1858
Langmuir 2006, 22, 1858-1862
Uniform Directional Alignment of Single-Walled Carbon Nanotubes in Viscous Polymer Flow Erin Camponeschi,† Bill Florkowski,† Richard Vance,†,‡ Glenn Garrett,† Hamid Garmestani,† and Rina Tannenbaum*,† School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, and NSF-SURF Student, Department of Materials Science and Engineering, Washington State UniVersity, Pulman, Washington ReceiVed October 6, 2005. In Final Form: December 7, 2005 In this work, we probed the effects of shear flow on the alignment of dispersed single-walled carbon nanotubes in polymer solutions. Two different systems were compared: Single-walled carbon nanotubes dispersed using an anionic surfactant and single-walled carbon nanotubes dispersed using an anionic surfactant and a weakly binding polymer. It was determined that the addition of the weakly binding polymer increased the degree of dispersion of the carbon nanotubes and the ability to induce their alignment when subjected to shear forces.
1. Introduction Carbon nanotubes have been shown to possess unique and superior properties that can be useful in a variety of materialsrelated applications.1-3 Since the nanotubes are essentially rolledup graphite sheets, they can exhibit either semiconducting or metallic properties4 and are expected to have a very high elastic modulus, high toughness, and good mechanical properties.4,5 Both multiwalled carbon nanotubes (MWNT) and singlewalled carbon nanotubes (SWNT) have been studied by various methods and were determined to exhibit high tensile strengths, on the order of 11 and 20 GPa, and Young’s moduli, on the order of 0.4 TPa and upward of 1 TPa, respectively.2,6-12 Carbon nanotubes have been used as a reinforcement in different matrices including polymer, ceramics, resins, metals, and other matrix materials.7,10,13-17 Most notably, the incorporation of carbon nanotubes in polymer matrices has generated a great deal of interest and high expectations regarding the possibility of a dramatic enhancement of the mechanical, electronic, and thermal properties of such nanocomposites, especially in cases where the * Corresponding author. E-mail:
[email protected]. † Georgia Institute of Technology. ‡ Washington State University. (1) Dresselhaus, M. S., Dresselhaus, G., Avouris, Ph., Eds.; Carbon Nanotubes; Topics in Applied Physics 80; Springer-Verlag: Berlin, 2001. (2) Kimura, T.; Ago, H.; Tobita, M.; Ohshima, S.; Kyotani, M.; Yumura, M. AdV. Mater. 2002, 14 (19), 1380-1383. (3) Odom, T. W.; Huang, J. L.; Kim, P.; Lieber, C. M. J. Phys. Chem. B 2000, 104, 2794-2809. (4) Huczko, A. Appl. Phys. A 2002, 74, 617-638. (5) Yamamoto, K.; Akita, S.; Nakayama, Y. J. Appl. Phys. (Jpn) 1996, 35, 917. (6) Kelly, B. T., Ed.; Physics of Graphite; Applied Science: London, 1981. (7) Baughman, R. H.; Zakhidov, A. A.; de Heer, W. A. Science 2002, 297 (5582), 787-792. (8) Lau, K. T. Composites B 2002, 33, 263-277. (9) Lu, Y.; Liaw, P. K. J. Occup. Med. 2001, 53 (3), 31-35. (10) Qian, D.; Dickey, E. C. J. Microscopy 2001, 204, 39-45. (11) Treacy, M. M. J.; Ebbesen, T. W.; Gibson, J. M. Nature 1996, 381, 6584. (12) Yu, M. F.; Lourie, O.; Dyer, M. J.; Noloni, K.; Kelly, T. F.; Ruoff, R. S. Science 2000, 287, 637. (13) Allaoui, A.; Bai, S.; Cheng, H. M.; Bai, J. B. Comput. Sci. Technol. 2002, 62, 1993-1998. (14) Cooper, A. C.; Diana, R.; David, L.; Joerg, M.; Wanger, H. D. Comput. Sci. Technol. 2002, 62, 1105-1112. (15) Jia, Z. J.; Wang, Z. Y.; Xu, C. L.; Liang, J.; Wei, B. Q.; Wu, D. H. Mater. Sci. Eng. A 1999, 271, 395-400. (16) Lourie, O.; Wagner, H. D. Comput. Sci. Technol. 1999, 55, 975-977. (17) Peigney, A.; Laurent, Ch.; Flahaut, E.; Rousset, A. Ceram. Int. 2000, 26, 677-683.
carbon nanotubes are aligned.1-3 It has been shown, however, that randomly oriented carbon nanotubes embedded in polymer matrices have failed to generate composites in which the full potential of the mechanical and conductive properties of the nanotubes is exhibited.7 It is believed that this phenomenon is due to the poor dispersion of the carbon nanotubes in the polymer matrix. Therefore, the use of carbon nanotubes as strengthening agents, even via a random distribution in the matrix, is hindered by their tendency to clump together as a result of van der Waals attractions due to their high surface energy.18 Hence, the main conclusion drawn from previous experiences with nanotubefilled polymer matrices is the crucial impact of the degree of dispersion of the carbon nanotubes on their ability to enhance the mechanical properties of the composites that they form. It has been found that SWNTs are insoluble both in water and organic solvents.18,19 Due to this, surfactants have been used to facilitate the dispersion and introduction into a polymer matrix, by reducing the van der Waals attraction between the individual tubes.18,19 The surfactants are able to disperse the carbon nanotubes by increasing the repulsive forces due to their hydrophilic and hydrophobic components. The addition of a surfactant creates a metastable array of carbon nanotubes, which allows the carbon nanotubes to remain dispersed in solution for long periods of time.18-21 In addition to the impact of the degree of carbon nanotube dispersion on composite properties, their degree of alignment in the respective matrix plays a crucial role as well. There are several different methods available to align carbon nanotubes, including slicing,22 chemical vapor depositon,23 melt processing,24 mechanical stretching,25 electrophoresis,5 application of magnetic fields,2,26,27 and electrospinning.28-31A problem with most (18) Bonard, J. M.; Stora, T.; Salvetat, J. P.; Maier, F.; Stockli, T.; Duschi, C.; Forro, L.; de Heer, W. A.; Chatelain, A. AdV. Mater. 1997, 9 (10), 827-833. (19) Matarredonna, O.; Rhoads, H.; Li, Z.; Harwell, J. H.; Balzano, L.; Resasco, E. J. Phys. Chem. B 2003, 107, 13357-13367. (20) Shvartzman-Cohen, R.; Levi-Kalishman, Y.; Nativ-Roth, E.; YerushalmiRozen, R. Langmuir 2004, 20 (15), 6085-6088. (21) Shvartzman-Cohen, R.; Nativ-Roth, E.; Baskaran, E.; Levi-Kalishman, Y.; Szleifer, I.; Yerushalmi-Rozen, R. J. Am. Chem. Soc. 2004, 126 (45), 1485014857. (22) Ajayan, P. M.; Stephan, O.; Colliex, C.; Trauth, D. Science 1994, 265, 1212. (23) Chen, Y.; Guo, L.; Patel, S.; Shaw, D. T. J. Mater. Sci. 2000, 35, 5517. (24) Mitchell, C. A.; Bahr, J. L.; Aerpalli, S.; Tour, J. M.; Krishnamoorti, R. Macromolecules 2002, 35, 8825-8830. (25) Jin, L.; Bower, C.; Zhoua, O. Appl. Phys. Lett. 1998, 73, 1197.
10.1021/la052714z CCC: $33.50 © 2006 American Chemical Society Published on Web 01/14/2006
Alignment of SWCN in Viscous Polymer Flow
alignment processes arises when combining the carbon nanotubes with the polymer matrix to form the composite materials. Under the experimental conditions, a well-aligned array of carbon nanotubes tends to become isotropic and cluster upon mixing into the composite. Hence, the alignment of the carbon nanotubes prior to mixing into a composite is reversed, and once again, they may disperse into random orientations.27 In this present work, shear flow was used to align the carbon nanotubes in a surfactant/polymer matrix. This alignment method has some advantages over electrospinning because it can generate polymer sheets rather than polymer fibers in a single step, providing a cheaper alternative. However, although it is possible to create a nanocomposite in which the SWNT are organized as 2D layers using the shear flow method, the highly confined 1D sequences generated by electrospinning can be further aligned to form 2D layers as well, with a higher degree of orientation. Nevertheless, a smaller shear stress will be required to orient the SWNT in a 2D polymer film than into a polymer fiber generated by electrospinning.32,33 The nanotubes were initially dispersed with sodium dodecyl benzene sulfonate, NaDDBS, an anionic surfactant, and placed in a polymer solution that was subjected to a circular shear flow, resulting in a well-dispersed, uncoiled, and aligned carbon nanotube structure within the matrix. The addition of a weakly binding polymer, such as carboxymethyl cellulose, CMC, serves, in this case, two purposes: constituting the polymer matrix in which the SWNT will be dispersed and aligned and providing a secondary mechanism for the promotion of carbon nanotube dispersion. Samples of the polymer-carbon nanotube suspensions were collected while under the shear flow, both parallel to the shear flow and perpendicular to it. Both the efficiency of the dispersion process and the carbon nanotube alignment in the polymer solution were analyzed by means of transmission electron microscopy (TEM). 2. Experimental Procedure The carbon nanotubes used in this work were used as purchased from Carbon Nanotechnology, Inc., in high purity form (>90%) synthesized by the laser ablation process. Sodium dodecyl benzene sulfonate (NaDDBS, molecular weight of 348.48 g/mol) (from TCI) was used in the same concentrations and methods as described in Matarredona, et al.19 Two different solutions were prepared, one solution made of 50 mL, 1.2 mM NaDDBS, with 0.4 mg carbon nanotubes and the second solution consisting of 25 mL, 1.2 mM NaDDBS, with 0.4 mg carbon nanotubes. Both solutions were then placed in a vibracell sonicator (Sonics and Materials, Inc., 20 kHz) at 14% amplification for 30 min. After sonication, 25 mL of a 1 wt % carboxyl methylcellulose (CMC, M h w ) 350 000 g/mol) solution was added to the NaDDBS/SWNT solution, and the new combined solution was again placed in the vibracell sonicator at 14% amplification for 30 min. After sonication, TEM samples of both solutions were taken to determine the extent of SWNT dispersion: a droplet of solution was placed onto a TEM grid (Ted Pella, Inc. carbon coated copper, PELCO Center-Marked Grids, 400 mesh, 3.0 (26) Al-Haik, M. S.; Garmestani, H.; Li, D. S.; Hussaini, M. Y.; Sablin, S. S.; Tannenbaum, R.; Dahmen, K. J. Polym. Sci. Polym. Phys. 2004, 42, 1586-1600. (27) Garmestani, H.; Al-Haik, M. S.; Dahmen, K.; Tannenbaum, R.; Li, D. S.; Sablin, S. S.; Hussaini, M. Y. AdV. Mater. 2003, 15 (22), 1918-1921. (28) Dror, Y.; Salalha, W.; Khalfin, R. L.; Choen, Y.; Yarin, A. L.; Zussman, E. Langmuir 2003, 19, 7012-7020. (29) Ko, F.; Gogotsi, Y.; Ali, A.; Naguib, N.; Ye, H. H.; Yang, G. L.; Li, C.; Willis, P. AdV. Mater. 2003, 15 (14), 1161-1165. (30) Salalha, W.; Dror, Y.; Khalfin, R. L.; Choen, Y.; Yarin, A. L.; Zussman, E. Langmuir 2004, 20, 9852-9855. (31) Sen, R.; Bin, Z.; Perea, D.; Itkis, M. E.; Hu, H.; Love, J.; Bekyarova, E.; Haddon, R. C. Nano Lett. 2004, 4 (3), 459-464. (32) Croce, V.; Cosgrove, T.; Dreiss, C. A. Langmuir 2005, 21, 6762-6768. (33) Kim, G.-M.; Wutzler, A.; Radusch, H.-J.; Michler, G. H.; Simon, P.; Sperling, R. A.; Parak, W. J. Chem. Mater. 2005, 17 (20), 4949-4957.
Langmuir, Vol. 22, No. 4, 2006 1859
Figure 1. Schematic representation of the experimental setup: (a) Concentric cylinder arrangement in the Brookfield viscometer; (b) TEM sample retrieval and preparation. mm O.D.) and allowed to dry for later analysis in the TEM. The solutions were then re-sonicated at 14% amplification for 30 min. The solutions were then placed in a 27.5 mm diameter (I.D.) stainless steel cylinder. A Brookfield DV-E viscometer with a stainless steel rotating spindle, having 19.0 mm diameter (O.D.) and 65.0 mm in length, was used for viscosity measurements. The sample solutions were placed in the 8.5 mm gap between the outer cylinder and the spindle, as shown in Figure 1a. In turn, the spindle was allowed to rotate for 1 week at several different angular velocities ranging from 12 to 100 rpm, corresponding to approximately 1-10 rad/s. A second set of TEM samples were taken in situ from the solutions flowing in circular motion in the gap between the outer cylinder and inner cylinder (spindle). The sample grids were affixed to a small diameter wire, and slowly dipped into the solution parallel and perpendicular to the flow, as illustrated in Figure 1b, to determine the position of maximum stress after analysis in the TEM. These samples were then analyzed using the JEOL 100CX II, 100 kV transmission electron microscope (TEM). Samples for Raman spectroscopy were prepared as follows: 20 mL of the CMC/NaDDBS/SWNT solution was mixed with 10 mL of epoxy. The mixture was placed it in a Brookfield viscometer equipped with a UL attachment, creating a very small gap (Rb/Rc e 0.1) that allowed the achievement of higher shears in viscous fluids. Once the shear flow was adequately developed, a hardener was added in order to promote cross-linking in the epoxy, thus rapidly increasing the solution viscosity as to prevent the de-alignment of the SWNT in the solution. The samples were tested on a, Holo Probe VPT system, with integrated fiber coupled Raman system (Kaiser Optical Systems, Inc.), using a 731 nm incident laser radiation and VV (parallel/parallel) configuration, to determine orientation of SWNTs.34-37 The samples were tested at various polarization angles raging from 0 to 90°, to determine the development of SWNT orientation after being subjected to shear flow.
3. Results and Discussion 3.1. Dispersion of SWNT. The high degree of aggregation of SWNT is due to the high cohesive energy of the tubes that has been estimated to be on the of order of 36 kT for each nanometer of length overlap between adjacent tubes, translating into several thousand kT for micron-long tubes.38 Since maintaining a stable dispersion is a necessary condition for the utilization of SWNTs in various composite applications, the promotion of exfoliation and dispersion of SWNT has been a very active field of research in recent years. Most notable (34) Anglaret, E.; Righi, A.; Sauvajol, J. L.; Bernier, P.; Vigolo, B.; Poulin, P. Physica B 2002, 323, 38-43. (35) Rao, A. M.; Dresselhaus, G.; Dresselhaus, M. S. Phys. ReV. Lett. 2000, 84, 1820. (36) Benoit, J. M.; Buisson, J. P.; Chauvet, O.; Gondon, C.; Lefrant, S. Phys. ReV. B 2002, 66, 073417. (37) Qing Zhao, Q.; Wagner, D. H. Philos. Trans. R. Soc. London A 2004, 362, 2407-2424. (38) Girifalco, L. A.; Hodak, M.; Lee, R. S. Phys. ReV. B 2000, 62, 13104.
1860 Langmuir, Vol. 22, No. 4, 2006
Camponeschi et al.
shown below: The hydrocarbon moiety interacts with the surface
Figure 2. Schematic representation of the stabilization mechanism of carbon nanotubes: Interaction between SWNT with NaDDBS, followed by the addition of CMC. It is important to note that the molecules were not drawn to scale, and the dimension of the polymer molecules is about 2 orders of magnitude larger that the dimension of the surfactant molecules (Rg ≈ 1000 Å for CMC as compared to ≈20 Å for NaDDBS).
Figure 3. TEM micrographs of SWNT at different levels of dispersion and stabilization efficiency: (a) Undispersed carbon nanotubes; (b) Carbon nanotubes dispersed with NaDDBS; (c) Carbon nanotubes dispersed with NaDDBS and CMC. Note that the black spherical aggregates present in the image are the remnant metallic oxide catalyst particles used in the synthesis of the SWNT.
approaches consisted of either the use of surfactants18-21,39-41 or polymers.42-46 Coupling the effects of both surfactants and polymers, such as treatment with NaDDBS and CMC, may result in a superior and more stable dispersion of SWNTs, as shown schematically in Figure 2. An example of an aqueous suspension of carbon nanotubes without proper dispersion is illustrated by the TEM image in Figure 3a. The addition of NaDDBS, an anionic surfactant, to the suspension results in the exfoliation and dispersion of the carbon nanotubes, as shown in Figure 3b. NaDDBS dissociates in an aqueous environment to generate a sulfonium group, as (39) Vigolo, B.; Penicaud, A.; Coulon, C.; Sauder, R.; Pailler, C.; Journet, P.; Bernier, P.; Poulin, P. Science 2000, 290, 1331. (40) O’Connell, M. J.; Bachilo, S. M.; Huffman, C. B.; Moore, V. C.; Strano, M. S.; Haroz, E. H.; Rialon, K. L.; Boul, P. J.; Noon, W. H.; Kittrell, C.; Ma, J.; Hauge, R. H.; Weisman, R. B.; Smalley, R. E. Science 2002, 297, 593-596. (41) Zheng, M.; Jagota, A.; Strano, M. S.; Santos, A. P.; Barone, P.; Chou, S. G.; Diner, B. A.; Dresselhaus, M. S.; McLean, R. C.; Onoa, G. B.; Samsonidze, G. G.; Semke, E. D.; Usrey, M.; Walls, D. J. Science 2003, 302, 1545-1548. (42) Moore, V. C.; Strano, M. S.; Haroz, E. H.; Schmidt, J.; Talmon, Y.; Hauge, R. H.; Smalley, R. E. Nano Lett. 2003, 3, 1379. (43) Strano, M. S.; Moore, V. C.; Miller, M. K.; Allen, M. J.; Haroz, E. H.; Kittrel, C.; Huage, R. H.; Smalley, R. E. J. Nanosci. Nanotechnol. 2003, 3, 8185. (44) O’Connell, M. J.; Boul, P. J.; Ericson, L. M.; Huffman, C. B.; Wang, Y. H.; Haroz, E. H.; Kuper, C.; Tour, J.; Ausman, K. D.; Smalley, R. E. Chem. Phys. Lett. 2001, 342, 265-271. (45) Chen. J.; Liu, H.; Weimer, W. A.; Halls, M. D.; Waldek, D. H.; Walker, G. C. J. Am. Chem. Soc. 2002, 31, 9034. (46) Bandyopadhyaya, R.; Nativ-Roth, E.; Regev, O.; Yerushalmi-Rozen, R. Nano Lett. 2002, 2, 25-28.
of the carbon nanotube, whereas the anionic group is solubilized by the surrounding water molecules. This increases the solubility of the carbon nanotubes and allows their exfoliation and dispersion in the aqueous medium. Conversely, since the experimental procedure involves the sonication of the SWNT suspensions, it is also quite likely that this generates considerable exfoliation of the nanotubes, followed by the adsorption of the surfactant molecules, which in turn, stabilizes the nanotubes due to steric repulsion.20,21,42,43 The addition of CMC to the SWNT/NaDDBS solution has a moderate effect on the resulting dispersion as shown in the TEM image in Figure 3c. CMC adsorbs weakly on the surface of the carbon nanotubes, most likely via the interactions of the β1 moiety with the SWNTs.20,21 Because at low concentration the CMC molecules are largely uncoiled (intrinsic persistence length Lp0 ) 160 Å, indicating a semi-flexible polymer47), this interaction is most likely to occur with the polymer end segments, e.g. as shown below:
The abundance of carboxylic functional groups along the polymer chain will ensure the presence of a large number of associated water molecules, effectively increasing the hydrodynamic radius of the individual chains (Rg ≈ 1000 Å,48 as compared to ≈20 Å, the size of the NaDDBS molecule), thus introducing a large steric hindrance into the system, that promotes additional dispersion and separation of the carbon nanotubes. 3.2. Orientation of SWNT in Shear Flow. A Brookfield viscometer was used to generate a shear flow for the alignment of the carbon nanotubes. Predictions for optimal shear forces generated by this setup (see Figure 1) were based on nonNewtonian flow characteristics of the CMC-containing solutions (the fluid constants for this system were established separately) and on a gap for which the ratio of the inner cylinder and outer cynlinder radii, Rb/Rc, respectively, is 0.69, i.e., between 0.5 and 0.99 for which the model holds.49 The measurement of the torque, T, necessary to maintain a constant angular velocity ω of the inner cylinder, is related to the shear stress τw according to the relationship τw ) T/2πRb2L, where L is the length of the inner cylinder in contact with the fluid. The shear rates and shear stresses obtained for the various systems that were examined are summarized in Table 1 and Figure 4a. In the absence of polymer, the shear stresses that are due to the presence of the carbon (47) Hoogendam, R. Adsorption and Desorption of Cellulose Derivatives. Ph.D. Dissertation Nr. 2487; Wageningen University and Research Center: The Netherlands, 1998. (48) Zauscher, S.; Klingenberg, D. J. Colloids Surf. A 2001, 178, 213-229. (49) Geankoplis, C. J. Transport Processes and Unit Operations, 3rd ed.; Prentice Hall: Englewood Cliffs, NJ, 1993.
Alignment of SWCN in Viscous Polymer Flow
Langmuir, Vol. 22, No. 4, 2006 1861
Figure 4. Summary of the calculated shear stresses of the various SWNT-containing solutions used in the various experiments: (a) Plots of shear stresses as a function of angular velocity; (b) Plots of the shear stresses originating from the contribution of pure SWNT in the various systems as a function of angular velocity.
nanotubes are negligible at low angular velocities, but become considerable at high angular velocities, as shown in Figure 4b. For example, at 6.28 rad/s (60 rpm), the shear stress at the wall, τw exerted by the carbon nanotubes is ∼30 kPa, whereas at 10.48 rad/s (100 rpm), it is ∼350 kPa. Conversely, the presence of CMC in the solution has the effect of increasing the shear stresses of the system at all angular velocities, due to the higher viscosities exhibited by the polymer medium. Moreover, the shear stresses that are due only to the presence of the carbon nanotubes in the solution are small, but increase linearly with the angular velocity, as shown in Figure 4b. It is important to note that CMC may be considered as a uniformly charged semiflexible cylinder with a radius of 0.95 nm and an average coiled length of ∼250 nm, and hence, each chain may be viewed as having comparable dimensions to those of SWNT. Under these circumstances, the presence of the CMC chains in solution “catalyzes” the dynamic behavior of the carbon nanotubes and in effect attenuates the development of the stresses associated with their movement in the solution.50 This is due to the fact that the CMC molecules are tethered to the carbon nanotubes (albeit via weak interactions) and, hence, create a network of entanglements that in effect lowers the change in shear stress of the system. The development of directional anisotropy in carbon nanotube solutions, when subjected to shear flow stresses, was possible only in systems in which the carbon nanotubes were dispersed
Figure 5. TEM micrographs of the orientation attempts of several systems containing SWNT: (a) Undispersed carbon nanotubes in a 1 wt % CMC suspension subjected to shear flow at 100 rpm; (b) Carbon nanotubes dispersed with NaDDBS and CMC and subjected to shear flow at 30 rpm; (c) Carbon nanotubes dispersed with NaDDBS and CMC and subjected to shear flow at 60 rpm; (d) Oriented carbon nanotubes dispersed with NaDDBS and CMC and subjected to shear flow at 100 rpm. The inset image is a 4-fold magnification of the larger image (same scale bar ) 5 nm) showing in more detail the local orientation of the surface-modified SWNT. Note that the black spherical aggregates present in these images are the remnant metallic oxide catalyst particles used in the synthesis of the SWNT.
by the cooperative surface interactions with both NaDDBS and CMC. Subjecting a SWNT suspension to shear flow at high angular velocities did neither contribute to their exfoliation nor to their orientation, as shown in Figure 5a. No noticeable orientation of the SWNT was observed in systems in which the dispersion was achieved with NaDDBS only. For systems in which effective dispersion of the carbon nanotubes was achieved by the combined action of both NaDDBS and CMC, no alignment was observed for lower angular velocities, as can be seen in Figure 5b,c. The only system in which tube alignment was observed was for the NaDDBS/CMC/SWNT solution that was subjected to shear stresses at the highest angular velocity used in the experiments (10.48 rad/s), as shown in Figure 5d. Additional confirmation for the results illustrated by TEM regarding the alignment of the SWNT was obtained with Raman spectroscopy. Since the Raman intensity of a vibration depends on the relative directions of the crystal axis and the electric wave polarization of the incident and scattered light, this technique may also be used to determine the orientation of nanotubes in polymer matrices.34-37,51-56 Figure 6a shows the orientationdependent Raman spectra of SWNT with different angles between the polarization of the incidence laser light and the nanotube axis using VV (parallel polarization of the incidence and scattered
Table 1. Summary of the Mechanical Properties of the Various SWNT-Containing Solutions conditions
NaDDBS
NaDDBS/SWNT
CMC/NaDDBS
CMC/NaDDBS/SWNT
angular velocity (rad/s)
shear rate (s-1)
shear stress (MPa)
shear rate (s-1)
shear stress (MPa)
shear rate (s-1)
shear stress (MPa)
shear rate (s-1)
shear stress (MPa)
1.26 3.14 5.24 6.28 10.48
4.27 10.67 17.78 21.34 35.57
0 0.20 0.33 0.45 0.74
4.70 11.75 19.58 23.49 39.15
0 0.12 0.33 0.48 1.10
4.70 11.75 19.58 23.49 39.15
0.35 1.14 1.86 2.21 3.65
4.70 11.75 19.58 23.49 39.15
0.36 1.15 1.87 2.22 3.68
1862 Langmuir, Vol. 22, No. 4, 2006
Camponeschi et al.
structure of the nanotubes and to electronic resonance effects. As shown in Figure 6, there appears to be some fluorescence due to the CMC matrix, but it also illustrates that the nanotubes are highly oriented in the polymer matrix. These results illustrate the fact that the orientation of carbon nanotubes in a polymer matrix by the application of shear forces can be achieved only if two requirements can be satisfied: (a) the carbon nanotubes are well dispersed in the polymer matrix and (b) the shear forces applied to the dispersed carbon nanotubes are large enough to induce their orientation. In general, the presence of the CMC molecules, given their size and their semiflexible characteristics, provided a molecular template that promoted carbon nanotube alignment at lower shear stresses.
4. Conclusion In this work, we have shown that the addition of an anionic surfactant to a carbon nanotube aqueous suspension facilitated their dispersion but was ineffective as a tool for enhancing the ability to orient them when subjected to shear forces. We have also determined that the addition of CMC, a weakly binding, semiflexible ionic polymer and mild detergent, to the surfactant/ nanotube system helped increase carbon nanotube dispersion and was a necessary condition for the onset of the orientation of the carbon nanotubes in the polymer solutions, provided the shear stresses that developed in the system were sufficiently high. Figure 6. Raman spectra of samples containing SWNT that were subjected to shear flow: (a) Orientation-dependent Raman spectra of SWNT with different angles (from 0 to 90°) between the polarization of the incidence laser light and the nanotube axis using VV (parallel polarization of the incidence and scattered light) configuration. (b) The direct comparison of the experimental relative intensities of the 1594 cm-1 G band as a function of the angle of polarization of the incident radiation, with theoretical calculations.
light) configuration. The G band at 1594 cm-1, corresponding to the resonantly excited metallic SWNT, shows a maximum intensity when the polarization of the incident radiation is parallel to the nanotube axis (i.e., R ) 0°) and is minimal when the polarization of the incident radiation is perpendicular to the nanotube axis (i.e., R ) 90°).52-56 The Raman spectra allow a direct comparison of experimental data with theoretical calculations, as shown in Figure 6b. The experimental angular dependencies in our system exhibit a nonnegligible deviation from the selection rules predicted by theoretical studies, where intensities scale with cos4 R.55 These differences may be attributed to depolarization effects generated by the pronounced anisotropic
Acknowledgment. This work was supported by the Air Force/ Bolling AFB/DC MURI project on Energetic Structural Materials, Grant No. F49620-02-1-0382. E.C. was supported by an IPSTGeorgia Institute of Paper Science and Technology Graduate Fellowship, and R.V. was supported by the Summer Undergraduate Research Fellowship at Georgia Tech through the NSFREU program Grant No. DMR-0139081. The authors also thank Mr. Michael Valikovsky for his help with the calculations of the TEM scale bars. LA052714Z (50) Smyth, S. F.; Liang, C.-H.; Mackaya, M. E.; Fuller, G. G. J. Rheol. 1995, 39 (4), 659-672. (51) Eklund, P. C.; Holden, J. M.; Jishi, R. A. Carbon 1995, 33 (7), 959-972. (52) Liu, T.; Kumar, S. Chem. Phys. Lett. 2003, 378, 257-262. (53) Wood, J. R.; Zhao, Q.; Wagner, H. D. Composites A 2001, 32, 391-399. (54) Frogley, M. D.; Zhao, Q.; Wagner, H. D. Phys. ReV. B 2002, 65 (11), 113413. (55) Duesberg, G. S.; Loa, I.; Burghard, M.; Syassen, K.; Roth, S. Phys. ReV. Lett. 2000, 85 (25), 5436. (56) Bhattacharyya, A. R.; Sreekumar, T. V.; Liu, T.; Kumar, S.; Ericson, L. M.; Hauge, R. H.; Smalley, R. E. Polymer 2003, 44, 2373-2377.
