Irradiation of CW-CO2 laser on a powder target. Formation of fullerene film from graphite powder

Article abstract of DOI:10.1016/S0009-2614(03)01326-5

A process was investigated to irradiate a continuous-wave CO₂ laser (4.5 kW) on graphite powder with a mean diameter of 5 μm in a flow of Ar or He gas at 30 or 200 Torr. The film produced on a Ni substrate in the Ar flow at 200 Torr showed the Raman bands of C₆₀ and C₇₀, and the C₆₀/C₇₀ ratio was estimated to be 5/1 by MALDI-TOF mass spectrometry. The film formation process is discussed on the basis of an analysis of laser-irradiated powder, which included 0.01 wt% of C₆₀, by UV-Vis absorption, and by observations of the plumes and SEM images.

Full text of DOI:10.1016/S0009-2614(03)01326-5

Chemical Physics Letters 378 (2003) 474–480
www.elsevier.com/locate/cplett
Irradiation of CW-CO2 laser on a powder target.
Formation of fullerene film from graphite powder
*
Seisuke Kano , Masamichi Kohno, Kaname Sakiyama, Shinya Sasaki,
Nobuhiro Aya, Hirofumi Shimura
Research Center for Advanced Manufacturing on Nanoscale Science and Engineering, National Institute of
Advanced Industrial Science and Technology (AIST), Namiki 1-2-1, Tsukuba, Ibaraki 305-8564, Japan
Received 27 December 2002; in final form 9 July 2003
Published online: 26 August 2003
Abstract
A process was investigated to irradiate a continuous-wave CO2 laser (4.5 kW) on graphite powder with a mean
diameter of 5 lm in a flow of Ar or He gas at 30 or 200 Torr. The film produced on a Ni substrate in the Ar flow at 200
Torr showed the Raman bands of C60 and C70, and the C60/C70 ratio was estimated to be 5/1 by MALDI-TOF mass
spectrometry. The film formation process is discussed on the basis of an analysis of laser-irradiated powder, which
included 0.01 wt% of C60, by UV–Vis absorption, and by observations of the plumes and SEM images.
Ó 2003 Elsevier B.V. All rights reserved.
1. Introduction
that assures stable and continuous synthesis of
nano-particles is awaited to achieve preparation of
particles or films with reproducible properties.
In this Letter, we propose a continuous laser
process using a continuous-wave (CW) CO2 laser
and fine particles as a target with inert gas. We
demonstrate this process using graphite powder
and explore processes for production of fullerene
films. Fullerenes can be synthesized by various
methods [6–8]. Laser ablation of a graphite bulk is
another popular method of synthesis [9,10]. Re-
cently, CO2 lasers have been used to synthesize
fullerenes [11] and other forms of carbon [12–16].
Kasuya et al. [11] synthesized C60 using pulsed
CO2 laser (pulse duration of 500 ms) to vaporize a
graphite bulk in the presence of He, Ar or N2 as
buffer gases. Almost all these methods used a solid
Laser processing is one of the popular methods
for producing nano-size particles and their depo-
sition films. This process, so called laser ablation
or deposition technique, usually employs a pulsed
beam and a bulk target [1–5]. The merit of this
process is that it enables direct production of pure
particles or films from various types of starting
materials in a low-pressure inert gas. However,
such a batch system needs replacement of the
target, and fluctuation of conditions caused by
target consumption is inevitable. Hence, a method
* Corresponding author. Fax: +81-298-61-7871.
E-mail address: s.kano@aist.go.jp (S. Kano).
0009-2614/$ - see front matter Ó 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0009-2614(03)01326-5

S. Kano et al. / Chemical Physics Letters 378 (2003) 474–480
475
graphite bulk. In our present method, graphite
powder is used instead and irradiated by a CW-
CO2 laser. This method enables a continuous
supply of a target material and minimizes fluctu-
ation of conditions.
before laser irradiation. The diameter of the focal
spot was ꢀ3.0 mm, and the distance between the
nozzle and the center of the beam was set to
3.0–7.0 mm. To form a thin film, a nickel substrate
(14 ꢁ 17 ꢁ 30 mm3) was placed 1.0–3.0 mm from
the center of the laser beam and made an angle
between the deposition surface and the laser beam
by 0–3.0°. The substrate was scanned in the di-
rection parallel to the laser beam with a speed of
0.3–1.0 mm/s. The powder deposited on the sub-
strate surface without heat control formed a thin
film.
The films were analyzed by optical microscopy
and micro-Raman spectroscopy in air at room
temperature. An Ar-ion laser beam (514.5 nm) was
adjusted to lower the power (ꢀ1 mW, ꢀ200 lm in
diameter) to avoid damage of the quality of the
sample. The film was dissolved in toluene, and its
chemical composition was analyzed by MALDI-
TOF spectrometry.
The laser-irradiated powder was also collected
on a piece of filter paper placed along the beam
line ꢀ100 mm apart from the center of the laser
beam. The collected powder was then dissolved in
toluene, and absorption spectra were recorded in
the UV–Vis (300–1100 nm) region. The surfaces of
the collected powder were compared by SEM ob-
servations of morphology. The plume of the
powder-gas flow after laser irradiation was moni-
tored in the vertical direction to the gas flow by a
CCD camera and analyzed spectroscopically at a
position of 10 mm from the laser beam in an area
of 2 mm in diameter.
2. Experimental
The experimental setup shown in Fig. 1, here
denoted as a bulk target system, was newly de-
signed for laser ablation using a powder target.
Graphite powder (Kojundo Chemical Laboratory)
with a mean diameter of 5 lm was stored in a
powder feeder with a buffer gas (Ar or He) at
2 atm. This powder was carried through a stainless
steel tube of 1.5 mm in diameter into a vacuum
chamber and injected from a nozzle, 0.5–1.5 mm in
inner diameter, with the buffer gas. The powder
feeder was vibrated mechanically during the ex-
periment to transport the powder smoothly. The
chamber was first evacuated using rotary and
mechanical booster pumps and then filled with
buffer gas up to 30 or 200 Torr by continuous in-
troduction with the powder keeping the gas pres-
sure constant. The gas types and process pressure
were taken from previous reports on the produc-
tion of fullerenes from graphite [9,11,17–19].
A focused CW-CO2 laser beam (Mitsubishi
ML-6050C, wavelength 10.6 lm, maximum power
at the focal point: 4.5 kW) was irradiated to
graphite powder from a side of the powder-gas
flow. The powder and buffer gas were not heated
Fig. 1. A schematic drawing of the experimental setup.

476
S. Kano et al. / Chemical Physics Letters 378 (2003) 474–480
3. Results
3.1. Film formation
toluene and analyzed by MALDI-TOF spec-
trometry. The spectrum shown in Fig. 2b exhibits
two clear peaks of C60 and C70. The ratio of C60
and C70 was estimated to be ꢀ5:1. No higher
fullerenes (Cn; n > 70) were detectable in this TOF
spectrum.
The surface of the Ni substrate was quite dif-
ferent depending on the setting position and
scanning speed. When the substrate was scanned
with 0.3 mm/s, the substrate surface was partly
melted. When the substrate was set more than
2 mm apart from the laser beam and at an angle of
less than 2.0°, the substrate surface was either
unchanged or covered with graphite powder. The
powder samples were probed with an optical mi-
croscope and analyzed by Raman spectroscopy.
The process conditions were as follows: scanning
speed, 1 mm/s; distance and angle between the
laser beam and the substrate, 2.0 mm and 3.0°,
respectively; and nozzle diameter, 1.5 mm.
Under these setup conditions and with Ar and
He gases at 30 and 200 Torr, brown-colored and
transparent areas were observed on the Ni surface
by the optical microscope. A micrograph of the
film is shown in Fig. 2a. This film was deposited in
Ar gas at 200 Torr and formed along the line on
which the substrate was moved. For identification
of the contents of this film, it was dissolved in
As the distributions of C60 in the films were
inhomogeneous, Raman shifts at 10 points in each
film were measured and compared (Fig. 3). The
differences in the noise levels observed in Figs. 3b–
d are attributed to those in the thickness, density,
crystallinity, and other conditions of the films. The
Raman shifts observed in crude graphite powder
(Fig. 3a) showed a broad peak at 1350 cmꢂ1
,
assignable to the ÔD-bandÕ, and a broad peak at
1590 cmꢂ1, assignable to the ÔG-bandÕ of graphite.
In the Raman shifts for the deposited films at 200
Torr of He gas and 30 Torr of Ar gas, Figs. 3b and
c, a small peak at 1469 cmꢂ1 observed between the
D- and G-bands (indicated by an asterisk) was
assigned to the Ag mode of C60 [20], indicating that
the film contained C60. The Raman shift for the
film synthesized at 200 Torr of Ar gas, Fig. 3d,
showed that this peak at 1469 cmꢂ1 was much
more enhanced relative to the D- and G-bands
than those in Figs. 3b and c so that this sample,
Fig. 2. (a) Optical micrograph of a deposited film synthesized in Ar gas at 200 Torr and (b) its MALDI-TOF spectrum.

S. Kano et al. / Chemical Physics Letters 378 (2003) 474–480
477
Fig. 4. Typical examples of the UV–Vis absorption spectrum of
toluene-extracted carbon powder: (a) crude graphite powder,
(b) laser-irradiated carbon powder in Argon gas at 200 Torr. A
peak at 335 nm (asterisk) is assigned to C60, and one at 378 nm
Fig. 3. Raman spectra of crude graphite powder (a) and syn-
thesized thin films (b–d). Thin films were synthesized in (b) He
gas at 200 Torr, (c) Ar gas at 30 Torr, and (d) Ar gas at 200
Torr. The peak with asterisk at 1469 cmꢂ1 is assigned to C60 and
(arrow) probably to C70
.
those with arrows at 1186, 1230, and 1568 cmꢂ1 to C70
.
peaks of fullerenes were observed in the reference
spectrum of crude graphite powder, Fig. 4a. Hence
we concluded that no fullerenes were contained in
the starting material in any amount beyond the
sensitivity of our spectrometer. In contrast, the
absorption spectrum of the laser-irradiated pow-
der prepared in Ar at 200 Torr, Fig. 4b, showed a
peak at 335 nm (indicated by an asterisk) corre-
sponding to the p ! pꢃ transition of C60 [22,23],
indicating that the laser-irradiated powder con-
tained C60. In addition, a small peak also observed
at 378 nm (indicated by an arrow) probably cor-
responds to C70 [24]. By comparison with pure C60
measured for reference, the yield of C60 from the
absorption spectrum in Fig. 4b was estimated to be
ꢀ0.01 wt%. The laser-irradiated powders prepared
under other gas conditions showed the absorption
peaks corresponding to C60 with similar concen-
trations.
Fig. 3d, contained C60 at a much higher concen-
tration.
In summary, 200 Torr of Ar gas was found to
be clearly the better operating condition than at 30
Torr of Ar and at 200 and 30 Torr of He for the
deposition of fullerenes. Several other peaks at
1186, 1230, and 1568 cmꢂ1 (indicated by arrows),
observed in Fig. 3d, were assigned to the vibra-
tional modes of C70 [21], indicating that C70 was
also present in the film.
3.2. Powder analysis and plume observation
The laser-irradiated powder dissolved in tolu-
ene was analyzed by UV–Vis absorption spec-
troscopy, as shown in Fig. 4. No characteristic

478
S. Kano et al. / Chemical Physics Letters 378 (2003) 474–480
Fig. 5. Typical images of the plumes observed during the process: (a) configuration of the nozzle, carbon powder jet and laser beam,
and plumes recorded in (b) He gas at 200 Torr, and Ar gas at (c) 30 Torr and (d) 200 Torr.
The surface morphology of the laser-irradiated
powder observed by SEM was compared with that
of crude graphite powder. The edges of the irra-
diated sample seemed to be smoother than that of
crude graphite powder. This surface change, ob-
served for all the buffer-gas conditions, may be
ascribed to the thermal effect of laser energy ab-
sorbed by the powder or powder cloud.
the carbon powder in the plume was roughly es-
timated from the plume length and calculated
powder-jet velocity to be several tens or hundreds
of microseconds.
4. Discussion
Fig. 5 shows typical plumes observed in our
process. The plume in Ar gas was brighter than
that observed for He gas, suggesting higher tem-
perature. We also compared the image observed at
30 Torr Ar gas (Fig. 5c) with that at 200 Torr,
Fig. 5d. The intensity of the light emission ap-
peared to be weaker at lower pressure. The tem-
perature of the plume was calculated from the
emission spectra assuming black-body radiation to
be ꢀ3500 °C at 200 Torr of Ar gas. The plume
temperatures at other buffer-gas conditions were
calculated to be ꢀ3000 °C. The residence time of
4.1. Formation of fullerenes
The film including fullerenes deposited on the
Ni substrate was formed by laser irradiation on
the graphite powder flowing with rare gas.
Fullerenes were identified in a brown-colored
transparent film prepared at 200 Torr of Ar gas.
In contrast, the laser-irradiated powder collected
on paper filter under the same condition con-
tained C60 with an estimated concentration of
0.01 wt%. The reason for the selective deposition
of fullerenes on the Ni substrate is discussed

S. Kano et al. / Chemical Physics Letters 378 (2003) 474–480
479
below with the aid of the experimental results of
ours and other groups.
activated in this environment. Here, we recall the
conditions required for fullerene formation: above
1000 °C and over 400 ls. The substrate was moved
in the direction parallel to the laser beam by 17 mm
at 1.0 mm/s during this process. Thus the substrate
surface was exposed to the plume flow of ꢀ3000 °C
for more than 17 s. Hence, the carbon species can be
maintained at high enough temperature and for
long enough time to form fullerenes on the Ni sub-
strate.
The substrate temperature can also be estimated
by optical observation of its surface, where no
melted area was observed. Since this indicates that
the surface temperature was lower than the melt-
ing temperature of Ni (1450 °C), it is likely to be
lower than the decomposition temperature of
fullerenes. Thus the hot plume and heated sub-
strate probably worked as source energies for
fullerene formation.
Previous reports suggested that carbon species
should be kept above 1000 °C for a period longer
than 400 ls to synthesize fullerenes using pulsed
lasers [9,11,18] or arc discharge [17]. In our ex-
periment, the plume temperature was estimated to
be higher than 3000 °C for Ar or He at 30 or
200 Torr. These data indicate that carbon species
are hot enough to form fullerenes. This idea has
been supported by our SEM observation of the
particle surface after laser irradiation. Under these
experimental conditions, graphite particles with a
mean diameter of 5 lm absorb laser energy (k ¼
10:6 lm), and sublimated carbon particles are
cooled slowly in the plume in several hundred
microseconds.
On the other hand, the deposited film is formed
on the Ni substrate, which is placed 2 mm from the
laser beam. Therefore, the plume may reach the
substrate surface within a few microseconds after
laser irradiation. This argument indicates that
fullerene synthesis before deposition is rather un-
likely. If fullerenes are formed on the substrate,
then one needs to explain how the sublimated
carbon species are deposited on the substrate.
From the viewpoint of aerosol science, nano-par-
ticles in a gas flow are deposited with a smaller
amount than micro-particles because of the dif-
ference in the mass or size. The particle shapes and
velocities are disregarded in this assumption, be-
cause their influences seem to be less important
than the particle masses or sizes. This implies that
graphite particles are deposited more easily on the
Ni surface than the vaporized carbon species.
Therefore, one of the possibilities is the deposition
of the carbon vapor with graphite particles.
4.2. Conditions of gas flow suitable for fullerene
formation
We have found in this study that synthesis at 200
Torr is more efficient than at 30 Torr, and so is Ar
gas than He. The differences in these gas conditions
probably affect the cooling of vaporized carbon
species and fullerenes and the gas flow for deposi-
tion of graphite particles. Since Ar gas has lower
thermal conductivity than He, Ar is expected to
confine thermal energy more efficiently than He. It is
also the case that a cooling gas with high density or
higher pressure deprives more thermal energy.
Among the four experimental arrangements of the
present study, Ar gas at 200 Torr seems to retain the
highest temperature. The plume at this temperature
may have provided a suitable condition to form
fullerenes efficiently on the Ni substrate, given
thermal energy to the adhering graphite particles,
and facilitated departure from the Ni surface.
In addition to the above-mentioned conditions
(i.e., high temperature and long enough time for
formation of fullerenes), another important re-
quirement for fullerene synthesis is the particular
density of vaporized carbon species. This density
depends on the type and pressure of the buffer gas
and the extent of evaporation of carbon species. If
a large quantity of vaporized carbon species is
Very hot carbon species are probably adhered
to larger particles of lower temperature. The carbon
species can then be adhered to the substrate because
the substrate surface is much colder than the laser-
irradiated graphite particles. A possible disadvan-
tage of this assumption is that the carbon species
may react with the graphite surface, by which ful-
lerene formation might be inhibited. However, the
carbon species and the graphite particles are con-
tained in the high-temperature plume (ꢀ3000 °C) or
exposed to it. The carbon species can thus remain

480
S. Kano et al. / Chemical Physics Letters 378 (2003) 474–480
[2] M. Hirasawa, T. Seto, T. Orii, N. Aya, H. Shimura, Appl.
Surf. Sci. 197–198 (2002) 661.
produced, e.g., in the continuous arc discharge
method [6], He gas is the most suitable. When only
a small quantities of carbon species is produced,
e.g., in nanosecond-pulsed Nd:YAG laser ablation
[7] or microsecond-pulsed arc discharge method
[22], Ar gas is more appropriate. In our experi-
ment, only a small amount is sublimated from the
graphite surface, as estimated from the surface
observation of particles. This finding also suggests
that Ar is a suitable gas for fullerene formation in
our process conditions.
[3] N. Suzuki, T. Makino, Y. Yamada, T. Yoshida, Appl.
Phys. Lett. 78 (2001) 2043.
[4] M. Oujja, E. Rebollar, M. Castillejo, Appl. Surf. Sci. 211
(2003) 128.
[5] A.E. Pichahchy, D.A. Cremers, M.J. Ferris, Spectrochim.
Acta B 52 (1997) 25.
[6] R.E. Haufler, J. Conceicao, L.P.F. Chibante, Y. Chai, N.E.
Byrne, S. Flanagan, M.M. Haley, S.C. OÕBrien, C. Pan, Z.
Xiao, W.E. Billups, M.A. Ciufolini, R.H. Hauge, J.L.
Margrave, L.J. Wilson, R.F. Curl, R.E. Smalley, J. Phys.
Chem. 94 (1990) 8634.
[7] R. Taylor, G.J. Langley, H.W. Kroto, D.R.M. Walton,
Nature 366 (1993) 728.
5. Conclusion
[8] J.B. Howard, J.T. McKinnon, Y. Makarovsky, A.L.
Lafleur, M.E. Johnson, Nature 352 (1991) 139.
[9] R.E. Haufler, Y. Chai, L.P.F. Chibante, J. Conceicao, C.
Jin, L.S. Wang, S. Maruyama, R.E. Smalley, Mater. Res.
Soc. Symp. Proc. 206 (1991) 627.
A process using CW-CO2 laser irradiation
(4.5 kW) to a target powder carried with buffer gas
is demonstrated using graphite powder with a
mean diameter of 5 lm for the synthesis of ful-
lerene film. We have clearly observed by Raman
spectroscopy a fullerene film on a Ni substrate
formed by the flow of Ar gas at 200 Torr without
temperature control. The powder collected on a
filter replacing Ni substrate contained 0.01 wt% of
C60. The following scheme is suggested for the
fullerene formation on the substrate from the es-
timated plume temperature during film formation,
ꢀ3500 °C, and smooth morphology of the powder
surface: The graphite powder in the flow absorbs
laser energy and causes partial sublimation of
carbon from the powder surface. The vaporized
carbon species are deposited on the Ni surface and
form fullerenes during exposure to the hot plume
in several hundred microseconds.
[10] T. Wakabayashi, D. Kasuya, H. Shiromaru, S. Suzuki,
K. Kikuchi, Y. Achiba, Z. Phys. D 40 (1997) 414.
[11] D. Kasuya, F. Kokai, K. Takahashi, M. Yudasaka,
S. Iijima, Chem. Phys. Lett. 337 (2001) 25.
ꢀ
[12] W.K. Maser, E. Munoz, A.M. Benito, M.T. Martinez,
G.F. de la Fuente, Y. Maniette, E. Anglaret, J.-L.
Sauvajol, Chem. Phys. Lett. 292 (1998) 587.
[13] S. Iijima, M. Yudasaka, R. Yamada, S. Bandow,
K. Suenaga, F. Kokai, K. Takahashi, Chem. Phys. Lett.
309 (1999) 165.
[14] F. Kokai, K. Takahashi, M. Yudasaka, R. Yamada,
T. Ichihashi, S. Iijima, J. Phys. Chem. B 103 (1999)
4346.
[15] S. Bandow, F. Kokai, K. Takahashi, M. Yudasaka, L.C.
Qin, S. Iijima, Chem. Phys. Lett. 321 (2000) 514.
[16] D. Kasuya, M. Yudasaka, K. Takahashi, F. Kokai,
S. Iijima, J. Phys. Chem. B 106 (2002) 4947.
[17] T. Sugai, H. Omote, H. Shinohara, Eur. Phys. J. D 9 (1999)
369.
ꢁ
[18] T. Ishigaki, S. Suzuki, H. Kataura, W. Kreschmer,
Y. Achiba, Appl. Phys. A 70 (2000) 121.
[19] Y. Saito, M. Inagaki, H. Shinohara, H. Nagashima,
M. Ohkochi, Y. Ando, Chem. Phys. Lett. 200 (1992)
643.
Acknowledgements
[20] A.M. Rao, P. Zhou, K.-A. Wang, G.T. Hager, J.M.
Holden, Y. Wang, W.-T. Lee, X.-X. Bi, P.C. Eklund, D.S.
Cornett, M.A. Duncan, I.J. Amster, Science 259 (1993)
955.
We are grateful to Professor S. Onari (Tsukuba
University), Drs. S. Suzuki (Tokyo Metro. Uni-
versity) and T. Orii (AIST) for their help with
measurements of UV–Vis absorption spectros-
copy, and to Applied Laser Engineering Center
Co. Ltd. for technical support.
[21] D.S. Bethune, G. Meijer, W.C. Tang, H.J. Rosen, Chem.
Phys. Lett. 174 (1990) 219.
[22] S. Leach, M. Vervloet, A. Despres, E. Breheret, J.P. Hare,
T.J. Dennis, H.W. Kroto, R. Taylor, D.R.M. Walton,
Chem. Phys. 160 (1992) 451.
References
[23] S. Kazaoui, R. Ross, N. Minami, Solid State Commun. 90
(1994) 623.
[24] J.P. Hare, H.W. Kroto, R. Taylor, Chem. Phys. 177 (1991)
394.
[1] M. Hirasawa, T. Seto, N. Aya, J. Nanosci. Nanotechnol. 1
(2001) 381.