Formation of single-wall carbon nanotubes by laser ablation of fullerenes at low temperature

Article abstract of DOI:10.1063/1.125239

Laser ablation products from fullerene materials have been studied by transmission electron microscopy and Raman spectroscopy. Using nickel and cobalt as a catalyst, single-wall carbon nanotubes were produced at an ambient temperature of 400°C. The results were compared with those using graphite as starting materials. It is suggested that the formation of single-wall carbon nanotubes is controlled by both the availability of proper precursors and the activity of the metal catalyst.

Full text of DOI:10.1063/1.125239

Formation of single-wall carbon nanotubes by laser ablation of fullerenes at low
temperature
Y. Zhang and S. Iijima
Citation: Applied Physics Letters 75, 3087 (1999); doi: 10.1063/1.125239
View online: http://dx.doi.org/10.1063/1.125239
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/75/20?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Growth of high-quality single-wall carbon nanotubes without amorphous carbon formation
Appl. Phys. Lett. 84, 269 (2004); 10.1063/1.1639509
Covalent Functionalization of Arc Discharge, Laser Ablation and HiPCO SingleWalled Carbon Nanotubes
AIP Conf. Proc. 685, 291 (2003); 10.1063/1.1628037
Ultraviolet laser treatment of multiwall carbon nanotubes grown at low temperature
Appl. Phys. Lett. 82, 1607 (2003); 10.1063/1.1559654
Gas dynamic and time resolved imaging studies of single-wall carbon nanotubes growth in the laser ablation
process
AIP Conf. Proc. 590, 27 (2001); 10.1063/1.1420049
Production of bundles of single walled nanotubes by a simple laser ablation technique
AIP Conf. Proc. 486, 245 (1999); 10.1063/1.59790
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.212.109.170 On: Tue, 16 Dec 2014 16:38:58

APPLIED PHYSICS LETTERS
VOLUME 75, NUMBER 20
15 NOVEMBER 1999
Formation of single-wall carbon nanotubes by laser ablation of fullerenes
at low temperature
Y. Zhanga)
Fundamental Research Laboratories, NEC Corporation, 34 Miyukigaoka, Tsukuba, Ibaraki 305-8501, Japan
S. Iijima
Fundamental Research Laboratories, NEC Corporation, 34 Miyukigaoka, Tsukuba, Ibaraki 305-8501
and Department of Physics, JST International Research Project on Nanotubulites, Meijo University,
Nagoya 468-8502, Japan
͑Received 28 July 1999; accepted for publication 15 September 1999͒
Laser ablation products from fullerene materials have been studied by transmission electron
microscopy and Raman spectroscopy. Using nickel and cobalt as a catalyst, single-wall carbon
nanotubes were produced at an ambient temperature of 400 °C. The results were compared with
those using graphite as starting materials. It is suggested that the formation of single-wall carbon
nanotubes is controlled by both the availability of proper precursors and the activity of the metal
catalyst. © 1999 American Institute of Physics. ͓S0003-6951͑99͒02146-4͔
Since the first successful synthesis of single-wall carbon
nanotubes ͑SWCNTs͒ by arcing the carbon rod together with
the metal catalyst in an inert gas atmosphere,1,2 many im-
provements have been made to realize high-yield production
of this interesting material. Compared with the arc-discharge
method,3 the laser ablation method can give more homog-
enous products and is thus suitable for the study of their
growth mechanism.4 However, formation of SWCNTs in the
laser ablation method requires relatively high ambient tem-
perature. The favorable temperature is over 1100 °C and the
yield decreases dramatically below 850 °C.5,6 Although the
detailed data at low temperature for SWCNT formation is
not available, it is generally accepted that there is no
SWCNT bundle growth below 600 °C when a graphite–
metal target was used. Interestingly, this temperature depen-
dence is very similar to that of fullerene production by laser
ablation of graphite.7–9 This similarity implies a strong struc-
tural and energetic correlation of precursors for growth of
fullerenes and SWCNTs. An understanding of this correla-
tion would help us to reveal the growth mechanism of
SWCNTs in laser ablation. For this purpose, we have de-
vised new laser ablation experiments using fullerenes as tar-
get materials, speculating that laser decomposition of
fullerenes may provide proper precursors for the formation
of SWCNTs. The experiments were done under the conven-
tional condition for SWCNT production, except at relatively
low temperature in order to suppress the direct sublimation
of fullerenes. We compared the result with that using the
graphite target and discussed their differences by considering
the structure and energy of carbon precursors produced by
the laser beam.
width Ϸ8 ns͒ was focused to get an energy density of about
3 J/cm2 per pulse on the target surface. The target was made
by pressing purified C60 polycrystalline powder, together
with 5 at. % ͑NiϩCo͒ powder as the catalyst, into a pellet of
1 cm in diameter. For purpose of comparison, a pure C60
͑99.98%͒ target and a graphite–metal target were also used.
Soot-like samples for investigation were collected from the
wall of the quartz reactor after laser ablation. Table I listed
the samples obtained using different targets and at different
temperatures, where M represents the metal catalyst. Trans-
mission electron microscopy ͑TEM͒ observation was per-
formed in a Topcon 002B transmission electron microscope
working at 200 kV. Raman spectra were measured using an
Arϩ ion laser with a wavelength of 488 nm. The spectra
resolution was ϳ5 cmϪ1
.
Figure 1 shows Raman spectra of the soot-like materials
formed in laser ablation under different conditions. Spectra
͑a͒–͑f͒ correspond to samples A–F in Table I, respectively.
In spectrum ͑a͒, which was measured from the soot produced
from the target containing C60 and metal catalyst at 400 °C,
there is a sharp peak at 1592 cmϪ1 and a shoulder at about
1569 cmϪ1 in the high-frequency region. This Raman dou-
blet can be assigned to the tangential C–C stretching modes
of SWCNTs.11 Another broad peak at 1346 cmϪ1, namely
the ‘‘D band,’’ represents the contribution of disordered gra-
phitic materials.12 In the low-frequency region, a broad peak
near 170–180 cmϪ1 can be identified, which is characteris-
TABLE I. Preparation conditions and TEM results for different samples.
Temperature
Laser ablation experiments were performed in an electri-
cal tube furnace.4,10 Argon gas was introduced in the quartz-
tube reactor with a flow rate of about 300 sccm. The pressure
in the reactor was kept at about 600 Torr during laser abla-
tion. The second harmonic beam of a Nd-YAG laser ͑pulse
Sample No.
Target
͑°C͒
SWCNT formation
A
B
C
D
E
F
C60ϩM
400
400
400
20
20
20
Yes
None
C60
GraphiteϩM
C60ϩM
C60
Almost nonea
Almost nonea
None
GraphiteϩM
Almost nonea
a͒
Author to whom correspondence should be addressed; electronic mail:
zhang@frl.cl.nec.co.jp
aIndividual SWCNT was observed in very few occasions.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
0003-6951/99/75(20)/3087/3/$15.00 3087 © 1999 American Institute of Physics
141.212.109.170 On: Tue, 16 Dec 2014 16:38:58

3088
Appl. Phys. Lett., Vol. 75, No. 20, 15 November 1999
Y. Zhang and S. Iijima
FIG. 1. Raman spectra measured from products of laser ablation using tar-
gets of ͑a͒ C60 and metal catalyst at 400 °C; ͑b͒ pure C60 at 400 °C; ͑c͒
graphite and metal catalyst at 400 °C; ͑d͒ C60 and metal catalyst at room
temperature; ͑e͒ pure C60 at room temperature; ͑f͒ graphite and metal cata-
lyst at room temperature.
tically associated with the breathing mode of SWCNTs.11
Spectra ͑b͒ and ͑c͒, measured from the soot samples pro-
duced from a pure C60 target and a graphite–metal target at
400 °C, are quite different from spectrum ͑a͒. They show
only two broad peaks near 1590 and 1340 cmϪ1, indicating
the absence of SWCNTs in the products. Similar Raman
spectra, as shown by ͑c͒–͑e͒, were obtained for soot samples
produced at room temperature using the three kinds of tar-
gets.
FIG. 2. High-resolution TEM micrographs of ͑a͒ SWCNT bundles and ͑b͒
individual SWCNTs in the laser ablated soot obtained from a C60–metal
target at 400 °C.
still dense and hot. The initial types and energies of carbon
species ablated from C60 and graphite are expected to be
different because of the different bonding environments. For
example, the C–C bonding energy on a pentagon of a C60
molecule is smaller than that on a hexagon,14 and breaking of
a C–C bond is more likely to happen on the pentagons. This
could lead to a preferential formation of some specific spe-
cies, such as dimmers, which would help the growth of nano-
tubes. Furthermore, if we assume a constant adsorption of
laser energy for carbon atoms,15 the kinetic energy of a spe-
cific species form C60 would be higher than that formed from
graphite, which contains only hexagons.14 The higher kinetic
energy could help the nanotube growth in two ways. First, it
would increase the collision rate between carbon species,
which is needed for mass transportation during nanotube
growth. Second, it would accelerate the decomposition of
some large fragments produced by laser ablation into proper
species for nanotube growth.16,17 The above arguments could
explain the different results obtained from C60 and graphite
targets at low temperature.
The temperature is an important factor to influence the
SWCNT formation. In conventional SWCNT production by
laser ablation of graphite,4,10 the high temperature ͑1200 °C͒
restricts the cooling rate of carbon species, and thus sustains
the necessary kinetic energy for them to decompose into
proper precursors and to arrive at the growth site of nano-
tubes. This argument, together with that in the previous para-
graph, can be used to explain the different results between
samples A and D. However, it should be noted that the high
temperature also favors the formation of fullerenes.7–9 The
The formation of a considerable amount of SWCNTs
using a C60–metal target at 400 °C was confirmed by high-
resolution TEM observation ͑Fig. 2͒. Here, the considerable
amount is characterized by the bundle formation as shown in
Fig. 2͑a͒. The total yield of SWCNTs in the soot is about
5%, much less than that in Refs. 4 and 10. The number of
SWCNTs in a bundle is usually less than ten. The diameter
of a single SWCNT is 1.2–1.3 nm, which is slightly smaller
than that produced at high temperature. The caps of indi-
vidual SWCNTs have hemisphere shapes ͓Fig. 2͑b͔͒. Bundle
formation was not found in other samples ͑Table I͒. How-
ever, individual SWCNT was occasionally observed in
samples C, D, and F. These samples were produced from the
targets containing a metal catalyst even at room temperature.
The yields of SWCNTs in these samples are so low that they
cannot be detected by Raman spectroscopy. No SWCNT was
found in the products ablated from the pure C60 target. This
is consistent with the result using a pure graphite target.13
In order to explain the experimental results, we use a
three-step growth model proposed for laser ablation of car-
bon materials.13 In this model, carbon species, mainly ex-
cited atoms and small clusters, are first ablated from the tar-
get and form a plasma zone. The plasma then begins to cool
down through expansion. Finally, carbon species are further
cooled down by collision with argon gas until a thermal dy-
namic equilibrium is established. The SWCNTs are sup-
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
posed to form in the second step where the plasma zone is
efficiency of nanotube formation may thus not be propor-
141.212.109.170 On: Tue, 16 Dec 2014 16:38:58

Appl. Phys. Lett., Vol. 75, No. 20, 15 November 1999
Y. Zhang and S. Iijima
3089
tional to the temperature due to the competing process of
fullerene growth that requires the same carbon precursors.
Another role of high temperature is to increase the activity of
the metal catalyst. The effect of metal catalyst on the
SWCNT formation is obvious from the comparison of
samples A and D with samples B and E. The occasionally
observed individual SWCNTs in samples C, D, and F may
be attributed to some carbon and metal species that stayed at
the center part of the plasma zone where the cooling rate is
relatively slow.
In conclusion, we have investigated the laser ablation of
C60 and graphite with or without a metal catalyst at relatively
low temperatures. The results showed SWCNT bundles can
form at 400 °C by using C60 and a metal catalyst. The ex-
perimental results were explained by a growth model of car-
bon materials in laser ablation. It is suggested that the for-
mation of single-wall carbon nanotubes is controlled by both
the availability of proper precursors and the activity of the
metal catalyst.
S. Lefrant, P. Deniard, R. Lee, and J. E. Fisher, Nature ͑London͒ 388, 756
͑1997͒.
4 A. Thess, R. Lee, P. Nikolaev, H. Dai, P. Petit, J. Robert, C. Xu, Y. H.
Lee, S. G. Kim, A. G. Rinzler, D. T. Colbert, G. E. Scuseria, D. Tomanek,
J. E. Fischer, and R. E. Smalley, Science 273, 483 ͑1996͒.
5 S. Bandow, S. Asaka, Y. Saito, A. M. Rao, L. Grigorian, E. Richter, and
P. C. Eklund, Phys. Rev. Lett. 80, 3779 ͑1998͒.
6 M. Yudasaka, T. Ichihashi, and S. Iijima, J. Phys. Chem. B 102, 10201
͑1998͒.
7 S. Iijima, T. Wakabayashi, and Y. Achiba, J. Chem. Phys. 100, 5839
͑1996͒.
8 R. E. Smalley, Acc. Chem. Res. 25, 98 ͑1992͒.
9 T. Wakabayashi, D. Kasuya, H. Shiromaru, S. Suzuki, K. Kikuchi, and Y.
Achiba, Z. Phys. D 40, 414 ͑1997͒.
10 Y. Zhang and S. Iijima, Philos. Mag. Lett. 78, 139 ͑1998͒.
11 A. M. Rao, E. Richter, S. Bandow, B. Chase, P. C. Eklund, K. A. Will-
iams, S. Fang, K. R. Subbaswamy, M. Menon, A. Thess, R. E. Smalley,
G. Dresselhaus, and M. S. Dresselhaus, Science 275, 187 ͑1997͒.
12 P. C. Eklund, J. M. Holden, and R. A. Jishi, in Carbon Nanotubes, edited
by M. Endo, S. Iijima, and M. S. Dresselhaus ͑Pergamon, Oxford, 1996͒,
p.129.
13 Y. Zhang, H. Gu, and S. Iijima, Appl. Phys. Lett. 73, 3827 ͑1998͒.
14 C–C bond length on a pentagon ͑0.146 nm͒ is larger than that on a hexa-
gon ͑0.140 nm͒. The binding energy per carbon atom for C60 is smaller
than that for graphite by 0.4–0.7 eV. See M. S. Dresselhaus, G. Dressel-
haus, and P. Eklund, Science of Fullerenes and Carbon Nanotubes ͑Aca-
demic, New York, 1996͒, p. 63.
This work is partially supported by the Special Coordi-
nation Funds of the Science and Technology Agency of the
Japanese Government.
15 This assumption is reasonable because the ablation rates of fullerene and
graphite targets are similar under the present experimental conditions.
16 K. Kaizu, M. Kohno, S. Suzuki, H. Shiromaru, T. Moriwaki, and Y.
Achiba, J. Chem. Phys. 106, 9954 ͑1997͒.
1 S. Iijima and T. Ichihashi, Nature ͑London͒ 363, 603 ͑1993͒.
2 D. S. Bethune, C. H. Kiang, M. S. de Vries, G. Gorman, R. Savoy, J.
Vazque, and R. Beyers, Nature ͑London͒ 363, 605 ͑1993͒.
17 T. Moriwaki, K. Kobayashi, M. Osaka, M. Ohara, H. Shiromaru, and Y.
Achiba, J. Chem. Phys. 107, 8927 ͑1997͒.
3 C. Journet, W. K. Maser, P. Bernier, A. Loiseau, M. Lamy de la Chapelle,
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
141.212.109.170 On: Tue, 16 Dec 2014 16:38:58