Raman characterization of aligned carbon nanotubes produced by thermal decomposition of hydrocarbon vapor

Article abstract of DOI:10.1063/1.118993

Raman characterization of aligned carbon nanotubes of average diameter 10-15 nm, produced by chemical vapor deposition on a mesoporous substrate, has been carried out. The resonance behavior and higher-order Raman bands up to fourth order have been observed and compared with those of carbon nanotubes produced by arc discharge and highly oriented pyrolytic graphite, as well as pyrolytic graphite. The phonon properties have been analyzed with the help of high-resolution transmission electron microscope studies.

Full text of DOI:10.1063/1.118993

Raman characterization of aligned carbon nanotubes produced by thermal
decomposition of hydrocarbon vapor
Wenzhi Li, Hao Zhang, Chaoying Wang, Yun Zhang, Liwen Xu, Ke Zhu, and Sishen Xie
Citation: Applied Physics Letters 70, 2684 (1997); doi: 10.1063/1.118993
View online: http://dx.doi.org/10.1063/1.118993
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/70/20?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Characterization of phosphorus-doped multiwalled carbon nanotubes
J. Appl. Phys. 111, 064315 (2012); 10.1063/1.3695452
Evaluation of metal-free carbon nanotubes formed by SiC thermal decomposition
J. Appl. Phys. 101, 104311 (2007); 10.1063/1.2732547
What causes the carbon nanotubes collapse in a chemical vapor deposition process
J. Chem. Phys. 118, 878 (2003); 10.1063/1.1527897
Raman characterization of boron carbonitride nanotubes
Appl. Phys. Lett. 80, 3590 (2002); 10.1063/1.1479207
Pure and aligned carbon nanotubes produced by the pyrolysis of benzene-based aerosols
AIP Conf. Proc. 591, 204 (2001); 10.1063/1.1426854
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:
128.42.202.150 On: Sun, 23 Nov 2014 05:42:20

Raman characterization of aligned carbon nanotubes produced by thermal
decomposition of hydrocarbon vapor
Wenzhi Li,a) Hao Zhang, Chaoying Wang, Yun Zhang, Liwen Xu, Ke Zhu,
and Sishen Xie
Institute of Physics and Center for Condensed Matter Physics, Chinese Academy of Sciences, Beijing
100080, People’s Republic of China
͑
Received 11 December 1996; accepted for publication 21 March 1997͒
Raman characterization of aligned carbon nanotubes of average diameter 10–15 nm, produced by
chemical vapor deposition on a mesoporous substrate, has been carried out. The resonance behavior
and higher-order Raman bands up to fourth order have been observed and compared with those of
carbon nanotubes produced by arc discharge and highly oriented pyrolytic graphite, as well as
pyrolytic graphite. The phonon properties have been analyzed with the help of high-resolution
transmission electron microscope studies. © 1997 American Institute of Physics.
͓
S0003-6951͑97͒02520-5͔
1
The discovery of carbon nanotubes ͑CNTs͒ has gener-
ated remarkable excitement in scientific communities.2
Large quantities of CNTs can now be produced by either arc
resolution transmission electron microscope ͑HRTEM͒ re-
sults.
–6
The aligned CCNTs were fabricated by chemical vapor
1
,2,4,5
3
discharge
or thermal decomposition of hydrocarbon
deposition of acetylene at 700 °C. The main feature of our
3
,6
vapor. However, CNTs produced by these methods are of
disordered orientation and mixed with carbon nanoparticles
or amorphous carbon, which restricts their experimental
characterization and applications. Recently, Ajayan et al.
and de Heer et al. have successfully produced aligned CNTs
by orienting the as-formed arc-discharge CNTs ͑DCNTs͒.
We have developed a new method to produce a pure, high
density, and aligned array of carbon nanotubes, which is
favorable for its property analysis. Raman characterization is
very sensitive to the breakdown in translational symmetry of
carbon materials and can provide detailed information about
the microstructure, hence, it offers a relatively easy and
unique tool to characterize the structure of CNTs without
destroying the sample. A considerable number of papers on
method is to use mesoporous silica embedded by iron nano-
particles as a substrate to produce aligned carbon nanotubes.
A more detailed process of the method has been described
4
7
elsewhere. Because the CCNTs were formed on the surface
5
of the silica substrate, we can, using a scanning electron
microscope ͑SEM͒ ͑S-4200, Hitachi͒, observe the original
morphology and distribution of the CCNTs without damag-
ing the samples. The HRTEM ͑JEM 200-cx͒ specimen was
prepared by grinding the sample in mortar and suspending in
ethanol before being dropped on a hole carbon copper grid.
Raman studies were performed directly on the as-formed
large area aligned CCNTs. A Renishaw Raman spectrometer
equipped with a charge coupled device detector ͑CCD͒ was
7
ϩ
employed. A 25 mW, 514.5 nm, Argon ion (Ar ) laser was
used as a light source, typically with 0.2 mW at the sample.
A 50ϫ objective was employed, giving an illuminated spot
of about 2 ␮m in diameter. The incident light was polarized
perpendicular to the orienting direction of the CCNTs. Typi-
cal acquisition times for the spectra were 90 s. The spectral
the Raman spectra and resonance behavior of DCNTs have
already been published,8
–13
revealing that this material has
unique Raman spectral features. Recently, single wall CNTs
have been studied by using a Raman scattering technique
with laser excitation wavelengths in the range from 514.5 to
320 nm, numerous Raman peaks were observed and identi-
fied with vibrational modes of armchair symmetry (n,n)
CNTs. However, Raman spectra of CNTs are often dis-
turbed by their disorder distribution and very low density, so
Raman studies of aligned and dense carbon nanotubes are
helpful in understanding the vibrational spectra and micro-
structure of the carbon nanotubes.
In this letter, the first- and higher-order Raman scattering
of an aligned array of carbon nanotubes produced by chemi-
cal vapor deposition ͑CVD͒ of hydrocarbon gas ͑CCNTs͒
Ϫ1
resolution was better than 2 cm , and no smoothing was
1
applied to the spectra.
The CVD process was manipulated under optimal
1
4
7
conditions; after reacting for about 2 h, a black layer of
CCNTs was formed on the substrate surface. The SEM im-
age ͓Fig. 1͑a͔͒ shows that the thickness of the tube layers is
about 50 ␮m. A high-magnification SEM image ͓Fig. 1͑b͔͒
illustrates that the layer is composed of aligned CCNTs
about 10–15 nm in diameter, most of which are approxi-
mately perpendicular to the substrate surface while a few are
slightly tilted. These CCNTs are isolated with spacings about
͑
Ref. 7͒ are studied and compared to that of DCNTs and
15
1
00 nm, and neither bundles of CCNTs, nor carbon
highly oriented pyrolytic graphite ͑HOPG͒ as well as pyro-
lytic graphite ͑PG͒. Higher-order Raman bands up to fourth
order of the CCNTs have been observed and assigned to the
corresponding fundamental frequencies, then the phonon
properties of CCNTs may be discussed with the help of high-
16
nanoparticles, have been detected in the array of aligned
CCNTs.
HRTEM images ͑Fig. 2͒ display the well-graphitized
CCNTs and only a little bit of carbonaceous material at-
tached to the surface of the CCNTs. These CCNTs consist of
more than ten concentric carbon shells with a spacing of 0.34
nm, which is consistent with that of graphite. Most aligned
a͒Electronic mail: user412@aphy01.iphy.ac.cn
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:
684 Appl. Phys. Lett. 70 (20), 19 May 1997 0003-6951/97/70(20)/2684/3/$10.00 © 1997 American Institute of Physics
28.42.202.150 On: Sun, 23 Nov 2014 05:42:20
2
1

FIG. 1. SEM images of aligned CCNTs formed on the substrate. ͑a͒ Low-
magnification image reveals that the aligned CCNTs are about 50 ␮m long,
͑b͒ high-magnification SEM image of aligned CCNTs shows that most
nanotubes are approximately perpendicular to the surface of the substrate.
The diameter of the CCNTs are of about 10–15 nm. These nanotubes are
separated and very pure.
CCNTs are very straight, as displayed in Fig. 2͑a͒, while
some slightly curved tubes caused by kinks or defects1 are
shown in Fig. 2͑b͒. The formation of the kinks and defects is
attributed to the low temperature ͑700 °C͒ of the CVD pro-
cess.
7,18
The first-order Raman spectrum of aligned CCNTs
Ϫ1
shows a strong sharp peak at 1581 cm ͑G line͒, which is
the high-frequency E2g first-order mode. One of the most
unusual properties of aligned CCNTs is that the Raman spec-
Ϫ1
trum shows an additional strong band at 1348 cm ͑D line͒
Ϫ1
and a weak band at around 1617 cm ͑DЈ line͒, as shown in
Fig. 3͑a͒. It is worth noting that the strong peak at
Ϫ1
1
348 cm is similar to that of PG and is not observed in the
8
–13
spectra of HOPG.
The origin of the D and DЈ lines in
other forms of carbon materials has been explained as
disorder-induced features due to the finite particle size effect
or lattice distortion.1
9–21
In other words, the relative intensity
FIG. 3. Micro-Raman spectra of aligned CCNTs excited by 514.5 nm. ͑a͒
Fundamental and second-order part, and ͑b͒ third- and fourth-order part.
of the 1348 cmϪ1 mode with respect to the 1581 cmϪ1 mode
depends on the crystal planar domain size of graphite. In our
case, SEM and HRTEM images of the present aligned
CCNTs reveal that the CCNTs only consist of many isolated
and well-graphitized tubes, and contain no carbon nanopar-
Ϫ1
ticles at all. Therefore, it is reasonable that the 1348 cm
mode origination may be attributed to defects in the curved
graphene sheets, tube ends, and finite size crystalline do-
2
2–24
mains of the tubes.
1
Compared with the weak line at
the appearance of the strong D
Ϫ1
8–13
350 cm of DCNTs,
line of CCNTs can be interpreted as being due to: ͑1͒ the
turbostratic structure of carbon sheets in tubes, namely the
finite size ͑nanometer order͒ of the crystalline domains; and
͑2͒ the high density of the aligned tubes. Thus, the large
amount of crystalline domains on the nanometer scale and
the surfaces of the tubes must account for the enhancement
Ϫ1
of the D line at 1348 cm . In principle, the position of the G
Ϫ1
line at 1581 cm is slightly dependent on domain size, but
the shift to high frequency is not observed within our experi-
mental accuracy, although the G line at 1581 cm appears
FIG. 2. HRTEM images of the well-graphitized nanotubes. The nanotube
consists of ten concentric graphene shells with a spacing of 0.34 nm. ͑a͒
Perfectly graphitized straight carbon nanotube, and ͑b͒ curved carbon nano-
tube resulting from kinks or defects.
Ϫ1
to be a doublet due to the finite size effect. The D line ͑at
Ϫ1
1348 cm ͒ has also been attributed to first-order scattering
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:
Appl. Phys. Lett., Vol. 70, No. 20, 19 May 1997 Wenzhi et al. 2685
128.42.202.150 On: Sun, 23 Nov 2014 05:42:20

TABLE I. Fundamental and higher-order bands of aligned CCNTs recorded
spectra. Undoubtedly, the high density of the aligned CCNTs
helps detection of the higher-order Raman spectra.
on a Renishaw Raman system with 514.5 nm laser line I /I ϭ1.3Ϯ0.02.
D
G
In summary, we have successfully synthesized large area
aligned CCNTs, which are perpendicular to the surface of the
substrate. The tubes are around 10–15 nm in diameter and 50
␮m long. Raman spectra up to fourth order of the aligned
CCNTs have been collected, indicating that CCNTs are
highly graphitized. The appearance of a strong D band shows
that a nanometer scale turbostratic structure may be formed
in carbon nanotubes due to the relatively low growth tem-
perature. The Raman studies are consistent with the results
from SEM and HRTEM. Resonance behavior shows that the
lattice properties of the aligned CCNTs are different from
that of other forms of carbon. Third- and fourth-order Raman
spectra have been detected.
⌬
cmϪ1
␯
⌬␯
͑cmϪ1
͒
͑
͒
Designation
Designation
1
094
348
581
617
453
697
930
3233
4043
4274
4511
4819
5385
5861
2␻
DЈ
1
1
1
2
2
2
␻D
␻G
3␻D
2␻ ϩ␻G
D
␻
␻ ϩ2␻G
D
DЈ
␻ ϩ2␻
G
DЈ
2␻D
4␻D
D
2␻ ϩ2␻G
␻ ϩ␻
D
G
from a zone–boundary phonon activated by the disorder as-
sociated with finite crystalline size. This assignment is cor-
roborated with the strong peak at ϳ2697 cm ͑i.e., ϳ2
ϫ1348 cm ͒ in the second-order Raman spectrum and the
weak peak at ϳ5385 cm ͑i.e., ϳ4ϫ1348 cm ͒ in the
Ϫ1
The authors thank Wei Liu for the TEM measurements
and Yulong Liu, Yanyong Yang, and David Pitt for the Ra-
man scattering measurements. This work was partially sup-
ported by the National Natural Science Foundation of China.
Ϫ1
Ϫ1
Ϫ1
fourth-order Raman spectrum. While the D line at
348 cm has a counterpart in the second-order spectrum, if
Ϫ1
1
its presence results from the finite crystalline size of nanom-
Ϫ1
eter order, it might be expected that the peak at 1617 cm
also has a counterpart in the second-order spectrum of
CCNTs, namely, 3233 cm . It is important to note that the
1
Ϫ1
S. Iijima, Nature ͑London͒ 354, 56 ͑1991͒.
T. W. Ebbesen and P. M. Ajayan, Nature ͑London͒ 358, 220 ͑1992͒.
M. Endo, K. Takeuchi, S. Igarashi, K. Kobori, M. Shiraishi, and H. W.
Kroto, J. Phys. Chem. Solids 54, 1841 ͑1993͒.
2
differences in frequency between the observed first-order
modes and half the frequency of the corresponding second-
order modes could be due to matrix element effects or to the
region of the k space strongly sampled by the breakdown of
wave–vector conservation.
3
4
P. M. Ajayan, O. Stephan, C. Colliex, and D. Trauth, Science 265, 1212
͑
1994͒.
5
W. A. de Heer, W. S. Bacsa, A. Ch aˆ telain, T. Gerfin, R. H. Baker, L.
Forro, and D. Ugarte, Science 268, 845 ͑1995͒.
H. Dai, E. W. Wong, and C. M. Lieber, Science 272, 523 ͑1996͒.
W. Z. Li, S. S. Xie, L. X. Qian, B. H. Chang, B. S. Zou, W. Y. Zhou, R.
A. Zhao, and G. Wang, Science 274, 1701 ͑1996͒.
H. Hiura, T. W. Ebbesen, K. Tanigaki, and H. Takahashi, Chem. Phys.
Lett. 202, 509 ͑1993͒.
6
7
The second-order Raman spectrum of CCNTs is shown
Ϫ1
in Fig. 3͑a͒. As mentioned above, the peak at 2697 cm
(
D*) is the overtone of the D line, and the peak at
8
9
Ϫ1
2930 cm is attributed to the combination of the D and G
lines. In addition, the second-order scattering at 2453 and
J. M. Holden, P. Zhou, X. X. Bi, P. C. Eklund, S. Bandow, R. A. Jishi, K.
D. Chowdhury, G. Dresselhaus, and M. S. Dresselhaus, Chem. Phys. Lett.
Ϫ1
3
233 cm has also been observed very clearly and attrib-
220, 186 ͑1994͒.
uted to the corresponding fundamental bands, as summarized
10
J. Kastner, T. Pichler, H. Kuzmany, S. Curran, W. Blau, D. N. Weldon,
M. Delamesiere, S. Draper, and H. Zandbergen, Chem. Phys. Lett. 221, 53
Ϫ1
in Table I. The observation of the line at 3233 cm is strong
͑
1994͒.
evidence for overtone scattering from the highest-frequency
11
Ϫ1
W. S. Bacsa, D. Ugarte, A. Ch aˆ telain, and W. A. de Heer, Phys. Rev. B
line at ϳ1617 cm in the density of states.
50, 15 473 ͑1994͒.
The third- and fourth-order Raman spectra are shown in
Fig. 3͑b͒. To our knowledge, this is the first report on the
detection of such higher-order Raman spectra of carbon
12
P. V. Huong, R. Cavagnat, P. M. Ajayan, and O. Stephan, Phys. Rev. B
51, 10 048 ͑1995͒.
1
1
3
4
P. C. Eklund, J. M. Holden, and R. A. Jishi, Carbon 33, 959 ͑1995͒.
A. M. Rao, E. Richter, Shunji Bandow, Bruce Chase, P. C. Eklund, K. A.
Williams, S. Fang, K. R. Subbaswamy, M. Menon, A. Thess, R. E. Smal-
ley, G. Dresselhaus, and M. S. Dresselhaus, Science 275, 187 ͑1997͒.
T. W. Ebbesen, H. Hiura, J. Fujita, Y. Ochiai, S. Matsui, and K. Tanigaki,
Chem. Phys. Lett. 209, 83 ͑1993͒.
Y. Saito, T. Yoshikawa, M. Inagaki, M. Tomita, and T. Hayashi, Chem.
Phys. Lett. 204, 277 ͑1993͒.
R. S. Ruoff and D. C. Lorents, Carbon 33, 925 ͑1995͒.
V. Ivanov, A. Fonseca, J. B. Nagy, A. Lucas, D. Lambin, D. Bernaerts,
and X. B. Zhang, Carbon 33, 1727 ͑1995͒.
nanotubes. The very clear and strong third-order peak at
Ϫ1
4
274 cm can be safely attributed to a combination of 2
Ϫ1
15
ϫ1348ϩ1581 cm . However, this peak shows a large
downshift in frequency with respect to both the 4320 cm
peak of HOPG and the 4305 cm peak of PG. Further-
Ϫ1
1
1
6
7
Ϫ1
25
Ϫ1
more, the third-order peaks at 4043, 4511, and 4819 cm
have also been observed and assigned to 3ϫ1348 cmϪ1,
18
Ϫ1
Ϫ1
1
348ϩ2ϫ1581 cm , and 1581ϩ2ϫ1617 cm , respec-
tively. The fourth-order Raman band at 5861 cm is also
1
2
2
9
0
1
G. Vitali, M. Rossi, M. L. Terranova, and V. Sessa, J. Appl. Phys. 77,
Ϫ1
4307 ͑1995͒.
observed clearly in Fig. 3͑b͒ and can be assigned to the com-
D. G. McCulloch, S. Prawer, and A. Hoffman, Phys. Rev. B 50, 5905
͑1994͒.
V. Barbarossa, F. Galluzzi, R. Tomaciello, and A. Zanobi, Chem. Phys.
Lett. 185, 53 ͑1991͒.
F. Tuinstra and J. L. Koenig, J. Chem. Phys. 53, 1126 ͑1970͒.
R. J. Nemanich and S. A. Solin, Phys. Rev. B 20, 392 ͑1979͒.
R. Tsu, J. H. Gonzalez, and I. G. Hernandez, Solid State Commun. 27,
Ϫ1
bination of 2ϫ1348ϩ2ϫ1581 cm , while the fourth-order
Ϫ1
Raman spectrum at 5392 cm seems to appear and can be
Ϫ1
assigned to 4ϫ1348 cm
.
22
23
Basically, all the Raman features of aligned CCNTs are
similar to those of PG.25 This indicates that the microstruc-
ture of individual tubes of CCNTs is composed of many
crystalline domains, which possess a high degree of crystal-
line order, as shown by the appearance of high-order Raman
24
507 ͑1978͒.
25
Y. Kawashima and G. Katagiri, in International Conference on Raman
Spectroscopy, 14th, Hong Kong, 1994, edited by N. T. Yu and X. Y. Li
͑Wiley, New York͒, p. 752.
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:
686 Appl. Phys. Lett., Vol. 70, No. 20, 19 May 1997 Wenzhi et al.
28.42.202.150 On: Sun, 23 Nov 2014 05:42:20
2
1