J. Phys. Chem. C 2007, 111, 1929-1934
1929
Number of Walls Controlled Synthesis of Millimeter-Long Vertically Aligned Brushlike Carbon Nanotubes Supriya Chakrabarti,† Hideki Kume,‡ Lujun Pan,§ Takeshi Nagasaka,| and Yoshikazu Nakayama*,§,⊥ Osaka Science and Technology Center, InnoVation Plaza Osaka, 3-1-10 Techno Stage, Izumi, Osaka 594-1144, Japan, Technology Research Institute of Osaka Prefecture, 2-7-1 Ayumino, Izumi, Osaka 594-1157, Japan, Department of Physics and Electronics, Graduate School of Engineering, Osaka Prefecture UniVersity, 1-1 Gakuen-cho, Naka-ku, Sakai, Osaka 599-8531, Japan, Taiyo Nippon Sanso Corporation, Toyo Building, 1-3-26, Koyama, Shinagawa-ku, Tokyo 142-8558, Japan, and Department of Mechanical Engineering, Osaka UniVersity, 2-1 Yamada-oka Suita, Osaka 565-0871, Japan ReceiVed: October 12, 2006; In Final Form: NoVember 22, 2006
Millimeter (mm) long vertically aligned carbon nanotubes (CNTs) were grown by the catalyst assisted thermal chemical vapor deposition (CVD) technique. The continuous growth of CNTs as long as 7 mm was observed after 12 h of deposition by adjusting the growth parameters for making the catalyst active for a long time. The direct dependence of the number of walls of mm-long CNTs on the Fe catalyst thickness was observed. The successful syntheses of single-walled nanotubes (SWNTs), double-walled nanotubes (DWNTs), and multiwalled nanotubes (MWNTs) with high percentages (∼80%) were achieved by varying the catalyst layer thickness. The effect of Al2O3 buffer layer was found to be critical for this controlled synthesis, which has been discussed in detail. The possible growth mechanism is also discussed to better understand this phenomenon.
Introduction Carbon nanotubes (CNTs) are currently the focus of intense research due to their unique properties and potential to impact broad areas of science and technology.1-4 The special characteristics of carbon nanotubes arise from their atomic structures, number of walls, diameters, and lengths. For example, CNTs can be either metals or semiconductors depending on chiralities and diameters.5,6 The alignment, diameter, and number of walls of a CNT significantly affect its mechanical and electrical properties, and thus can impact a wide range of applications such as probe microscopy tips,7,8 field emission devices,9,10 field effect transistors,11 fuel cells,12,13 electromechanical devices,14 and structural composites.15,16 Recently, new technology based on the properties of very long aligned CNTs has been developing rapidly. The long nanotubes can be spun into ropes/fibers2,17 that are much stronger than any current structural materials. This will allow revolutionary advances in lightweight and high-strength applications. Intense research efforts have been undertaken to synthesize aligned long CNTs;18-21 nevertheless, many limitations to the synthesis of very long aligned CNTs remain. Recently, the Iijima group succeeded in growing 2.5-mm-long aligned single-walled carbon nanotubes of high purity using the water-assisted chemical vapor deposition (CVD) technique.22 For the synthesis of very long aligned CNTs by CVD technique, the catalyst activity and lifetime are very important factors. The coating of the catalyst particles by amorphous carbon during * Corresponding author. Telephone: +81-6-6879-7307. Fax: +81-66879-7307. E-mail:
[email protected]. † Osaka Science and Technology Center. ‡ Technology Research Institute of Osaka Prefecture. § Osaka Prefecture University. | Taiyo Nippon Sanso Corp. ⊥ Osaka University.
CVD reduces the catalyst activity and stops the growth of CNTs. The successful synthesis of very long CNTs can be achieved if the catalyst activity and lifetime can be enhanced.22,23 The control of the diameter and number of walls of CNTs represents one of the most basic issues in developing nanotube growth methods. There are some reports that the size of the catalyst used in thermal catalytic CVD can define the diameter of as-grown nanotubes.24-26 This hypothesis has been supported by the observation that catalytic particles at the ends of CVDgrown CNTs have sizes commensurate with the nanotube diameters.27,28 The direct growth of milllimeter (mm) long CNTs with a tunable number of walls by adjusting the catalyst thickness can be very promising due to its simplicity and the possibility of wide applications, which have not been yet reported. Accordingly, the developments of synthesis techniques that can control the length, diameter, alignment, and number of walls of the nanotubes have become the most important part of intense research. In the present study, we report the successful synthesis of mm-long aligned CNTs with controlled numbers of walls by the water-assisted22,23 catalytic thermal CVD technique. Experimental Methods The millimeter-long aligned CNTs were synthesized using a conventional thermal catalytic CVD technique. CNT growth was performed in a single-zone atmospheric pressure quartz tube furnace having the inner diameter of 26 mm. High-purity ethylene (99.5%) was used as the source gas of carbon. Helium (99.9999%) was used as a carrier gas, and H2 (99.9999%) was introduced with He at 1 atm pressure. A controlled amount of water vapor with the concentration of 350 ppm was also introduced with the carrier gas during the deposition process. The small and controlled amount of water vapor was supplied from the water bubbler, and the schematic diagram of the water
10.1021/jp0666986 CCC: $37.00 © 2007 American Chemical Society Published on Web 01/18/2007
1930 J. Phys. Chem. C, Vol. 111, No. 5, 2007
Chakrabarti et al.
Figure 1. SEM images of mm-long aligned brushlike CNTs synthesized by using Fe catalyst of different thicknesses: (a) 0.5 nm Fe layer, (b) 1.0 nm Fe layer, and (c) 1.5 nm Fe layer. The insets show the corresponding high magnification SEM images of aligned CNT strands. (d) Digital photographic image of ∼7 mm long aligned brushlike CNTs deposited for 12 h by using 1.0 nm thick Fe catalyst layer.
introduction system is given in Figure S1 of the Supporting Information. The water concentration was measured by using a moisture analyzer (GE Panametrics; Model No. MMS35-3111-100). The optimized concentration of water vapor was determined by studying the dependence of the height of CNTs on different water concentrations (4-500 ppm) during CVD, and the plot of water concentration vs CNT height is given in Figure S2 of the Supporting Information. Typical CVD growth was carried out at 750 °C for 30 min as the standard growth time. During heating of the reactor, a steady flow of 120 sccm helium in two streams (105 and 15 sccm) and 80 sccm H2 was maintained into the chamber. After the growth temperature (750 °C) was attained, during CNT deposition the 15 sccm He stream was switched to 15 sccm ethylene by keep flowing the other 105 sccm He stream with 80 sccm H2; 350 ppm water vapor was also introduced. This quick exchange of helium gas by ethylene rapidly increased the concentration of the source gas reaching the substrate and produced a very high growth rate of CNTs during the first few seconds of deposition.23 The total gas flow rate throughout the deposition process was maintained at 200 sccm. The optimum CVD condition was determined by balancing the relative levels of ethylene and water, as well as those of ethylene and H2. This sharp optimum condition was achieved for the growth of super-long aligned brushlike CNTs by studying the dependence of the height of CNTs on ethylene-to-water and ethylene-to-H2 ratios, which has been reported earlier.23 In contrast to the report of Hata et al.,22 where the catalyst was active only for 30 min, we achieved very long time (ca. 12 h) catalyst activity by optimizing the CVD parameters very accurately, which resulted in the growth of aligned CNTs as long as 7 mm. Pieces of Si wafers (2 cm × 2 cm) with 100 nm thick SiO2 layer were used as the substrate for the growth of mm-long CNTs. The Al2O3 layer of different thicknesses (10, 20, and 30 nm) was used as a buffer layer between the Si/SiO2 substrate and the catalyst. The Al2O3 layer was deposited on the substrate on which three different thicknesses (0.5, 1.0, and 1.5 nm) of Fe catalyst layer were deposited to grow CNTs. Both the catalyst and buffer layers were deposited by a sputtering technique. The
sputtering was normally for few seconds (3 s for 0.5 nm, 6 s for 1.0 nm, and 9 s for 1.5 nm thick Fe catalyst) to achieve catalyst thicknesses. The critical point of our study is to get control over the number of walls of aligned brushlike mm-long CNTs. To achieve this target, the CNTs were grown by using Fe layers of different thicknesses. The variation of Fe catalyst thickness was found to be most effective for controlling the number of walls of the aligned CNTs. The surface morphology of the Al2O3 buffer layer was observed by atomic force microscopy (SPI3800N (Seiko Instruments Inc.)). The microstructures of the deposited nanotubes were studied by scanning electron microscopy (Hitachi; S-4300) and high-resolution transmission electron microscopy (Hitachi; HF-2000). Raman spectroscopy of the synthesized CNTs was performed by using a monochromator (250IS_3) equipped with a detection device (cooling CCD camera DV401-FIS). The 632.8 nm laser line of a He-Ne laser (Seki Technotron Corp.) was used for excitation with an output power of 15 mW. Results and Discussion The experimental results showed that the synthesis of mmlong or super-long aligned CNTs is very sensitive to the CVD parameters. The small amount of water vapor during CVD plays a critical role in synthesizing long aligned CNTs. The water vapor acts like a weak oxidizer and helps to remove the amorphous carbon deposited on the catalyst surface during CVD. The deposition of amorphous carbon layer on the catalyst surface reduces the activity and lifetime of the catalyst; as a result long continuous CNTs cannot be synthesized.22,23 By optimizing the CVD parameters and the water concentration, it is possible to grow mm-long or super-long CNTs with high reproducibility. Scanning electron microscope (SEM) images of the aligned CNTs deposited for 30 min by using different thicknesses of Fe catalyst layer (0.5, 1.0, and 1.5 nm) are shown in Figure 1a-c. Fe catalyst of 0.2 nm was also used to grow aligned long CNTs; however, no aligned nanotube was observed. Due to the very low thickness of the catalyst layer, the density of the catalyst particles was too low to grow aligned CNTs. Again, for higher thicknesses (2-10 nm) of Fe catalyst, the growth of
Synthesis of mm-Long Vertically Aligned CNTs
Figure 2. Plot of variation of growth rate of CNTs with time for 12 h of deposition.
mm-long aligned CNTs was barely found because during annealing the formation of big nanoclusters prevented the growth of long nanotubes. For the same deposition time a variation in the height of deposited CNTs was observed for different catalyst thicknesses, which indicates that the growth rate of CNTs is slightly different for different thicknesses of the Fe catalyst layer. The SEM image of CNTs deposited by using a 0.5 nm thick Fe layer is shown in Figure 1a, which indicates aligned high-density CNTs with an average height of 0.93 mm. The average growth rate is ∼0.52 µm/s. In the case of the aligned CNTs grown from a 1.0 nm thick Fe layer, the average height is 1.13 mm (Figure 1b). Here the average growth rate is ∼0.63 µm/s. Figure 1c shows the aligned CNTs grown from a 1.5 nm thick Fe layer, and the average height is 1.09 mm with an average growth rate of ∼0.60 µm/s. The maximum growth rate (0.63 µm/s) of CNTs was obtained for a 1 nm thick Fe catalyst layer. The different catalyst thicknesses lead to different sizes of catalyst particles. The activity of the catalyst seems to depend on the active catalyst surface area which is commensurate with the catalyst particle size. The slight variation of the growth rate of CNTs with catalyst layer thicknesses can be attributed to the variation of catalyst activity depending on the different catalyst thicknesses. The insets of Figure 1a-c show the high magnification SEM images of vertically aligned high-density CNTs, which reveal that the CNTs are arranged in strands consisting of closely packed CNTs and that there are significant interconnections between nearby strands. The continuous growth of CNTs was observed for 12 h and the height was 7 mm, and the corresponding photograph is shown in Figure 1d. The growth rate of CNTs was found to vary for long-time deposition. The plot of variation of growth rate with time for 12 h deposition is shown in Figure 2. The plot indicates gradual decrease in the growth rate of CNTs with time. The growth rate is directly related to the catalyst activity during deposition. The balance between the rate of amorphous carbon deposition on the catalyst surface and removal of that amorphous carbon layer by water vapor and H2 is the factor keeping the catalyst active for a long time. However, with increasing time the rate of amorphous carbon deposition increases compared to the rate of removal, which results in the decrease in the catalyst activity; the growth rate decreases followed by saturation and finally the growth terminates due to the termination of catalyst activity. No metal clusters were observed at the tip and the side body of the CNTs by thorough transmission electron microscope (TEM) observations which indicated that the CNTs were grown by a base growth mechanism.29 In this experiment, it was observed that the Al2O3 buffer layer significantly contributes to the synthesis of mm-long aligned CNTs. Without the Al2O3 layer no deposition of mm-long aligned CNTs was observed. The buffer layer helps to increase the efficiency of the CNT growth process by avoiding undesired
J. Phys. Chem. C, Vol. 111, No. 5, 2007 1931 chemical interaction between the catalyst and the substrate.32 It also alters the catalyst-support interactions, therefore modifying the characteristics and growth rate of the CNTs. The effect of different buffer layers between the catalyst and the substrate in the growth of CNTs by CVD has been studied widely.30,31 Among many buffer layers to grow CNTs with high yield and low diameter, an Al2O3 layer was found to be the most effective.33 It can enhance the decomposition rate of hydrocarbons34 on its rough surface. Furthermore, it prevents the interdiffusion of the catalyst metal into the substrate, by which the formation of metal silicide (in case of Si substrate) can be avoided. Well-dispersed catalyst nanoparticles can be formed by using an Al2O3 buffer layer between the substrate and the catalyst layer. Atomic force microscope (AFM) images of the surface of 10 nm thick Al2O3 buffer layer before and after annealing in H2 and He at 750 °C for 30 min are shown in parts a and c, respectively, of Figure 3. Figure 3a shows the grainlike morphology of the Al2O3 surface with an average surface roughness of ∼3 nm. Figure 3c indicates that, after annealing in H2 and He, the surface became more rough with (average surface roughness ∼ 60 nm) many holes (traps) on the surface. The large area surface morphology of the Al2O3 surface after annealing in H2 and He is shown in the inset of Figure 3c. Figure 3b shows the AFM image of the Al2O3 buffer layer after annealing in He only at 750 °C for 30 min. Here (in Figure 3b) also the grainlike morphology similar to Figure 3a can be clearly observed with an average roughness of ∼5 nm. By comparing Figure 3a-c, it can be mentioned that the annealing in H2 affected the Al2O3 surface very much and the surface roughness increased significantly due to the etching effect of H2 at high temperature. It seems that upon heating to CNT growth temperature the Fe nanocluster formation occurred from the thin Fe layer. These particles were then trapped in the holes of the Al2O3 surface, and as a result agglomeration of the nanoclusters could not occur. The shape of the metal clusters was also guided by the dip traps of the Al2O3 surface, and the appropriate shape of the nanoclusters is very important for growing high-quality CNTs, especially for single-walled nanotubes (SWNTs).35 The surface morphology of the Fe-coated Al2O3 surface was also studied by AFM; however, no Fe nanoclusters were distinguished by the AFM images due to their very small size. The growth of CNTs was performed by varying the Al2O3 buffer layer thickness. Millimeter-long aligned CNTs were also obtained by using 20 and 30 nm thick Al2O3 layers. The variation of CNT height with Al2O3 thickness for different catalyst thicknesses is shown in Figure 4. It can be clearly observed that no significant variation of CNT height occurred for thicker (20 and 30 nm) Al2O3 layers. The X-ray diffraction studies (shown in Figure S3 of the Supporting Information) of the Al2O3 buffer layer and the Fe catalyst coated Al2O3 layer indicated the amorphous structure of Al2O3, and also no signal for the Fe catalyst was observed in the spectra due to its extremely low thickness. This structurally defective amorphous Al2O3 layer allows high dispersion of the Fe catalyst into and along the surface of the substrate, resulting in a very fine dispersion and high surface area of the catalyst.36 Also, the high surface roughness improves the diffusion of reactant gas to the catalyst clusters, which favors the growth of CNTs. The proposed mechanism of CNT growth in our case is schematically presented in Figure 5, which summarizes the most important and relevant steps for the growth of vertically aligned CNTs.
1932 J. Phys. Chem. C, Vol. 111, No. 5, 2007
Chakrabarti et al.
Figure 4. Plot of CNT height vs Al2O3 thickness for different catalyst thicknesses.
Figure 3. AFM images of the surface of Al2O3 buffer layer (a) just after coating by sputtering technique, (b) after annealing in He only for 30 min at 750 °C, and (c) after annealing in He and H2 for 30 min at 750 °C (same condition as CVD condition of mm-long aligned CNT growth). The inset in (c) shows the large area AFM image.
Here, the first step of the growth of CNTs is the formation of nanocluster of Fe metal, where the size and shape of the nanoclusters are guided by the traps (holes) of the Al2O3 buffer layer. Then the Fe nanoclusters are supersaturated with carbon37 by diffusion and dissolution of C atoms on the nanocluster surface. The growth process is driven by a supersaturation of carbon-metal solution followed by carbon segregation and emergence of carbonaceous structures from the catalyst surface.38 The nature of these structures, ranging from amorphous carbon to SWNTs, double-walled nanotubes (DWNTs), and multiwalled nanotubes (MWNTs), depends on the size of the Fe catalyst nanoclusters and specific thermodynamic conditions. Figure 6a shows a representative high-resolution TEM image of aligned mm-long SWNTs deposited by using a 0.5 nm thick
Figure 5. Schematic diagram describing the growth mechanism of aligned CNTs by CVD and effect of Al2O3 buffer layer on the catalyst layer.
Fe catalyst layer. After a thorough observation by TEM it is confirmed that most of the nanotubes are single-walled nanotubes with inner diameters between 2 and 3 nm. The statistical calculation indicates that about 80% of the nanotubes are singlewalled and the rest are a mixture of DWNTs and MWNTs. The TEM image of CNTs grown from a 1.0 nm Fe catalyst layer is shown in Figure 6b, which indicates the presence of DWNTs. Here also detailed TEM study indicates that most of the nanotubes (∼80%) are DWNTs with inner diameters between 3 and 4 nm and the rest are MWNTs and a very low percentage of SWNTs. The TEM observation of CNTs grown from a 1.5 nm thick Fe catalyst layer indicated the presence of MWNTs (varied in the range of three to six walls) with inner diameters between 3 and 5 nm. The representative TEM image is shown in Figure 6c, and the inset shows one carbon nanotube with three walls. In this case no SWNTs are observed; however, a few percentage DWNTs is observed. As the layer thickness of Fe catalyst increases from 0.5 to 1.5 nm, the number of walls
Synthesis of mm-Long Vertically Aligned CNTs
J. Phys. Chem. C, Vol. 111, No. 5, 2007 1933
Figure 6. TEM images and Raman spectra of CNTs with different numbers of walls synthesized by using Fe catalyst layer of different thicknesses: (a, d) 0.5 nm Fe layer produced SWNTs; (b, e) 1.0 nm Fe layer produced DWNTs; (c, f) 1.5 nm Fe layer produced MWNTs. The inset in (c) shows one carbon nanotube with three walls.
of the synthesized nanotubes increases, which confirms the direct relationship between the number of walls and catalyst thickness. The Fe nanocluster size is determined by the reaction parameters such as the temperature, gas flow rate, and pressure. To eliminate the variations of CVD parameters and create a growth environment in which the film thickness is the sole variable, all the CVD parameters were kept constant for every thickness of the catalyst layer. This confirms that the size of the Fe nanocluster is strongly dependent on the thickness of the deposited Fe catalyst layer. It seems that the number of walls of mm-long CNTs is dependent on the size of the Fe nanoclusters. Some previous studies25,26,39 reported the relation between the size of catalyst nanocluster and the thicknesses of the deposited catalyst layer, which also support our experimental findings. The Fe nanoclusters are produced before the formation of CNTs, and their size depends on the Fe layer thickness deposited on the substrate. The dispersion and the shape of the Fe nanoclusters depend on the effect of the Al2O3 buffer layer. These results thus suggest that it should be possible to control the number of walls of mm-long aligned CNTs by controlling the thickness of the catalyst layer. Raman spectroscopy has been applied to all the samples. The Raman spectra of SWNTs, DWNTs, and MWNTs are shown in parts d, and e, and f, respectively, of Figure 6. All the spectra indicate the presence of G-band (at ∼1590 cm-1) and D-band (at ∼1320 cm-1) with different G-band/D-band ratios. The G-band is representative of the amount of graphitization associated with the nanotube growth, while the D-band represents the amount of defects (open ends, disorder, amorphous
deposit, etc.). For the sample where 80% of the nanotubes are single-walled, the G-band/D-band ratio is about 4.68. The G-band/D-band ratios of DWNT-rich sample and MWNT-rich sample are 2.68 and 2.44, respectively. The G-band/D-band ratio can be used as a rough measure of carbon nanotube sample quality, because it is the relative response of graphitic carbon to defective carbon from intrinsic defects in the CNTs or amorphous carbon on the CNT surfaces. Figure 6d-f reveals that the G-band/D-band ratio decreases with increasing number of walls and clearly indicates that the defect content increases with increasing number of walls. The radial breathing mode (RBM) at ∼189 cm-1 in the Raman spectrum (Figure 6d) was found only for the CNTs grown from the 0.5 nm Fe catalyst layer. The RBM in the Raman spectra and higher G-band/Dband ratio (∼4.68) also support the presence of SWNTs as observed by TEM (Figure 6a). However, the nanotube diameter (∼1.3 nm) calculated from the RBM frequency does not match well with the TEM observation. It seems that the CNTs with very small diameters tend to be bundled and are very difficult to be well-dispersed for TEM observation. Conclusions In summary, well-aligned brushlike high-density mm-long carbon nanotubes were synthesized successfully. By optimizing the CVD parameters, it was possible to synthesize CNTs with lengths of up to 7 mm after 12 h of deposition. Precise control of the gas parameters (flow rates of hydrogen, water, and ethylene) prevented the accumulation of amorphous carbon on the Fe catalyst surface. This increased the lifetime of the catalyst
1934 J. Phys. Chem. C, Vol. 111, No. 5, 2007 and prolonged the growth time of the CNTs. We have shown that different thicknesses of the Fe catalyst layer can be used to define the number of walls of the mm-long aligned CNTs. The direct dependence of the number of walls of the nanotubes on the catalyst thickness has been demonstrated in detail. The effect of the Al2O3 buffer layer was found to be most effective to gain control over the shape and size of the Fe nanocluster, which is a very important parameter to control the number of walls of the nanotubes. The possible growth mechanism has been discussed for better understanding of this phenomenon. The well-controlled synthesis of mm-long SWNTs, DWNTs, and MWNTs with a high percentage (∼80%) has been achieved, which will be promising for future technological developments. Acknowledgment. This work was carried out for the Osaka Prefecture Collaboration of Regional Entities for the Advancement of Technological Excellence, JST. Supporting Information Available: Schematic diagram of water introduction part with total gas flow system (Figure S1), plot of the variation of CNT height with water concentration for all thicknesses of Fe catalyst (Figure S2), and X-ray diffraction patterns of 10 nm Al2O3 buffer layer and 1 nm thick Fe catalyst (Figure S3). This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Choi, H. C.; Kundaria, S.; Wang, D.; Javey, A.; Wang, Q.; Rolandi, M.; Dai, H. Nano Lett. 2003, 3, 157. (2) Zhang, M.; Fang, S.; Zakhidov, A. A.; Lee, S. B.; Aliev, A. E.; Williams, C. D.; Atkinson, K. R.; Baughman, R. H. Science 2005, 309, 1215. (3) Jarillo-Herrero, P.; VanDam, J. A.; Kouwenhoren, L. P. Nature 2006, 439, 953. (4) Teo, K. B. K.; Minoux, E.; Hudanski, L.; Peauger, F.; Schnell, J.; Gangloff, L.; Legagneux, P.; Dieumegard, D.; Amaratunga, G. A. J.; Milne, W. I. Nature 2005, 437, 968. (5) Odom, T. W.; Huang, J.-L.; Kim, P.; Lieber, C. M. Nature 1998, 391, 62. (6) Wildoer, T. W.; Venema, L. C.; Rinzler, A. G.; Smalley, R. E.; Dekker, C. Nature 1998, 391, 59. (7) Nishijima, H.; Kamo, S.; Akita, S.; Nakayama, Y.; Hohmura, K. I.; Yoshimura, S. H.; Takeyasu, K. Appl. Phys. Lett. 1999, 74, 4061. (8) Akita, S.; Nakayama, Y.; Mizooka, S.; Takano, Y.; Okawa, T.; Miyatake, Y.; Yamanaka, S.; Tsuji, M.; Nosaka, T. Appl. Phys. Lett. 2001, 79, 1691. (9) Tanaka, H.; Akita, S.; Pan, L.; Nakayama, Y. Jpn. J. Appl. Phys. 2004, 43, L197. (10) Fan, S.; Chapline, M. G.; Franklin, N. R.; Tombler, T. W.; Cassell, A. M.; Dai, H. Science 1999, 283, 512.
Chakrabarti et al. (11) Misewich, J. A.; Martel, R.; Avouris, P.; Tsang, J. C.; Heinze, S.; Tersoff, J. Science 2003, 300, 783. (12) Li, W.; Wang, X.; Chen, Z.; Waje, M.; Yan, Y. J. Phys. Chem. B 2006, 110, 15353. (13) Wang, X.; Li, W.; Chen, Z.; Waje, M.; Yan, Y. J. Power Sources 2006, 158, 154. (14) Sazanova, V.; Yaish, Y.; Ustu¨nel, H.; Roundy, D.; Arias, T. A.; McEuen, P. L. Nature 2004, 431, 284. Suekane, O.; Nagataki, A.; Nakayama, Y. Appl. Phys. Lett. 2006, 89, 183110. (15) Suhr, J.; Koratkar, N.; Keblinski, P.; Ajayan, P. Nat. Mater. 2005, 4, 134. (16) Hinds, B. J.; Chopra, N.; Rantell, T.; Andrews, R.; Gavalas, V.; Bachas, L. G. Science 2004, 303, 62. (17) Li, Y.; Kinloch, I. A.; Windle, A. H. Science 2004, 304, 276. (18) Christen, H. M.; Puretzky, A. A.; Cui, H.; Belay, K.; Fleming, P. H.; Geohegan, D. B.; Lowndes, D. H. Nano Lett. 2004, 4, 1939. (19) Li, W. Z.; Xie, S. S.; Qian, L. X.; Chang, B. H.; Zou, B. S.; Zhou, W. Y.; Zhao, R. A.; Wang, G. Science 1996, 274, 1701. (20) Xiong, G.; Wang, D. Z.; Ren, Z. F. Carbon 2006, 44, 969. (21) Hurt, A. J.; Slocum, A. H. J. Phys. Chem. B 2006, 110, 8250. (22) Hata, K.; Futaba, D. N.; Mizuno, K.; Namai, T.; Yumura, M.; Iijima, S. Science 2004, 306, 1362. (23) Chakrabarti, S.; Nagasaka, T.; Yoshikawa, Y.; Pan, L.; Nakayama, Y. Jpn. J. Appl. Phys. Exp. Lett. 2006, 45, L720. (24) Dai, H.; Rinzler, A. G.; Nikolaev, P.; Thess, A.; Colbert, D. T.; Smalley, R. E. Chem. Phys. Lett. 1996, 260, 471. (25) Chopra, N.; Hinds, B. Inorg. Chim. Acta 2004, 357, 3920. (26) Wei, Y. Y.; Eres, G.; Merkulov, V. I.; Lowndes, D. H. Appl. Phys. Lett. 2001, 78, 1394. (27) Anderson, P. E.; Rodriguez, N. M. Chem. Mater. 2000, 12, 823. (28) Sinnott, S. B.; Andrews, R.; Quian, D.; Rao, A. M.; Mao, Z.; Dicky, E. C.; Derbyshire, F. Chem. Phys. Lett. 1999, 315, 25. (29) Dresselhaus, M. S.; Dresselhaus, G.; Avouris, P. Carbon nanotubes: Synthesis, Structure, Properties and Applications; Springer: New York, 2001; Vol. 80. (30) De los Arcos, T.; Garnier, M. G.; Seo, J. W.; Oelhafan, P.; Thommen, V.; Mathys, D. J. Phys. Chem. B 2004, 108, 7728. (31) Kong, J.; Soh, H. T.; Cassel, A. M.; Quate, C. F.; Dai, H. Nature 1998, 395, 878. (32) De los Arcos, T.; Vonau, F.; Garnier, M. G.; Thommen, V.; Boyen, H.-G.; Oelhafen, P.; Du¨ggelin, M.; Mathis, D.; Guggenheim, R. Appl. Phys. Lett. 2002, 80, 2383. (33) De los Arcos, T.; Wu, Z. M.; Oelhafen, P. Chem. Phys. Lett. 2003, 380, 419. (34) Vander Wal, R. L.; Ticich, T. M.; Cartis, V. E. Carbon 2001, 39, 2277. (35) Seidel, R.; Duesberg, G. S.; Unger, E.; Graham, A. P.; Leibau, M.; Freupl, F. J. Phys. Chem. B 2004, 108, 1888. (36) Ward, J. W.; Wei, B. Q.; Ajayan, P. M. Chem. Phys. Lett. 2003, 376, 717. (37) Ding, F.; Rosen, A.; Bolton, K. Phys. ReV. B 2004, 70, 075416. (38) Raty, J.-Y.; Gygi, F.; Galli, G. Phys. ReV. Lett. 2005, 95, 0961031. (39) Hoffmann, S.; Cantoro, M.; Kleinsorge, B.; Casiraghi, C.; Parvez, A.; Robertson, J.; Ducati, C. J. Appl. Phys. 2005, 98, 034308-1.
