Growth and Structure of Carbon Nanotube Y-Junctions - The Journal

Molecular simulations of the carbon nanotubes intramolecular junctions under mechanical loading. Sanjib C. Chowdhury , Bazle Z. (Gama) Haque , John W...
0 downloads 0 Views 535KB Size
Download PDF
23694

J. Phys. Chem. B 2006, 110, 23694-23700

Growth and Structure of Carbon Nanotube Y-Junctions W. Z. Li* and B. Pandey Department of Physics, Florida International UniVersity, Miami, Florida 33199

Y. Q. Liu AdVanced Materials and Engineering Research Institute, Florida International UniVersity, Miami, Florida 33174 ReceiVed: July 5, 2006; In Final Form: September 17, 2006

The effect of a catalyst on the growth and structure of carbon nanotube Y-junctions (CNTYs) using chemical vapor deposition has been investigated. Cobalt-, magnesium-, and calcium-nitrates are utilized as precursors of catalysts Co, Mg, Ca, Co/Mg, Co/Ca, and Mg/Ca for CNTY synthesis. Experimental result shows that Co/Mg or Co/Ca can grow CNTYs with straight branches while Co, Mg, Ca, and Mg/Ca will not grow any CNTYs, indicating that only combinations of Co with Mg or Ca will facilitate the formation of CNTYs. In addition, the effect of the carbon source on the formation of CNTYs has also been studied. It is found that thiophene (C4H4S) can promote the formation of CNTYs, while other sources such as methane (CH4) and acetylene (C2H2) cannot. The result shows that both the catalyst and the carbon source substantially affect the formation of CNTYs.

Introduction Since the discovery of carbon nanotubes (CNTs),1 there has been a great deal of interest in this type of materials due to their potential applications as electron emitters, biological and chemical sensors, quantum wires, field effect transistors, etc.2-8 To construct nanodevices, a formidable challenge is to interconnect individual CNTs at atomic levels to form functional circuits. As a self-assembled architecture of straight carbon nanotubes, carbon nanotube Y-junctions (CNTYs) may provide a short cut for wide applications of CNTs in electronic device design and fabrication. Theoretical and experimental studies show that CNTYs have possible rectifying and amplifying behaviors.9-13 They may be used not only as a connection of linear tubes but also as new type of diode and triode transistor. Undoubtedly, a successful and controllable synthesis of CNTYs is highly desirable in the development of novel nanotube-based electronic devices. After the discovery of CNTYs from carbon arc-discharge deposits,14 various techniques including chemical vapor deposition (CVD) have been developed for synthesizing CNTYs.15-21 In the CVD technique, the growth of CNTYs could depend on several factors such as the catalysts, the type of carbon source, the growth temperature, etc. To date, it is not understood which factor (or combination of several factors) plays a major role in the growth of CNTYs. In this paper, we report the effect of the catalyst and carbon source on the synthesis and structure of CNTYs grown by CVD. Catalysts are prepared by simply mixing cobalt nitrate, magnesium nitrate, and calcium nitrate with desired weight ratios. Various carbon sources, including thiophene (C4H4S), methane (CH4), and acetylene (C2H2), are tested for the growth of CNTYs. It is found that both the catalyst composition and the carbon source strongly affect the formation of CNTYs. The * Corresponding author. E-mail: [email protected].

findings should be a valuable addition to the database of CNTY research, which is necessary to understand the growth mechanism of CNTYs and to control their synthesis. Experimental Procedures The CVD method was adapted to synthesize CNTYs on catalysts prepared by mixing nitrates Co(NO3)2‚6H2O (SigmaAldrich 98+%), Mg(NO3)2‚6H2O (Sigma-Aldrich 99%), and Ca(NO3)2‚4H2O (Sigma-Aldrich 99%) in different weight ratios. The mixtures were melted in glass vials placed in a muffle furnace at a temperature of 130 °C for 12 h to achieve uniform solution. About 1 g of the molten mixture was transferred to a clean quartz boat, which was inserted into a quartz tube reaction chamber. The reaction chamber was pumped down to ∼10-3 Torr prior to raising the temperature to 1000 °C. Hydrogen gas (H2, 100 sccm) was introduced into the reaction chamber for 10 min to reduce the catalyst from its oxide base to metal base. Then H2 gas was switched to bubble through thiophene (C4H4S, liquid) for nanotube growth. During the growth, the pressure inside the reaction chamber was adjusted to 200 Torr, and the growth lasted for 10 min. After the growth, the chamber was pumped down to 10-3 Torr and cooled to room temperature. To investigate the role of the carbon source, besides C4H4S, different carbon precursors, such as CH4 and C2H2, were also tested for growing CNTYs. Scanning electron microscopy (SEM, JEOL JSM-6330F) was employed to examine the morphology and fraction of carbon nanotubes in the synthesized materials. Transmission electron microscopy (TEM, Philips CM 200) was used to characterize the microstructure of the nanotubes. The specimens for TEM analysis were prepared by dispersing the synthesized samples in ethanol followed by sonication for 10 min. A few drops of the suspension were dripped onto a microgrid covered with a holey carbon thin film.

10.1021/jp064233+ CCC: $33.50 © 2006 American Chemical Society Published on Web 10/31/2006

Growth and Structure of Carbon Nanotube Y-Junctions

J. Phys. Chem. B, Vol. 110, No. 47, 2006 23695

Figure 1. SEM images of CNTYs grown on a catalyst of Co/Mg with a ratio of 20 wt %. (a) High yield of multiple branched CNTYs. (b) CNTY dispersed on silicon substrate, showing its neat surface and multiple-branch structure.

Results and Discussion Growth and Structure of CNTYs on Co/Mg and Co/Ca Catalysts. Initially, a mixture of cobalt nitrate and magnesium nitrate was prepared with a ratio of Co to Mg of 20 wt % and was used as a catalyst (denoted as Co/Mg catalyst). CNTYs were synthesized by pyrolysis of C4H4S on the Co/Mg catalysts in a hydrogen atmosphere. SEM examination of the synthesized materials shows a high yield of CNTYs with straight branches (Figure 1a). Some CNTYs have multiple branches with lengths of about one hundred nanometers to micrometers, and all the branches in an individual Y-junction have the same diameter. The diameters of the synthesized CNTYs are in the range of 20-80 nm. To observe the entire structure of an isolated CNTY, the sample was dispersed on a silicon substrate. About 5 mg of the synthesized materials was dispersed in 20 mL of dichloromethane (CH2Cl2) by sonicating for 1 h. The solution was put aside for 12 h to let large particles settle down, and then a few drops of the top clear solution were dripped onto a clean silicon substrate. Figure 1b shows a typical isolated multiple CNTY; its branches have lengths of 200 nm to 2 µm and diameters of 60 nm. To study the effect of the composition of Co/Mg catalysts on the formation of Y-junctions, the Co/Mg catalysts with weight ratios of 5, 10, 400, and 2000% were prepared by mixing cobalt nitrate with magnesium nitrate. The growth procedure was the same as that for growing CNTYs on Co/Mg of 20 wt %. Figure 2a-d contains micrographs of the samples grown on the four catalysts. CNTYs have been observed in all of the four samples, and they have straight branches and diameters of 20-300 nm depending on the composition of the catalysts. Multiple CNTYs were occasionally observed. The yield of CNTYs from Co/Mg of 10 wt % (Figure 2b) is higher than that from the other three catalysts. However, the yield of CNTYs from the Co/Mg of 10 wt % (Figure 2b) is still much lower than that from the Co/Mg of 20 wt % (Figure 1a). To quantify the yield of the CNTYs, the numbers of CNTYs and straight CNTs were counted in SEM images. Five different regions of 4 µm2 for each sample were used to count the average numbers of the two types of nanotubes. Figure 3 shows a graph of the average numbers of CNTYs and straight CNTs in an

Figure 2. CNTYs synthesized from Co/Mg catalysts with different weight ratios (a) 5.0%, (b) 10%, (c) 400%, and (d) 2000%.

Figure 3. Graph of the average numbers of CNTYs and straight CNTs in an area of 4 µm2 in each sample vs the weight ratio of Co/Mg catalysts (with error bars). In the sample grown on the catalyst Co/Mg of 20 wt %, the average number of CNTYs is higher than the number of straight ones. If the weight ratio of Co/Mg is increased or decreased from 20%, the number of CNTYs will decrease and the number of straight CNTs will increase.

area of 4 µm2 in different samples. It indicates that the sample synthesized from the catalyst Co/Mg of 20 wt % has the highest yield of CNTYs and lowest yield of straight CNTs. As the

23696 J. Phys. Chem. B, Vol. 110, No. 47, 2006

Li et al.

Figure 4. Graph of average diameter of CNTYs vs Co/Mg catalysts of different weight ratios (with error bars). It shows that CNTYs grown on Co/Mg of 20% have the smallest average diameter.

weight ratio of the Co/Mg increases or decreases, the fraction of CNTYs decreases, and the fraction of straight CNTs increases. These results show that catalyst Co/Mg of 20 wt % is the most effective catalyst among all the tested catalysts for the growth of CNTYs at the described growth conditions. The average diameter of CNTYs in each sample is measured on 25 different CNTYs. It is found that CNTYs synthesized on Co/Mg catalysts with weight ratios of 5, 10, 20, 400, and 2000% have diameters in the range of 34-119, 31-106, 22-98, 52-123, and 35-298 nm, respectively. The average diameter of CNTYs in each sample versus the weight ratio of Co/Mg is plotted in Figure 4. One can see that the average diameter of CNTYs first decreases slightly to a minimum of 63 ( 18 nm at Co/Mg ) 20 wt % and then increases as the weight ratio of Co/Mg increases from 5 to 2000%. CNTYs grown on Co/Mg with weight ratios of 5% 10, 400, and 2000% have average diameters of 78 ( 19, 74 ( 20, 84 ( 18, and 86 ( 21 nm, respectively. The previous results indicate that the ratio of Co to Mg in the catalyst Co/Mg has a dramatic impact on the growth of Y-junctions. CNTYs synthesized using catalysts containing 20 wt % of Co to Mg have the highest yield and smallest average diameter. If the weight ratio of Co to Mg is either increased or decreased from 20 wt %, the yield of Y-junctions will decrease and the average diameter of Y-junctions will increase. It is clear from the previous results that Co/Mg is a reliable catalyst for synthesizing CNTYs. However, it is unknown whether Mg is critical in the formation of CNTYs. To investigate the role of Mg in the growth of CNTYs, calcium nitrate (Ca(NO3)2‚4H2O) is used to replace magnesium nitrate to prepare catalyst Co/Ca with ratios of 10, 20, and 2000 wt %. Then, these catalysts are used to synthesize carbon nanotubes by pyrolysis of thiophene in a procedure similar to that described previously. It is interesting that all the Co/Ca catalysts can grow CNTYs, although the size and yield of the CNTYs from these catalysts are different, as shown in Figure 5. CNTYs grown on Co/Ca of 10 wt % have diameters of about 100 nm (Figure 5a), while the CNTYs grown on Co/Ca of 20 wt % have diameters of about 70 nm (Figure 5b). When Co/Ca of 2000 wt % is used as a catalyst, the synthesized CNTYs have diameters close to 300 nm (Figure 5c). To quantify the yield of nanotubes, five areas of 4 µm2 in SEM images of each sample are selected to count the average numbers of CNTYs and straight CNTs per 4 µm2. Figure 6

Figure 5. SEM images of CNTYs grown on Co/Ca with weight ratios of (a) 10%, (b) 20%, and (c) 2000%.

Figure 6. Graph of the average numbers per 4 µm2 of CNTYs and straight CNTs vs catalyst Co/Ca with different weight ratios (with error bars). For all the catalysts, the number of straight CNTs is higher than that of CNTYs. In other words, straight CNTs are a majority in the samples. Among the three samples, the sample synthesized on Co/Ca of 20 wt % contains the highest number of CNTYs.

shows the average numbers of CNTYs and straight CNTs in three samples grown on different catalysts. In all the samples, the number of straight CNTs is higher than that of CNTYs. The catalyst Co/Ca of 20 wt % yields the highest fraction of CNTYs among all tested Co/Ca catalysts. If the ratio of Co to Ca is either increased or decreased, the number of CNTYs will decrease and the amount of straight CNTs will increase. It should be noted that the yield of CNTYs synthesized by using Co/Ca as a catalyst is lower than that synthesized by using Co/Mg as a catalyst with the same weight ratios.

Growth and Structure of Carbon Nanotube Y-Junctions

Figure 7. Graph of average diameters per 4 µm2 of CNTYs vs Co/Ca catalysts of different weight ratios (with error bars). CNTYs grown on catalyst Co/Ca of 20 wt % have the smallest average diameter. When the weight ratio of Co/Ca either increases or decreases, the average diameter of the CNTYs increases.

As shown in the SEM images (Figure 5), the diameters of CNTYs synthesized on Co/Ca catalysts distribute in a wide range of 50-300 nm. The average diameter of CNTYs has been measured on CNTYs found in five areas of 4 µm2 in each sample. Figure 7 is a graph of the average diameters of the CNTYs versus different Co/Ca catalysts. It shows that the average diameter of the CNTYs grown on the catalyst Co/Mg of 20 wt % is the smallest; its value is 66 ( 15 nm. For catalysts Co/Mg of 10 and 2000 wt %, the average diameters are 81 ( 21 and 92 ( 23 nm, respectively. Since Mg and Ca can be substituted for each other in the catalyst Co/M (M ) Mg or Ca) for synthesizing CNTYs, magnesium and calcium play similar roles in the growth of CNTYs. To further investigate the role of each catalyst component, individual cobalt nitrate, calcium nitrate, and magnesium nitrate as well as a mixture of magnesium nitrate and calcium nitrate are used as catalyst precursors for synthesizing CNTYs as described next. Attempt of Synthesis of CNTYs on Co, Mg, Ca, and Mg/Ca Catalysts. Individual calcium nitrate, magnesium nitrate, and cobalt nitrate and mixtures of calcium nitrate and magnesium nitrate are used as precursors of catalysts to synthesize CNTYs under the same synthesis conditions as described in the first section. When Ca/Mg with weight ratios of 10, 20, and 400% are used as catalysts, only nanoparticles are formed, as shown in Figure 8a. A similar result is obtained when calcium nitrate or magnesium nitrate is used as a catalyst precursor for the synthesis of nanotubes. When Co nitrate alone is used as a catalyst, regular carbon nanotubes and large amorphous carbon particles are formed (Figure 8b), but a Y-junction is not observed in the synthesized materials. From the previous experiments, it is found that CNTYs cannot be synthesized on Co, Mg, Ca, and Ca/Mg catalysts by pyrolysis of thiophene. Cobalt as a catalyst is very important in the formation of CNTYs; however, cobalt itself alone cannot catalyze the growth of CNTYs, and it must be mixed with other metals such as calcium or magnesium to grow Y-junctions. Effect of Carbon Source on the Growth of CNTYs. Reported results indicate that various carbon sources can be used for growing Y-junctions in the CVD method. CH421 and C2H222 have been used as carbon precursors to synthesize CNTYs on pure iron (Fe) powders21 and electrochemically deposited Fe nanoparticles,22 respectively. Fullerene (C60) has

J. Phys. Chem. B, Vol. 110, No. 47, 2006 23697

Figure 8. (a) SEM image of sample synthesized on catalyst Ca/Mg with ratio of 20 wt %, showing the growth of nanoparticles and (b) sample grown on Co catalyst, showing short regular tubes and nanoparticles.

been decomposed on Fe, Ni, and Ti particles to grow Y-junctions.23 It is noteworthy that Y-junctions have also been grown by pyrolysis of acetone on in situ evaporated copper (Cu) catalyst particles in hot-filament CVD.24 The previous results indicate that Y-junctions can be synthesized by using various carbon precursors under the catalysis of certain transition metals. In fact, these carbon precursors have been widely used for synthesizing straight carbon nanotubes, such as single- and multi-walled tubes.7,25-27 It was reported that CNTYs could be formed by vapor phase pyrolysis of metallocenes (e.g., cobaltocene and ferrocene) with thiophene.19,28 Metallocences were used as both catalyst precursor and carbon source, and thiophene was carried into the reaction chamber by hydrogen gas. The experiment was carried out in a two-stage furnace. The metallocenes were first sublimed in the first furnace at 623 K and then carried by argon (Ar) gas into the second furnace at 1273 K. The CNTYs were grown on in situ formed catalytic metal nanoparticles (e.g., Co and Fe), and they were deposited at the inlet and outlet ends of the second furnace. It was expected that thiophene had a catalytic effect on the formation of the CNTYs. In our experiment, cobalt nitrate, magnesium nitrate, and calcium nitrate were used as only catalyst precursors; carbon was from the thiophene. The catalyst precursors were preloaded directly into the single-zone reaction chamber, and the nanotubes were synthesized by pyrolyzing thiophene at 1000 °C and formed on the quartz boat. To reveal the role of thiophene in the formation of Y-junctions in our experiment, non-sulfur-containing hydrocarbon gases such as CH4 and C2H2 were used as carbon precursors to try the growth of CNTYs on Co/Mg and Co/Ca catalysts. First, CH4 was used as a carbon source to synthesize carbon nanotubes on Co/Mg and Co/Ca catalysts by keeping the other growth parameters the same as described in the first section. When Co/Mg with ratios of 20 or 400 wt % was used as a catalyst, the regular carbon nanotubes with small diameters of 10 nm were formed (Figure 9a,b). These thin nanotubes have lengths of up to several tens of micrometers and are intended to form bundles similar to the results reported in refs 29 and 30. We have carefully examined the sample by using SEM, and we have not observed any CNTYs. In addition, when Co/Ca

23698 J. Phys. Chem. B, Vol. 110, No. 47, 2006

Figure 9. (a) Carbon nanotubes synthesized by decomposing CH4 on Co/Mg with a weight ratio of 20%. The synthesized nanotubes are regular nanotubes rather than branched ones, and they intend to form bundles. (b) SEM micrograph of sample synthesized by decomposing CH4 on Co/Mg with a weight ratio of 400%. The nanotubes are regular tubes with a very small diameter and aggregate to form bundles. (c) Sample synthesized by decomposing CH4 on Co/Ca with a weight ratio of 20%, indicating the formation of amorphous particles. (d) Nanotubes grown by using C2H2 as a carbon source on Co/Mg with a weight ratio of 20%. A high yield of regular carbon nanotubes is formed, but CNTY has not been observed.

with a weight ratio of 20% was used, only porous carbon was formed, as shown in Figure 9c. Co/Mg and Co/Ca with other ratios have also been prepared for growing carbon nanotubes by decomposing CH4, and no CNTYs have been observed in the synthesized materials. Second, C2H2 was used as carbon source to grow carbon nanotubes on the catalysts of Co/Mg and Co/Ca with various ratios. Figure 9d shows the materials synthesized on Co/Mg with a ratio of 20 wt %. As one can see, the large amount of regular nanotubes is presented, but no CNTY was observed in the sample. A similar result is obtained from catalysts of Co/Mg and Co/Ca with different weight ratios.

Li et al. These results show that CH4 and C2H2 are not suitable for synthesizing CNTYs on Co/Mg or Co/Ca catalysts. It is clear that sulfur contained in thiophene (C4H4S) plays a critical role in the formation of CNTYs. In fact, sulfur decomposed from sulfur-containing materials such as thiophene (C4H4S) and octadecanethiol (C18H38S) has been used as additive to metal catalysts to promote the growth of Y-junctions by pyrolysis of organometallic precursors (for example, nickelocene ((C8H12)2Ni),28 cobaltocene ((C5H5)2Co), ferrocence ((C5H5)2Fe), Nickel phthalocyanines (C32H16NiN8), iron phthalocyanines (C32H16FeN8), and iron pentacarbonyl (Fe(CO)5)19,31). Metal particles formed from the decomposition of organometallic precursors act as catalysts and are responsible for the growth of the Y-junctions. On the other hand, sulfur resulted from the decomposition of sulfur-containing reagents is expected to enhance the opportunity of CNTY formation.32 Sulfur has been considered as a catalyst poison; however, controlled poisoning of metal catalysts is expected to improve both the reactivity and the selectivity of the catalysts.33 There are different explanations to the role of the sulfur in Y-junction formation. One explanation is that sulfur will induce the fragmentation of the catalyst particles during the growth of carbon nanotubes, which will in turn result in the branches of carbon nanotubes.34 In other words, the Y-junction is grown by splitting one catalyst particle into two. The other explanation is that sulfur changes the active site on the surface of the catalyst particle, which will lead to the change of precipitation direction of carbon on the surface of the catalyst particles. The active site on the surface of the catalyst particle can change back and forth, and as a result, branched carbon nanotubes will be formed.31,35 Unfortunately, in our experiments, catalyst particles attached to the ends of CNTYs have not been observed in SEM and, especially, in TEM examinations. The chemical composition of the catalyst particles responsible for the formation of CNTYs is unavailable to provide any evidence for CNTY growth. However, since every observed CNTY has the same diameter for its three branches, it is unlikely that CNTY is formed by splitting one catalyst particle into two during the growth; otherwise, the CNTY should have one thicker stem and two thinner branches since the diameter of the nanotubes is proportional to the size of the catalysts. The current experimental results support the second explanation of the role of sulfur and the growing model of the CNTYs (i.e., CNTYs are formed due to the change of the carbon precipitation site on the catalyst particles). TEM Examination of Microstructures of CNTYs. The structure of CNTYs synthesized by pyrolysis of C4H4S on Co/Mg and Co/Ca with a ratio of 20 wt % has been examined by TEM. Figure 10a shows the typical structure of CNTYs synthesized on Co/Mg. Most of the CNTYs have three long branches of several micrometers, while some have two long branches and one short branch. All the branches of a CNTY have hollow channels and are uniform in diameter, although the diameters are different from CNTY to CNTY. The CNTY shown at the center of image Figure 10a has an outer diameter of 80 nm and an inner diameter of 30 nm, implying that this CNTY has about 70 layers. It is often observed that CNTY branches many times to form multiple Y-junctions, as shown in Figure 10b. In a multiple Y-junction, all the branches have the same inner and outer diameters. CNTYs grown on Co/Ca with a ratio of 20 wt % have similar structures to that synthesized on a Co/Mg catalyst, but these CNTYs have smaller inner diameters. As shown in Figure 10c,

Growth and Structure of Carbon Nanotube Y-Junctions

J. Phys. Chem. B, Vol. 110, No. 47, 2006 23699 precursors and C4H4S, CH4, and C2H2 as carbon sources in the CVD process. CNTYs can be synthesized by using catalysts Co/Mg and Co/Ca with different weight ratios and a carbon source C4H4S. Catalysts Co/Mg and Co/Ca with concentrations of 20 wt % grow CNTYs with the highest yield. When C4H4S is replaced by CH4 and C2H2, CNTYs cannot grow; instead, regular carbon nanotubes or amorphous carbon particles will form. In addition, if Co alone or a combination of Ca and Mg is used as a catalyst, no CNTYs can be formed. The experimental results indicate that a combination of Co and Mg or Co and Ca is in favor of the growth of CNTYs. Sulfur decomposed from C4H4S also plays a very critical role in the formation of CNTYs. Sulfur may cause a change of active site on the surface of catalyst particles, which will result in the branches of carbon nanotubes. Acknowledgment. This work is supported by the NSF Career Grant DMR-0548061 and the FIU 2005 faculty research award. References and Notes

Figure 10. (a) TEM image of CNTYs synthesized on a catalyst Co/Mg of 20 wt %, showing the clean and hollow structure of the CNTYs. (b) Multiple branched CNTYs grown on Co/Mg with a ratio of 20 wt %. (c) CNTY synthesized on catalyst Co/Ca with a ratio of 20 wt %. The CNTY has a smaller inner diameter and thicker wall.

the inner diameter of this CNTY is about 15 nm, and the outer diameter is about 80 nm, and it has a thicker wall. We have reported that catalyst particles composed of Co, Ca, Si, and O may promote the growth of CNTYs by decomposing CH4. It is speculated that the Ca, Si, and O impurities modulate the catalytic properties of Co in favor of the formation of branched nanotubes.20 In the present research, sulfur may play a similar role as Ca, Si, and O in the formation of CNTYs. However, in the present experiments, catalyst nanoparticles attached to the tips of CNTYs have not been found, and hence, the chemical composition of catalyst particles cannot be examined. The disappearance of the catalyst nanoparticles from the tips of CNTYs is attributed to the sonication process during the preparation of the TEM specimen, which causes the detachment of the nanoparticles from the nanotube tips. Further TEM examinations of the as-grown CNTYs need to be designed to reveal the chemical composition, which should shed some light on the understanding of the behavior of the catalyst particles during the CNTY growth and the growth mechanism of the CNTYs. Conclusion The synthesis of CNTYs has been studied by using cobalt nitrate, magnesium nitrate, and calcium nitrate as catalyst

(1) Iijima, S. Nature (London) 1991, 56, 354. (2) Tans, S. L.; Devoret, M. H.; Dai, H.; Thess, A.; Smalley, R. E.; Geerlings, L. J.; Dekker, C. Nature (London) 1997, 386, 474. (3) Kim, P.; Lieber, C. M. Science 1999, 286, 2148. (4) Dai, H.; Hafner, J. H.; Rinzler, A. G.; Colbert, D. T.; Smalley, R. E. Nature 1996, 348, 147. (5) Bockrath, M.; Cobden, D. H.; McEuen, P. L.; Chopra, N. G.; Zettl, A.; Thess, A.; Smalley, R. E. Science 1997, 275, 1922. (6) Rinzler, A. G.; Hafner, J. H.; Nikolaev, P.; Lou, L.; Kim, S. G.; Tomanek, D.; Nordlander, P.; Colbert, D. T.; Smalley, R. E. Science 1995, 269, 1550. (7) Kong, J.; Franklin, N. R.; Zhou, C.; Chapline, M. G.; Peng, S.; Cho, K.; Dai, H. Science 2000, 287, 622. (8) Dillon, A. C.; Jones, K. M.; Bekkedahl, T. A.; Kiang, C. H.; Bethune, D. S.; Heben, M. J. Nature (London) 1997, 386, 377. (9) Bandaru, P. R.; Daraio, C.; Jin, S.; Rao, A. M. Nat. Mater. 2005, 4, 663. (10) Papadopoulos, C.; Rakitin, A.; Li, J.; Vedeneev, A. S.; Xu, J. M. Phys. ReV. Lett. 2000, 85, 3476. (11) Meunier, V.; Buongi, M.; Nardelli, O.; Bernholc, J.; Zacharia, T.; Charlier, J. C. Appl. Phys. Lett. 2002, 18, 5234. (12) Del Valle, M.; Tejedor, C.; Cuniberti, G. Phys. ReV. B 2005, 71, 125306. (13) Latge, A.; Grimm, D.; Venezuela, P. J. Mol. Catal. 2005, 228, 125. (14) Zhou, D.; Seraphin, S. Chem. Phys. Lett. 1995, 238, 286. (15) Sui, Y. C.; Gonzalez-Leon, J. A.; Bermudez, A.; Saniger, J. M. Carbon 2001, 39, 1709. (16) Biro, L. P.; Ehlich, R.; Tellgmann, R.; Gromov, A.; Krawez, N.; Tschaplyguine, M.; Pohl, M.; Zsoldos, E.; Vertesy, Z.; Horvath, Z. E.; Campbell, E. E. B. Chem. Phys. Lett. 1999, 306 155. (17) Li, J.; Papadopoulos, C.; Xu, J. Nature (London) 1999, 402, 253. (18) Tsai, S. H.; Shiu, C. T.; Jong, W. J.; Shih, H. C. Carbon 2000, 38, 1879. (19) Deepak, F. L.; Govindaraj, A.; Rao, C. N. R. Chem. Phys. Lett. 2001, 354, 5. (20) Li, W. Z.; Wen, J. G.; Ren, Z. F. Appl. Phys. Lett. 2001, 79, 1879. (21) Gan, B.; Ahn, J.; Zhang, Q.; Rusli; Yoon, S. F.; Yu, J.; Huang, Q. F.; Chew, K.; Ligatchev, V. A.; Zhang, X. B.; Li, W. Z. Chem. Phys. Lett. 2001, 333, 23. (22) Ting, J. M.; Chang, C. C. Appl. Phys. Lett. 2002, 80, 324. (23) Liu, D.; Chen, J.; Deng, W.; Zhou, H.; Kuang, Y. Mater. Lett. 2004, 58, 2764. (24) Nagy, P.; Ehlich, R.; Biro, L. P.; Gyulai, J. Appl. Phys. A 2000, 70, 481. (25) Collins, P. G.; Bradley, K.; Ishigami, M.; Zettl, A. Science 2000, 287, 1801. (26) Baughman, R. H.; Zakhidov, A. A.; de Heer, W. A. Science 2002, 297, 787. (27) de Jonge, N.; Lamy, Y.; Schoots, K.; Oosterkamp, T. H. Nature 2002, 420, 393.

23700 J. Phys. Chem. B, Vol. 110, No. 47, 2006 (28) Satishkumar, B. C.; Thomas, P. J.; Govindaraj, A.; Rao, C. N. R. Appl. Phys. Lett. 2000, 77, 2530. (29) Li, W. Z.; Wen, J. G.; Sennett, M.; Ren, Z. F. Chem. Phys. Lett. 2003, 368, 299. (30) Colomer, J. F.; Stephan, C.; Lefrant, S.; Van Tendeloo, G.; Willems, I.; Konya, Z.; Fonseca, A.; Laurent, C.; Nagy, J. B. Chem. Phys. Lett. 2000, 83, 317. (31) Zhu, H.; Ci, L.; Xu, C.; Liang, J.; Wu, D. Diamond Relat. Mater. 2002, 11, 1349.

Li et al. (32) Biro, L. P.; Horvath, Z. E.; Mark, G. I.; Osvath, Z.; Koos, A. A.; Benito, A. M.; Maser, W.; Lambin, P. Diamond Relat. Mater. 2004, 13, 241. (33) Kim, M. S.; Rodriguez, N. M.; Baker, R. T. K. J. Catal. 1993, 143, 449. (34) Huang, S.; Dai, L.; Mau, A. Phys. B 2002, 323, 336. (35) Heyning, O. T.; Bernier, P.; Glerup, M. Chem. Phys. Lett. 2005, 409, 43.