NANO LETTERS
Solubilization of Oxidized Single-Walled Carbon Nanotubes in Organic and Aqueous Solvents through Organic Derivatization
2002 Vol. 2, No. 11 1215-1218
Michael G. C. Kahn,† Sarbajit Banerjee,† and Stanislaus S. Wong*,†,‡ Department of Chemistry, State UniVersity of New York at Stony Brook, Stony Brook, New York 11794, and Materials and Chemical Sciences Department, BrookhaVen National Laboratory, Building 480, Upton, New York 11973 Received August 15, 2002; Revised Manuscript Received September 10, 2002
ABSTRACT The solubilization of oxidized carbon nanotubes has been achieved through derivatization using a functionalized organic crown ether. The resultant synthesized adduct yielded concentrations of dissolved nanotubes on the order of ∼1 g/L in water as well as in methanol, according to optical measurements. The nanotube−crown ether adduct can be readily redissolved in 10 different organic solvents at substantially high concentrations. Characterization of these solubilized adducts was performed with 1H NMR spectroscopy; 7Li NMR was also used to examine the ability of the crown ether’s macrocyclic ring to bind Li+ ions. The solutions were further analyzed using UV−visible, photoluminescence, and FT-IR spectroscopies and were structurally characterized using atomic force microscopy (AFM) and transmission electron microscopy (TEM). Adduct formation likely results from a noncovalent chemical interaction between carboxylic groups on the oxidized tubes and amine moieties attached to the side chain of the crown ether derivative.
Introduction. Understanding the chemistry of single-walled carbon nanotubes (SWNTs)1 is critical to the rational manipulation of their unique structure-dependent electronic and mechanical properties.2 The ability3-6 to disperse, solubilize, and separate carbon nanotubes would not only aid in their purification but also open up new prospects in aligning and forming molecular devices7 as well as in generating nanoscale architectures.8 Acid-shortened SWNTs9 have been rendered soluble in common organic solvents with thionyl chloride and octadecylamine.10,11 Solubilization has also been achieved by attaching tubes to highly soluble poly(propionylethylenimine-co-ethylenimine)12 as well as by noncovalent functionalization of the carboxylic groups present in purified SWNTs.4 Sidewall functionalization with fluorine and alkanes also appears to render tubes soluble in a number of different organic solvents including chloroform and methylene chloride.13,14 Recently, water solubilization has been achieved by functionalization of SWNTs with glucosamine and gum arabic.15,16 We have previously demonstrated that nanotubes can be solubilized through coordination using metal-containing complexes.17,18 Herein, we extend and optimize SWNT solubilization through * Corresponding author. E-mail:
[email protected]; sswong@ bnl.gov. Phone: 631-632-1703; 631-344-3178. † State University of New York at Stony Brook. ‡ Brookhaven National Laboratory. 10.1021/nl025755d CCC: $22.00 Published on Web 10/02/2002
© 2002 American Chemical Society
organic derivatization, combining a facile synthetic approach with the ability to complex cations in the resultant adduct. In this letter, we report on a straightforward means of solubilizing oxidized SWNTs in a variety of organic solvents as well as in aqueous media through derivatization with functionalized crown ethers. Why crown ethers? Crown ethers, because of their hydrophobic exteriors and hydrophilic cores, are typically attached to larger globular organic molecules to facilitate dissolution in solvents of varying polarity. We exploit this chemistry in an analogous fashion with SWNTs. Derivatized adducts were characterized by 1H NMR (250 MHz Bruker multinuclear FT-NMR), 7Li NMR (700 MHz Bruker multinuclear FT-NMR), FT-IR (Mattson Galaxy Series 3000 FTIR), UV-visible (ThermoSpectronics UV1), and fluorescence (Jobin Yvon Spex) spectroscopy. Heights and morphological data were obtained using AFM (Digital Instruments Multimode Nanoscope IIIa) and TEM (Philips CM12). These findings should broaden the biochemical and biomedical applications of nanotubes as well as allow for more facile photophysical analyses and chemical manipulation. Moreover, because the presence of ionic, attractive electrostatic forces likely enables the solubilization of the synthesized adduct, this work has implications for rationally designing biocompatibility in artificial nanostructures. Experimental Section. Raw SWNTs (HiPco: average diameters of 0.7 to 1.1 nm) were purified by a mild nitric
Table 1. Concentrations of SWNTs in Solubilized Adducts for Selected Solventsa
a
solvent
concentration of SWNTs in solubilized adduct (in mg/L)
THF acetone DMSO ODCB DMF water methanol
270 280 290 300 610 1100 1600
Values are within a (10% error range.
acid reflux followed by filtration using a polycarbonate membrane with a pore diameter of 0.2 µm. This process generates surface functionalities, particularly carboxylic acids at nanotube ends and sidewall defect sites. The bucky paper mat thereby obtained was then redispersed in 12.1 N HCl and briefly sonicated to remove the metal catalyst. Upon the second filtration, the precipitate was washed thoroughly with large amounts of deionized water and placed in a vacuum oven at 180 °C. To create the derivatized adduct, the purified bucky paper was initially ground up with a 3:1 mass excess of 2-aminomethyl-18-crown-6 ether (Aldrich) (CE), a clear yellow, viscous liquid, to form a black paste; CE readily moistened and permeated the bucky paper. Next, 1 mL of distilled deionized water was added. The mixture was swirled, sonicated for 1 s, and then allowed to stand for 1 h, after which an additional 9 mL of distilled water was added followed by vigorous stirring. The resultant mixture was filtered by a polycarbonate membrane to separate out unfunctionalized or partially functionalized SWNT precipitate, yielding a dark-brown solution, which could be further dried by heating under an Ar flow to form a black paste, the SWNT-CE adduct. It was found that the paste could then be dissolved in many organic solvents such as methanol, ethanol, 2-propanol, acetone, o-dichlorobenzene (ODCB), dimethylformamide (DMF), tetrahydrofuran (THF), dimethyl sulfoxide (DMSO), ethyl acetate, and benzene. Excess, unreacted CE could be removed from the adduct by washing with diethyl ether. Prior to characterization, the solutions were passed through a column packed with glass wool to remove excess solid particulate matter in order to obtain an optically clear solution. The resultant solutions were visually nonscattering and were appropriately diluted for optical measurements. The optical characteristics of SWNTs in solution were monitored19 by the absorbance at 500 nm; the derivatized crown ether does not absorb beyond that value. Quantitative concentrations were calculated using optical absorption data fitted to a Beer-Lambert plot. Representative solubility values are listed in Table 1. Results and Discussion. Stable solutions of SWNTs could not be formed with simply neat 15-crown-5, 18-crown-6, and dibenzo-30-crown-10 ethers. Hence, we theorized that a noncovalent interaction was needed to attach the crown ether to the nanotube. We used CE, a crown ether containing 1216
Figure 1. NMR spectroscopy of functionalized adducts. 1H NMR data: (a) CE (2-(aminomethyl)-18-crown-6 ether) in deuterated methanol. (b) SWCNT-CE adduct. Inset: 7Li NMR data. (i) CELi+ complex. (ii). SWNT-CE-Li+ complex adduct. (iii) LiCl standard.
an amino functionality attached to a pendant methylene side chain on the macrocyclic ring. The amino group of 2-aminomethyl-18-crown-6 likely interacts with oxygenated groups, particularly the carboxylic acid sites, at the ends of the purified tubes as well as at oxidized defect sites scattered along the sidewalls. Sterically, it follows that the crown ether’s macrocyclic ring should be expected to dangle from the SWNT. Evidence that it does comes from 1H NMR data (Figure 1). The protons on the macrocyclic ring contribute to a broad resonance in the 3.6 to 3.8 ppm range, whereas the methylene side chain resonances appear in the 2.6 ppm region, where a pair of quartets is evident. The macrocyclic proton resonances remain strong in the SWNT-CE adduct, whereas the methylene side chain resonances are widened to the point of almost disappearing. This signal attenuation for protons in functionalizing moieties in close physical proximity to SWNTs has been previously reported.3,10,20 Indeed, the observed broadening of these latter resonances occurs because of localization of the methylene side chains onto the SWNTs coupled with slow tumbling of the adduct in solution, preventing rotational averaging, as well as the presence of large diamagnetic ring currents in the tubes.3,10 To further ascertain the conformational nature of SWNTCE bonding within the adduct, the adduct was incubated with a solution of lithium chloride to observe the Li cation movement; 7Li NMR was performed on the adduct mixture in MeOH. In a solution of pure crown ether, because of the fast exchange kinetics21 between the Li+ complexed within the cavity of the crown ether and the free, solvated cation, only one 7Li peak is visible. In a solution of the SWNTCE adduct, a single narrow Li peak was observed, similarly implying the presence of fast exchange between Li cations Nano Lett., Vol. 2, No. 11, 2002
We have also observed that the emission peak is wider in the SWNT-CE adduct, an effect likely arising from slower tumbling of the larger nanotubes in solution, which would thereby slow self-quenching of the fluorescence due to molecular motion.22 These results can be taken as further evidence of SWNT-CE adduct formation.
Figure 2. Optical characterization of functionalized adducts. (a) Mid-IR of the SWNT-CE adduct. (b) Emission spectra of CE (2(aminomethyl)-18-crown-6) (red) and functionalized SWNT-CE adduct (blue). (c) Excitation spectrum of SWNT-CE adduct.
residing in the crown ether cavities within the SWNT-CE adduct and those free in solution. Whereas exchange between CE and the oxygenated sites on the SWNTs may be possible, the fact that only one Li peak was observed is highly suggestive of our hypothesis, supported by 1H NMR, that the macrocyclic ring is tethered to the SWNT through an interaction involving the aminomethyl side chain and that the ring freely dangles. Thus, the crown ether can readily complex with the Li cation. The proposed configuration in which CE is chemically bonded to the SWNT contrasts with a scenario in which the organic crown ether molecules sheath or wrap around SWNTs in an analogous manner to polymers. Such a scenario would have rendered the macrocyclic ring spatially inaccessible for Li cation complexation. Optically, the presence of a sharp peak at 1105 cm-1 in the mid-IR range (Figure 2a) for dried SWNT-CE adducts indicates C-O-C ether bonds originating from the crown ether; this band is slightly shifted from the ether peaks observed in free CE. Fluorescence spectra (Figure 2b, c) show that the functionalized adduct fluoresces strongly with an emission maximum at 455 nm. The excitation spectra suggest that the fluorescing moiety is the crown ether chromophore because the free unreacted crown ether absorbs in the 360- to 370-nm region and fluoresces with an emission maximum near 408 nm. Indeed, the fact that the fluorescence signal is not quenched but remains undiminished is further evidence that the macrocyclic ring freely dangles from the tubes as opposed to wrapping directly around the SWNTs. One implication is that the electronic structure of the chromophore is unlikely to be strongly coupled with that of the SWNT itself, suggesting that there is no disruption of the π conjugation within the nanotube electronic structure. Nano Lett., Vol. 2, No. 11, 2002
Hence, the SWNT-CE adduct likely arises from a zwitterionic interaction between a protonated amine on CE and an oxyanion from a carboxylic acid group, creating a COO-NH3+ ionic bond. This option may more reasonably account for the high yield of SWNTs in solution; in fact, the ease with which the adduct can be precipitated out upon the addition of high ionic strength solution suggests that the tubes are charged to some extent in solution and that solubility occurs to a certain degree by means of electric double-layer stabilization.4,18 Raw SWNTs cannot be solubilized, suggesting that the oxygenated moieties present on the surfaces of purified SWNTs are necessary to tether the crown ether and that they are present in significant quantities. We note that because of the mild conditions used for the reaction, it is unlikely that amide bond formation plays any significant role in the adduct formation. We have operationally defined4,15 solubilization thus far as the absence of precipitation upon prolonged standing (in our case, as long as several months) of a visually nonscattering solution. Although our adducts are clearly soluble by this set of criteria, whether solvation is associated with the exfoliation of nanotube ropes into individual tubes can be ascertained by means of microscopy. Even with the most dilute samples, upon removal of solvent, there was a degree of tube clumping and coalescence observed in the sample. Hence, in this system, solubilization does not completely exfoliate SWNT bundles to yield individual tubes. Indeed, individual SWNTs were often very difficult to observe. Compared with the raw tubes, the AFM and TEM images (Figure 3; Supporting Information) of the adducts show several bundles of SWNT aggregates between diameters of ∼30 and ∼200 nm. The lengths of these adduct bundles are shorter than those of the raw SWNT material, likely because of the etching effects of HNO3.9 The observed aggregation, upon solvent removal, may be a result of increased noncovalent (such as van der Waals) interactions between the dangling macrocycles on neighboring adducts. We believe that the macrocycles may preferentially orient and align 1217
also acknowledged for a Grant-in-Aid of Research. We also thank Dr. James Marecek and Dr. James Quinn for their guidance with the NMR and TEM work, respectively, as well as Dr. Martine Ziliox for her help with the Li NMR analyses. S.S.W. also thanks 3M for a Nontenured Faculty Award. Supporting Information Available: TEM image of SWNT-CE adduct bundles deposited from a methanol solution onto a lacey carbon grid. Scale bar represents 77 nm. This material is available free of charge via the Internet at http://pubs.acs.org. References
Figure 3. Three-dimensional AFM height image of functionalized SWNT-CE adduct bundles, adsorbed onto a flat mica substrate.
relative to each other, projecting outward and parallel to the tubular axis, thereby facilitating the formation of larger bundles. In addition, it has been shown that oxidative derivatization can cause smaller tubes to associate into larger bundles because of H-bonding between carboxyl groups attached to the walls of functionalizing tubes, thereby adding to the overall stacking effect.3,5 Functionalization of oxidized SWNTs with an organic crown ether derivative renders them reversibly soluble in a vast array of organic and aqueous solvents, enabling further exploitation of their solution chemistry for photophysical analyses as well as for the generation of novel nanoscale architectures. The derivatization process occurs as a result of salt formation initiated by a complementary attractive, zwitterionic interaction between carboxylic groups located at the ends, sidewalls, and defect sites of the oxidized SWNTs and amine moieties dangling from the side chain of the crown ether. This so-called ionic (charge-transfer) functionalization4,11,23 enhances the stability of SWNT solutions by effectively preventing nanotubes from aggregating in the solution state, though it does not necessarily prevent them from clumping together upon drying. Acknowledgment. We acknowledge support of this work through start-up funds provided by the State University of New York at Stony Brook as well as Brookhaven National Laboratory. Acknowledgment is also given to the donors of the Petroleum Research Fund, administered by the American Chemical Society, for support of this research. Sigma Xi is
1218
(1) Iijima, S.; Ichihashi, T. Nature (London) 1993, 363, 603. (2) Dresselhaus, M. S.; Dresselhaus, G.; Eklund, P. C. Science of Fullerenes and Carbon Nanotubes; Academic Press: New York, 1996. (3) Georgakilas, V.; Kordatos, K.; Prato, M.; Guldi, D. M.; Holzinger, M.; Hirsch, A. J. Am. Chem. Soc. 2002, 124, 760. (4) Chen, J.; Rao, A. M.; Lyuksyutov, S.; Itkis, M. E.; Hamon, M. A.; Hui, H.; Cohn, R. W.; Eklund, P. C.; Colbert, D. T.; Smalley, R. E.; Haddon, R. C. J. Phys. Chem. B 2001, 105, 2525. (5) Kukovecz, A.; Kramberger, C.; Holzinger, M.; Kuzmany, H.; Schalko, J.; Mannsberger, M.; Hirsch, A. J. Phys. Chem. B 2002, 106, 6374. (6) Bahr, J. L.; Tour, J. M. Chem. Mater. 2001, 13, 3823. (7) Banerjee, S.; Wong, S. S. Nano Lett. 2002, 2, 195. (8) Sano, M.; Kamino, A.; Okamura, J.; Shinkai, S. Science (Washington, D.C.) 2001, 293, 1299. (9) Liu, J.; Rinzler, A. G.; Dai, H.; Hafner, J. H.; Bradley, R. K.; Boul, P. J.; Lu, A.; Iverson, T.; Shelimov, K.; Huffman, C. B.; RodriguezMacias, F.; Shon, Y. S.; Lee, T. R.; Colbert, D. T.; Smalley, R. E. Science (Washington, D.C.) 1998, 280, 1253. (10) Chen, J.; Hamon, M. A.; Hu, H.; Chen, Y.; Rao, A. M.; Eklund, P. C.; Haddon, R. C. Science (Washington, D.C.) 1998, 282, 95. (11) Hamon, M. A.; Chen, J.; Hu, H.; Yongsheng, C.; Itkis, M. E.; Rao, A. M.; Eklund, P. C.; Haddon, R. C. AdV. Mater. (Weinheim, Ger.) 1999, 11, 834. (12) Riggs, J. E.; Guo, Z.; Carroll, D. L.; Sun, Y.-P. J. Am. Chem. Soc. 2000, 122, 5879. (13) Mickelson, E. T.; Huffman, C. B.; Rinzler, A. G.; Smalley, R. E.; Hauge, R. H.; Margrave, J. L. Chem. Phys. Lett. 1998, 296, 188. (14) Boul, P. J.; Liu, J.; Mickelson, E. T.; Huffman, C. B.; Ericson, L. M.; Chiang, I. W.; Smith, K. A.; Colbert, D. T.; Margrave, J. L.; Smalley, R. E. Chem. Phys. Lett. 1999, 310, 367. (15) Pompeo, F.; Resasco, D. E. Nano Lett. 2002, 2, 369. (16) Bandyopadhyaya, R.; Nativ-Roth, E.; Regev, O.; Yerushalmi-Rozen, R. Nano Lett. 2002, 2, 25. (17) Banerjee, S.; Wong, S. S. Nano Lett. 2002, 2, 49. (18) Banerjee, S.; Wong, S. S. J. Am. Chem. Soc. 2002, 124, 8940. (19) Bahr, J. L.; Mickelson, E. T.; Bronikowski, M. J.; Smalley, R. E.; Tour, J. M. Chem. Commun. 2001, 2001, 193. (20) O’Connell, M. J.; Boul, P.; Ericson, L. M.; Huffman, C.; Wang, Y.; Haroz, E.; Kuper, C.; Tour, J.; Ausman, K. D.; Smalley, R. E. Chem. Phys. Lett. 2001, 342, 265. (21) Karkhaneei, E.; Zebajadian, M. H.; Shamsipur, M. J. Solution Chem. 2001, 30, 323. (22) Sun, Y.; Wilson, S. R.; Schuster, D. I. J. Am. Chem. Soc. 2001, 123, 5348. (23) Chattopadhyay, D.; Lastella, S.; Kim, S.; Papadimitrakopoulos, F. J. Am. Chem. Soc. 2002, 124, 728.
NL025755D
Nano Lett., Vol. 2, No. 11, 2002
