NANO LETTERS
Length-Dependent Extraction of Single-Walled Carbon Nanotubes
2005 Vol. 5, No. 12 2355-2359
Kirk J. Ziegler,† Daniel J. Schmidt, Urs Rauwald, Kunal N. Shah, Erica L. Flor, Robert H. Hauge, and Richard E. Smalley* Department of Chemistry, Center for Nanoscale Science and Technology, Rice UniVersity, Houston, Texas 77005 Received June 1, 2005; Revised Manuscript Received October 27, 2005
ABSTRACT A two-phase liquid−liquid extraction process is presented which is capable of extracting water-soluble single-walled carbon nanotubes into an organic phase. The extraction utilizes electrostatic interactions between a common phase transfer agent and the sidewall functional groups on the nanotubes. Large length-dependent van der Waals forces for nanotubes allow the ability to control the length of nanotubes extracted into the organic phase as demonstrated by atomic force microscopy.
Single-walled carbon nanotubes (SWNTs) are typically synthesized with polydisperse micrometer lengths and are bound into macroscopic entangled ropes precluding their use in some applications. However, the availability of nanotube samples with uniform length is essential to many specialized applications. Such applications include biological imaging and sensing applications where shorter nanotube lengths may be required to penetrate cells and to serve as biological markers. Finally, the ability to control the length of the nanotubes will allow control over their solubility, which will aid in the development of many applications such as the dispersion of SWNTs into composites. Several approaches have been developed for cutting SWNTs into shorter segments. The first reported method1 used concentrated H2SO4/HNO3 mixtures to cut the SWNTs into short, open-ended pipes. More recently, fluorination followed by room-temperature piranha solutions2 has demonstrated high yields of short, cut SWNTs with lengths in the range of 30-150 nm having an average length of approximately 100 nm.3 Other researchers have shown that ultrasonication4-6 or ion etching7 can also prepare shorter nanotubes. While these processes yield shorter length SWNTs, they often still have significant polydispersity. To date, length-based separations of SWNTs have been limited to small-scale (microgram scale) techniques based on the preparation of surfactant-based suspensions. These surfactant-based separations are severely limited by the small concentrations of nanotubes that can be processed but have been utilized in techniques such as size exclusion chromatography,8 high-performance liquid chromatography,9 ion exchange chromatography,10 capillary electrophoresis,11 and * Corresponding author: tel, 713-348-3250; fax, 713-348-5320; e-mail,
[email protected]. † Current address: Department of Chemical Engineering, University of Florida, Gainesville, FL 32611. 10.1021/nl0510208 CCC: $30.25 Published on Web 11/25/2005
© 2005 American Chemical Society
field flow fractionation.1,12 Many of these techniques are also limited by nanotubes plugging the narrow channels causing relatively wide length distributions. The problems with plugging and dilute solutions render these length separation procedures unlikely techniques for large-scale separations. Extraction is a common separation technique that can be easily scaled up. Recently, researchers have demonstrated the extraction of gold nanoparticles from an aqueous solution to an organic phase using a common phase transfer catalyst, tetraoctylammonium bromide (TOAB).13 The extraction involved the electrostatic interaction of the TOAB with either carboxylic acid13-15 or sulfonate functional groups.14 This work has since been extended to show that size-selective extractions of the gold nanoparticles can be achieved.15,16 Here, we describe a scalable liquid-liquid extraction process for the length-based separation of SWNTs. HiPco17,18 single-walled carbon nanotubes (lot 130.4) were purified using SF6/O2 oxidation as described by Xu et al.19 The metal content after purification was approximately 1.6 wt %. Concentrated nanotube solutions in water were prepared without the use of surfactants by functionalizing the sidewalls of the purified nanotube with chloroaniline sulfonate using previously described methods.20 To obtain a higher degree of functionality, the nanotubes were functionalized multiple times to obtain a functionality ratio of approximately 1:18 as determined by thermal gravimetric analysis (TA Instruments). These functionalized nanotubes have high solubility in water and reasonable solubility in other polar solvents (e.g., methanol, ethanol). The nanotubes were dissolved in water (0.04-0.2 mg/mL) and placed in a vial. Then, a solution of TOAB in an organic solvent (either ethyl acetate or toluene) is added to the vial at a 1:1 volume ratio resulting in a liquid-liquid phase system. The vial is shaken vigorously by hand for 30 s to increase interfacial
Figure 1. Optical microscope image of the emulsions formed with 0.2 mg/mL aqueous SWNT solution and a 0.001 M TOAB in toluene solution.
Figure 3. (a) The UV-vis absorbance spectrum and (b) the integrated area for the upper organic phase as a function of TOA+: SO3- ratio.
Figure 2. The effect of TOAB concentration on the extraction of SWNTs into an ethyl acetate phase (a) before and (b) after extraction. The TOA+:SO3- ratio for vials 1-9 are 0, 0.15, 0.3, 0.45, 0.6, 0.75, 0.9, 1.0, and 2.0, respectively.
area and assist transfer across the interface. After the twophase mixture is shaken, the system is filled with gray emulsions. With an excess of TOAB, the emulsions settle and complete extraction of the nanotubes into the organic layer occurs. The emulsions formed in toluene systems coalesced within a few minutes, while those formed in ethyl acetate systems took considerably longer. These differences in stability can be related to the differences in colloidal interactions of each system. Previous approaches that have involved sulfonate14 or other groups13-15 interacting with TOAB have shown an electrostatic interaction. Therefore, it is expected that the extraction of the nanotubes occurs via stoichiometric electrostatic interactions between the cationic TOA+ and the anionic SO3moiety on the functional groups on the nanotube sidewall. The TOA+ is a positively charged surfactant with four alkane chains. Therefore, ion pairing increases the organophilicity of the SWNT allowing the transfer of the nanotube complex into the organic layer. No extraction of nanotubes occurred without the presence of the functional groups on the SWNTs or without the TOAB. The collected SWNTs in the organic layer could not be resuspended in water but could be easily transferred back to 2356
an aqueous or alcohol solution by reversing the reaction. The TOAB ligands are decomplexed by the addition of an excess of 99.8% acetic acid to the organic phase, which apparently causes the TOA+ to preferentially bind to the acetate anions. The solution was then stirred for 20-30 s resulting in the immediate formation of aggregated particles, filtered through a 0.2 µm PTFE filter membrane, and washed three to five times with ethanol. The nanotube powder was then collected and resuspended in water, methanol, or ethanol after ultrasonication for ∼1 min. Length-dependent extractions are obtained by utilizing a substoichiometric ratio of TOAB. Under these TOAB-starved conditions, colloidal interactions of the nanotubes and their solubility can be controlled, allowing separations based on the length of the nanotubes. Emulsion formation became more problematic at these conditions. The emulsions settle slowly (over ∼2 h) still giving a water phase swollen with emulsions that were stable for days as seen in Figure 1. When ethyl acetate is used as the organic solvent, very fine emulsions fill the swollen water phase occupying almost the entire liquid volume while toluene systems form emulsions that are coarser. The high conductivity of the swollen emulsion phase suggests that it is a water continuous phase. The high surface activity of the partially complexed SWNTs is likely due to the significant number of charged sites that remain as well as the significant number of hydrophobic sites rendering the SWNT an amphiphobic surfactant. Due to the significant number of charges remaining on partially complexed SWNTs, the SWNTs could not be Nano Lett., Vol. 5, No. 12, 2005
Figure 4. AFM images of the functionalized SWNTs (a) before extraction and after partial extraction with TOAB corresponding to ionpairing of (b) 30%, (c) 60%, and (d) 75%. Note that only individual nanotubes shown in green are utilized for length measurements.
extracted into the organic phase. Therefore, salts such as ammonium chloride were used to charge neutralize the remaining sulfonate anions. The ammonium cations will allow charge neutrality to be achieved without increasing the organophilicity of the nanotubes. While emulsion stability was significantly diminished, dilute TOAB solutions still failed to phase separate completely. Several approaches were attempted to breakup the formation of emulsions including pH changes, temperature changes, the addition of salts and cosolvents, and centrifugation. Of these approaches, only freezing was found to effectively break the emulsions. After the two-phase system was thawed, the carbon nanotubes are transferred to the organic phase after gentle agitation. Figure 2 shows the extraction of nanotubes with varying concentrations of TOAB. No extraction of the SWNTs occurs without TOAB (vial 1). At low TOAB concentrations (vial 2), there is insufficient complexation to render the nanotubes organophilic and very few nanotubes are extracted. However, the complexation of the TOAB to the nanotubes has rendered the SWNTs hydrophobic. Therefore, these solutions tend to form a very stable emulsion/interphase that cannot be broken up even after freezing. An increase in the concentration of Nano Lett., Vol. 5, No. 12, 2005
TOAB shown in vials 3 and 4 results in the extraction of a small amount of nanotubes into the organic layer. Continued increases in the TOAB concentration (vials 5 through 7) causes enough complexation that a significant portion of the nanotubes have become organophilic enough to be extracted into the organic layer. Finally, complete complexation of the nanotubes results in the extraction of all nanotubes (vials 8 and 9). UV-vis-NIR absorbance spectra (Shimadzu UV3101PC spectrometer) of the upper organic phases were recorded by carefully transferring the solution to a cuvette. The transfer of each phase often resulted in a turbid solution that required several minutes to settle before accurate spectra could be measured. As seen in Figure 3, the increase in absorbance intensity confirms that there is a gradient in nanotube extraction with TOAB concentration. Furthermore, the concentration of TOAB that extracts all of the nanotubes corresponds to the theoretical amount of TOA+ required to complex with every SO3- group confirming a 1:1 stoichiometric electrostatic interaction. For length measurements of the extracted organic layer, the organic phase was collected, washed with acetic acid as described above, and resuspended in methanol after ultra2357
Figure 6. The average length of nanotube extracted as a function of ion pairing (circles). The starting material is shown on the plot at 100% (square) to demonstrate that the lengths extracted approach complete extraction.
Figure 5. Length distributions of the functionalized SWNTs (a) before extraction and after partial extraction with TOAB corresponding to ion pairing of (b) 30%, (c) 40%, (d) 60%, and (e) 75%. The dashed lines indicate the average length of the sample and N is the number of SWNTs measured. 2358
sonication for ∼1 min. The extracted nanotube solutions were then spin-coated onto freshly cleaved mica substrates yielding a high quantity of individual nanotubes. Figure 4 shows the tapping mode atomic force microscopy (AFM) images (Digital Instruments Nanoscope IIIA), and Figure 5 shows the length distributions using SIMAGIS software21 for the starting material and for extractions at stoichiometric ratios (TOA+:SO3-) of 0.3, 0.4, 0.6, and 0.75. Only individual nanotubes (shown in green) were utilized for length measurements. The original collected AFM images and details of the measurements are available as Supporting Information. Typically, over 1500 nanotubes were measured to obtain statistically meaningful results. The starting material (Figures 4a and 5a) has a broad distribution of nanotube lengths with an average of 275 nm. The addition of enough TOAB to complex 30% of all SO3- moieties with TOA+ results in only very short nanotubes being extracted into the organic layer as seen by AFM and corresponding histograms (Figures 4b and 5b). As can be seen, the length distribution narrows considerably with 81% below 100 nm and an average length of 73 nm. Increased ion pairing results in longer nanotubes being extracted. After 75% of the SO3- moieties are ion paired, the length distribution begins to approach that of the starting material. Figure 6 shows the average length of the extracted nanotubes as a function of ion pairing. It is evident that the average length increases monotonically with TOAB concentration and approaches the average length of the starting material (square symbol in figure). The length-dependent phase transfer can be understood by considering the attractive interactions between colloidal particles in solution. Colloidal dispersions in organic solvents are typically characterized by the attractive van der Waals interactions between particles and the steric repulsion interaction caused by the brush layer attached to the colloids.22,23 Pairwise summation of the van der Waals attractions results in significantly stronger interactions for larger colloidal particles. This summation results in strong van der Waals interactions for nanotubes, which have been shown to have attractive energies on the order of 36 kBT/nm.23 At a given TOAB concentration, the shortest length nanotubes will always have less attractive van der Waals interactions. Nano Lett., Vol. 5, No. 12, 2005
However, the presence of the organophilic chains creates a steric barrier sufficient to overcome the attractive van der Waals interactions for the smallest length nanotubes.23 Those nanotubes that cannot be stabilized in the organic layer succumb to the attractive van der Waals interactions, aggregate, and settle to the interface. The addition of more TOA+ stabilizes and extracts longer nanotubes. Due to the role of the functionalization in the steric stabilization, the degree of functionalization will play an important role in the extraction process. Indeed, SWNTs could only be extracted when high degrees of functionalization were present. In conclusion, an extraction process has been developed which takes advantage of the considerable attractive force existing between nanotubes. With control of the concentration of the steric barrier, the length-dependent stabilization of nanotubes in the organic phase can be readily manipulated. The simplicity of the extraction process allows this separation to be easily scaled-up with relatively narrow length distributions. Acknowledgment. We gratefully acknowledge financial support from the National Science Foundation (Grant No. 0139202), the Office of Naval Research, the Robert A. Welch Foundation, the Center for Biological and Environmental Nanotechnology, and the Air Force Office of Scientific Research. We thank J. L. Hudson and J. M. Tour for supplying the functionalized nanotubes and S. R. P. da Rocha for helpful discussions. Supporting Information Available: Detailed description of the length measurements and supporting AFM images to demonstrate the length-dependent extraction of SWNTs. This material is available free of charge via the Internet at http:// pubs.acs.org. References (1) Liu, J.; Rinzler, A. G.; Dai, H. J.; Hafner, J. H.; Bradley, R. K.; Boul, P. J.; Lu, A.; Iverson, T.; Shelimov, K.; Huffman, C. B.;
Nano Lett., Vol. 5, No. 12, 2005
(2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23)
Rodriguez-Macias, F.; Shon, Y. S.; Lee, T. R.; Colbert, D. T.; Smalley, R. E. Science 1998, 280, 1253. Ziegler, K. J.; Gu, Z. N.; Peng, H. Q.; Flor, E. L.; Hauge, R. H.; Smalley, R. E. J. Am. Chem. Soc. 2005, 127, 1541. Ziegler, K. J.; Gu, Z.; Shaver, J.; Chen, Z.; Flor, E. L.; Schmidt, D. J.; Chan, C.; Hauge, R. H.; Smalley, R. E. Nanotechnology 2005, 16, S539. Yudasaka, M.; Zhang, M.; Jabs, C.; Iijima, S. Appl. Phys. A 2000, 71, 449. Stepanek, I.; Maurin, G.; Bernier, P.; Gavillet, J.; Loiseau, A.; Edwards, R.; Jaschinski, O. Chem. Phys. Lett. 2000, 331, 125. Maurin, G.; Stepanek, I.; Bernier, P.; Colomer, J. F.; Nagy, J. B.; Henn, F. Carbon 2001, 39, 1273. Lustig, S. R.; Boyes, E. D.; French, R. H.; Gierke, T. D.; Harmer, M. A.; Hietpas, P. B.; Jagota, A.; McLean, R. S.; Mitchell, G. P.; Onoa, G. B.; Sams, K. D. Nano Lett. 2003, 3, 1007. Duesberg, G. S.; Muster, J.; Krstic, V.; Burghard, M.; Roth, S. Appl. Phys. A 1998, 67, 117. Farkas, E.; Anderson, M. E.; Chen, Z. H.; Rinzler, A. G. Chem. Phys. Lett. 2002, 363, 111. Zheng, M.; Jagota, A.; Semke, E. D.; Diner, B. A.; McLean, R. S.; Lustig, S. R.; Richardson, R. E.; Tassi, N. G. Nat. Mater. 2003, 2, 338. Doorn, S. K.; Fields, R. E.; Hu, H.; Hamon, M. A.; Haddon, R. C.; Selegue, J. P.; Majidi, V. J. Am. Chem. Soc. 2002, 124, 3169. Chen, B.; Selegue, J. P. Anal. Chem. 2002, 74, 4774. Chen, S.; Yao, H.; Kimura, K. Langmuir 2001, 17, 733. Mayya, K. S.; Caruso, F. Langmuir 2003, 19, 6987. Yao, H.; Momozawa, O.; Hamatani, T.; Kimura, K. Chem. Mater. 2001, 13, 4692. Cheng, W.; Wang, E. J. Phys. Chem. B 2004, 108, 24. Nikolaev, P.; Bronikowski, M. J.; Bradley, R. K.; Rohmund, F.; Colbert, D. T.; Smith, K. A.; Smalley, R. E. Chem. Phys. Lett. 1999, 313, 91. Bronikowski, M. J.; Willis, P. A.; Colbert, D. T.; Smith, K. A.; Smalley, R. E. J. Vac. Sci. Technol., A 2001, 19, 1800. Xu, Y.; Peng, H.; Hauge, R. H.; Smalley, R. E. Nano Lett. 2005, 5, 163. Hudson, J. L.; Casavant, M. J.; Tour, J. M. J. Am. Chem. Soc. 2004, 126, 11158. Ziegler, K. J.; Rauwald, U.; Gu, Z.; Liang, F.; Billups, W. E.; Hauge, R. H.; Smalley, R. E. Submitted for publication. Shah, P. S.; Holmes, J. D.; Johnston, K. P.; Korgel, B. A. J. Phys. Chem. B 2002, 106, 2545. Shvartzman-Cohen, R.; Nativ-Roth, E.; Baskaran, E.; Levi-Kalisman, Y.; Szleifer, I.; Yerushalmi-Rozen, R. J. Am. Chem. Soc. 2004, 126, 14850.
NL0510208
2359
