ChemComm Published on 24 September 2015. Downloaded by Universitat Erlangen Nurnberg on 24/01/2017 11:18:53. COMMUNICATION Cite this: Chem. Commun., 2015, 51, 16553 Received 10th July 2015, Accepted 20th September 2015 View Article Online View Journal | View Issue Atomic layer deposition on 2D transition metal chalcogenides: layer dependent reactivity and seeding with organic ad-layers Christian Wirtz,ab Toby Hallam,b Conor Patrick Cullen,ab Nina C. Berner,ab Maria O’Brien,ab Mario Marcia,c Andreas Hirschc and Georg S. Duesberg*ab DOI: 10.1039/c5cc05726d www.rsc.org/chemcomm This commmunication presents a study of atomic layer deposition of Al2O3 on transition metal dichalcogenide (TMD) two-dimensional films which is crucial for use of these promising materials for electronic applications. Deposition of Al2O3 on pristine chemical gate oxides in integrated circuits.8 The most common ALD process is the deposition of Al2O3 from alternating exposures of trimethylaluminium (TMA, Al(CH3)3) and water according to the reaction:9 vapour deposited MoS2 and WS2 crystals is demonstrated. This deposition is dependent on the number of TMD layers as there is 2Al(CH3)3 + 3H2O - Al2O3 + 6CH4 DH = 376 kcal no deposition on pristine monolayers. In addition, we show that it is possible to reliably seed the deposition, even on the monolayer, using non-covalent functionalisation with perylene derivatives as anchor unit. The integration of transition metal dichalcogenides (TMDs) into existing semiconductor technology is of major interest.1,2 The synthesis of these materials has been vastly improved over the last few years and many possible devices have been proposed and realised.3–6 But to finally achieve their large-scale integration and production, it is necessary to make TMDs fully CMOS processable. One prerequisite for this is the deposition of subsequent layers on top of the TMD for gating and passivation as the performance and stability of TMD based devices hugely depends on their dielectric environment. This task is nontrivial as any impact on the surface of the TMD will result in the destruction of its electronic properties. It has been shown that encapsulation of the 2D material by mechanical deposition of hexagonal boron nitride results in the best preservation of its electronic properties but this approach is not scalable.7 Other frequently used deposition methods for oxides such as sputtering or plasma-enhanced chemical vapour deposition (PECVD) are not suitable as their application will cause damage to the monolayer. Atomic layer deposition (ALD) is a mild and highly precise technique for thin film deposition, mainly used for depositing a School of Chemistry, Trinity College Dublin, Dublin, Ireland. E-mail: duesberg@tcd.ie; Tel: +353 1 896 3035 b CRANN & AMBER Institutes, Trinity College, Dublin, Ireland c Department of Chemistry and Pharmacy, Organische Chemie II, Friedrich-Alexander Universität Erlangen-Nürnberg, Henkestrasse 42, 91054 Erlangen, Germany This journal is © The Royal Society of Chemistry 2015 This reaction is thermodynamically highly favourable and works over a large range of temperatures with temperatures between 33 1C and 500 1C demonstrated, making it very reliable and common in the silicon and III–V semiconductor industries.10,11 However, in the initial step the TMA needs a surface hydroxyl group with which it reacts and the lack of such groups on the TMD’s basal plane makes starting the deposition non-trivial; a challenge also encountered with graphene.12–15 Using ozone instead of water may prove harmful to the oxidation-sensitive TMD layers, though there have been some recent successes.16 An initial, purely adsorptionbased deposition can be achieved but tends to be dependent on temperature and other factors like underlying electronic structure and is therefore often not entirely reproducible; it has been shown several times that studies may not reproduce results under apparently similar conditions.12,17,18 In this study we used single crystalline layers of MoS2 and WS2 grown via chemical vapour deposition (CVD) as previously demonstrated.19 The ALD on those layers was performed at the relatively low temperature of 80 1C with 27 cycles of TMA and H2O using 0.5 second pulses with 20 seconds purge time (60 sccm N2) and should yield B3 nm of Al2O3. Those samples were analysed with atomic force microscopy (AFM) to determine step heights and scanning Raman spectroscopy. Al2O3 as deposited by ALD does not exhibit a Raman signal but both MoS2 and WS2 have characteristic peaks in the region from 370 cm 1 to 420 cm 1. The MoS2 and WS2 layers have the characteristic triangular shape and consisted of mostly monolayer except for the flake centre which occasionally exhibited an onset of multilayer growth. This can be seen in the different contrast regions shown in Fig. 1a. An AFM scan (Fig. 1b), along with the line Chem. Commun., 2015, 51, 16553--16556 | 16553 View Article Online Published on 24 September 2015. Downloaded by Universitat Erlangen Nurnberg on 24/01/2017 11:18:53. Communication ChemComm Fig. 1 Representation of the MoS2 flakes used in the experiment: (a) optical image of flakes consisting of crystalline monolayers and some multilayers in the flake centres. (b) AFM image of a MoS2 flake; (c and d) show the line profiles along the marked lines in (b) with a step height of 0.7 nm. scans in Fig. 1c and d, shows the ideal step height of 0.7 nm per TMD layer. This proved to be very useful as it allowed for side-byside comparison of Al2O3 ALD on mono- and bilayer samples. In Fig. 2a we present an AFM image of a triangular WS2 flake on SiO2 after 3 nm Al2O3 deposition. It is apparent that the triangular flake (dark region) is lower than the surrounding substrate. This becomes better visible in the line profile across the flake as shown in Fig. 2b. The step height between the flake and substrate is B2.3 nm after deposition, perfectly corresponding to the difference of 3 nm Al2O3 deposition minus the flake height of 0.7 nm. Thus there is no deposition of Al2O3 on the monolayer. The double-layered centre of the flake increased in height to B3.7 nm with respect to the monolayer, perfectly corresponding to 0.7 nm flake height plus 3 nm Al2O3. At the edge of the monolayer Al2O3 was deposited which we attribute to dangling bonds and defects that are more reactive. This high selectivity in the deposition process was found in both both WS2 and MoS2 single crystals grown via CVD. This is to our knowledge the first observation of selective chemical behaviour between mono- and doublelayered or multilayered TMDs. We attribute this selectivity to two factors: firstly, our films must be extremely clean with no defects in the basal plane as otherwise the ALD would occur at those centres like at the flake edge. As the films came straight out of the oven and there was no transfer, no polymer residue could seed any ALD, something that may be responsible for the inconsistent results in the field so far. Secondly, TMDs undergo a significant change in electronic structure when going from multi-layered structures to single layers.20,21 There is a change from indirect bandgap of 1.2 eV and 1.3 eV to a direct bandgap of 1.9 eV and 2 eV for MoS2 and WS2, respectively.22,23 Hence the initially adsorption-based ALD may be so heavily influenced by this difference in underlying electronic structure that it is significantly different for monolayers in comparison to bulk. However, theoretical modelling is required 16554 | Chem. Commun., 2015, 51, 16553--16556 Fig. 2 (a) AFM analysis of TMDs after ALD. (a) Topography of a WS2 triangle after 27 cycles of TMA and H2O. The monolayer part of the triangle lies lower than the surrounding SiO2. (b) Line profile along the marked line in (a). The step height between monolayer WS2 and SiO2 is 2.3 nm and between monolayer WS2 and bilayer WS2 is 3.7 nm. Only the monolayer edges were covered with the expected 3 nm Al2O3. (c) Schematic representation of the surface structure in (a) as indicated by the AFM scan. (d) Topography of MoS2 triangles after 27 cycles of ALD after seeding with a perylene bisimide derivative. The flakes are higher than the surrounding substrate. (b) Line profile along the line marked in (d). The step height is 1.9 nm. (f) Schematic representation of the structure of (d) as indicated by the scan. to further investigate the underlying mechanism. This selectivity can be a significant advantage: covering all layers but the monolayer with a dielectric leaves only the material with the direct bandgap exposed. This provides a novel pathway to select monolayer regions and can be used for vertical device fabrication and selective chemistry. However, deposition of uniform dielectrics on monolayer TMDs is desired for passivation and electronic device fabrication. We therefore explore non-covalent functionalisation of TMD basal planes with molecules that contain –OH and –COOH units which can react with TMA and thereby seed the reaction.8,24 We use a perylene-based anchor unit on our TMDs.25 This perlyene, shown in Fig. 3a, has a large aromatic core that can non-covalently attach to the TMD and end-groups with hydroxy and carboxyl functionalities. It is similar to molecules that have been employed to seed these depositions on graphene by Alaboson et al. with the difference that our molecule has longer end-group chains with carboxylic acid groups.26 This kind of end-group has been shown to seed Al2O3 ALD from TMA and H2O.27 The perylene adsorbs very strongly on TMDs and is deposited from an aqueous pH7 buffer solution by simple dipcoating. Washing the sample with water after deposition leaves only a thin layer of perylenes on MoS2 and WS2. As can be seen in the AFM scan in Fig. 2d, ALD on a perylene-covered sample This journal is © The Royal Society of Chemistry 2015 View Article Online Published on 24 September 2015. Downloaded by Universitat Erlangen Nurnberg on 24/01/2017 11:18:53. ChemComm Communication Fig. 4 Mapped out Raman spectrographs of MoS2 flakes. Shown is the MoS2 A1g peak position whose shift indicates different layer numbers and doping levels. (a) Pristine MoS2, the central area of the flake shows multilayer signal; (b) the same MoS2 flake after exposure to 27 cycles of TMA and H2O, yielding 3 nm thick Al2O3; there is no appreciable change, indicating the exposure had no impact on electronic structure. (c) A different MoS2 flake after perylene deposition and exposure to ALD shows a slight blueshift, possibly indicating some p-doping. (d) Average spectra of the different regions of the flake in (a). Fig. 3 (a) Chemical structure of the perylene bisimide used for non-covalent functionalisation and ALD seeding. (b) Normalised Raman spectra of the samples at various stages: the SiO2 spectrum only shows the standard strong peak at 521 cm 1. Upon functionalisation with perylene bisimide a strong fluorescent background is observed. This remains after ALD. On the TMD the perylene shows several intense peaks in the region of 1300 cm 1 to 1600 cm 1 and around 2700 cm 1 while the MoS2 has its peaks at 4400 cm 1. All these peaks remain after ALD. leads to a perfectly homogeneous deposition of Al2O3 on all layers of MoS2. The topological cross-section in Fig. 2e reveals that the step height between MoS2 flake and SiO2 substrate is not 0.7 nm any more but around 1.9 nm. The additional height difference of 1.2 nm implies that the perylene adsorbs better on the TMDs than on SiO2 and that it may be in an upright conformation rather than laying flat. To investigate the effect of ALD on the TMD we utilised scanning Raman spectroscopy. The Raman spectra of both MoS2 and WS2 are very similar, with two peaks closely spaced in the region around 400 cm 1, shown for MoS2 in Fig. 4d. Change in molecular or electronic structure due to Al2O3 deposition should result in an alteration of those signals.28 The out-of-plane Raman active vibration of MoS2 at 405–410 cm 1 has previously been shown to shift with layer number and doping.29 A shift with layer number is expectedly observed as shown in Fig. 4a but no further shift upon Al2O3 deposition occurs as shown in Fig. 4b. Hence the electronic structure of the multilayer (coated) and monolayer (uncoated) TMD appears unperturbed upon Al2O3 deposition. Raman mapping of the TMD’s ‘‘A’’ peak (Fig. 4c) shows that the MoS2 and WS2 are not damaged by the deposition and retain their characteristics although there is a minor blueshift of B1 cm 1, potentially indicating p-doping.24,30 As this is at our spectrometer’s resolution limit it will require further investigation. This journal is © The Royal Society of Chemistry 2015 Investigation of the perylene bisimide’s Raman spectrum, shown in Fig. 3b, shows several features. On just SiO2 it has a very strong fluorescent background and does not show any features. After ALD this fluorescence remains, indicating the continued presence of the perylene. On TMDs it shows several peaks in the regions around 1500 cm 1 and 2700 cm 1 which remain after ALD, confirming its presence at all stages. This makes the interpretation of the 1.9 nm step height from Fig. 2e difficult and indicates different molecular orientations on SiO2 and MoS2. In summary, we demonstrated that there exists selective chemistry of monolayer TMD surfaces over their multilayer counterparts. We have shown that this can lead to the selective deposition of Al2O3 which is essentially a method to selectively mask all areas but monolayer of MoS2 and WS2. This will hopefully lead to a whole variety of layer number selective chemistry in future. Furthermore we demonstrated that an easily accessible, non-covalent functionalisation with perylene derivatives allows for reliable ALD of Al2O3 on monolayer TMDs without damage to their electronic integrity as observed by Raman spectroscopy. These findings are an important step forward toward integration of TMDs in real devices. The authors thank Science Foundation Ireland for funding under the grant SFI-Pica:Pi_10/IN.1/I3030. References 1 D. Jariwala, V. K. Sangwan, L. J. Lauhon, T. J. Marks and M. C. Hersam, ACS Nano, 2014, 8, 1102–1120. 2 A. C. Ferrari, F. Bonaccorso, V. Fal’ko, K. S. Novoselov, S. Roche, P. Boggild, S. Borini, F. H. L. Koppens, V. Palermo, N. Pugno, J. A. Garrido, R. Sordan, A. Bianco, L. Ballerini, M. Prato, E. Lidorikis, J. Kivioja, C. Marinelli, T. Ryhanen, A. Morpurgo, J. N. Coleman, Chem. Commun., 2015, 51, 16553--16556 | 16555 View Article Online Published on 24 September 2015. Downloaded by Universitat Erlangen Nurnberg on 24/01/2017 11:18:53. Communication 3 4 5 6 7 8 9 10 11 12 13 14 V. Nicolosi, L. Colombo, A. Fert, M. Garcia-Hernandez, A. Bachtold, G. F. Schneider, F. Guinea, C. Dekker, M. Barbone, Z. Sun, C. Galiotis, A. N. Grigorenko, G. Konstantatos, A. Kis, M. Katsnelson, L. Vandersypen, A. Loiseau, V. Morandi, D. Neumaier, E. Treossi, V. Pellegrini, M. Polini, A. Tredicucci, G. M. Williams, B. Hee Hong, J.-H. Ahn, J. Min Kim, H. Zirath, B. J. van Wees, H. van der Zant, L. Occhipinti, A. Di Matteo, I. A. Kinloch, T. Seyller, E. Quesnel, X. Feng, K. Teo, N. Rupesinghe, P. Hakonen, S. R. T. Neil, Q. Tannock, T. Lofwander and J. Kinaret, Nanoscale, 2015, 7, 4598–4810. H. R. Gutiérrez, N. Perea-López, A. L. Elı́as, A. Berkdemir, B. Wang, R. Lv, F. López-Urı́as, V. H. Crespi, H. Terrones and M. Terrones, Nano Lett., 2013, 13, 3447–3454. Y. Gong, J. Lin, X. Wang, G. Shi, S. Lei, Z. Lin, X. Zou, G. Ye, R. Vajtai, B. I. Yakobson, H. Terrones, M. Terrones, B. Tay, J. Lou, S. T. Pantelides, Z. Liu, W. Zhou and P. M. Ajayan, Nat. Mater., 2014, 13, 1135–1142. X. Duan, C. Wang, J. C. Shaw, R. Cheng, Y. Chen, H. Li, X. Wu, Y. Tang, Q. Zhang, A. Pan, J. Jiang, R. Yu, Y. Huang and X. Duan, Nat. Nanotechnol., 2014, 9, 1024–1030. A. M. van der Zande, P. Y. Huang, D. A. Chenet, T. C. Berkelbach, Y. You, G.-H. Lee, T. F. Heinz, D. R. Reichman, D. A. Muller and J. C. Hone, Nat. Mater., 2013, 12, 554–561. T. Roy, M. Tosun, J. S. Kang, A. B. Sachid, S. B. Desai, M. Hettick, C. C. Hu and A. Javey, ACS Nano, 2014, 8, 6259–6264. S. M. George, Chem. Rev., 2010, 110, 111–131. R. L. Puurunen, J. Appl. Phys., 2005, 97, 121301. M. D. Groner, F. H. Fabreguette, J. W. Elam and S. M. George, Chem. Mater., 2004, 16, 639–645. R. Matero, A. Rahtu, M. Ritala, M. Leskelä and T. Sajavaara, Thin Solid Films, 2000, 368, 1–7. H. Liu, K. Xu, X. J. Zhang and P. D. Ye, Appl. Phys. Lett., 2012, 100, 152115. B. Dlubak, P. R. Kidambi, R. S. Weatherup, S. Hofmann and J. Robertson, Appl. Phys. Lett., 2012, 100, 173113. J. Kim and S. Jandhyala, Thin Solid Films, 2013, 546, 85–93. 16556 | Chem. Commun., 2015, 51, 16553--16556 ChemComm 15 S. McDonnell, B. Brennan, A. Azcatl, N. Lu, H. Dong, C. Buie, J. Kim, C. L. Hinkle, M. J. Kim and R. M. Wallace, ACS Nano, 2013, 7, 10354–10361. 16 L. X. Cheng, X. Y. Qin, A. T. Lucero, A. Azcatl, J. Huang, R. M. Wallace, K. Cho and J. Kim, ACS Appl. Mater. Interfaces, 2014, 6, 11834–11838. 17 S. Brunauer, P. H. Emmett and E. Teller, J. Am. Chem. Soc., 1938, 60, 309–319. 18 J. Yang, S. Kim, W. Choi, S. H. Park, Y. Jung, M. H. Cho and H. Kim, ACS Appl. Mater. Interfaces, 2013, 5, 4739–4744. 19 M. O’Brien, N. McEvoy, T. Hallam, H.-Y. Kim, N. C. Berner, D. Hanlon, K. Lee, J. N. Coleman and G. S. Duesberg, Sci. Rep., 2014, 4, 7374. 20 K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805. 21 W. Zhao, Z. Ghorannevis, L. Chu, M. Toh, C. Kloc, P.-H. Tan and G. Eda, ACS Nano, 2013, 7, 791–797. 22 A. M. Goldberg, A. R. Beal, F. A. Lévy and E. A. Davis, Philos. Mag., 1975, 32, 367–378. 23 A. L. Elı́as, N. Perea-López, A. Castro-Beltrán, A. Berkdemir, R. Lv, S. Feng, A. D. Long, T. Hayashi, Y. A. Kim, M. Endo, H. R. Gutiérrez, N. R. Pradhan, L. Balicas, T. E. Mallouk, F. López-Urı́as, H. Terrones and M. Terrones, ACS Nano, 2013, 7, 5235–5242. 24 X. R. Wang, S. M. Tabakman and H. J. Dai, J. Am. Chem. Soc., 2008, 130, 8152–8153. 25 M. Marcia, P. Singh, F. Hauke, M. Maggini and A. Hirsch, Org. Biomol. Chem., 2014, 12, 7045–7058. 26 J. M. P. Alaboson, Q. H. Wang, J. D. Emery, A. L. Lipson, M. J. Bedzyk, J. W. Elam, M. J. Pellin and M. C. Hersam, ACS Nano, 2011, 5, 5223–5232. 27 M. Li, M. Dai and Y. J. Chabal, Langmuir, 2009, 25, 1911–1914. 28 B. Chakraborty, A. Bera, D. V. S. Muthu, S. Bhowmick, U. V. Waghmare and A. K. Sood, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 85, 161403. 29 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. 30 Y. Shi, J.-K. Huang, L. Jin, Y.-T. Hsu, S. F. Yu, L.-J. Li and H. Y. Yang, Sci. Rep., 2013, 3, 1839. This journal is © The Royal Society of Chemistry 2015