Patterned growth and contact transfer of well-aligned carbon nanotube films

Article abstract of DOI:10.1021/jp990342v

The preparation of aligned and/or micropatterned carbon nanotubes is important to applications ranging from nanocomposites to field-emitting displays. By pyrolysis of iron (II) phthalocyanine under Ar/H₂ at 800-1100 °C, we have synthesized large arrays of vertically carbon nanotubes on various substrates, including quartz glass plates, from which substrate-free films were obtained simply by immersing the nanotube-deposited quartz plates into an aqueous hydrofluoric acid solution. Micropatterns of the aligned nanotubes suitable for device fabrication were generated either by patterned growth of the nanotubes on a partially masked/prepatterned surface or through a contact printing process involving region-specific transfer of the substrate-free nanotube films to other substrates (e.g., polymer films), which otherwise may not be suitable for nanotube growth at high temperatures.

Full text of DOI:10.1021/jp990342v

J. Phys. Chem. B 1999, 103, 4223-4227
4223
Patterned Growth and Contact Transfer of Well-Aligned Carbon Nanotube Films
Shaoming Huang,† Liming Dai,* and Albert W. H. Mau
CSIRO Molecular Science, Bag 10, Clayton South, Victoria 3169, Australia
ReceiVed: January 28, 1999
The preparation of aligned and/or micropatterned carbon nanotubes is important to applications ranging from
nanocomposites to field-emitting displays. By pyrolysis of iron (II) phthalocyanine under Ar/H2 at 800-
1100 °C, we have synthesized large arrays of vertically aligned carbon nanotubes on various substrates,
including quartz glass plates, from which substrate-free films were obtained simply by immersing the nanotube-
deposited quartz plates into an aqueous hydrofluoric acid solution. Micropatterns of the aligned nanotubes
suitable for device fabrication were generated either by patterned growth of the nanotubes on a partially
masked/prepatterned surface or through a contact printing process involving region-specific transfer of the
substrate-free nanotube films to other substrates (e.g., polymer films), which otherwise may not be suitable
for nanotube growth at high temperatures.
Carbon nanotubes are promising as new optoelectronic
materials,1-4 for example, as new electron field emitters in panel
displays5 and single-molecular transistors in microelectronics.6
For most of these applications, it is highly desirable to prepare
aligned and/or micropatterned carbon nanotubes so that the
properties of individual nanotubes can be easily assessed and
they can be incorporated effectively into devices.3,7 Although
carbon nanotubes synthesised by most of the common tech-
niques, such as arc discharge1,2 and catalytic pyrolysis,8 often
exist in a randomly entangled state,2,8 aligned carbon nanotubes
have been prepared either by postsynthesis manipulation9,10 or
by synthesis-induced alignment.11-17 However, the preparation
of micropatterned and/or free-standing films of aligned nano-
tubes remains both scientifically and technically challenging.
Here, we report a novel method for synthesis of perpendicularly
aligned carbon nanotubes over large areas. More importantly,
we also describe the first preparation of micropatterns and
substrate-free films of the aligned nanotubes, which can be
readily transferred onto various substrates including those that
otherwise may not be suitable for nanotube growth at high
temperatures (e.g., polymer films).
We prepared aligned carbon nanotubes by pyrolysis of iron
(II) phthalocyanine, FeC32N8H16 (Aldrich, designated as FePc
hereafter), which, like Ni phthalocyanine,18 contains both the
metal catalyst and carbon source required for the nanotube
growth. The pyrolysis of FePc was performed under Ar/H2 at
800-1100 °C using an appropriate substrate in a flow reactor
consisting of a quartz glass tube and a dual furnace fitted with
independent temperature controllers (Figure 1). The resulting
carbon nanotubes appeared on the substrate (e.g., a quartz glass
plate) as a black layer, which could be scraped off from the
substrate as powder. More importantly, the black deposit could
be easily separated from the quartz glass as a substrate-free,
floating film simply by immersing the nanotube-deposited quartz
plate into an aqueous hydrofluoric acid solution (10-40%
w/w).19 As we discuss later, this technique allows a patternwise/
nonpatternwise transfer of the nanotube films onto various other
substrates of particular interest (e.g., electrodes for electro-
chemistry, polymer films for organic optoelectronic devices)
with the integrity of the aligned nanotubes being largely
maintained in the transferred films. Alternatively, micropatterns
of the aligned nanotubes have also been prepared by patterned
growth of the nanotubes on a prepatterned surface or using a
mask in the same manner as for patterned plasma polymeriza-
tion.20
Figure 2a represents a typical scanning electron microscopic
(SEM, XL-30 FEG SEM, Philips, at 5 KV) image of the carbon
nanotubes, showing that the as-synthesized nanotubes align
almost normal to the substrate surface. The aligned nanotubes
grow homogeneously over the whole quartz glass plate so that
the nanotube growing area is essentially only limited by the
size of the substrate. Figure 2b shows a photograph of a film
(ca. 4 × 1 cm2) of the aligned nanotubes, floating on the HF/
H2O solution (vide supra). As can be seen in Figure 2a, the
aligned nanotubes produced under normal experimental condi-
tions used in this study (see caption of Figure 1) are almost
free from pyrolytic impurities (e.g., carbon particles and other
carbonaceous materials). Prolonged pyrolysis, however, may
generate some amorphous carbon on the top of the aligned
nanotube layer. Inspection of the film edge in Figure 2a at a
higher magnification (Figure 2c) shows that the aligned nano-
tubes are densely packed with a fairly uniform tubular length
of ca. 9 µm. However, the length of the aligned nanotubes can
be varied over a wide range (up to several tens of micrometers)
in a controllable fashion by changing the experimental condi-
tions (e.g., the pyrolysis time, flow rate). A well-graphitized
structure with ca. 40 concentric carbon shells and an outer
diameter of ca. 40 nm is illustrated in the high-resolution
transmission electron microscopic (HR-TEM, CM30, Philips,
at 300 KV) image of an individual nanotube (Figure 2d).
Electron energy loss spectroscopy (EELS attached on the HR-
TEM, CM 30 Philips) reveals carbon K-edges at 284 and 291
eV, attributable to the π* and δ* features of the sp2 hybridized
carbons, along with a signal at 400 eV due to nitrogen,13
indicating that the nanotubes contain nitrogen.13,18 This is further
confirmed by X-ray photoelectron spectroscopic (XPS, Kratos
* Corresponding author. E-mail: Liming.Dai@molsci.csiro.au.
† On leave of absence from Nanjing University, Nanjing 210093, P.R.
China.
10.1021/jp990342v CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/07/1999

4224 J. Phys. Chem. B, Vol. 103, No. 21, 1999
Letters
Figure 1. Apparatus for the generation of aligned carbon nanotubes by pyrolysis of FePc. In typical experiments, a predetermined amount of FePc
(0.1-0.3 g) and a clean quartz glass plate (4 × 1 × 0.125 cm, ultrasonicated in acetone) were placed in the quartz glass tube (as shown). A flow
of Ar/H2 (1:1 to 1:2 v/v, 20-40 cm3/min) mixture was then introduced into the quartz tube during heating. After the second furnace reached
800-1100 °C, the first furnace was heated to 500-600 °C for 2-15 min. Thereafter, both furnaces were kept at the pyrolysis temperature (800-
1100 °C) for an additional 20-30 min for the deposition of nanotubes to be completed.
Analytical, monochromatised Al KR at 200 W) measurements.
While the XPS N1s spectrum is similar to that observed for
N-substituted graphite, showing doublets at the binding energies
(BE) of 398.4 and 400.7 eV,21,18 the corresponding BE value
for the C1s peak is almost the same as that of graphite (284.7
eV).
prepared is given in Figure 4a, which shows a close replication
of the mask structure. As can be seen in Figure 4a, micropatterns
consisting of the densely packed, perpendicularly aligned
nanotubes are clearly evident. In this case, the spatial resolution
of the patterns is mainly limited by the resolution of the mask
used. Furthermore, our preliminary results indicate that the
aligned nanotube patterns can also be produced by first
depositing a thin, patterned ITO or Au layer onto an appropriate
substrate (e.g., a quartz glass plate) using a mask (e.g., a TEM
grid) followed by deposition of the pyrolyzed FePc and removal
of the ITO- or Au-covered regions by selective dissolution in
an aqueous hydrochloride or cyanide solution. This method
allowed us to generate a positive image of the mask.
Further investigation on the nanotube microfabrication showed
that both a positive and a negative image of the aligned
nanotubes can be produced simultaneously simply by lifting up
a substrate-free nanotube film floating on the HF/H2O solution
(vide supra) onto a TEM grid, followed by region-specific
transfer of it to another substrate (e.g., polystyrene film) through
physical contact between the TEM grid and the substrate surface.
This contact transfer process is much like the widely used
microcontact printing (µCP) technology for generating self-
assembled monolayer patterns.24 In the present study, the
nanotubes suspended over the windows were selectively trans-
ferred onto the substrate surface as a negative image (Figure
4b) while those on the grid bars remained there as a positive
pattern (Figure 4c). The most important feature to note in parts
b and c of Figure 4 is that the constituent nanotubes in the µCP-
generated patterns are still aligned normal to the substrate
surfaces.
The growth mechanism for aligned nanotubes deposited either
in porous matrixes11,12 or metal-coated surfaces13,14 has been
proposed previously. Our SEM images taken at different stages
during the pyrolysis (Figure 3) are consistent with the following
process. Upon the thermal decomposition of FePc, Fe particles,
surrounded by carbon, form on the substrate surface (Figure
3a). Segregation of Fe then occurs, leading to an increase in
the size of the catalytic center. Once the Fe particle reaches an
optimal size for carbon nanotube nucleation,22 the surrounding
carbon transforms into a graphite tube. Further decomposition
of FePc provides carbon to the contact region between the Fe
particle and the already formed tubule segment, allowing
continuous growth of the carbon nanotube in the direction
normal to the substrate surface (Figure 3b). Iron atoms may be
preferentially left at the first part of the reactor, where the
temperature was above the FePc decomposition point during
this stage of pyrolysis. This “base-growth” process23 is supported
by the presence of metal particles at the bottom of the nanotubes,
as shown in Figure 3c. The high surface density of the growing
nanotubes serves as an additional advantage for the constituent
nanotubes to be “uncoiled” and allows the as-grown aligned
nanotubes to be removed as a substrate-free film.
The perpendicular alignment of the nanotubes should facilitate
device construction. In this regard, we have tried to grow aligned
carbon nanotubes on various conducting surfaces including Au,
Pt, Cu, and ITO (indium tin oxide). Such constructs have many
potential applications, especially in electrochemistry. We have
also developed techniques for patterning the aligned nanotubes
either through patterned growths or by postsynthesis microfab-
rications. For example, the patterned growth can be achieved
by placing a mask (e.g., a TEM grid consisting of hexagonal
windows) on the quartz glass substrate. The partially masked
quartz glass plate was then coated with aligned nanotubes by
the pyrolysis of FePc, as described above. The resulting
nanotube films were studied by SEM after careful removal of
the TEM grid. An example of the nanotube patterns thus
In summary, we have demonstrated that vertically aligned
nanotubes can be prepared over large areas by pyrolysis of iron
phthalocyanine under appropriate conditions. We have also
developed simple methods for obtaining large-area, substrate-
free nanotube films and for micropatterning vertically aligned
nanotubes. While the substrate-free nanotube films can be
transferred onto various substrates of particular interest, the
techniques for micropattern formation described in this paper
should also be applicable to many other systems. With these
methods, we can deposit the aligned nanotubes, in either a
patterned or nonpatterned fashion, onto various substrates
including those that otherwise cannot be used for nanotube

Letters
J. Phys. Chem. B, Vol. 103, No. 21, 1999 4225
Figure 2. (a) Typical SEM image of the aligned nanotubes. Part of the carbon nanotube film was peeled off from the quartz substrate with
tweezers for examining the orientation and alignment of the nanotubes. (b) Typical photographic image of the large-area, substrate-free, aligned
nanotube films floating on a HF/H2O solution (30% w/w) contained in a polystyrene petri dish. (c) Enlarged view of (a) along the peeled edge. The
misalignment seen for some of the carbon nanotubes was caused by the peeling action. (d) Typical HR-TEM image of the individual constituent
nanotubes showing ca. 40 concentric graphitic layers.
growth (e.g., polymer films). Therefore, the combination of the
simple methods for production of large-area aligned nanotubes
and the simple techniques for micropatterning of them should
not only facilitate the characterization of physical properties of

4226 J. Phys. Chem. B, Vol. 103, No. 21, 1999
Letters
Figure 4. Typical SEM micrographs of patterned films of aligned
nanotubes. (a) Film produced by patterned growth using a TEM grid
consisting of hexagonal windows as the mask. (b) The microcontact-
printed negative image on a Au surface. (Although the negative pattern
could also be deposited onto various other substrates including polymers
(e.g., polystyrene), Au was chosen for better SEM imaging). (c) The
microcontact-printed positive image on a TEM grid used as the stamp.
Note that the TEM grides used for (a) and (b, c) have grid bars of ca.
8 and 30 µm wide and hexagonal windows with a width of 50 and 80
µm, respectively.
Figure 3. SEM micrographs showing the growth of the aligned
nanotubes during the pyrolysis. (a) Carbon-surrounded iron nanopar-
ticles formed at the initial stage of the pyrolysis. (b) The initial growth
of the nanotubes. (c) The resultant aligned nanotubes normal to the
substrate surface.
nanotubes but also allow their effective incorporation into
devices for practical applications.

Letters
Note Added in Proof. After the submission of our manu-
script, we noted that Fan et al. recently reported the synthesis
of regular arrays of oriented nanotubes on Fe-patterned porous
silicon by pyrolysis of ethylene (Fan, S.; Chapline, M. G.;
Franklin, N. R.; Tomber, T. W.; Cassell, A. M.; Dai, H. Science
1999, 283, 512).
J. Phys. Chem. B, Vol. 103, No. 21, 1999 4227
(10) De Heer, W. A.; Bacsa, W. S.; Chaˆtelain, A.; Gerfin, T.; Humphrey-
baker, R.; Forro´, L.; Ugarte, D. Science 1995, 268, 845.
(11) Li, W. Z.; Xie, S. S.; Qian, L. X.; Chang, B. H.; Zhous, B. H.;
Zhou, W. Y.; Zhao, R. A.; Wang, G. Science 1996, 274, 1701. Pan, Z. W.;
Xie, S. S.; Chang, B. H.; Wang, C. Y.; Lu, L.; Liu, W.; Zhou, W. Y.; Li,
W. Z. Nature 1998, 394, 631.
(12) See, for example: Kyotani, T.; Tsai, L. F.; Tomita, A. Chem. Mater.
1996, 8, 2109. Che, G.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Nature
1998, 393, 346.
Acknowledgment. S.H. is grateful for a fellowship from
the Department of Industry, Science, and Technology (DIST),
Australia.
(13) Terrones, M.; Grobert, N.; Olivares, J.; Zhang, J. P.; Terrones, H.;
Kordatos, K.; Hsu, W. K.; Hare, P. P.; Townsend, P. D.; Prassides, K.;
Cheetham, A. K.; Kroto, H. W.; Walton, D. R. M. Nature 1997, 388, 52.
(14) Ren, Z. F.; Huang, Z. P.; Xu, J. H.; Wang, P. B.; Siegal, M. P.;
Provencio, P. N. Science 1998, 282, 1105.
(15) Kusunoki, M. K.; Rokkaku, M.; Suzuki, T. Appl. Phys. Lett. 1997,
71, 2620. Kusunoki, M.; Shibata, J.; Rokkaku, M.; Hirayama, T. Jpn. J.
Appl. Phys. 1998, 37, L605.
References and Notes
(1) Iijima, S. Nature 1991, 354, 56.
(2) Ebbesen, T. W.; Ajayan, P. M. Nature 1992, 358, 220.
(3) See, for example: De Heer, W. A.; Chaˆtelain, A.; Ugarte, D.
Science 1995, 270, 1179. Dresselhaus, M. S.; Dresselhaus, G.; Eklund, P.
C. Science of Fullerenes and Carbon Nanotubes; Academic Press: New
York, 1996. Yakobson, B. I.; Smalley, R. E. American Scientist 1997, 85,
324. Ebbsen, T. Carbon Nanotubes; CRC Press: Boca Raton, FL, 1997.
Ajayan, P. M.; Ebbesen, T. W. Rep. Prog. Phys. 1997, 60, 1025. Wong, E.
W.; Sheehan, P. E.; Lieber, C. M. Science 1997, 277, 1971. Collins, P. G.;
Babdo, H.; Zettl, A. Nanotechnol. 1998, 9, 153. Frank, S.; Poncharal, P.;
Wang, Z. L.; de Heer, W. A. Science, 1998, 280, 1744.
(4) See, for example: Ebbesen, T. W. Phys. Today 1996, 26. Collins,
P. G.; Zettl, A.; Bando, H.; Thess, A.; Smalley, R. E. Science 1997, 278,
100.
(5) Wang, Q. H.; Setlur, A. A.; Lauerhaas, J. M.; Dai, J. Y.; Seelig, E.
W.; Chang, P. R. H. Appl. Phys. Lett. 1998, 72, 2912.
(6) See, for example: Kouwenhoven, L. Science 1997, 275, 1896. Tans,
S. J.; Verscheren, A. M. R.; Dekker, C. Nature 1998, 393, 49.
(7) Kong, J.; Soh, H. T.; Cassell, A. M.; Quate, C. F.; Dai, H. Nature
1998, 395, 878.
(16) Takikawa, H.; Miyano, R.; Yatsuki, M.; Sakakibara, T. Jpn. J. Appl.
Phys. 1998, 37, L187.
(17) Rao, C. N.; Sen, R.; Satishkumar, B. C.; Govindaraj, A. J. Chem.
Soc., Chem. Commun. 1998, 1525.
(18) Yudasaka, M.; Kikuchi, R.; Ohki, Y.; Yoshimura, S. Carbon 1997,
35, 195.
(19) Upon the immersion of the nanotube-deposited quartz glass plate
into the HF/H2O solution, the nanotube film was found to spontaneously
drift away from the quartz glass surface due, most probably, to acid etching
of the quartz glass surface underneath the nanotubes. The floating nanotube
film was very quickly transferred into pure water to minimize any possible
side effect. Aqueous solutions of HF, however, have been commonly used
to dissolve aluminium oxide membranes for recovering the template-
synthesised carbon nanotubes without any noticeable effect on the nanotube
properties (see ref 12).
(20) Dai, L.; Griesser, H. J.; Mau, A. W. H. J. Phys. Chem. B 1997,
101, 9548.
(21) Matsui, T.; Yudasaka, R.; Kikuchi, R.; Ohki, Y.; Yoshimura, S.
Appl. Phys. Lett. 1994, 65, 2145.
(22) Yudasaka, M.; Kikuchi, R.; Matsui, T.; Ohki, Y.; Yoshimura, S.
(8) See, for example: Endo, M.; Takeuchi, K.; Kobori, K.; Takahashi,
K.; Kroto, H. W.; Sarkar, A. J. Phys. Chem. Solids 1994, 54, 1841. Ivanov,
V.; Nagy, J. B.; Lambin, Ph.; Lucas, A.; Zhang, X. B.; Zhang, X. F.;
Bernaerts, D.; Tendeloo, G. V.; Amelinckx, S.; Landuyt, J. V. Chem. Phys.
Lett. 1994, 223, 329.
(9) Ajayan, P. M.; Stephan, O.; Collix, C.; Trauth, D. Science 1994,
265, 1212.
Appl. Phys. Lett. 1995, 67, 2477.
(23) Baker, R. T. K. Carbon 1989, 27, 315.
(24) See, for example: Xia, Y.; Whitesides, G. M. Ann. ReV. Mater.
Sci. 1998, 28, 153. Xia, Y.; Qin, D.; Whitesides, G. M. AdV. Mater. 1996,
8, 1015 and references cited therein.