 3. A uniform contrast and strong intensity is observed in the Raman (Figures 3a, d) and PL mapping plots (Figures 3c, f), implying that the MS2 exhibits a high crystallinity and good uniformity. The A1g Raman mode is very sensitive to layer number, and the peak frequency difference between the E2g and A1g modes can be used to identify the layer number of MoS2.28 In Figure 3g, the E2g and A1g modes of the Raman band of single-layer MoS2 are located at 385 and 403 cm−1, respectively, with full-width-halfmaximum (fwhm) values of 3.5 and 6.6 cm−1, while those of the bulk MoS2 are at 383 and 408 cm−1 with fwhm values of 4.1 and 3.3 cm−1. In contrast, the Raman E2g and A1g energies of WS2 are less sensitive to layer thickness, where the E2g and A1g 1855

dx.doi.org/10.1021/nl400687n | Nano Lett. 2013, 13, 1852−1857

Nano Letters

Letter

PTAS in water and the hydrophobic surfaces of MoS2. Figure S6 demonstrates the mass production of single-layer MoS2 nanosheets in DI water. After an as-grown MoS2 sample is put into a container with of DI water, the MoS2 monolayer rapidly peels off and breaks into small flakes that are suspended in the solution. After redeposit of the solution on another fresh SiO2/ Si substrate, a strong PL and Raman intensity of the transferred MoS2 indicates high-quality LTMD nanosheets in liquid solution are achievable. The transfer of the entire monolayer is also demonstrated using polydimethylsiloxane (PDMS) and water, as shown in Figure 5b. A clean and continuous MoS2 monolayer on PDMS is observed (Figure 5c,d). Using PDMS as a stamp, the transfer of MS2 monolayers to other substrates can be implemented. In Figure 5e,f, single-layer MoS2 can be well transferred to HOPG and a flexible PET substrate with direct stamping, which may enhance the developments in flexible optoelectronics and STM-related studies. A strong photoluminescence of the transferred MoS2 monolayer on PDMS and PET surfaces are observed in Figure 5g, illustrating the maintenance of quality for the MoS2 monolayer. Since only a drop of water is involved in the transfer process, contamination is avoided. Moreover, hybrid structures based on LTMD monolayer and functional materials, including conductive graphene,31 insulating h-BN,32 and multiferroic BiFeO3,33 are successfully fabricated using a direct stamping, as shown in Figure 5h−j. Thus it is anticipated that the development demonstrated in the present work will stimulate progress toward developing various novel hybrid structures and functional materials based on LTMD monolayers.

Figure 5. (a) Evolution of a drop of H2O, IPA, and acetone on the surfaces of an as-grown MoS2 monolayer. The three samples are cut from a uniform sample. (b) Procedures of the sample transfer with H2O and PDMS for monolayer MoS2 onto a SiO2/Si substrate (1−4). (c) Transferred MoS2 on PDMS (right) and on the original substrate (left), which is empty now. (d) OM image of the transferred MoS2 monolayer on (d) PDMS, (e) HOPG, and (f) PET. The inset in f shows a photo of MoS2 transferred on PET via direct stamping, (g) PL spectra of transferred MoS2 on PET and PDMS surfaces. The PL experiments were performed using a 532 nm excitation laser. (h−j) OM image of various hybrid structures: (h) MoS2 transferred to graphene on SiO2/Si substrate; (i) MoS2 transferred to h-BN flakes onto a SiO2/Si substrate; (j) MoS2 on BiFeO3 with an inset showing the SEM image of the clear interface between MoS2 on BiFeO3 and the BiFeO3 substrate only.

■

ASSOCIATED CONTENT

S Supporting Information *

Growth mechanism on the seed-initiated growth, experiment setup, growth flexibility, identification of the crystal structure with TEM, chemical configurations, and production of singlelayer LTMD nanosheets solutions, Figures S1−S6, Table S1. This material is available free of charge via the Internet at http://pubs.acs.org.

respectively. From the output characteristics of as-grown MoS2 (Figure 4b), n2D is extracted to be ∼5.2 × 1010 cm−2. Figure 4c, d shows the electrical characterizations of WS2 FETs. The on− off ratio is approximately 105, and the mobility is around 0.01 cm2/V·s, which is relatively low compared to that of the MoS2 based FET. It is worth pointing out this is the first FET based on CVD-grown WS2 and the metal electrodes should be optimized in the future to improve the performances. As the growth temperature of MS2 monolayers are relatively high, temperature-sensitive substrates (such as polymer-based substrates) could not be used in the synthetic process. It is essential to develop a transfer technique to implement largearea MS2 on even more versatile types of substrates. Here, we report a transfer technique that maintains the quality of the asgrown monolayer. In Figure 5a, the as-grown MoS2 sample is cut into three pieces and is respectively treated with DI water, isopropyl alcohol (IPA), and acetone for 30 s. The surface of the as-grown monolayer is hydrophobic, so that IPA and acetone are spread out on MoS2, whereas water remains as a droplet. During the 30 s, the as-grown MoS2 monolayer starts breaking into small pieces and floating on the water droplet. Thus the as-grown MoS2 monolayer can be easily removed from the substrate with DI water. We do not observe such obvious lift-off behaviors for organic solvents. It is suspected that the DI water in the MoS2-substrate interface assists the liftoff of the MoS2 monolayer because of the high solubility of

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected], [email protected]. Author Contributions

Y.H.L. carried out most experiments and analyzed the data. L.Y., H.W., and T.P. contributed to device fabrications and electronic measurements. W.F., X.L., and Y.S. contributed to the synthesis of MoS2 and Raman analysis. C.T.L. and J.K.H. contributed to TGA and XPS. M.T.C. and C.S.C. contributed to SEM and TEM. J.K., L.J.L., and Y.H.L. designed the experiment and cowrote the paper. M.D. provided suggestions on the manuscript. All of the authors discussed the results and provided comments. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by the National Science Foundation under award number NSF DMR 0845358. W.F. ackowledges the support from Materials, Structures and Device (MSD) Center, one of the five programs in the focus center research program (FCRP), a Semiconductor Research Corporation program. J.K. acknowledges the Graphene Approaches to 1856

dx.doi.org/10.1021/nl400687n | Nano Lett. 2013, 13, 1852−1857

Nano Letters

Letter

(32) Lee, Y. H.; Liu, K. K.; Lu, A. Y.; Wu, C. Y.; Lin, C. T.; Zhang, W.; Su, C. Y.; Lin, T. W.; Wei, K. H.; Shi, Y.; Li, L. J. RSC Adv. 2012, 2, 111−115. (33) Lee, Y. H.; Wu, J. M.; Lai, C. H. Appl. Phys. Lett. 2006, 88, 042908.

Terahertz Electronics (GATE)MURI grant N00014-09-11063. The authors acknowledge financial support from the Office of Naval Research (ONR) Young Investigator Program.

■

■

REFERENCES

NOTE ADDED AFTER ASAP PUBLICATION Figure 4 was incorrect in the version published ASAP March 29, 2013. The corrected version was re-posted on April 10, 2013.

(1) Frindt, R. F.; Yoffe, A. D. Proc. R. Soc. London, A 1963, 273, 69− 83. (2) Wilson, J. A.; Yoffe, A. D. Adv. Phys. 1969, 18, 193−335. (3) Divigalpitiya, W. M. R.; Frindt, R. F.; Morrison, S. R. Science 1989, 246, 369−371. (4) Tenne, R.; Margulis, L.; Genut, M.; Hodes, G. Nature 1992, 360, 444−446. (5) Rapoport, L.; Bilik, Y.; Feldman, Y.; Homyonfer, M.; Cohen, S. R.; Tenne, R. Nature 1997, 387, 791−793. (6) Chhowalla, M.; Amaratunga, G. A. Nature 2000, 407, 164−167. (7) Tenne, R. Nat. Nanotechnol. 2006, 1, 103−111. (8) Lauritsen, J. V.; Kibsgaard, J.; Helveg, S.; Topsoe, H.; Clausen, B. S.; Lagsgaard, E.; Besenbacher, F. Nat. Nanotechnol. 2007, 2, 53−58. (9) Jaramillo, T. F. Science 2007, 317, 100−102. (10) Radisavljevic, B.; Radenovic, A.; Brivio, J.; Giacometti, V.; Kis, A. Nat. Nanotechnol. 2011, 6, 147−150. (11) Wang, H.; Yu, L.; Lee, Y. H.; Shi, Y.; Hsu, A.; Chin, M. L.; Li, L. J.; Dubey, M.; Kong, J.; Palacios, T. Nano Lett. 2012, DOI: 10.1021/ nl302015v. (12) Novoselov, K. S.; Geim, A. K. Nat. Mater. 2007, 6, 183−191. (13) Frey, G. L.; Tenne, R.; Matthews, M. J; Dresselhaus, M. S.; Dresselhaus, G. J. Mater. Res. 1998, 13, 2412−2417. (14) Mak, K. F.; Lee, C.; Hone, J.; Shan, J.; Heinz, T. F. Phys. Rev. Lett. 2010, 105, 136805. (15) Splendiani, A.; Sun, L.; Zhang, Y.; Li, T.; Kim, J.; Chim, C. Y.; Galli, G.; Wang, F. Nano Lett. 2010, 10, 1271−1275. (16) Ataca, C.; Ciraci, S. Phys. Rev. B 2012, 85, 195410. (17) Mak, K. F.; He, K.; Shan, J.; Heinz, T. F. Nat. Nanotechnol. 2012, DOI: 10.1038/NNANO.2012.96. (18) Zeng, H.; Dai, J.; Yao, W.; Xiao, D.; Cui, X. Nat. Nanotechnol. 2012, DOI: 10.1038/NNANO.2012.95. (19) Cao, T.; Wang, G.; Han, W.; Ye, H.; Zhu, C.; Shi, J.; Niu, Q.; Tan, P.; Wang, E.; Liu, B.; Feng, J. Nat. Commun. 2012, DOI: 10.1038/ncomms1882. (20) Zeng, H.; Liu, G. B.; Dai, J.; Yan, Y.; Zhu, B.; He, R.; Xie, L.; Xu, S.; Chen, X.; Yao, W.; Cui, X. arXiv:1208.5864. (21) Kuc, A.; Zibouche, N.; Heine, T. Phys. Rev. B 2011, 83, 245213. (22) Shi, Y.; Zhou, W.; Lu, A. Y.; Fang, W.; Lee, Y. H.; Hsu, A. L.; Kim, S. M.; Kim, K. K.; Yang, H. Y.; Li, L. J.; Idrobo, J. C.; Kong, J. Nano Lett. 2012, 12, 2784−2791. (23) Coleman, J. N.; et al. Science 2011, 331, 568−571. (24) Helveg, S.; Lauritsen, J. V.; Lægsgaard, E.; Stensgaard, I.; Nørskov, J. K.; Clausen, B. S.; Topsøe, H.; Besenbacher, F. Phys. Rev. Lett. 2000, 84, 951−954. (25) Lee, Y. H.; Zhang, X. Q.; Zhang, W.; Chang, M. T.; Lin, C. T.; Chang, K. D.; Yu, Y.-C.; Wang, T. W.; Chang, C.-S.; Li, L. J.; Lin, T. W. Adv. Mater. 2012, 24, 2320−2325. (26) Zhan, Y.; Liu, Z.; Najmaei, S.; Ajayan, P. M.; Lou, J. Small 2012, 8, 966−971. (27) Liu, K. K.; Zhang, W.; Lee, Y. H.; Lin, Y. C.; Chang, M. T.; Su, C. Y.; Chang, C. S.; Li, H.; Shi, Y.; Zhang, H.; Lai, C. S.; Li, L. J. Nano Lett. 2012, 12, 1538−1544. (28) Li, X. L.; Li, Y. D. Chem.Eur. J. 2003, 9, 2726−2731. (29) Seo, J. W; Jun, Y. W.; Park, S W.; Nah, H.; Moon, T.; Park, B.; Kim, J.; Kim, Y. J.; Cheon, J. Angew. Chem., Int. Ed. 2007, 46, 8828− 8831. (30) Li, H.; Yin, Z. Y.; He, Q. Y.; Li, H.; Huang, X.; Lu, G.; Fam, D. W. H.; Tok, A. I. Y.; Zhang, Q.; Zhang, H. Small, DOI: 10.1002/ smll.201101016. (31) Reina, A.; Jia, X.; Ho, J.; Nezich, D.l; Son, H.; Bulovic, V.; Dresselhaus, M. S.; Kong, J. Nano Lett. 2009, 9, 30−35. 1857

dx.doi.org/10.1021/nl400687n | Nano Lett. 2013, 13, 1852−1857