Synthesis of metallofullerenes by electric-arc method with a surface-modified electrode

Article abstract of DOI:10.1007/s11167-005-0341-9

A method for synthesis of metallofullerenes by electric-arc sputtering of graphite, in which lanthanum compounds are deposited on the surface of a graphite electrode, was suggested. The influence exerted by the amount of deposited lanthanum on the yield of lanthanum was studied.

Full text of DOI:10.1007/s11167-005-0341-9

Russian Journal of Applied Chemistry, Vol. 78, No. 4, 2005, pp. 559 564. Translated from Zhurnal Prikladnoi Khimii, Vol. 78, No. 4, 2005,
pp. 566 571.
Original Russian Text Copyright
2005 by Kuz’min, Borovkov, Kolker, Brazhyunas.
APPLIED ELECTROCHEMISTRY
AND CORROSION PROTECTION OF METALS
Synthesis of Metallofullerenes by Electric-Arc Method
with a Surface-Modified Electrode
S. M. Kuz’min, N. Yu. Borovkov, A. M. Kolker, and V. L. Brazhyunas
Institute of Chemistry of Solutions, Russian Academy of Sciences, Ivanovo, Russia
Ivanovo State University of Chemical Technology, Ivanovo, Russia
Received September 30, 2004; in final form, February 2005
Abstract A method for synthesis of metallofullerenes by electric-arc sputtering of graphite, in which lan-
thanum compounds are deposited on the surface of a graphite electrode, was suggested. The influence exerted
by the amount of deposited lanthanum on the yield of lanthanum was studied.
At present, endohedral complexes of fullerenes [1]
and higher fullerenes attract researchers’ attention as
potential objects of nanotechnologies [2], which pos-
sess nonlinear optical characteristics [3] and unique
physicochemical properties [4]. Synthesis and identi-
fication of new forms of endofullerenes (various iso-
mers, bi- and trinuclear complexes in encapsulation
of the whole variety of the available elements in the
void in the fullerene molecule) have been extensively
studied [5 18]. The known methods for synthesis of
metallofullerenes consist in bombardment of a ful-
lerene-containing substrate with metal ions [19] or
in evaporation of a metal carbon composite in the
atmosphere of an inert gas [16 23]. The composites
are evaporated with lasers [21, 22] or by electric-arc
sputtering [16, 22, 23].
Use of a combined electrode results in that metal
atoms are introduced into the discharge zone in sput-
tering of the electrode. Having a low ionization po-
tential, metal atoms markedly change the discharge
parameters, and, primarily, make lower the tempera-
ture of electrons in the arc and of the gas [25]. Using
a combined electrode, as a rule, leads to a decrease
in the yield of fullerenes [16, 25, 26].
The yield of metallofullerenes in synthesis is low
and the procedure for preparing a composite is pro-
longed and complicated. This results in that the prob-
lem of obtaining industrial samples of metallofuller-
enes remains unsolved. The efficiency of synthesis
of metallofullerenes is raised by varying the compos-
ition of the composite [22] and by improving the de-
sign of the reactors used for this purpose [23].
Commonly, the composite is prepared as follows.
A hole bored in a graphite rod is filled with a mixture
of a metal oxide, powdered graphite, and binder res-
ins. Then the rod (electrode) is annealed in a vacuum
at a temperature of about 2000 C. As a result, the
carbide of the starting metal is formed and fixed in
the bulk of the electrode.
The aim of this study was to examine specific
features of synthesis of metallofullerenes with a sur-
face-modified electrode.
EXPERIMENTAL
The experimental installation for synthesis of ful-
lerenes by electric-arc sputtering of graphite is shown
schematically in Fig. 1. A combined electrode was
fabricated from 10-cm-long rectangular graphite rods
with a cross-section of 2 4 to 4 6 mm. The elec-
trode was prepared by depositing an aqueous suspen-
sion of lanthanum carbonate on its surface. The de-
posited layer was dried in air to constant mass and
then the electrode mounted into the reactor was an-
It should be noted that C60 and C70 are formed in
synthesis of metallofullerenes. There is evidence [16,
21, 23] that the amount of a metal oxide introduced
into the electrode affects the total yield of fullerenes.
The ratio of yields of hollow fullerenes and metallo-
fullerenes [16, 21, 24] suggests that the formation of
an endocomplex is, in fact, the side process in syn-
thesis of hollow fullerenes.
1070-4272/05/7804-0559 2005 Pleiades Publishing, Inc.

560
KUZ’MIN et al.
across the discharge gap equal to 20 25 V. The com-
pletion of the sputtering process was judged by visual
inspection.
After the sputtering of the electrode, the carbon
black obtained in the course of synthesis was col-
lected and its mass was determined. Also weighed
were the residue of the electrode being sputtered and
the cathode deposit (carbon layer formed on the cath-
ode in synthesis).
The fullerenes contained in the carbon black were
extracted with ortho-xylene at a temperature of 40 C
under agitation for 24 h. Then the solution was fil-
tered on a glass filter and the carbon black was addi-
tionally washed with ortho-xylene to a colorless ex-
tract. The resulting filtrate was brought to a mark of
100 ml in the volumetric flask with ortho-xylene.
The concentration of individual fullerenes in solu-
tion was analyzed by HPLC and spectrophotometry.
Chromatograms were obtained on a LIQUOCHROME
2010 chromatograph with a UV detector (analytical
wavelength 330 nm), with hexane as eluent (residual
3
content of water
6
10 %) and a column packed
with SGX Separon adsorbent (grain size 7 m).
The chromatograms of reference solutions and those
being analyzed were processed using Multichrom 1.5
software. The spectra were recorded with a Hitachi
U2001 spectrometer.
Fig. 1. Schematic of the experimental installation for elec-
tricarc synthesis of fullerenes: (1) reactor, (2) reflector,
(3) cooling circuit, (4) graphite rods, (5) microlift,
(6) bell-jar, (7) vacuum station, (8) vacuum gage, (9) leak
valve, (10) discharge feeding unit, (11) voltmeter, (12) am-
meter, (13) current-carrying bar, (14) microlift-feeding unit,
(15) vacuum cock, and (16) cylinder with helium.
A theoretical description of the synthesis of ful-
lerenes has resulted in model concepts of the assembly
of a carbon skeleton [28 31]. Also represented in the
literature are the results of quantum-chemical calcula-
tions of the stability of various endocomplexes [32,
33]. An analysis of published data made it possible
to maintain that both the C60 fullerene and the endo-
complex are formed in the jet of a carbon vapor pro-
duced by sputtering in the electric-arc synthesis of
fullerenes [30].
nealed by passing a dc current of 50 70 A through
short-circuited electrodes at a residual pressure in
2
the reactor of 10 Pa or in the atmosphere of helium.
The cycle of reactor preparation to synthesis in-
cluded the following stages: (1) mounting of graphite
electrodes in the reactor, with their mass preliminarily
determined; (2) evacuation of the reactor to a pressure
3
of about 10 Pa; (3) heating of the electrode by pass-
It is known that [34], in synthesis of the fullerene,
the ions bombarding the cathode do not sputter it, but,
conversely, are incorporated into its structure to give
a cathode deposit. In this case, both the amount of
the deposit formed and its structural features depend
on the composition and energy of ions bombarding
the cathode. Under the conditions of the synthesis
performed, three sources are responsible for the com-
position of the near-cathode layer of plasma: buffer
gas, sputtered graphite, and sputtered lanthanum. The
lower the ionization energy, the larger the fraction
of an ionized component. The ionization energy of
helium is 24.587 eV [35]; that for components of the
carbon vapor: C 11.260, C2 12.10, C3 11.65, and C4
11.28 eV [36]; the energies of successive ionization
ing an electric current; (4) filling of the reactor with
helium to the working pressure and heating of the elec-
trodes in this atmosphere; (5) repeated evacuation of
3
the reactor to a pressure of 10 Pa; and (6) letting-in
helium to a working pressure of 2 104 Pa.
The working pressure in synthesis of metallofuller-
enes has been experimentally determined previously
[27]. This pressure corresponds to the maximum yield
of hollow fullerenes.
To sputter the combined electrode, a voltage suf-
ficient for arc initiation was applied across the elec-
trodes and microlift displacement at a preliminarily
adjusted velocity was switched on. In the course of
sputtering, the arc current was 100 150 A at a voltage
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 4 2005

SYNTHESIS OF METALLOFULLERENES BY ELECTRIC-ARC METHOD
for lanthanum ions are 5.577, 11.06, and 19.17 eV
561
[35]. It should be noted that the existence of high-en-
ergy metastable states of helium leads to ionization of
easily ionized components of plasma by the Penning
mechanism [36]:
A* + B
A + B+ + e,
where A* stands for helium in excited state, and B,
for the easily ionized component.
Thus, it would be expected that the ionic composi-
tion of plasma is mainly represented by C+n and La+
ions. As the discharge current in the cathode region
is carried by positive ions, the mass of the deposit is
proportional to the quantity of electricity consumed
in the process and to the accommodation coefficient
of the incident ion. The amount of graphite being
evaporated is also proportional to the charge passed
[37]. As a result, the ratio of the mass of sputtered
graphite to that of the deposit can be used to charac-
terize an experiment.
The shape of the deposit obtained in the experi-
ment (Figs. 2a and 2b) corresponds to that of the elec-
trode being sputtered. The cross-section of the cathode
deposit exceeds that of the initial electrode and a cra-
ter is always present in the central part of the cathode
deposit. Use of both graphite and surface-modified
electrodes leads to the same distribution of sputtered
graphite among the fullerene black and the cathode
deposit (Fig. 3). The lanthanum being evaporated
does not change the nature of the ion flow toward
the cathode surface and, consequently, does not enter
into the near-cathode zone of the discharge. In the
authors’ opinion, this is an indication of the separate
evaporation of graphite and lanthanum.
Fig. 2. Photographs of a typical cathode deposit upon
completion of the synthesis. (a) Correspondence between
the electrode being sputtered and the deposit formed, and
(b) characteristic crater at the center of the deposit.
The decrease in the amount of the cathode deposit
with respect to the amount of sputtered graphite on
making smaller the cross-section of the electrode is
understandable. There exists a considerable tempera-
ture gradient at the cathode from the axis of the arc
toward its periphery, as well as a temperature gradient
in the gas phase [37, 38]. As a result, the gas phase
contains larger carbon clusters in the peripheral parts
of the near-cathode region; the lower surface temper-
ature leads to a higher accommodation coefficient,
compared with that in the central parts of the cathode.
Apparently, the area of the peripheral regions of the
electrodes increases in proportion to a squared char-
acteristic size (equivalent electrode diameter). The
factors mentioned lead to an increase in the fraction
of the cathode deposit on raising the cross-section area
of the electrode being sputtered and provide the ob-
served shape of the cathode deposit.
Fig. 3. Influence exerted by the mass m of the electrode
being sputtered on the ratio M /M of the mass of the sput-
tered graphite to that of the cathode deposit for (1) graphite
and (2) combined electrodes.
s
d
The flow of evaporated lanthanum is virtually
not involved in the formation of the plasma composi-
tion in the arc, but affects the synthesis of fullerenes
in the carbon vapor jet. This influence is primarily
manifested in a change in the composition of the
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 4 2005

562
KUZ’MIN et al.
Fig. 4. Typical chromatogram of an extract of the carbon
black synthesized. (A) Signal from the UV detector and
(t) time. Peak assignment: (1) ortho-xylene, (2) C
,
60
(3) C , (4, 5) C
C , and (6) La@C .
70
72 86 n
Fig. 5. Typical absorption spectrum of the extract.
Absorption and ( ) wavelength. Extraction with
(
)
Fig. 6. Effect of the content n of lanthanum carbonate
in the electrode on (a) yield of C and yield ratios,
(1) ortho-xylene and (2) DMFA; (3) published La@C
82
60
spectrum.
(b) C /C , and (c) La@C /C .
60 70 n 60
carbon black being synthesized and extracts obtained
(Figs. 4 6).
in the chromatogram to metallofullerenes. The shapes
of the chromatographic peak 6 and optical absorption
peak at around 1000 nm indicate that several forms
of endohedral fullerenes exist in the substrate syn-
thesized.
A typical chromatogram of an extract of the car-
bon black obtained using the combined electrode un-
der study differs from chromatograms of samples ob-
tained with a graphite electrode in that peak 6 appears
(Fig. 4). The spectrum of the extract obtained with
the combined electrode contains absorption bands
coinciding with the characteristic absorption bands of
metallofullerenes [39] (Fig. 5). The yields of the sub-
stance (Fig. 6c), calculated from the absorption band
at around 1000 nm and from peak 6 of the chromato-
gram (Fig. 4) are in a good agreement for all syn-
theses, which makes it possible to attribute peak 6
The yield of the fullerene C60 increases as the frac-
tion of lanthanum in the combined electrode becomes
higher (Fig. 6a). It is noteworthy that a similar ten-
dency has been observed previously in synthesis of
metallofullerenes by electric-arc sputtering of a com-
bined electrode [24]. The maximum total yield of ful-
lerenes was observed at a C/La ratio in the combined
electrode equal to 50. Surprisingly, the maximum
yield of La@C82 has been observed at the same C/La
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 4 2005

SYNTHESIS OF METALLOFULLERENES BY ELECTRIC-ARC METHOD
ratio in synthesis of metallofullerenes by laser evap- CONCLUSIONS
563
oration of a combined electrode [21], although the in-
fluence exerted by lanthanum-doping of the electrode
on the yield of hollow fullerenes was quite the oppo-
site. It may be assumed that this is due to the exis-
tence of a temperature gradient in the carbon vapor
jet. It has been shown in a study of how the synthesis
temperature affects the yield of fullerenes that the
yield of La@C82 is the highest at 1200 K, whereas
the maximum yield of C60 is obtained at lower tem-
peratures. Apparently, the optimal conditions for stab-
ilization of a certain structure of the carbon cluster
can be satisfied in an expanding carbon vapor jet only
locally.
(1) Use of a surface-modified electrode simplifies
the electrode preparation procedure and makes it pos-
sible to synthesize by electric-arc sputtering a mixture
of metallofullerenes in amounts comparable with
those reported by other authors.
(2) Raising the amount of lanthanum carbonate
contained in the electrode does not change the C60/C70
ratio in the samples synthesized and leads to an in-
crease in the total amount of fullerenes obtained at
a mass fraction of lanthanum within the range studied.
In addition, use of the electrodes suggested leaves
unchanged the amount of graphite utilized into the
cathode deposit. This fact supports the assumption
that lanthanum and carbon evaporate separately in
the case when surface-modified electrodes are used.
The relative yield of C60 (Fig. 6b) remained virtu-
ally unchanged in a set of experiments, with the ex-
ception of that with a mass fraction of lanthanum car-
bonate equal to 0.75% (C60/C70 of about 6.6). At the
same time, the yield of metallofullerenes shows a sig-
nificant scatter (Fig. 6c). This scatter is possibly due,
as mentioned above, to presence of a mixture of var-
ious forms of metallofullerenes in a sample being
analyzed, with the concentration of any particular
form of metallofullerenes in the extract being low.
REFERENCES
1. Chai, Y., Guo, T., Jin, C., et al., J. Phys. Chem.,
1991, vol. 95, pp. 7560 7564.
2. Neeb, M., Klingeler, R., Bechthold, P.S., et al., Appl.
Phys. A, 2001, vol. 72, pp. 289 295.
3. Gu, G., Huang, H., Yang, S., et al., Chem. Phys. Lett.,
Experiments with a mass fraction of lanthanum
carbonate equal to 0.75 and 2% (Fig. 6) cannot be
explained in the same way as other experiments. Of
a similar nature are the data of [16, 21, 23]. The pos-
sible reason for this phenomenon is that different
steady states of the arc and structures of the gas-dy-
namic flows in the reactor are possible in the course
of synthesis, even if the synthesis procedure is strictly
observed. In the case under study, there exist at least
two different dependences of the yield of fullerenes
on the mass fraction of lanthanum in the combined
electrode (Fig. 6a).
1998, vol. 289, pp. 167 173.
4. Liu, S. and Sun, S., J. Organomet. Chem., 2000,
vol. 599, pp. 74 86.
5. Wang, C.R., Inakuma, M., and Shinohara, H., Chem.
Phys. Lett., 1999, vol. 300, pp. 379 384.
6. Nishibori, E., Takata, M., Inakuma, M., and Shino-
hara, H., Chem. Phys. Lett., 1998, vol. 298, pp. 79 84.
7. Wang, C.R., Kai, T., Tomiyama, T., et al., Nature,
2000, vol. 408, pp. 426 427.
8. Takata, M., Nishibori, E., Sakata, M., et al., Phys.
Rev. Lett., 1999, vol. 83, pp. 2214 2216.
9. Yamamoto, K., Saunders, M., Khong, A., et al., J. Am.
Chem. Soc., 1999, vol. 121, pp. 1591 1603.
Apparently, it is necessary to take into account, in
determining the yield of fullerenes and in comparing
the results obtained with published experimental data,
that studies of extracts provide only a limited knowl-
edge of the structure and composition of the carbon
black. It is known that the carbon substrate obtained
by electric-arc sputtering of graphite contains a large
amount of fullerene-like fragments [40]. In the course
of extraction, the imperfect carbon structures are sep-
arated from the fullerenes synthesized. The aggrega-
tion susceptibility of fullerenes, including that with
incomplete fullerene fragments, hinders extraction of
fullerenes from the substrate. Therefore, the yield of
the target product can be raised severalfold by chang-
ing the solvent used for extraction and the method
for recovery of fullerenes from the substrate [42, 43].
10. Murphy, T.A., Pawlik, T., Weidinger, A., et al., Phys.
Rev. Lett., 1996, vol. 77, pp. 1075 1078.
11. Knapp, C., Dinse, K.-P., Piezak, B., et al., Chem.
Phys. Lett., 1997, vol. 272, pp. 433 437.
12. Stevenson, S., Rice, G., Glass, T., et al., Nature,
1999, vol. 401, pp. 55 57.
13. Olmstead, M.M., de Bettencourt-Dias, A., Duch-
amp, J.C., et al., J. Am. Chem. Soc., 2000, vol. 122,
pp. 12200 12206.
14. Olmstead, M.M., de Bettencourt-Dias, A., Duch-
amp, J.C., et al., Angew. Chem. Int. Ed., 2001,
vol. 40, pp. 1223 1225.
15. Wang, C.R., Kai, T., Tomiyama, T., et al., Angew.
Chem. Int. Ed., 2001, vol. 40, pp. 397 399.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 4 2005

564
KUZ’MIN et al.
16. Bubnov, V.P., Kol’tover, V.K., Laukhina, E.E., et al.,
Izv. Akad. Nauk, Ser. Khim., 1997, no. 2, pp. 254 258.
29. Gao, X. and Gao, J., J. Phys. Chem. B, 1994, vol. 98,
pp. 5618 5620.
30. Krestinin, A.V., Moravskii, A.P., and Tesner, P.A.,
17. Tagmatarchis, N., Aslanis, E., Shinohara, H., and
Prassides, K., J. Phys. Chem. B, 2000, vol. 104,
pp. 11010 11012.
Khim. Fiz., 1998, vol. 17, no. 9, pp. 70 84.
31. Churilov, G.N., Novikov, P.V., Tarabanko, V.E., et al.,
Carbon, 2002, vol. 40, pp. 891 896.
18. Okazaki, T., Lian, Y., Gu, Z., et al., Chem. Phys.
Lett., 2000, vol. 320, pp. 435 440.
32. Mandziuk, M. and Bacic, Z., J. Phys. Chem. B, 1994,
vol. 101, pp. 2126 2140.
33. Kobayashi, K., Nagase, S., and Akasaka, T., Chem.
Phys. Lett., 1996, vol. 261, pp. 502 506.
19. Hirata, T., Otomo, Y., and Hatakeyama, R., Thin
Solid Films, 2002, vol. 407, pp. 32 37.
20. Ying, Z.C., Jin, C., Hettich, R.L., et al., Fullerenes:
Recent Advances in the Chemistry and Physics of
Fullerenes and Related Materials, vol. 1, Kadish, K.
and Ruoff, R., Eds., Pennington: Electrochem. Soc.,
1994, pp. 1402 1412.
34. Ellacott, M.V., Pang, L.S.K., and Prochazka, L.,
Carbon, 1994, vol. 32, pp. 542 544.
35. Fizicheskaya entsiklopediya (Physical Encyclopedia),
Prokhorov, A.M., Ed., Moscow: Bol’sh Sov. Entsikl.,
1988, vol. 1; 1990, vol. 2.
36. Egorov, V.S., Tolmachev, V.A., Klyucharev, A.N.,
et al., Spravochnik konstant elementarnykh protsessov
s uchastiem atomov, ionov, elektronov, fotonov (Ref-
erence Book og Constants of Elementary Processes
Involving Atoms, Ions, Electrons and Phonons),
Zhiglinskii, A.G., Ed., St. Petersburg: SPb. Gos.
Univ., 1994.
21. Suganuma, T., Kasuya, D., Ishigaki, T., et al., Eur.
Phys. J. D, 1999, vol. 9, pp. 373 376.
22. Lian, Y.F., Shi, Z.J., Zhou, X.H., et al., J. Phys.
Chem. Solids, 2000, vol. 61, pp. 1037 1040.
23. Tran, N.E. and Lagowski, J.J., Carbon, 2002, vol. 40,
pp. 939 948.
24. Sueki, K., Kikuchi, K., Akiyama, K., et al., Chem.
Phys. Lett., 1999, vol. 300, pp. 140 144.
37. Byszewski, P., Lange, H., Huczko, A., and Behnke, J.,
J. Phys. Chem. Solids, 1997, vol. 58, pp. 1679 1683.
38. Lange, H., Huczko, A., and Starski, A., J. Phys.
Chem. A., 1997, vol. 101, pp. 1267 1269.
25. Lange, H., Huczko, A., Byszewski, P., et al., Chem.
Phys. Lett., 1998, vol. 289, pp. 174 180.
26. Wilson, M.A., Moy, A., Rose, H., et al., Fuel, 2000,
39. Shinohara, H., Rep. Prog. Phys., 2000, vol. 63,
vol. 79, pp. 47 56.
pp. 843 892.
27. Kuz’min, S.M., Kozlov, A.V., and Leonov, A.A.,
Abstracts of Papers, VII Mezhdunarodnaya konferen-
tsiya Problemy sol’vatatsii i kompleksoobrazovaniya
v rastvorakh Ivanovo, 8 11 okt. 2001 g, (VII Int.
Conf. Problems of Solvation and Complexation in
Solutions, Ivanovo, October 8 11, 2001), p. 185.
40. Kanowski, M., Vieth, H.-M., Luders, K., et al., Car-
bon, 1997, vol. 35, pp. 685 695.
41. Nath, S., Pal, H., and Sapre, A.V., Chem. Phys. Lett.,
2000, vol. 327, pp. 143 148.
42. Sun, D., Liu, Z., Guo, X., et al., J. Phys. Chem. B,
1997, vol. 101, pp. 3927 3930.
28. Yamaguchi, Y. and Maruyama, S., Chem. Phys. Lett.,
43. Sun, B., Feng, L., Shi, Z., and Gu, Z., Carbon, 2002,
1998, vol. 286, pp. 336 342.
vol. 40, pp. 1591 1595.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 4 2005