J Mater Sci: Mater Electron (2009) 20:223–229 DOI 10.1007/s10854-008-9706-1 Dispersion of single-walled carbon nanotubes in aqueous and organic solvents through a polymer wrapping functionalization S. Manivannan Æ Il Ok Jeong Æ Je Hwang Ryu Æ Chang Seok Lee Æ Ki Seo Kim Æ Jin Jang Æ Kyu Chang Park Received: 25 November 2007 / Accepted: 14 March 2008 / Published online: 27 March 2008 Ó Springer Science+Business Media, LLC 2008 Abstract The processing and application of single-walled carbon nanotubes (SWNTs) is limited by their purity and dispersion in most common solvents. Non-covalent polymer wrapping functionalization of SWNTs provides an excellent route to improve their property and thus the dispersion in aqueous and organic solvents occurred simultaneously. This paper reports the purification and a methodology for obtaining dispersion of SWNTs, obtained by arc-discharge, using a biocompatible water soluble polymer which, presumably, wraps around nanotubes. Scanning electron microscope (SEM) and high resolution transmission electron microscope (HR-TEM) were employed to examine the dispersion. UVVis-NIR spectroscopy further substantiates the presence of dispersed SWNTs in aqueous solution. Raman spectroscopy reveals the impact of chemical and ultrasonic processing on the structural order of the nanotubes. No trace for structural damage has been observed from the Raman studies and so, our treatment is considered as a physical rather chemical process. The polymer treated tubes become more hydrophilic which was confirmed by contact angle measurement. 1 Introduction Single-walled carbon nanotubes (SWNTs) have a great deal of potential for nanoscale device applications S. Manivannan
I. O. Jeong
J. H. Ryu
C. S. Lee
J. Jang
K. C. Park (&) Department of Information Display and Advanced Display Research Center, Kyung Hee University, Seoul 130-701, Korea e-mail: kyupark@khu.ac.kr K. S. Kim Department of Physics, Kyung Hee University, Seoul 130-701, Korea including large area transparent electrodes, [1, 2] solar cells, [3] flat and flexible displays [4] owing to their novel electronic, structural and mechanical properties. On the other hand, a number of challenges must be overcome before SWNTs can be exploited for most of these envisioned applications. One of the crucial problems in realizing these applications is that, while the majority require highly pure, individually dispersed nanotubes, SWNTs tend to aggregate forming bundles due to their long aspect ratios ([1,000) and the strong anisotropic interactions between them (0.5 eV nm-1) [5]. The unavailability of methods to produce individually dispersed highly pure SWNTs in aqueous and organic mixtures by a unique treatment further limited the applications. The mixture of solvents containing individual tubes will bring solution to many problems in the device fabrication processes with enhanced performance at where different parameters such as viscosity, volatility has to be taken into account at different concentrations/ratios. In addition, the SWNTs in mixed solvents have potential application for surface treatments with paints and polymers. There have been a number of attempts to develop an effective method to debundle and discretely disperse the SWNTs. These strategies have involved covalent chemical functionalization of the tubes’ sidewalls and scission of tubes by mechanical milling [6]. It is to be noted that a number of surfactants and polymers have been identified that create dispersions of SWNTs in various media [7–12]. Generally, a medium for the dispersion of SWNTs has to be capable of both wetting the hydrophobic tubes’ surfaces and then modifying these surfaces to decrease the interaction between tubes. Previously, high concentration surfactant solutions [13] and concentrated superacid solutions [14] have been found effective in making tube suspensions since the tubes’ surfaces can be wetted and 123 224 J Mater Sci: Mater Electron (2009) 20:223–229 charged by the adsorption of surfactant molecules or protonation of SWNTs in superacids. Both methods can damage the nanotubes’ structure and the treated tubes are only dispersible either in aqueous media or in organic solutions. Among the widely used polymers for this purpose, polyvinylpyrrolidone (PVP) has been identified as one the best for wrapping functionalization of the tubes and already used to disperse [15, 16] or stabilize [17] SWNTs in aqueous environment. Recently, PVP has also been used to stabilize the debundled SWNTs in N-methyl-2-pyrrolidone (NMP) [18]. However, to our knowledge, dispersion of SWNTs in both aqueous and organic media by a unique treatment has not yet been attained so far. We use sodium hydroxide (NaOH)–ethanol (EtOH) solution to remove the iron catalyst present in the commercial tubes and thus the purity of SWNTs was enhanced. The use of NaOH to remove debris from CNTs and carbon based fibers is already known and exploited for wetting of hydrophobic nanotubes [19]. In this paper, we present results from a systematic study on the effects of NaOH treatment and a biocompatible, water soluble polymer wrapping functionalization of arcderived bundles of SWNTs to generate solutions containing individual SWNTs. The impact of our chemical and ultrasonic processing on the nanotubes was examined by Raman and UV-Vis-NIR spectroscopy. Electron microscopic techniques were used to study the dispersion of the tubes. In addition, our treatment yields more hydrophilic nanotubes which gains importance in biological and electronic applications. All the results are presented and discussed here. 2 Experimental details Arc-discharge SWNTs purchased from Iljin Nanotech (ASP-100F) was used for the present investigation. Though the commercial sample reaches purity of 90 volume percent after removing the majority of amorphous carbon, carbon nanoparticles and iron metal catalysts, it contains still a considerable amount of metal catalysts and amorphous carbon as shown in Fig. 1. The as delivered material was found to contain bundles typically with *20 s of SWNTs with a range of nanotube diameter (dt) *1.2– 1.4 nm, as determined by high resolution transmission electron microscope (HR-TEM) and Raman-active radial breathing modes (RBM). We followed a new process to remove the amorphous carbon and metal catalysts without damaging the tubes. In a typical experiment, 20 mg of SWNTs were immersed in 50 mL of absolute ethanol (99.99% purity) and ultrasound was applied for 30 min (9 Watts) with an interruption of 10 sec for every 30 sec using an ultrasonic 123 Fig. 1 HR-TEM images of raw SWNTs on Cu grid (a) and NaOH treated SWNTs on carbon grid (b) obtained by dispersing them in ethanol. The treated tubes seem to be consisting of large bundles with a considerable amount of small bundles. The thick and dark band on both sides in (b) indicates the carbon grid probe (Sonics Vibra, model VCX 750). The solution was then filtered by a 3 lm membrane filter (TSTP, isopore) and a thick mat of SWNTs film (bucky paper) was formed and peeled off from the filter after 3 h drying with vacuum pump. The film was kept in a furnace at 350 °C in an atmospheric pressure for 3 h and thus the amorphous carbon was removed through oxidation. The ultrasonic treatment gives the separation of amorphous carbon (mostly attached to the metal catalyst) from the sidewalls of the tubes and thus the separation protects the excess heating of tubes during dry oxidation. The oxidation temperature was so chosen as 350 °C from TG analysis (Fig. 2) to avoid the possible oxidation of tubes. The J Mater Sci: Mater Electron (2009) 20:223–229 225 Fig. 2 TGA curve of SWNTs after the ultrasonic treatment and washing with ethanol presence of iron metal catalyst was subsequently removed using 50 mL of saturated solution of NaOH in ethanol and DI water (18.2 MX cm) mixture in the volume ratio 80:20 for EtOH and DI water, respectively. The dry oxidized bucky paper (BP) was immersed in the prepared base solution which seems to be homogeneous and black slurry was formed after 4 h of ultrasonic treatment. The slurry was filtered (3 lm TSTP membrane filter), washed several times (until the mother liquid reaches the pH = 7) with ethanol–water (3:1 volume ratio) mixture to remove NaOH, iron catalyst and finally washed with absolute ethanol to remove the water. After washing, the SWNTs on filter were dried using vacuum pump for 3 h at room temperature and thus the BP of pure nanotubes was formed. The filtered mother liquid appears light yellow in colour suggests the presence of iron catalyst. We use polyvinylpyrrolidone (PVP) (Sigma-Aldrich, typical Mw 29,000) to debundle and disperse the NaOH treated SWNT bundles in aqueous solution. In this process, 10 mg of SWNTs, 2 mg of PVP were mixed in 50 mL of DI water. Ultrasonic dispersion was carried out for 4 h (8 Watts) with an interruption of 10 sec for every 30 sec. 3 Results and discussion The purity of SWNTs was confirmed by Raman (Fig. 3), TEM (Fig. 1b) and the change in sheet resistance of the formed BPs’ before and after the treatment. Interestingly, the sheet resistance of the BP after the NaOH treatment was measured as 10–12 X/h (four point probe) which was four times lower than that of the sheet resistance (40–43 X/h) before the treatment for the same quantity of SWNTs. This clearly indicates the enhancement in purity of NaOH treated tubes. The TEM (Fig. 1b) and SEM (Fig. 4) images of the treated tubes further illustrate the enrichment of the purity. Fig. 3 Raman spectra of SWNTs before and after the treatments; (a) D and G-band, (b) RBM region. The inlay of (a) shows a photo image of a uniform, 4.7 cm diameter, polymer treated and dispersed carbon nanotube film (bucky paper) on a 200 nm GTTP membrane filter (black region) We believe that the removal of metal catalyst was due to the significant wetting by the ethanol–NaOH–water solution. It is known that the alcohol–NaOH solutions are generally used to clean grease and fatty or oily surfaces in laboratory by effective wetting of highly hydrophobic surfaces. Recently, Verdejo et al. [19] have used NaOH to remove debris from multi walled carbon nanotubes (MWNTs) and carbon based fibers. The SEM image (Fig. 4) reveals the average bundle size as *50–100 nm, become larger (*3– 5 times) after the NaOH treatment than the pretreated 123 226 J Mater Sci: Mater Electron (2009) 20:223–229 Fig. 4 SEM image of NaOH treated SWNTs aggregates densely showing the average bundle size as *50–100 nm sample. No remarkable increment was noticed on disorder peak at *1,350 cm-1 in Raman spectra (514.5 nm excitation) (Fig. 3) of treated SWNTs that confirms the improved purity without damaging the tube. However, the NaOH treatment leads the tubes become more hydrophobic and so difficult to disperse in aqueous solution. The hydrophobic nature of tubes was confirmed through contact angle measurements (Fig. 5a, b). After the PVP treatment for dispersion, the ultrasonic treated SWNTs solution appears black and homogeneous without any precipitation even after 3 weeks. During the ultrasonic process, small air bubbles were released from the polymer added solution indicating the bundles become softened and began to break. The debundling is being made due to effective wetting by polymer wrapping on the tubes. We believe that the possible mechanism behind the dispersion process is the formation of hydrogen bonds between polymer and water molecules where SWNTs are wrapped by PVP through van der Waals forces. Figure 6 illustrates the hydrogen bonding (C=O


H) between C=O of PVP and hydroxyl group (OH) of water. The hydrogen bonding interaction between C=O and water molecule leads to effective wetting of hydrophobic tubes, causing reduction of anisotropic interaction between the tubes and thus dispersion occurs. The absence of electronegative atom in SWNTs overruled the possibility of forming Hbonds with PVP and so wrapping functionalization occurs through van der Waals interaction. Here, the PVP acts as a hydrophilic carrier for debundling and dispersion in water of the resulting system. It is known that in the closed-shell interactions (hydrogen-bonds and van der Waal’s interactions) the charge density is contracted towards each of the interacting nuclei and the electronic charge is depleted in the interatomic surface. Thus, the effective wetting process reduces the anisotropic interactions between the tubes and 123 Fig. 5 Contact angle measurement images of SWNTs showing variation of angles between the water droplets (black portion) and the carbon nanotube films on glass before and after the treatments. The measured contact angles for raw, NaOH treated and PVP treated SWNTs are 50°, 72° and 26°, respectively hence the debundling and dispersion occurs. This phenomenon is similar to the solid dispersion systems of drugs into inert polymer matrices, aiming at the optimization of the dissolution rate of poorly soluble drugs, where the drug molecules bind to the polymer by means of van der Waals interactions or hydrogen bonds. This idea can further be extended to other hydrophilic carriers like polyethyleneglycol which is also widely used in solid dispersion systems of drugs [20]. J Mater Sci: Mater Electron (2009) 20:223–229 A bucky paper was formed from the dispersed solution on a membrane filter (200 nm pore size, GTTP) after washing with DI water and absolute ethanol. Then, the treated SWNTs were re-dispersed in DI water, EtOH, N-methyl-2-pyrrolidone (NMP) and 1,2-dichloroethane (DCE) by transferring them from the filter after a few minutes of gentle sonication. This was done before drying the tubes to avoid the possible formation of small bundles. More interestingly, it was found that the SWNTs redispersed in aqueous and organic solutions were stable with a little sediment after 1 week (Fig. 7). Figure 8 shows the UV-Vis-NIR absorption spectra (Scinco, Model S-4100) of commercial SWNTs, after NaOH and PVP treatments suspended in DI water. The PVP treated tubes redispersed in DI water was used for this experiment after a week. The presence of resolved peaks from polymer treated tubes solution, indicating the existence of nanotubes of a range of diameters as individual and/or very small bundles, as well as nanotubes whose individual transition frequencies are Fig. 6 A model for the hydrogen bonding (C=O


H) between PVP and water Fig. 7 Vials containing redispersed SWNTs 0.1 mg mL-1 in DI water (a), 0.15 mg ml-1 in NMP (b), EtOH (c), and DCE (d) imaged after 1 week. A small quantity of black powder was precipitated at the bottom with the remaining supernatant stable and homogeneous 227 Fig. 8 UV-Vis-NIR absorption spectra of 0.1 mg mL-1 SWNTs suspended in DI water after a gentle ultrasonic process; (a) commercial SWNTs (b) NaOH treated SWNTs (c) PVP treated SWNTs affected by intermolecular interactions. The broadening of peaks arising from small bundles of nanotubes formed after a week with significant amount of individual tubes suspended on the top of the solution. The quality of the dispersion in DI water was examined through HR-TEM (Fig. 9a) by dropping the solution on a ultra thin lacey carbon film (\3 nm thick) supported by a mesh copper grid and the average bundle size was measured as *3–5 nm. The aggregation of tubes in the solution after a week is due to the anisotropic interactions between the sidewalls of the tubes and the PVP wrapped around the tubes is weekly interacted with the SWNTs. For such a system, after centrifugation, UV–Vis absorption spectra of nanotubes contained in the supernatant were often found to exhibit sharper peaks, resulting from interband transitions for isolated tubes and a decrease in intermolecular interactions [13, 21]. However, the appearance of resolved peaks in the PVP treated tubes compared to NaOH and raw SWNTs, is a good sign for the occurrence of dispersion in DI water. Further, the PVP treated tubes become hydrophilic (contact angle = 26°) (Fig. 5c) than the commercial sample (contact angle = 50°) (Fig. 5a) and so the dispersion in aqueous solution appeared. HR-TEM image of the treated nanotubes redispersed in EtOH (Fig. 9b) shows that the average diameter/size of the individual tubes/bundles was in the range 1.2–3 nm, revealing that the treated tubes could be dispersed as individuals in the organic solvents. Besides, the dispersion in mixture of solvents such as DI water–ethanol and DI water–ethylene glycol was successfully made. These mixtures can be used as ink for inkjet printing to prepare nanotube films. Furthermore, the sheet resistance of the dried BP (inlay of Fig. 3a) after the polymer treatment was measured as 2 X/h. This value is nearly 5 times lower than the NaOH treated sample and nearly 20 times lower than the commercial sample. This 123 228 J Mater Sci: Mater Electron (2009) 20:223–229 functionalization on the tubes. The observed Raman spectra of PVP treated tubes in the present investigation (Fig. 3b) look like a Gaussian diameter distribution centred on a mean diameter of 1.295 nm. This is the mere indication that our sample composed of all the possible (n, m) nanotubes (real samples) which usually exhibit a Gaussian diameter distribution around some mean diameter [22]. 4 Conclusion In summary, our results show that the details of the oxidation and metal digestion by which the amorphous carbon and metal catalyst were removed. This is a crucial step for debundling and dispersion. The polymer wrapping functionalization significantly impacts the tubes character and the SWNTs become hydrophilic after the treatment. This is perhaps a little surprising, as the treated tubes are dispersed in aqueous and many organic solvents as well as mixture of solvents with no evidence for damage. The hydrophilic nature of the treated tubes is the added advantage of this method and is expected to play a vital role in biological and electronic applications. The dispersed SWNTs in mixed solvents have potential for surface treatments with paints and polymers. Above all, this process is less expensive, simple and potentially scalable. Acknowledgements This work was supported by Ministry of Education through BK21 program and Seoul Research and Business Development Program (Grant No.10853). References Fig. 9 HR-TEM images of debundled SWNTs redispersed in DI water (a) and ethanol (b) after the NaOH and PVP treatments low resistance, highly uniform and homogeneous BP can be used to form films on glass or other flexible substrates using imprint techniques. The reduction in the sheet resistance is a good sign of dispersion of tubes in the BP. Room temperature Raman spectra of SWNTs before and after the NaOH and polymer treatments are compared in Fig. 3. From the overall observation, there is no remarkable change in radial breathing mode (RBM) as well as in single high-frequency band or graphite band (G-band). The observation of characteristic multi peak features in G-band region around 1,570 and 1,590 cm-1 provides a signature of SWNTs (Fig. 3a). Similarly, no significant increase in the intensity of disorder-induced band (D-band) was observed. The spectroscopic evidence indicates the dispersion is a physical rather than a chemical process since there is no evidence of any wall disorder or covalent chemical 123 1. Z. Wu, Z. Chen, X. Du, J.M. Logan, J. Sippel, M. Nikolou, K. Kamaras, J.R. Reynolds, D.B. Tanner, A.F. Hebard, A.G. Rinzler, Science 305, 1273 (2004) Medline. doi:10.1126/science. 1101243 2. B.B. Parekh, G. Fanchini, G. Eda, M. Chhowalla, Appl. Phys. Lett. 90, 121913 (2007) doi:10.1063/1.2715027 3. J.H. Lehman, C. Engtrakul, T. Gennett, A.C. Dillon, Appl. Opt. 44, 483 (2005) Medline. doi: 10.1364/AO.44.000483 4. A.G. Rinzler, J.H. Hafner, P. Nikolaev, K. Lou, S.G. Kim, K. Tomanek, D. Colbert, R.E. Smalley, Science 269, 1550 (1995) Medline. doi: 10.1126/science.269.5230.1550 5. Q. Li, I.A. Kinloch, A.H. Windle, Chem. Commun. 26, 3283 (2005) Medline. doi: 10.1039/b419039d 6. J. Chen, M. Dyer, M. Yu, J. Am. Chem. Soc. 123, 6201 (2001) Medline. doi: 10.1021/ja015766t 7. S. Banerjee, S.S. Wong, Nano Lett. 4, 1445 (2004) doi:10.1021/ nl049261n 8. S.M. Keogh, T.G. Hedderman, E. Gregan, G. Farrell, G. Chambers, H. Byne, J. Phys. Chem. B 108, 6233 (2004) doi:10.1021/jp0369387 9. A.B. Dalton, C. Stephan, J.N. Coleman, B. McCarthy, P.M. Ajayan, S. Lefrant, P. Bernier, W.J. Blau, H.J. Byrne, J. Phys. Chem. B 104, 10012 (2000) doi:10.1021/jp002857o 10. A. Star, J.F. Stoddart, D. Steuerman, M. Diehl, A. Boukai, E.W. Wong, X. Yang, S.W. Chung, H. Choi, J.R. Heath, Angew. J Mater Sci: Mater Electron (2009) 20:223–229 11. 12. 13. 14. 15. 16. Chem. Int. Ed. 40, 1721 (2001) doi:10.1002/1521-3773 (20010504)40:9\1721::AID-ANIE17210[3.0.CO;2-F M.J. Connell, P. Boul, C.B. Huffman, R.E. Smalley, Chem. Phys. Lett. 342, 265 (2001) doi:10.1016/S0009-2614(01)00490-0 A. Ortiz-Acevedo, H. Xie, V. Zorbas, W.M. Sampson, A.B. Dalton, R.H. Baughman, R.K. Draper, I.H. Musselman, G.R. Dieckmann, J. Am. Chem. Soc. 127, 9512 (2005) Medline. doi: 10.1021/ ja050507f M.F. Islam, E. Rojas, A.T. Johnson, A.G. Yodh, Nano Lett. 3, 269 (2003) doi:10.1021/nl025924u S. Ramesh, L.M. Ericson, V.A. Davis, R.K. Saini, C. Kittrell, M. Pasquali, W.E. Billups, W.W. Adams, R.H. Hauge, R.E. Smalley, J. Phys. Chem. B 108, 8794 (2004) doi:10.1021/ jp036971t M.J. O’Connell, P. Boul, L.M. Ericson, C. Huffman, Y. Wang, E. Haroz, C. Kuper, J. Tour, K.D. Ausman, R.E. Smalley, Chem. Phys. Lett. 342, 265 (2001) doi:10.1016/S0009-2614(01)00490-0 X. Zhang, T. Liu, T.V. Sreekumar, S. Kumar, V.C. Moore, R.H. Hauge, R.E. Smalley, Nano Lett. 3, 1285 (2003) doi: 10.1021/nl034336t 229 17. M.J. O’Connell, S.M. Bachilo, C.B. Huffman, V.C. Moore, M.S. Strano, E.H. Haroz, K.L. Rialon, P.J. Boul, W.H. Noon, C. Kittrell, J. Ma, R.H. Hauge, R.B. Weisman, R.E. Smalley, Science 297, 593 (2002) Medline. doi:10.1126/science.1072631 18. T. Hasan, V. Scardaci, P.H. Tan, A.G. Rozhin, W.I. Milne, A.C. Ferrari, J. Phys. Chem. C. 111, 12594 (2007) doi:10.1021/ jp0723012 19. R. Verdejo, S. Lamoriniere, B. Cottam, A. Bismarck, M. Shaffer, Chem. Commun. 513 (2007) Medline. doi:10.1039/b611930a 20. V.I. Teberekidis, M.P. Sigalas, J. Mol. Struct.: THEOCHEM. 803, 29 (2007) doi:10.1016/j.theochem.2006.09.016 21. V.C. Moore, M.S. Strano, E.H. Haroz, R.H. Hauge, R.E. Smalley, J. Schmidt, Y. Talmon, Nano Lett. 3, 1379 (2003) doi: 10.1021/nl034524j 22. A. Jorio, M.A. Pimenta, A.G. Souza Filho, R. Saito, G. Dresselhaus, M.S. Dresselhaus, New J. Phys. 5, 139.1 (2003) 123