Fullerene macro structures

Article abstract of DOI:10.1016/S0009-2614(97)01460-7

At an arc discharge with CH₄ present in the gas ambient, conditions were created stimulating the simultaneous formation of single layer fullerene structures, which after being covered by many layers, yield multiple layer structures. As a result well formed regions containing numerous spherical, conical and polyhedral macro forms of the order of several micrometers were grown. Also obtained were regions containing large quantities of nanotubes over 10 μm in length. A model is proposed to explain the structure built up. To aid in the pyrolytic growth of cover layers the temperature of the deposit was maintained high by the inverse method .

Full text of DOI:10.1016/S0009-2614(97)01460-7

13 March 1998
Ž
.
Chemical Physics Letters 285 1998 1–6
Fullerene macro structures
N.S. Koprinarov, M.V. Marinov, G.V. Pchelarov, M.A. Konstantinova
Bulgarian Academy of Sciences, Central Laboratory on Solar Energy and New Energy Sources, 72 Tsarigradsko shose, 1784 - Sofia,
Bulgaria
Received 31 October 1997; in final form 9 December 1997
Abstract
At an arc discharge with CH4 present in the gas ambient, conditions were created stimulating the simultaneous formation
of single layer fullerene structures, which after being covered by many layers, yield multiple layer structures. As a result
well formed regions containing numerous spherical, conical and polyhedral macro forms of the order of several micrometers
were grown. Also obtained were regions containing large quantities of nanotubes over 10 mm in length. A model is proposed
to explain the structure built up. To aid in the pyrolytic growth of cover layers the temperature of the deposit was maintained
high by the ‘‘inverse method’’. q 1998 Elsevier Science B.V.
1. Introduction
sent in an environment affording coverage by several
layers and the conditions will rarely be suitable for
the growth of many layers. Even though the spheri-
cal multi-layer forms obtained so far only have
several layers, they are known as gigantic fullerenes.
When special measures are taken during synthesis
w x
Following the discovery of C60 1 it was found
that a fullerene type of carbon net consisting of 5
and 6 atom rings, can close in different spatial form
w
x
structures 2–5 . Once built, the basic structure can
be enveloped by multiple concentric layers as a
result of which forms resembling unions or Russian
dolls are grown, these exhibit a minimum surface of
circumscription and a maximum volume to surface
ratio. In principle the number of enveloped layers is
unlimited, however in order to build these up it is
necessary for the growth process to be maintained
for a long time and to secure a constant flux of a
large quantity of C atoms. It is almost impossible to
maintain such growth conditions via the methods
presently used for obtaining fullerenes, since during
synthesis the structures formed, quickly leave the arc
discharge region and the influx of carbon atoms
required for their growth is halted. Only a small part
of the fullerene structures can be considered as pre-
w x
6 , the number of multiple layer forms can be
increased, while the quantity of atoms that are a part
of these forms may reach a few thousand. The new
type of layered structures are interesting mainly be-
cause of their large mass at a minimum surface of
circumscription. As an example, if different chemical
elements are introduced between the layers and also
various methods are employed for the extraction of
these elements in reverse, these structures may to be
utilized as sites for storage and transfer of the respec-
tive chemical elements in adsorption and catalysis
w
x
7,8 . However, before a sufficient volume of experi-
ence is gained with the giant fullerenes, the problems
related to elemental extraction, grading and their
synthesis in large quantities must be addressed.
0009-2614r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII S0009-2614 97 01460-7

(
)
2
N.S. KoprinaroÕ et al.rChemical Physics Letters 285 1998 1–6
2. Experiment and results
w x
In distinction from Ref. 9 , where CH4 is utilized
as the work gas, with the aim of determining the
extent to which it may affect the formation of nan-
otubes and C60 , the initial aim of our experiment
was to study the supply of C to the sublimating
electrode by pyrolytic dissociation of CH4. For us
the end task was to obtain a work mode such that a
balance can be reached between the quantity of C
that deposits on the sublimating electrode and the
quantity of C which sublimates from this electrode.
This secured a constant regeneration of the sublimat-
ing electrode and process continuity. We used a gas
mix of 90% Ar and 10% CH4 with safety considera-
tions in mind and because of the ability to maintain a
stable arc discharge. In order that the pyrolytic de-
composition of CH4 on the sublimating electrode is
more effective, we employed the ‘‘inverse method’’
Ž
.
Fig. 1. Part of the anode deposit obtained in CH4 10% and Ar
Ž
.
90% , zone A – spherical forms, zone B – nanotubes, zone C –
polyhedral structures.
the smallest forms obtained after deposit milling in a
ball mill, boiling in xylene and ultrasound treatment
as observed by a TEM are commensurate in size
with C60 . The large diversity of diameters of the
spherical particles and their simultaneous existence
as stand alone structures within the sample, as well
as their perfect forms and large density, is reason for
us to suppose that all these spherical forms are
obtained from single layer fullerene types, such as
C60 and C70. We expect that during the discharge in
an Ar atmosphere these single layer fullerene types
can gain the C atoms necessary for their growth only
during sublimation and that they will quickly exit the
discharge region after losing their ability to grow
further. When CH4 is utilized as a work gas compo-
nent, the ability for pyrolytic growth is high and
while the particles are still found within the dis-
charge region, these become covered with a multi-
tude of concentric envelop layers. The larger the
particles become, the lower their rate of propagation,
the lower the speed with which these will leave the
region of intense growth, the more these will in-
crease in size. These observed structural forms, pos-
sessing hundreds of layers consisting of millions of
C atoms that are many times larger in size and
magnitude from the so-called gigantic fullerenes ob-
w
x
10 , where the electrodes are different in size, such
that the cathode sublimates, while a deposit covers
the anode. Under these conditions the electrode tem-
peratures are higher than for the classical method of
deposition on the cathode. In the experiment de-
scribed below, the size of the anode was 10 mm in
length and 64 mm2 in cross-section while the cath-
ode was 30 mm in length and 6 mm2 in cross-sec-
tion. The work pressure was 3=104 Pa and the
discharge current was 75 A. While maintaining this
current the potential was about 20 V. The distance
between the electrodes was visually maintained con-
stant at 0.2 mm. The diameter of the deposit ob-
tained is commensurate with the diameter of the thin
electrode – the cathode.
In distinction from the structures of the deposits
observed during arc discharge in an inert ambient for
cathodic or anodic deposits, the well expressed pecu-
liarities connected with the utilization of CH4 are
illustrated by SEM photographs of part of the deposit
obtained, see Fig. 1. Three regions can be clearly
discerned in which macro forms of different sizes
Ž
.
and forms are grown. Region A Fig. 1 is closest to
the deposit periphery. This consists of a multitude of
spherical forms, grown enveloping one another. Ap-
proaching the boundary between regions A and B, a
multiplicity of small and large disjoined spherical
w x
tained by other research groups 6 , we have reason
to believe can best be termed as ‘‘macro fullerene
forms’’. Only for experiments based on the immisci-
bility of C into copper are reports to be found of
Buckyballs that are commensurate in size, but cer-
Ž
Ž ..
forms were observed by SEM Fig. 2 a . The largest
of these are several micrometers in diameter, while
w
x
tainly not in quantity 11 .

(
)
N.S. KoprinaroÕ et al.rChemical Physics Letters 285 1998 1–6
3
Ž .
Ž .
Fig. 2. a Spherical forms region. b Region with a prevalent growth of nanotubes.
Due to the constantly decreasing speed of move-
ment, most of the macro structural forms are never
diverse forms and sizesrare so many, that these may
be observed on the surface of the growing deposit as
being stand alone structures with the aid of a SEM
Ž
.
able to leave region A in Fig. 1 and are eventually
incorporated within the growing deposit, which as
can be seen from the photograph appears monolithic
and comprised of larger forms the further away we
go from the boundary between A and B. At first
sight the macro structures merge and can no longer
be discerned as separate structures after growing in
size, however following the destruction of the de-
posit by milling in a ball mill, boiling in xylene and
ultrasound treatment the material surrounding the
structures is removed and the macro structures are
reveled as separate structures which allows their
observation by a TEM.
Ž Ž ..
see Fig. 2 a .
Large quantities of nanotubes with lengths ex-
Ž
ceeding 10 mm were observed within zone B Fig.
.
1 . Their prevalent direction in orientation coincides
with the direction of the arc. Within some regions,
where there were more nanotubes to be found whose
growth directions cross at close distances, cone bun-
Ž
dles were obtained see the left hand side in Fig.
Ž ..
2 b .
The large nanotube lengths, their small thick-
Ž
.
nesses some of these are mono layer and their
stand alone nature show that within this zone, as well
as along the boundary between the regions A and B
Except while utilizing CH4, we have also dis-
cussed the observed spherical macro structures ob-
tained previously at different synthesis conditions
Ž
.
see Fig. 1 the conditions of growth are favorable.
Multiple layer nanotubes were also observed,
w
x
12 . There the macro structures were also found
characteristic for these is the well defined stand
alone nature of the growth of the various concentric
layers. Often the structures observed of this type are
of the kind shown in Fig. 3. In all likelihood the side
layers start to grow on the already built nanotubes
and continue to grow far beyond the ends of the
nanotubes maintaining their diameters as constant. In
the standard arc discharge case we have observed the
different stages of growth of the layers on ready built
mainly along the periphery of the deposit, but they
were smaller in size and quantity. We assumed a
similar mechanism of growth for these. The principle
difference in utilizing an arc discharge in highly pure
Ar as compared to a gas mix of ArrCH4 is that in
the first case the closer the synthesized fullerene
structures are to the periphery of the discharge cham-
ber the more the C atom flux is decreased and
growth conditions deteriorate, while in the second
case in approaching the periphery the possibility for
pyrolytic accumulation is increased. As a result the
quantity and size of the macro structures synthesized
are many times larger, the macro forms themselves
during different stages of growth rmacro forms of
Ž
w x.
structures for example in Ref. 10 , but at the lower
scale the layers never grow beyond the nucleation
site. The explanation why the latter grown layers are
formed outside the base area is that either this is due
to the direct impact of the C atoms on the surface, or
due to the migration of the atoms in contact with the

(
)
4
N.S. KoprinaroÕ et al.rChemical Physics Letters 285 1998 1–6
We built the model presented below to explain the
reasons leading to the synthesis of the observed
macro structural forms along the periphery of the arc
discharge. The model is based on the investigations
of the plasma-chemical processes taking place in
w
x
carbon structure built up 13 . From these it is obvi-
ous that as a result of the plasma activity within the
work chamber, as well as the gases Ar and CH4
which enter the work chamber, there also arise com-
pounds of the types C2 H2 , C2 H4, C2 H6 and C4 H6.
Ions of the types CHq1 , CHq2 , CHq3 , CHq5 , C2 Hq2 ,
C2 Hq3 and C2 Hq5 are also formed. Together with
these, there are C atoms and ions sublimating from
the electrodes. The quantity of the different types of
ions and particles within the discharge space is a
function of their location with respect to the elec-
trodes. This is also changed in the direction from the
central part of the plasma space towards the periph-
ery and is strongly dependent on a number of techno-
logical factors.
Fig. 3. A typical multi-wall nanotube in which the internal and
external layers are grown independently of one another reaching
different sizes.
surface, the C flux towards the newly grown layers is
higher and these grow much faster.
Internally, towards the center of the deposit within
The presence of different kinds of particles is
illustrated in Fig. 5. The diagram describes the pro-
cess of dissociation of CH4 from the periphery
towards the center of the discharge region. We have
shown for simplicity only the most widely found
ions with 1 or 2 C atoms. The sublimating C atoms
are shown schematically close to the sublimating
electrode. Argon is not shown so as not to further
Ž
.
the region C Fig. 1 , a multitude of polyhedral
macro structures identical in size were observed. The
sharp borders visible between the separate regions,
within which the different types of macro structures
are grown, we regard as a sign of the strong influ-
ence of the discharge mode on the nature of the
structures grown. Along the border between regions
A and B we also observed conical and cylindrical
Ž
.
macro formations Fig. 4 , which show that as well
as the spherical, tubular and other fullerene micro
w
x
forms described in Ref. 12 macro sized forms are
also grown.
Fig. 5. Gas distribution in the vicinity of the arc. The deposit
shown in Fig. 1 is taken from the region indicated with a broken
line.
Fig. 4. Macro structures with different forms observed on the
boundary between the spherical macro forms and nanotubes.

(
)
N.S. KoprinaroÕ et al.rChemical Physics Letters 285 1998 1–6
5
complicate the drawing. On the electrode, on which
region and exits together with other C clusters, mov-
ing from the inside out. Critical for the conversion
into macro forms is the speedy increase in size,
which makes the forms less manoeuvrable and forces
them to remain within the region of the growing
deposit.
Ž
.
a deposit is grown in this case the anode a broken
line indicates the approximate location of the region
of the deposit shown in Fig. 1. Within this region
close to the arc discharge periphery, the particle
diversity is highest and their concentration is con-
stantly varied. This renders it impossible at this stage
to define which are the best conditions for macro
form synthesis but generally we conclude that along
the periphery of the arc discharge, the probability for
pyrolytic dissociation of the hydrocarbons, on the
surface of the growing fullerene forms, is greater
than at the center of the arc. The C atoms obtained
on the growing surface, directly as a result of pyroly-
sis, can maintain the process of layer envelopment of
the structure already built effectively, for a suffi-
ciently long time and in a much wider temperature
interval. As known, fullerene structures can be ob-
tained even only by pyrolytic decomposition at a
lower temperature than the temperature of the de-
w x
In distinction from Ref. 9 , where the presence of
C60 has not been observed, in our experiments we
Ž
.
observed C60 in small quantities only a few percent
in the soot obtained. The quantity of soot obtained
was also small. We are not able to say whether the
synthesis of C60 in our case originates during the
sublimation of a pyrolytically grown electrode in
accordance with the same mechanisms giving rise to
C60 as a result of the sublimation of the electrode in
an inert ambient, however in this process also, the
principle role is played by the gases introduced in
the discharge chamber.
w
x
posit grown under the present conditions 14 . Doubt-
less, combining nucleation under arc discharge con-
ditions with the possibilities of a prolonged envelop-
ment of these nucleation sites pyrolytically, is the
reason for the synthesis of fullerene type macro
forms. Each of these processes left on its own would
not be able to assure the build up of large size
structures.
Hydrogen is no less important in macro structure
build up. As a result of the constant supply of C
atoms from the dissociating hydrocarbons, the con-
centration of H in an atomic and ionized state is
high. It is about 1.5 times higher than that of CH4
3. Conclusion
As a result of the experiments carried out, we
came to the conclusion that the combination of the
mechanism of a quick synthesis of fullerene struc-
tures with a more protracted growth, layer by layer,
of these structures and also the constant etching with
active hydrogen can lead to the build up of forms of
significant sizes in large quantities and may prove to
be a promising method for obtaining such macro
forms.
w
x
13 . In the ionized or atomic state H is most active
Acknowledgements
and has a constant etching effect on all the C atoms
falling on the growing surface which are not strongly
attached to it. This aids the build up of macro
formations without defects, not only in every sepa-
rate layer, but also in terms of the arrangement of the
layers with respect to each other. The strong etching
effect of H is also maintained by the fact that in the
three regions of Fig. 1 all the macro forms are
clearly distinct from each other and there is a lack of
amorphous Carbon.
The support of the Bulgarian Science Foundation
under contract No. X 627 is appreciated.
References
w x
1
H.W. Kroto, J.R. Heath, S.C. O’Brein, R.F. Curl, R.E.
Smally, Nature 318 1985 162.
Ž
.
w x
Ž .
2
w x
3
S. Iijima, Nature 354 1992 56–58.
Ž .
D. Ugarte, Chem. Phys. Lett. 207 1993 473.
Y. Saito, T. Yoshikawa, J. Cryst. Growth 134 1993 154–
156.
It is logical that the C60 type spherical phase,
could be the base for the spherical macro forms
located along the periphery. Normally C60 is mobile
and spends the shortest time within the discharge
w x
4
Ž
.
w x
5
Ž
E. Osawa, M. Yoshida, M. Fujita, MRS Bull. November
.
1994 33.

(
)
6
N.S. KoprinaroÕ et al.rChemical Physics Letters 285 1998 1–6
w x
w
w
w
w
x
Ž
.
6
D.H. Parker, K. Chatterjee, P. Wurz, K.R. Lykke, M.J.
11 T. Cabioc’h, J.P. Riviere, J. Delafond, J. Mater. Sci. UK 30
Ž
.
Ž
.
Pellin, L.M. Stock, J.C. Hemminger, Carbon 30 1992 1167.
1995 4787.
w x
7
x
A.C. Dillon, K.M. Jones, T.A. Bekkedahl, C.H. Kiang, D.S.
12 N. Koprinarov, M. Marinov, G. Pchelarov, M. Konstanti-
Ž
.
Ž .
nova, R. Stefanov, J. Phys. Chem. 99 1995 2042.
Bethune, M.J. Heben, Nature 386 1997 377–379.
w x
8
x
G. Pchelarov, I. Topalova, N. Koprinarov, M. Konstantinova,
13 H. Yasuda, Plasma Polymerization, Academic Press, 1985
Ž
.
Ž
.
R. Stefanov, Carbon 35 1997 755–758.
Harcourt Brace Jovanovich publishers .
w x
Ž
.
x
9
Y. Ando, S. Iijima, Jpn. J. Appl. Phys. 32 1993 L–107.
14 M. Endo, K. Takeuchi, S. Igarashi, K. Kobari, M. Shiraishi,
w
x
Ž . Ž .
H.W. Kroto, J. Phys. Chem. Solids UK 54 1993 1841.
10 N. Koprinarov, M. Marinov, G. Pchelarov, M. Konstanti-
Ž
.
nova, R. Stefanov, Z. Phys. B 99 1996 353.