Control of Hole Opening in Single-Wall Carbon Nanotubes and Single

Saeki , Yoshio Yamada , Noriko Yoshizawa , Osamu Tanaike , Junya Yamashita , Futaba Don , Kenji Hata and Hiroaki Hatori ..... Vito Sgobba , Dirk M...
0 downloads 0 Views 681KB Size
Download PDF
J. Phys. Chem. B 2006, 110, 1587-1591

1587

Control of Hole Opening in Single-Wall Carbon Nanotubes and Single-Wall Carbon Nanohorns Using Oxygen Jing Fan,† Masako Yudasaka,*,†,‡ Jin Miyawaki,† Kumiko Ajima,† Katsuyuki Murata,† and Sumio Iijima†,‡,§ SORST, Japan Science and Technology Agency, c/o NEC, 34 Miyukigaoka, Tsukuba, Ibaraki 305-8501, Japan, NEC Corporation, 34 Miyukigaoka, Tsukuba, Ibaraki 305-8501, Japan, and Meijo UniVersity, 1-501 Shiogamaguchi, Tenpaku-ku, Nagoya, Aichi 468-8502, Japan ReceiVed: July 14, 2005; In Final Form: NoVember 18, 2005

Due to the simplicity of the process, holes in the graphene walls of single-wall carbon nanotubes (SWNTs) and single-wall carbon nanohorns (SWNHs) have often been opened using O2 gas at high temperatures, even though this contaminates the nanotubes with carbonaceous dust (C-dust). To open holes with less C-dust contamination, we found that a slow temperature increase of 1 °C/min or less, in air, was effective. We also found that SWNHs having little C-dust could store a large quantity of materials inside the tubes. We infer that the local temperature increase due to the exothermic reaction of combustion may have been suppressed in the slow combustion process, which was effective in reducing the C-dust.

1. Introduction Various materials have been incorporated inside carbon nanotubes (CNTs),1-14 suggesting that CNTs are potentially useful for various purposes such as gas storage,9,13 catalyst support,12 and drug delivery.14,15 To enable materials to be incorporated, the opening of holes through the CNT wall is generally the most important process. To date, holes have been made through wet-chemical processes using oxidative acids15 or dry processes using O2 gas.17 The former method is suitable not only for opening the holes but also for removing metal catalysts, but a problem arises: the chemical byproducts often contaminate the nanotubes.18 On the other hand, the latter method is simple and the only byproduct contamination is carbonaceous fragments (C-dust) produced as a result of combustion of the tube walls. When the CNTs do not contain metal catalysts, the latter process is much better than the former one. Currently, there are four kinds of metal-free single-wall carbon nanotubes (SWNTs): (i) ones from which the metals were removed by wet-chemical processes, (ii) ones from which the metals were removed by heat treatment at high temperature,19,20 (iii) super-growth SWNTs grown by water-assisted chemical vapor depositions,21 and (iv) single-wall carbon nanohorns (SWNHs) formed by CO2 laser ablation without using any metal catalysts.22 In this report, we show that C-dust-free hole-opening in SWNHs and heat-treated SWNTs could be easily achieved by slow combustion and discuss its mechanism. 2. Experimental Section HiPco SWNTs were purchased from Nanotechnology Inc. The major impurity was Fe, which we decreased by evaporation: HiPco SWNTs were heat treated at 1780 °C for 2 h in a * Corresponding author. Tel.: +81 29 850 1190. Fax: +81 29 850 1366. E-mail address: [email protected]. † SORST, Japan Science and Technology Agency. ‡ NEC Corporation. § Meijo University.

vacuum of 10-6 Torr (these are referred to hereafter as HTHiPco).19 As a result of this heat treatment, the Fe quantity was reduced from about 20% to 2-3 wt % and there was an accompanying enlargement of the SWNT diameters from the initial values of 0.8-1.2 nm to about 1.5 nm, as estimated from the peak energies of the radial breathing modes in the Raman spectra.19 The SWNH preparation method was previously reported.22 Summarizing briefly, they were formed by CO2 laser ablation (power 3 kW; beam diameter 3 mm) of graphite in an Ar atmosphere (760 Torr) at room temperature. No metal catalyst was used in this process. We opened the holes of the HT-HiPco and SWNHs either by slow or quick combustion methods and compared their results. In the slow combustion method, the specimens were heated from room temperature to a target temperature (T(target)) at a rate (Rate(rise)) of 1 °C/min and cooled naturally in the furnace. T(target) was changed from 350 to 600 °C. The period for which the specimen temperature was held at T(target) was set to zero. We chose to use dry air as the atmosphere for the slow combustion, expecting that the low oxygen concentration in air (21%) would be effective for reducing the combustion rate. To investigate the mechanism of the combustion of the HT-HiPco and SWNHs, we exceptionally used Rate(rise)’s of 4, 2, and 0.5 °C/min and separately extended the specimen holding time at T(target) to 5 h. In the quick combustion method, high-temperature combustion was performed in a 100% oxygen atmosphere. In detail, HT-HiPco and SWNHs were heated to T(target) in Ar and, then, Ar was replaced by 100% O2 soon after the temperature reached T(target). The temperature was kept at T(target) for 10 min in 100% O2 and lowered to room temperature in O2. The HTHiPco and SWNH samples with holes opened or caps removed are denoted as HTox-HiPco and SWNHox, respectively, for short in this report. The structures of HTox-HiPco and SWNHox were studied by transmission electron microscopy (TEM) and Raman spectroscopy. The weight decrease of a SWNH caused by slow

10.1021/jp0538870 CCC: $33.50 © 2006 American Chemical Society Published on Web 01/07/2006

1588 J. Phys. Chem. B, Vol. 110, No. 4, 2006

Fan et al.

combustion was estimated by thermogravimetric (TG) analyses carried out under the same conditions as for slow combustion, namely, a Rate(rise) of 1 °C/min and a dry air atmosphere. The material-adsorption capacities of the SWNHox and HTox-HiPco were first examined using C60. The nanocondensation method was used to examine the C60 incorporation inside the nanotubes,23 that is, 10 µL of C60-saturated toluene solution was dropped onto the HTox-HiPco or SWHox placed on a grid with a filter paper beneath them, after which they were observed by TEM without any further processes.23 The material-adsorption capacity of the SWNHox was examined a second time using m-xylene. To estimate the quantity of xylene adsorbed by SWNHox, we exposed the SWNHox to air saturated with m-xylene vapor in a closed container for 1 h at room temperature.24,25 Then, SWNHs with adsorbed xylene were placed in a thermogravimetric analysis (TGA) apparatus with a He gas flow (100 cm3/min) for about 30 min before the measurement was started. The weight decrease due to xylene desorption was measured while the temperature was increased from room temperature to 600 °C at 5 °C/min in a He atmosphere. The adsorption surface area and porosity of the SWNHox were estimated through measurements of the adsorption and desorption isotherms of nitrogen (N2) at 77 K with an apparatus (AS-1-MP, Quantachrome). Before the isotherm measurements, the samples were pretreated at 150 °C for 2 h in a vacuum. 3. Results and Discussion Effects of Slow and Quick Combustion on HT-HiPco and SWNHs. When holes were opened in HT-HiPco (Figure 1a) and SWNHs (Figure 1b) by quick combustion (T(target) 580 °C; O2 100%), the obtained HTox-HiPco and SWNHox were contaminated with C-dust (Figures 1c and d). The C-dust could mostly be removed by post-treatments in a vacuum of 5 × 10-6 Torr at 500 °C for 2 h; however, a considerable amount still remained (Figures 1e and f). On the other hand, the quantity of C-dust was very small when slow combustion (Rate(rise) of 1 °C/min in dry air with T(target) of 550 °C) was applied (Figures 1g and h). The confirmation that slow combustion opened holes for the HToxHiPco and SWNHox is visually presented in Figures 1i and j, which show that C60 molecules were incorporated inside the HTox-HiPco and SWNHox. Structures of SWNHox Obtained by Slow Combustion. Hereafter, we present the results of the structure analysis and material-adsorption capacities of the SWNHox obtained by slow combustion with various values of T(target). Based on these data, we discuss how slow combustion worked to reduce the amount of C-dust. We used SWNHox rather than HTox-HiPco because the structure and adsorption phenomena of SWNHox have been studied in more detail.5,6,11,13,17,26 As-grown SWNH did not have any C-dust (Figure 2a), but the SWNHox obtained by slow combustion with a T(target) of 350 °C had a little C-dust outside the SWNHox (Figure 2b). The 350 °C-T(target) slow combustion caused weight loss of 2-3% (Figure 3). The relative intensity of the D-band (∼1350 cm-1) to the G-band (∼1590 cm-1) in the Raman spectrum of as-grown SWNH (Figure 4a) slightly increased after 350 °CT(target) slow combustion (Figure 4b), reflecting the increase in the number of defects in the graphene sheets constructing the SWNHox or the C-dust.27 The C-dust attached to the 350 °C-T(target) SWNHox disappeared when T(target) in slow combustion was increased to 400 °C: little C-dust was seen in the TEM images of 400

Figure 1. TEM images. HT-HiPco (a) and as-grown SWNH (b). HToxHiPco (c) and SWNHox (d) were obtained by quick combustion (T(target) 580 °C; 10 min.; O2 100%). After the quick combustion, heat treatment (500 °C, 2 h, 5 × 10-6 Torr) was performed for HToxHiPco (e) and SWNHox (f). HTox-HiPco (g) and SWNHox (h) were obtained by slow combustion (Rate(rise) 1 °C/min; T(target) 550 °C; holding period at T(target) 0 min; dry air). After slow combustion, C60 molecules were incorporated into HTox-HiPco (i) and SWNHox (j).

°C-T(target) SWNHox (Figure 2c). The weight loss was 4-5% (Figure 3), and the Raman D-band increased in intensity relative to the G-band (Figure 4c). The D-band increase reflects the increase in the number of holes. A little C-dust again appeared for 450 °C-T(target) slow combustion (Figure 2d). This C-dust was often located inside the SWNHox (Figure 2d). Although the weight loss was still 7% in total at 450 °C in the TG result (Figure 3), the Raman spectrum showed a remarkable change: a shoulder appeared at 1620 cm-1 (Figure 4d). The 1620-cm-1 peak indicates that there were CdO groups,26,28 and we infer that the CdO groups existed at the edges of the holes.29 The SWNHox obtained by the slow combustions with a 500 °C and 550 °C T(target) had small quantities of C-dust (Figures 2e and f). The weight decreases estimated from TG results were large, i.e., 12 and 30%, respectively (Figure 3). The Raman

Hole Opening in SWNTs and SWNHs Using O2

J. Phys. Chem. B, Vol. 110, No. 4, 2006 1589

Figure 4. Raman spectra of SWNH (a) and the SWNHox treated by slow combustion with various values of T(target) while keeping Rise(rate) at 1 °C/min in dry air. The T(target) holding period was 0 min. Values of T(target) were 350 (b), 400 (c), 450 (d), 500 (e), 550 (f), and 600 °C (g).

Figure 2. TEM images of as-grown SWNHs (a) and SWNHox treated by slow combustion with various values of T(target) while keeping Rate(rise) at 1 °C/min in dry air. The T(target) holding period was 0 min. Values of T(target) were 350 (b), 400 (c), 450 (d), 500 (e), 550 (f), and 600 °C (g).

Figure 3. Thermogravimetric analysis of the SWNHox measured in dry air with a temperature increase rate of 1 °C/min, that is, the same conditions as for slow combustion. The solid line represents a weighttemperature curve, and the broken line is the derivative of it.

spectra showed the intensity increases of the 1620-cm-1 shoulders (Figures 4e and f) and the D-bands. These results reflect the fact that the number of holes and the hole sizes both abruptly increased when T(target) in slow combustion was elevated to 500 or 550 °C.

After slow combustion with a 600 °C T(Target), a considerable amount of the SWNHox was combusted (Figure 3), and the SWNHox without horn-shaped tips remained (Figure 2g). The Raman G-band and 1620-cm-1 shoulder became even more distinct (Figure 4g), indicating that the main component of the residue was graphite-like impurities called giant graphite-balls.30 (Giant graphitic balls correspond to materials burnt at 650750 °C in the TG profiles (Figure 3).)30 Adsorption Capacities of SWNHox Obtained by Slow Combustion. It is known that the sizes and numbers of holes of the SWNHox opened by quick combustion increase with T(target).16 We expected the same to happen to the SWNHox obtained by slow combustion. We also expected the materialadsorption capacities inside the SWNHox obtained by slow combustion to be larger than those obtained by quick combustion, because the former had less C-dust inside the tubes. These expectations were reasonable, as shown below. C60 molecules were incorporated inside the SWNHox with holes opened by slow combustion by the “nanocondensation” method23 and were observed by TEM (Figure 5). Almost no C60 molecules were incorporated when T(target) was 400 °C (Figure 5a) or lower (not shown). The 500 °C-T(target) SWNHox was seen to encapsulate C60 (Figure 5b), and the amount of the encapsulated C60 appeared to become larger for 550 °C-T(target) SWNHox (Figure 5c). These results again indicated that the holes increased in both number and size with T(target). Since the diameter of the C60 molecule is 0.92 nm,31,32 the SWNHox treated by slow combustion at T(target) above 450 °C had holes with diameters of about 0.92 nm or larger. The adsorption phenomena of the SWNHox with holes opened by slow combustion were studied quantitatively using xylene and nitrogen. The quantities of xylene adsorbed by SWNHox were estimated from the TGA of xylene-adsorbing SWNHox in a He atmosphere. This method is described in detail in our previous reports.24,25 Typical TG data are shown in Figure 6a, and the adsorbed-xylene quantity per 1 g of SWNHox

1590 J. Phys. Chem. B, Vol. 110, No. 4, 2006

Fan et al. TABLE 1. Surface Area and Pore Volume Estimated from Quantities of Nitrogen that Was Isothermally (77 K) Adsorbed by SWNHox T(target) (°C)

BET surface areaa (m2 g-1)

pore volumeb (cm-3 g-1)

Quick Combustionc (O2 keeping period ) 10 min) as-grown 320 0.21 350 550 0.35 400 990 0.63 450 1200 0.75 500 1300 0.82 550 1400 0.92 580 1270 0.85

Figure 5. TEM images of C60 incorporated in the SWNHox treated by slow combustion with various values of T(target) while keeping Rise(rate) at 1 °C/min in dry air. The T(target) holding period was 0 min. Values of T(target) were 400 (a), 450 (b), 500 (c), and 550 °C (d).

Figure 6. (a) Weight-decrease profile and its derivative indicating xylene desorption from the SWNHox treated by slow combustion (Rate(rise) 1 °C/min; T(target) 500 °C; air). (b) Quantities of xylene adsorbed by the SWNHox treated by slow combustion with various values of T(target) while keeping Rise(rate) at 1 °C/min in dry air. The xylene adsorption quantities are indicated in units of gram per 1 g of SWNHox in part b.

obtained by slow combustion with a 500 °C T(target) was estimated from a weight loss of 28% as 28/(100 - 28) ) 0.39 g/g. The quantities of adsorbed xylene for various values of T(target) are shown in Figure 6b, indicating that the maximum value of 0.39 g/g was attained at a T(target) of 500 °C. This maximum value attained as a result of slow combustion was

Slow Combustiond (air; Rate(rise) ) 1 °C/min; keeping period ) 0 min) 400 1120 0.71 450 1300 0.81 500 1450 0.92 550 1360 0.89 600 540 0.41 a Calculated in the pressure region from P/P0 ) 0.001-0.1. Estimated from adsorption amount at P/P0 ) 0.7. c SWNHox were treated by quick combustion performed at various values of T(target) in 100% oxygen gas for 10 min. d SWNHox were treated by slow combustion with various values of T(target) while keeping Rate(rise) at 1 °C/min in dry air. The T(target) holding period was 0 min.

b

about 14% larger than the maximum value, 0.34 g/g,25 attained by quick combustion at a T(target) of 550 °C in 100% oxygen. The surface areas and pore volumes calculated from the isothermal nitrogen-adsorption quantity according to the BET equation are shown in Table 1. The SWNHox with holes opened by slow combustion (Rate(rise) 1 °C/min in dry air) attained the largest surface area of 1450 m2/g and a pore volume of 0.92 cm3/g at a T(target) of 500 °C. In the case of the SWNHox obtained by quick combustion (atmosphere 100% oxygen), the largest adsorption-values were attained at a T(target) of 550 °C; these values were 1400 m2/g and 0.92 cm3/g (Table 1). These values were not very different from those for the SWNHox obtained by slow combustion at a T(target) of 500 °C in dry air. We consider that the xylene adsorption was more sensitive to the C-dust inside the SWNHox than the nitrogen adsorption. The large molecule size of xylene (kinetic diameter 0.68 nm)31,32 had trouble entering and occupying the inner nanospaces of the SWNHox if the C-dust was enclosed inside the SWNHox. On the other hand, the nitrogen molecules were small enough (kinetic diameter 0.363 nm)31,32 that they were not obstructed by the C-dust in entering and occupying the SWNHox. Thus, we think that the adsorption capacity for xylene was reduced by the C-dust, while that for nitrogen was not influenced much. Mechanims of Slow Combustion. To consider how slow combustion worked to suppress C-dust generation, we tested various combustion conditions. First, the Rise(rate) was changed to 0.5, 2, or 4 °C/min, while T(Target) was set at 500 °C in dry air. Through TEM observations (images not shown), the C-dust contamination was apparent for the 2 or 4 °C/min cases, but the quantity of C-dust for 0.5 °C/min was as small as that for the 1 °C/min case. This result indicated that the quicker the combustion rate, the more C-dust was generated, which was consistent with the results that quick combustion generated a lot of C-dust (Figures 1c and d). It, also, became apparent that a Rise(rate) of 1 °C/min or slower was effective for reducing the quantity of C-dust. Second, the period for keeping the temperature constant at T(target) was extended from 0 to 5 h, while the Rise(rate) was kept at 1 °C/min in dry air. These experiments were carried

Hole Opening in SWNTs and SWNHs Using O2 out for various values of T(target): 350, 400, and 450 °C. As a result, the C-dust seen in TEM observation did not change in quantity even when the holding period was extended to 5 h. This implies that the defects that should burn at and below T(target) were entirely burnt out while the temperature was slowly elevated with rates of 1 °C/min or less from room temperature to T(target). Here, we discuss how slow combustion worked to reduce the C-dust contamination. We think that, in slow combustion, various defects in SWNTs and SWNHs were ignited gradually and the defects and their surroundings burnt slowly. This was effective in generating little C-dust because the local temperature increase caused by the exothermic reaction of combustion may have been suppressed. These phenomena were not expected in the case of quick combustion, that is, ignition started simultaneously at many defects, leading to a local temperature increase, which enhanced the combustion rates. In such combustion, the huge generation of C-dust and the opening of holes of large sizes occurred at lower temperatures, and a lot of the C-dust could easily be sucked inside the SWNTox and remained there as a contaminant. 4. Conclusion When holes were opened in SWNTs and SWNHs by quick combustion carried out by holding a high temperature in 100% oxygen gas for 10 min, they were often contaminated with C-dust, and a lot of C-dust was sucked inside the nanotubes. We found that the quantity of C-dust was greatly reduced by slow combustion using a slow temperature-increase, of 1 °C/ min or less and a low oxygen concentration, of 21%, in air. The structure and material-adsorption studies for the SWNHox with the holes opened by slow combustion clarified that the sizes and number of holes increased as T(target) increased. The quantity of materials adsorbed by the SWNHox with holes opened by slow combustion was larger than that for the SWNHox with the holes opened by quick combustion. We think that, in slow combustion, various defects in SWNTs and SWNHs were ignited gradually and the defects and their surroundings burnt slowly. This is effective in generating little C-dust because the local temperature increase caused by the exothermic reaction of combustion was suppressed in the slow combustion process. These phenomena are not expected in the case of quick combustion. References and Notes (1) Smith, B. W.; Monthioux, M.; Luzzi, D. Nature 1998, 396, 323. (2) Ajayan, P. M.; Iijima, S. Nature 1993, 361, 333. (3) Meyer, R. R.; Sloan, J.; Dunin-Borkowski, R.; Kirkland, A. I.; Novotny, M. C.; Bailey, S. R.; Hutchison, J. L.; Green, M. L. H. Science 2000, 289, 1324.

J. Phys. Chem. B, Vol. 110, No. 4, 2006 1591 (4) Sloan, J.; Wright, D. M.; Woo, H. G.; Bailey, S.; Brown, G.; York, A. P. E.; Coleman, K. S.; Hutchison, J. L.; Green, M. L. H. J. Chem. Soc., Chem. Commun. 1999, 699. (5) Hirahara, K.; Suenaga, K.; Bandow, S.; Kato, H.; Okazaki, T.; Shinohara, H.; Iijima, S. Phys. ReV. Lett. 2000, 85, 5384. (6) Hirahara, K.; Bandow, S.; Suenaga, K.; Kato, H.; Okazaki, T.; Shinohara, H.; Iijima, S. Phys. ReV. B 2001, 64, 115420. (7) Takenobu, T.; Takano, T.; Shiraishi, M.; Murakami, Y.; Ata, M.; Kataura, H.; Achiba, Y.; Iwasa, Y. Nat. Mater. 2003, 2, 683. (8) Maniwa, Y.; Kataura, H.; Abe, M.; Suzuki, S.; Achiba, Y.; Kira, H.; Matsuda, K. J. Phys. Soc. Jpn. 2002, 71, 2863. (9) Bekyarova, E.; Kaneko, K.; Yudasaka, M.; Kasuya, D.; Iijima, S.; Huidobro, A.; Rodriguez-Reinoso, F. J. Phys. Chem. B 2003, 107, 4681. (10) Bekyarova, E.; Hanzawa, Y.; Kaneko, K.; Silvestre-Albero, J.; Sepulveda-Escribano, A.; Rodriguez-Reinoso, F.; Kasuya, D.; Yudasaka, M.; Iijima, S. Chem. Phys. Lett. 2002, 366, 463. (11) Hashimoto, A.; Yorimitsu, H.; Ajima, K.; Suenaga, K.; Isobe, H.; Miyawaki, J.; Yudasaka, M.; Iijima, S. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 8527. (12) Yuge, R.; Ichihashi, T.; Shimakawa, Y.; Kubo, Y.; Yudasaka, M.; Iijima, S. AdV. Mater. 2004, 16, 1420. (13) Murata, K.; Hashimoto, A.; Yudasaka, M.; Kasuya, D.; Kaneko, K.; Iijima, S. AdV. Mater. 2004, 16, 1520. (14) Murakami, T.; Ajima, K.; Miyawaki, J.; Yudasaka, M.; Iijima, S.; Shiba, K. Mol. Pharm. 2004, 1, 399. (15) Ajima, K.; Yudasak, M.; Murakami, T.; Maigne, A.; Shiba, K.; Iijima, S. Mol. Pharm., in press. (16) Dujardin, E.; Ebbesen, T. W.; Krishnan, A.; Treacy, M. M. J. AdV. Mater. 1998, 10, 611. (17) Murata, K.; Hirahara, K.; Yudasaka, M.; Iijima, S. J. Phys. Chem. B 2002, 106, 12668. (18) Nagasawa, S.; Yudasaka, M.; Hirahara, K.; Ichihashi, T.; Iijima, S. Chem. Phys. Lett. 2000, 328, 374. (19) Yudasaka, M.; Kataura, H.; Ichihashi, T.; Qin, L. C.; Kar, S.; Iijima, S. Nano Lett. 2001, 1, 487. (20) Yudasaka, M.; Ichihashi, T.; Kasuya, D.; Kataura, H.; Iijima, S. Carbon 2003, 41, 1273. (21) Hata, K.; Futaba, D. N.; Mizuno, K.; Namai, T.; Yumura, M.; Iijima, S. Science 2004, 306, 1362. (22) Iijima, S.; Yudasaka, M.; Yamada, R.; Bandow, S.; Suenaga, K.; Kokai, F.; Takahashi, K. Chem. Phys. Lett. 1999, 309, 165. (23) Yudasaka, M.; Ajima, K.; Suenaga, K.; Ichihashi, T.; Hashimoto, A.; Iijima, S. Chem. Phys. Lett. 2003, 380, 42. (24) Fan, J.; Yudasaka, M.; Kasuya, Y.; Kasuya, D.; Iijima, S. Chem. Phys. Lett. 2004, 397, 5. (25) Yudasaka, M.; Fan, J.; Miyawaki, J.; Iijima, S. J. Phys. Chem. B 2005, 109, 8909. (26) Ajima, K.; Yudasaka, M.; Suenaga, K.; Kasuya, D.; Azami, T.; Iijima, S. AdV. Mater. 2004, 16, 397. (27) Nakamizo, M.; Honda, H.; Inagaki, M. Carbon 1978, 16, 281. (28) Nakamizo, M.; Tamai, K. Carbon 1984, 22, 197. (29) Miyawaki, J.; Yudasaka, M.; Iijima, S. J. Phys. Chem. B 2004, 108, 10732. (30) Fan, J.; Yudasaka, M.; Kasuya, D.; Azami, T.; Yuge, R.; Imai, H.; Kubo, Y.; Iijima, S. J. Phys. Chem. B 2005, 109, 10756. (31) Rigby, M.; Smith, E. B.; Wakeham, W. A.; Maitland, G. C. The forces between molecules; Clarendon Press: Oxford, U.K., 1986. (32) Maitland, G. C.; Rigby, M.; Smith, E. B.; Wakaham, W. A. Intermolecular forces: Their origin and determination; Clarendon Press: Oxford, U.K., 1987.