Synthesis, structure, and properties of novel open-cage fullerenes having heteroatom(s) on the rim of the orifice

Article abstract of DOI:10.1002/chem.200390184

A thermal liquid-phase reaction of fullerene C₆₀ with 3-(2-pyridyl)-5,6-diphenyl-1,2,4-triazine afforded aza-open-cage fullerene derivative 5 having an eight-membered-ring orifice on the fullerene cage. Compound 5 was found to undergo oxidative

Full text of DOI:10.1002/chem.200390184

FULL PAPER
Synthesis, Structure, and Properties of Novel Open-Cage Fullerenes Having
Heteroatom(s) on the Rim of the Orifice
[a]
Yasujiro Murata, Michihisa Murata, and Koichi Komatsu*
Dedicated to Emeritus Professor Soichi Misumi on the occasion of his 77th birthday
Abstract: A thermal liquid-phase reac-
tion of fullerene C60 with 3-(2-pyridyl)-
5,6-diphenyl-1,2,4-triazine afforded aza-
open-cage fullerene derivative 5 having
an eight-membered-ring orifice on the
fullerene cage. Compound 5 was found
to undergo oxidative ring-enlargement
reactions with singlet oxygen under
photo-irradiation to give azadioxo-
open-cage fullerene derivatives 9 and
10, which have a 12-membered-ring
orifice, in addition to a small amount
of azadioxa-open-cage fullerene deriva-
tive 11, which has a 10-membered-ring
orifice. A thermal reaction of 9 with
elemental sulfur in the presence of
tetrakis(dimethylamino)ethylene result-
ed in further ring-enlargement to give
azadioxothia-open-cage fullerene deriv-
ative 15, which has a 13-membered-ring
orifice. The structures of 5 and 15 were
determined by X-ray crystallography,
while those of 9, 10, and 11 were
confirmed by the agreement of observed
13C NMR spectra with those obtained by
DFT-GIAO calculations. These reac-
tions were rationalized based on the
results of molecular orbital calculations.
Following electrochemical measure-
ments, compounds 9 and 10, which have
two carbonyl groups on the rim of the
orifice, were found to be more readily
reduced than C60 itself (the first reduc-
tion potential was found to be 0.2 V
lower than that of C60), while the first
reduction potentials of other open-cage
fullerene derivatives, 5, 11, and 15, were
nearly the same as that of C60.
Keywords:
calculations
fullerenes
density
electrochemistry
insertion sulfur
functional
¥
¥
¥
¥
¥
X-ray diffraction
was subsequently employed as a precursor to novel azafuller-
Introduction
[4]
enes,[3] (C59N)2 and C59NH.[5]
Compared to the abundant studies of chemical modification
of the exterior of fullerene cages utilizing the reactivity of the
p bond,[1] those concerning transformation of the s framework
of fullerene cage leading to the formation of open-cage
fullerenes are quite limited. Thus, in his pioneering work,
Wudl and co-workers utilized the reaction of C60 with organic
azide to create an azafulleroid derivative for the synthesis of
the first open-cage fullerene 1,[2] which has an 11-membered-
ring orifice with a ketolactam functionality. The compound 1
On the other hand, Rubin and co-workers have discovered
a versatile method leading to a bisfulleroid 2, which is looked
upon as an open-cage fullerene with an eight-membered-ring
orifice. They demonstrated that when a 1,3-cyclohexadiene
moiety is fused to C60, a photochemical intramolecular [4‡4]
cycloaddition followed by a retro [2‡2‡2] reaction can cause
the formation of an eight-membered-ring orifice on the C60
cage.[6] According to essentially the same reaction pathway,
several derivatives of open-cage fullerene 2 have been
prepared by the use of the reactions of C60 with nickela-
cyclopentadiene derivatives,[7] palladacyclopentadiene deriv-
atives,[8] and phthalazine.[9] Subsequently, the eight-mem-
bered-ring orifice on the C60 cage in these derivatives was
found to be enlarged into the 12-membered-ring by oxidative
cleavage of the double bond with photochemically generated
singlet oxygen.[10]
These transformations of the s framework of fullerene
cages are of great importance because of the possibility in
organic synthesis of endohedral fullerene complexes encap-
sulating materials such as rare gases, molecular hydrogen, and
particularly metals. It has been proposed by Rubin and co-
workers that organic syntheses of these endohedral fullerene
complexes could be realized if we can establish the method-
[a] Prof. K. Komatsu, Dr. Y. Murata, M. Murata
Institute for Chemical Research, Kyoto University Uji
Kyoto 611-0011 (Japan)
Fax : (‡81)774 38 3178
E-mail: komatsu@scl.kyoto-u.ac.jp
Supporting information for this article is available on the WWW under
http://www.chemeurj.org or from the author.
1600
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1600 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 7

1600 1609
ologies for ™cracking open fullerenes∫ and ™zipping up
fullerene precursors∫.[11] In connection with the latter ™zip-
ping up∫ process, the formation of C60 has been detected by
mass spectrometry of various precursors,[12] and, most impor-
tantly, the actual organic synthesis of C60 has recently been
achieved by Scott and co-workers using flash vacuum
pyrolysis as the final step of the synthesis.[13] Although this
strategy appears quite promising for providing access to
various kinds of known or as yet unknown fullerenes, it has
not been applied to the synthesis of endohedral fullerene
complexes. Some limited success has, however, been attained
in the ™cracking open∫ strategy for the endohedral complexes.
As the simplest example, rare gases such as helium, neon,
argon, krypton, and xenon have been incorporated into
pristine C60 under forced conditions (such as 6008C, 3000 atm)
to afford C60 containing these gases, albeit in yields typically as
low as 0.1%,[14] that is only one molecule of C60 out of
approximately 1000 molecules contains the gas atom. It is
supposed that in the transition state, a bond on the C60 cage is
cleaved to form a window that can incorporate these gas
atoms under such conditions.[15] In a recent study, Rubin and
co-workers synthesized a bislactam derivative of C60 having a
14-membered-ring orifice 3,[16] and succeeded in the encap-
sulation of a helium atom (1.5%) or molecular hydrogen
(5%) under the conditions of 3008C and 475 atm or 4008C
and 100 atm, respectively.[17]
In our previous work, we demonstrated that the course of
the reaction of C60 with nitrogen-containing aromatic com-
pounds, such as phthalazine,[9] 3,6-di(2-pyridyl)-1,2,4,5-tetra-
zine,[18] and 4,6-dimethyl-1,2,3-triazine,[10c] shows a large
dependence on the reaction phase (solid state or liquid phase)
and the liquid-phase thermal reactions particularly with
phthalazine[9] and with 4,6-dimethyl-1,2,3-triazine[10c] afford
new derivatives of open-cage fullerenes. Interestingly, the
above-mentioned intramolecular [4‡4] and retro [2‡2‡2]
reactions can also take place under the thermal conditions
employed in our work, probably by a radical mechanism.
With the continuation of this study, we became interested in
a thermal reaction of C60 with a 1,2,4-triazine derivative. This
reaction is expected to give a novel aza-open-cage fullerene,
that is, the one containing an eight-membered-ring orifice
incorporating an unsaturated nitrogen atom on the rim of the
orifice. It appeared interesting to clarify the electronic effects
of the presence of this nitrogen atom on the properties and
reactivity of the fullerenyl p system of this new derivative of
an open-cage fullerene. In the present study we succeeded in
obtaining a series of new open-cage fullerene derivatives with
10-, 12-, and 13-membered-ring orifice containing hetero-
atom(s) on the rim from the first-formed aza-open-cage
fullerene derivatives. Here we report on the synthesis,
structure, and properties of a series of these newly formed
derivatives of open-cage fullerenes.
(Scheme 1) having a 2-pyridyl group at the 3-position and
two phenyl groups at the 5- and 6-positions. We conducted a
thermal reaction with one equivalent of C60 in o-dichloro-
benzene (ODCB) at 1808C for 17 h. The reaction was found
Scheme 1. Synthesis of open-cage fullerene derivative 5.
to proceed smoothly, giving almost a single product, which
exhibits a purple color in solution, together with unreacted C60
in 41% recovery. The product, which was highly soluble in
common organic solvents such as CHCl3, toluene, CS2, and
ODCB, was determined as an aza-open-cage fullerene
derivative 5 (Scheme 1a) based on the characterization
described below. Its yield was 50%, or 85% based on
consumed C60, and no bisfunctionalized product was isolated.
The structure determination of open-cage fullerene 5 was
first attempted by spectroscopy. The FAB mass spectrum
showed the molecular ion peak at m/z 1003 ([M‡1])
corresponding to M ˆ C80H14N2, which clearly indicates that
5 is formed by addition of triazine 4 to C60 followed by
1
extrusion of N2. While the H NMR spectrum displayed the
signals for aromatic protons only, the 13C NMR spectrum
showed 65 signals in the sp2-carbon region between d ˆ 167.96
and 122.47 ppm, in which nine signals are apparently over-
lapped, in addition to two signals at d ˆ 73.02 and 56.35 ppm
for the sp3-carbon atoms, suggesting that 5 has C1 symmetry.
The purple color observed for the solution of 5 in CHCl3 and
its maximum absorption at 533 nm in the electronic spectrum
(closely resembling the spectrum of C60), indicate that the 60
original fullerene carbon atoms retain their sp2 hybridization
9]
in a p-conjugated system.[6
The structure of 5 was unambiguously determined by X-ray
crystallography of a single crystal obtained from a solution in
benzene. As shown in Figure 1, an eight-membered-ring
orifice composed of C1 C7 and N1 is present on the fullerene
À
cage of 5. In the eight-membered ring, the C2 C3
À
À
(1.406(4) ä), C4 C5 (1.403(4) ä), and C7 N1 (1.270(4) ä)
bonds have double bond character, whereas the others are
single bonds. The eight-membered ring is a tub-form with the
C1-C2-C3-C4and C3-C-C45-C6 dihedral angles being
À39.0(4) and 38.8(4)8, respectively, deviating significantly
from 08. The butadiene system, C2-C3-C4-C5, is almost planar
Results and Discussion
Open-cage fullerene derivatives having an eight-membered-
ring orifice: To avoid the problem of obtaining a product with
low solubility, we chose
a 1,2,4-triazine derivative 4
Chem. Eur. J. 2003, 9, No. 7
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1601 $ 20.00+.50/0
1601

K. Komatsu et al.
FULL PAPER
Figure 1. The X-ray structure of open-cage fullerene derivative 5 with
displacement ellipsoids drawn at the 50% probability level. Selected
Figure 2. The optimized structure of open-cage fullerene derivative 5 with
the contour of the HOMO, calculated at the B3LYP/6 311G** level of
theory.
À
distances [ä], bond angles [8], and dihedral angles [8]: C1 C2 1.521(4),
À
À
À
À
À
C2 C3 1.406(4), C3 C41.511(4), C4 C5 1.403(4), C5 C6 1.517(4), C6 C7
À
À
À
À
1.540(4), C7 N1 1.270(4), N1 C1 1.457(3), C2 C13 2.274(4), C5 C14
2.279(4); C1-C2-C3 124.0(3), C2-C3-C4 129.8(2), C3-C4-C5 128.7(2), C4-
C5-C6 124.6(2); C1-C2-C3-C4 À39.0(4) C3-C4-C5-C6 38.8(4), C2-C3-C4-
C5 0.4(5).
with the addition of O2 as shown by FAB MS, in 60%, 31%,
and 2% yields, respectively. The IR spectra of 9 and 10
exhibited strong carbonyl stretching bands at 1747 and
1748 cmÀ1, respectively, whereas no carbonyl band was
observed for 11.
with the dihedral angle between the two double bonds being
only 0.4(5)8.
A probable reaction mechanism for the formation of 5 is
shown in Scheme 1b. A [4‡2] cycloaddition of triazine 4 to
C60 affords 6, and the following extrusion of nitrogen gives
2-aza-1,3-cyclohexadiene-fused C60 derivative 7, which would
undergo a formal intramolecular [4‡4] cycloaddition to give
8, and then a retro [2‡2‡2] reaction affords 5.
Because comparison of the experimental and calculated
13C NMR spectra seemed to be crucial for the structural
determination of 9, 10, and 11, we first examined the validity
of such a comparison using the spectrum of compound 5, for
which the structure was confirmed by X-ray crystallography.
To determine which level of theory is the most accurate for
reproducing the experimental 13C NMR spectrum of 5, the
gauge-independant atomic orbital (GIAO)[20,21] calculations
on 5 were conducted using Hartree Fock (HF) and density
functional theory (B3LYP[22] and B3PW91[23]) methods with
both the 6 311G** and TZV[24] basis sets, based on the
structure optimized at the B3LYP/6 311G** level of theo-
ry.[25] All chemical shifts are given as the values relative to that
of tetramethylsilane, also calculated using the same method at
the same level of theory as those used for the corresponding
GIAO calculations.
Enlargement of the orifice in open-cage fullerene 5: It has
been reported both by us and another group that one of the
double bonds in the eight-membered-ring orifice in an open-
cage fullerene can undergo oxidative cleavage by the action of
photochemically generated singlet oxygen.[10]
To clarify the reactivity of 5, DFT calculations were
conducted. The geometry of 5 was fully optimized at the
B3LYP/6 311G** level of theory[19] to give a structure with
the HOMO shape shown in Figure 2. The HOMO was found
À
to be localized primarily at the two double bonds, C2 C3 and
As shown in Figure 3a , the experimentally observed
spectrum of 5 is characterized by two sp2-carbon signals at
d ˆ 167.96 and 164.81 ppm, which are notably downfield
shifted, the rest of the sp2-carbon signals appearing in the
range d ˆ 149.10 122.47 ppm (63 signals observed out of 76
expected signals for fullerenyl and aryl carbon atoms), and
two sp3 carbon signals at d ˆ 73.02 and 56.35 ppm. As shown
in Figures 3b g, all six GIAO calculations, conducted with
B3PW91/6 311G**, B3LYP/6 311G**, B3PW91/TZV,
B3LYP/TZV, HF/6 311G**, and HF/TZV, reproduced the
general characteristics of the experimental 13C NMR spec-
trum fairly well. Of the six calculations, DFT methods were
shown to be generally superior to HF methods in reproducing
À
C4 C5, in the same manner as reported for the carbon
analogues,[10] whereas the LUMO was found to be spread over
almost all the sp2-carbon atoms on the fullerene cage. The
absolute values of the coefficients of HOMO were 0.33 for
C2, 0.22 for C3, 0.21 for C4, and 0.32 for C5. Thus, the
electrophilic addition of the singlet oxygen was expected to
take place on these double bonds.
In fact, when the photochemical reaction of 5 with oxygen
was conducted in a CCl4 solution under irradiation with a
high-pressure mercury lamp for 6 h, the HPLC separation of
the reaction mixture afforded three oxidation products 9, 10,
and 11, all having the molecular formula corresponding to 5
1602
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1602 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 7

Open-Cage Fullerenes
1600 1609
shown in Figure 4together with the experimentally observed
spectra. From the fairly good agreement of the observed
spectra with the calculated ones, the structural assignment of 9
and 10 is considered to be correct.[28]
Figure 3. The experimental and calculated 13C NMR spectra of 5. The
GIAO calculations were conducted at different levels of theory and basis
sets using the optimized structure at the B3LYP/6 311G** level of theory:
a) experimental spectrum, b) GIAO-B3PW91/6 311G**, c) GIAO-
B3LYP/6 311G**, d) GIAO-B3PW91/TZV, e) GIAO-B3LYP/TZV,
f) GIAO-HF/6 311G**, and g) GIAO-HF/TZV.
Figure 4. The experimental and calculated 13C NMR spectra of 9 and 10.
The GIAO calculations were conducted at the B3PW91/6 311G** level of
theory: a) observed for 9, b) calculated for 9, c) observed for 10, and
d) calculated for 10.
the experimental spectrum, regardless of the basis sets. The
contribution of electron correlations[26] seems to play an
important role in the properties of compound 5. A similar
trend has also been observed in the calculated 13C NMR
chemical shifts on a naphthalene derivative.[27] As shown by
the comparison of Figure 3a with Figure 3b, the GIAO-
B3PW91/6 311G** calculation was found to be the best
method for reproducing the experimental 13C NMR spectrum
for compound 5.
For the minor product 11, obtained in 2% yield from the
photochemical oxidation of 5, the experimental 13C NMR
spectrum shown in Figure 5a displayed no peak correspond-
ing to a carbonyl carbon, while the FAB MS data indicated
that two oxygen atoms were incorporated in 5. In comparison
Based on the results of the FAB MS and IR spectroscopy
described above, possible structures of the two major photo-
oxidation products of 5 are supposed to be diketo-compounds
9 and 10, which could be formed by oxidative cleavage of the
À
À
C2 C3 bond and C4 C5 bond, respectively. The GIAO-
B3PW91/6 311G** calculations conducted for 9 and 10 are
Chem. Eur. J. 2003, 9, No. 7
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1603 $ 20.00+.50/0
1603

K. Komatsu et al.
FULL PAPER
with the 13C NMR spectrum of starting material 5, a notable
downfield shift was observed for one of the sp3-carbon signals
(a signal at d ˆ 73.02 ppm for 5 was shifted to d ˆ 96.47 ppm
for 11), whereas another sp3-carbon signal remained relatively
unchanged (d ˆ 56.35 ppm for 5 and d ˆ 53.26 ppm for 11).
Because the signals at d ˆ 73.02 and 56.35 ppm for 5 are
assigned to the C1 and C6 carbon atoms, respectively, by the
GIAO calculations described above, it is implied that the
reaction with oxygen has taken place near the C1 carbon atom
of 5 without any formation of additional sp3-carbon atom(s).
Thus the most probable structure for this minor product is
considered to be 11, shown below, with a 10-membered-ring
orifice on the fullerene cage, resulting from insertion of each
À
À
of the two oxygen atoms into C1 C2 and C3 C4single bonds
of 5. This assignment was verified by comparison of the
13C NMR spectra calculated by the GIAO B3PW91/6
311G** method for 11 and its positional isomer 12 with the
experimental 13C NMR spectrum. Clearly, a better agreement
was observed for compound 11, as shown in Figure 5.
The formation of three oxidation products 9, 10, and 11 by
the reaction of 5 with photochemically generated singlet
oxygen is explained as follows. As shown in Scheme 2, singlet
Figure 5. The experimental 13C NMR spectrum of the minor product and
calculated 13C NMR spectra of 11 and 12. The GIAO calculations were
conducted at the B3PW91/6 311G** level of theory: a) observed for the
minor product, b) calculated for 11, and c) calculated for 12.
À
oxygen is considered to add electrophilically to the C2 C3
À
À
À
and C4 C5 double bonds having relatively high coefficients of
each oxygen atom into the C1 C2 bond and the C3 C4bond
À
the HOMO to form intermediate dioxetanes 13 and 14,
respectively. The optimized structures and relative energies of
dioxetanes 13 and 14 as well as diethers 11 and 12 were also
calculated at the same level of theory (B3LYP/6 311G**).
with the concomitant cleavage of the O O bond. A similar
reaction pathway can be supposed for the formation of
diketone 10 from dioxetane 14, but the formation of diether
12, which is calculated to be 1.8 kcalmolÀ1 less stable than 11,
was not observed under the present reaction conditions.
À
Although the addition of singlet oxygen to the C2 C3 bond
would be slightly unfavorable as judged from the relative
energy of dioxetanes 13 and 14 calculated at the B3LYP/6
311G** level of theory, it would be kinetically favorable
compared with the addition to the C4 C5 bond when the
presence of two bulky phenyl groups in close proximity is
An open-cage fullerene derivative having a 13-membered-
ring orifice: To further enlarge the 12-membered-ring orifice
of 9, a reaction to cause insertion of a heteroatom at the rim of
the orifice appeared quite appealing. It has been shown that a
À
À
considered. Diketone 9 is formed from dioxetane 13 by the
sulfur atom can be inserted into an activated C C single
cleavage of the C2 C3 single bond and the O O bond,
bond.[29] We therefore attempted a thermal reaction of 9 with
À
À
whereas diether 11 is produced by means of the insertion of
elemental sulfur in ODCB and found that the desired reaction
Scheme 2. Formation of open-cage fullerene derivatives 9, 10, and 11. The relative energies calculated at the B3LYP/6 311G** level of theory with
reference to 11 are shown in parentheses.
1604
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1604 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 7

Open-Cage Fullerenes
1600 1609
actually takes place in the presence of tetrakis(dimethylami-
no)ethylene (TDAE). Heating a mixture of 9, elemental
sulfur (8 equiv), and TDAE (1 equiv) in refluxing ODCB for
30 min successfully afforded a single product with a FAB MS
data exhibiting a molecular ion peak at m/z 1067 correspond-
ing to 9‡S‡H, in 77% yield.
fullerenyl and aryl sp2-carbon atoms in the range d ˆ 166.40 to
122.68 ppm, and two signals corresponding to the sp3-carbons
at d ˆ 74.42 and 52.33 ppm (Figure 7a). This spectrum showed
good agreement with the spectrum calculated for the structure
In this reaction, TDAE, a typical p-electron donor,[30] is
supposed to activate 9 either by one-electron transfer or by
complexation so that the electrophilic addition of elemental
sulfur to 9 can readily take place. The theoretical calculations
for 9 at the B3LYP/6 311G** level of theory demonstrated
that the LUMO of 9 is relatively localized at the conjugated
À
À
À
butadiene part, C9 C10 C11 C12, on the rim of the orifice
(Figure 6); the numbering is shown in Scheme 3. The absolute
Figure 7. The experimental and calculated 13C NMR spectra of 15. The
GIAO calculations were conducted at the B3PW91/6 311G** level of
theory: a) observed for 15 and b) calculated for 15.
15 by the GIAO B3PW91/6 311G** method using the
optimized structure at the B3LYP/6 311G** level of theory
(Figure 7b).[28] Finally, the validity of the assignment of the
structure 15 to this product was proved by the X-ray
crystallography for the single crystal grown from a solution
in toluene (Figure 8).
As shown by the X-ray crystal structure given in Figure 8a,
compound 15 has a 13-membered-ring orifice on the fullerene
cage (shown in red) containing sulfur and nitrogen atoms. The
À
À
À
À
bond lengths of C9 C10, C10 S1, S1 C11, and C11 C12 are
1.398(4), 1.780(3), 1.754(3), and 1.367(4) ä, respectively. Thus
a divinyl sulfide moiety was formed in the 13-membered ring
À
as a result of the sulfur-atom insertion into the C10 C11 bond
of 9. Determination of this X-ray structure also confirmed the
validity of the structural assignment of the starting material 9,
which was made based on the comparison with the DFT
GIAO calculated spectrum, as described above.
Figure 6. The optimized structure of open-cage fullerene derivative 9 with
the contour of the LUMO, calculated at the B3LYP/6 311G** level of
theory.
From the top view of the X-ray structure shown in
Figure 8b, it can be seen that the 13-membered-ring orifice
in 15 has a somewhat more circular shape as compared to the
rather elliptic shape observed for the 14-membered-ring
orifice (counting the fused benzene-ring carbon atoms) in
the bislactam derivative of an open-cage fullerene 3 reported
by Rubin and co-workers.[16, 17] This is the largest hole
constructed thus far on the surface of C60. Thus, the insertion
of a small molecule or atom is expected to be possible through
the 13-membered-ring orifice in 15.
Scheme 3. Synthesis of open-cage fullerene derivative 15.
values of the coefficients of the LUMO at C9, C10, C11, and
C12 of 9 are 0.29, 0.35, 0.32, and 0.25, respectively. Thus, it is
assumed that the sulfur atom has been inserted into the
Properties of open-cage fullerene derivatives: The UV/Vis
spectra of the obtained open-cage fullerene derivatives, 5, 9,
10, 11, and 15, taken in CHCl3 are shown in Figure 9. As
mentioned previously, compound 5, having an eight-mem-
bered-ring orifice, displayed a similar absorption pattern to
that of C60 itself, reflecting the fact that 5 retains the
À
C10 C11 bond to give the novel open-cage fullerene deriv-
ative 15 having a 13-membered-ring orifice (Scheme 3). The
13C NMR spectrum of this product exhibited two signals for
the carbonyl carbon atoms at d ˆ 193.06 and 185.02 ppm, 52
signals (out of possible 76 signals) corresponding to the
9]
conjugated system composed of 60 sp2-carbon atoms.[6
Chem. Eur. J. 2003, 9, No. 7
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1605 $ 20.00+.50/0
1605

K. Komatsu et al.
FULL PAPER
Figure 9. UV/Vis spectra of open-cage fullerene derivatives: blue line (5),
black (9), green (10), red (11), and brown (15).
tions at 253, 317, and 442 nm. Compound 15, which has the
largest orifice among the compounds examined in this study,
showed maximum absorptions at 257 and 319 nm as well as a
broad absorption at 430 nm. With the increase in the size of
the orifice in the order 5, 9 and 10, and 15, the absorption at
the visible region lost their structure. The colors of the
solutions of 5, 9, 10, 11, and 15 in CHCl3 were purple, brown,
brown, red-brown, and bright red, respectively.
To examine the electronic properties of the open-cage
fullerene derivatives 5, 9, 10, 11, and 15, their redox behaviors
were studied by cyclic voltammetry (CV) and differential
pulse voltammetry (DPV) in ODCB using Bu4NBF4 as a
supporting electrolyte. Despite the significant change in the p-
electronic system resulting from the transformation in the
fullerene skeleton made for these derivatives, all compounds
showed four quasi-reversible redox waves in the negative
scans upon CV. These were quite similar to those observed for
C60 itself. The corresponding four one-electron reduction
peaks were observed upon DPV measurements. As an
example, the CV and DPV results for compound 10 are
shown in Figure 10, together with those for C60.
Figure 8. The X-ray structure of open-cage fullerene derivative 15 with the
displacement ellipsoids model drawn at the 50% probability level: a) side
À
À
view and b) top view. Selected distances [ä]: N1 C1 1.445(3), C1 C2
À
À
À
À
1.584(4), C2 O1 1.209(3), C2 C9 1.492(4), C9 C10 1.398(4), C10 S1
À
À
À
À
1.780(3), S1 C11 1.754(3), C10 C11 2.546(4), C11 C12 1.367(4), C12 C3
À
À
À
À
1.518(4), C3 O2 1.202(3), C3 C41.553()4, C4
C5 1.380(4), C5 C6
À
À
1.552(3), C6 C7 1.545(4), C7 N1 1.281(3).
The reduction potentials determined by DPV are summar-
ized in Table 1, together with the LUMO energy levels
calculated by the B3LYP/6 31G* level of theory. As shown
by the values in Table 1, the reduction potentials of com-
pounds 5 and 11 were found to be rather similar to those of
C60, indicating that the fully conjugated p-electron systems
made of 60 sp2 carbons in 5 and 11 retain the high electron
affinities characteristic of C60. Most noteworthy is the
observation that compounds 9 and 10 exhibited first reduction
waves at potentials as low as À0.97 and À0.98 V versus the
ferrocene/ferrocenium couple, respectively, which are ap-
proximately 0.2 V lower than that of C60 measured under the
Compounds 9 and 10, which are isomeric, showed quite
similar absorptions, that is, strong maximum absorptions in
the UV region (257 and 325 nm for 9; 259 and 324nm for 10),
which are commonly observed in C60 derivatives, and a small
absorption in the visible region (424 nm for 9; 423 nm for 10)
as well as the long-wavelength visible absorption extending to
approximately 700 nm. These absorptions are different from
those of an enol-ketone derivative 16, which has a similar 12-
membered-ring orifice. This showed maximum absorptions at
453 and 751 nm.[10a] Compound 11 showed maximum absorp-
1606
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1606 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 7

Open-Cage Fullerenes
1600 1609
with the low levels of LUMO predicted for these compounds
by theoretical calculations at the B3LYP/6 31G* level of
theory.[32]
Conclusion
In summary, an open-cage fullerene derivative having a
nitrogen-containing, eight-membered-ring orifice (5) was
synthesized in high yield (85% based on consumed C60 at
59% conversion) by a liquid-phase thermal reaction of C60
with 3-(2-pyridyl)-5,6-diphenyl-1,2,4-triazine (4) in one pot.
The eight-membered-ring orifice was further enlarged to a 10-
membered ring containing ether linkages (11) and a 12-
membered ring containing two carbonyl carbon atoms (9 and
10) by photochemical oxidation with singlet oxygen. The
structures of these compounds were characterized by spec-
troscopy and DFT calculations, as well as by the X-ray
crystallography for 5. In particular, the GIAO calculation was
found to be a powerful method for structural determination of
derivatives such as 9, 10, and 11 by comparison of the
calculated with the observed 13C NMR spectra.
À
A novel sulfur-atom insertion into the central C C bond in
the butadiene unit of the rim of the 12-membered-ring orifice
was found. It gave a novel fullerene derivative 15 with a 13-
membered-ring orifice containing both nitrogen and sulfur
atoms on the rim. The X-ray crystallography on this com-
pound indicated that the shape of the orifice is closer to a
circle than to an ellipse compared with compound 3, which has
the largest orifice reported thus far.[17] The energy barrier for
insertion of a helium atom through the orifice was calculated
to be approximately 5.5 kcalmolÀ1 lower for compound 15
than for 3.[33] This finding is a good indicator that the
preparation of endohedral fullerene complexes by means of
organic synthesis, which was initiated by Rubin and co-
workers,[17] is indeed possible. Work in this field is now under
way using compound 15.
Figure 10. CV and DPV of C60 and 10: 1mM in ODCB, 0.05M TBABF4,
scan 0.02 VsÀ1; a) CV of C60, b) DPV of C60, c) CV of 10, and d) DPV of 10.
Table 1. Reduction potentials[a] and calculated LUMO levels.[b]
1[a]
2[a]
3[a]
4[a]
Compound
Ered
Ered
Ered
Ered
LUMO [eV][b]
C60
5
9
10
11
15
16[10a]
À 1.17
À 1.21
À 0.97
À 0.98
À 1.18
À 1.15
À 1.07
À 1.59
À 1.62
À 1.40
À 1.39
À 1.58
À 1.56
À 1.43
À 2.06
À 2.11
À 2.01
À 2.00
À 2.20
À 2.05
À 1.99
À 2.56
À 2.62
À 2.43
À 2.44
À 2.70
À 2.52
À 2.29
À 3.23
À 3.03
À 3.32
À 3.34
À 3.13
À 3.16
À 3.26
[a] V versus ferrocene/ferrocenium: All reduction potentials were meas-
ured by DPV (1 mM sample, 0.05 m Bu4NBF4 in ODCB, scan rate
0.02 VsÀ1). [b] Calculated at the B3LYP/6 31G* level of theory.
Experimental Section
same conditions. The second to fourth reduction potentials
were also lower than those of C60. This enhanced reducibility
could be ascribed to the presence of two carbonyl groups
directly connected to the partially broken fullerenyl p system.
Actually, the reduction potentials observed for 9 and 10 are
lower than those of the monocarbonyl analogue of the similar
open-cage fullerene 16[10a] by about 0.1 V. Thus, one carbonyl
group connected to the fullerenyl p system is considered to
decrease the reduction potential by approximately 0.1 V. This
effect is nearly comparable to that of the two cyano groups
connected by sp3-carbon atoms to the C58 fullerenyl p system
in C60(CN)2.[31] In contrast, the sulfur-containing derivative 15
displayed its first reduction at À1.15 V, which is more negative
than that of 9 and 10 and close to that of C60 despite the
presence of two carbonyl groups. This is considered to reflect
the strong electron-donating effect of a sulfur atom directly
bonded to the p system. As shown in Table 1, the lowered
reduction potentials observed for 9 and 10 are in agreement
General: The 1H and 13C NMR measurements were carried out on Varian
Mercury 300 and JEOL AL-400 instruments and referenced to TMS or to
the solvent signals: dH ˆ 2.05 ppm for [D6]acetone and dC ˆ 132.35 ppm for
[D4]o-dichlorobenzene ([D4]ODCB). UV/Vis spectra were recorded on a
Shimadzu UV-2100PC spectrometer. IR spectra were recorded with a
Shimadzu FTIR-8600 spectrometer. MS spectra were recorded on a JEOL
MStation JMS-700. Cyclic voltammetry and differential pulse voltammetry
were conducted on a BAS Electrochemical Analyzer CV-100 W using a
three-electrode cell with a glassy carbon working electrode, a platinum wire
counter electrode, and a Ag/0.01 m AgNO3 reference electrode. The
potentials were corrected against ferrocene used as an internal standard
which was added after each measurement. Fullerene C60 was purchased
from Matsubo Co. 3-(2-Pyridyl)-5,6-diphenyl-1,2,4-triazine and tetrakis(-
dimethylamino)ethylene were purchased from Aldrich Co. and used as
received, while elemental sulfur was purchased from Nacalai Tesque and
used as received. o-Dichlorobenzene (ODCB) was distilled over CaH2 and
stored over 4ä molecular sieves.
Computations: All calculations were conducted by using the Gaussian98
series of electronic structure programs.[19] The geometries were fully
optimized with the restricted Becke hybrid (B3LYP) method[22] for all
compounds. The GIAO calculations were performed at HF/6 311G**,
Chem. Eur. J. 2003, 9, No. 7
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1607 $ 20.00+.50/0
1607

K. Komatsu et al.
FULL PAPER
B3LYP/6 311G**, B3PW91/6 31G**, HF/TZV, B3LYP/TZV, and
B3PW91/TZV level of theory using the optimized structure at the
B3LYP/6 311G** level of theory.
(100 MHz, [D4]ODCB): d ˆ 167.97, 162.52, 162.16, 156.68, 151.84, 149.55,
147.73, 147.37, 147.09, 146.84, 146.82, 146.56, 145.89, 145.75, 145.67, 145.54,
145.45, 145.34, 145.30, 145.18, 145.03, 144.99, 144.67, 144.65, 144.60, 144.41,
144.23, 144.00, 143.81, 143.54, 143.27, 142.68, 142.50, 142.40, 142.00, 140.61,
139.99, 139.91, 139.77, 139.69, 139.54, 139.30, 138.62, 138.52, 138.47, 138.31,
138.18, 137.67, 137.48, 137.15, 137.04, 136.99, 136.91, 136.31, 134.37, 134.30,
132.94, 126.58, 126.15, 123.01, 120.04, 96.47, 53.26 ppm (some signals in the
region d ˆ 132.7 126.9 ppm overlap with the [D4]ODCB peaks); HRMS
(‡FAB): calcd for C80H15N2O2 (M‡1): 1035.1134, found 1035.1121.
Open-cage fullerene derivative 5: A mixture of fullerene C60 (50 mg,
0.069 mmol) and 3-(2-pyridyl)-5,6-diphenyl-1,2,4-triazine (4) (21 mg,
0.068 mmol) in o-dichlorobenzene (ODCB, 4mL) was refluxed at 180 8C
for 17 h under an argon atmosphere. The resulting dark purple solution was
directly subjected to flash column chromatography over silica gel. Elution
with CS2 gave unreacted C60 (20 mg, 41%) while the following elution with
CS2-ethyl acetate (20:1) gave open-cage fullerene derivative 2 (35 mg,
0.035 mmol, 50%) as a brown powder. M.p. >3008C; IR (KBr): nÄ ˆ
Open-cage fullerene derivative 15: To a heated and stirred solution of
compound 9 (32 mg, 0.031 mmol) and elemental sulfur (8 mg, 0.031 mmol
as S8) in ODCB (15 mL) was added tetrakis(dimethylamino)ethylene
(7.1 mL, 0.031 mmol) at 1808C. The solution was refluxed at 1808C for
30 min then the resulting dark red-brown solution was concentrated by
evaporation to about 3 mL. This was added to pentane (30 mL) with
vigorous stirring to give brown precipitates. The precipitates, collected by
centrifuge, were dissolved in ODCB (2 mL). The resulting solution was
subjected to flash chromatography on silica gel eluted with toluene/ethyl
acetate (30:1) to give open-cage fullerene derivative 15 (25 mg,
0.023 mmol, 77%) as a brown powder. M.p. >3008C; IR (KBr): nÄ ˆ
1749 cmÀ1 (C N); UV/Vis (CHCl3): lmax (log e) ˆ 262 nm (5.05), 329
ˆ
1
(4.62), 431 (3.29), 533 (3.07); H NMR (300 MHz, CS2:[D6]acetone (7:1)):
d ˆ 8.68 (m, 1H), 8.00 7.90 (m, 2H), 7.76 7.69 (m, 2H), 7.44 7.37 (m,
2H), 7.22 7.10 ppm (m, 7H); 13C NMR (100 MHz, CS2:CDCl3 (1:2)): d ˆ
167.96, 164.81, 149.10, 148.10, 147.96, 145.40, 145.21, 145.14, 145.12, 145.06,
144.85, 144.37, 144.34, 144.20, 144.15, 144.02, 143.96, 143.94, 143.80, 143.77,
143.71, 143.65, 143.59, 143.54, 143.50, 143.40, 143.35, 143.31, 143.21, 143.19,
143.16, 143.13, 143.02, 142.86, 141.29, 140.53, 140.49, 140.32, 140.23, 140.20,
139.96, 139.18, 138.77, 138.09, 137.92, 136.73, 136.69, 136.57, 136.48, 136.35,
135.18, 135.02, 134.88, 134.60, 134.48, 130.78, 130.65, 128.36, 128.21, 128.00,
127.18, 127.04, 125.17, 125.08, 122.47, 73.02, 56.35 ppm; HRMS (‡FAB):
calcd for C80H15N2 ([M‡1]): 1003.1235, found 1003.1238.
1748 cmÀ1 (C O); UV/Vis (CHCl3): lmax (log e) ˆ 257 (5.14), 319 (4.69),
ˆ
430 nm (3.78); 1H NMR (300 MHz, CS2:[D6]acetone (7:1)): d ˆ 8.60 (m,
1H), 8.23 (m, 1H), 8.18 8.15 (m, 2H), 8.02 (m, 1H), 7.53 (m, 1H), 7.36 (m,
1H), 7.30 7.25 (m, 3H), 7.04 6.88 ppm (m, 4H); 13C NMR (75 MHz,
[D4]ODCB): d ˆ 193.06, 185.02, 166.40, 164.00, 150.18, 149.19, 149.06,
148.64, 148.57, 148.53, 148.43, 148.23, 148.03, 147.71, 147.59, 147.56, 147.49,
147.40, 147.35, 147.33, 147.20, 146.82, 146.75, 146.56, 145.81, 145.60, 145.28,
144.67, 144.14, 143.73, 142.12, 141.65, 141.32, 141.09, 140.73, 140.30, 140.15,
139.93, 139.13, 138.91, 138.58, 138.54, 138.50, 138.32, 137.97, 137.88, 137.54,
137.39, 137.31, 135.44, 135.19, 135.10, 132.91, 122.68, 74.42, 52.33 ppm
(signals in the range d ˆ 132.4 126.9 ppm overlap with the signals of
[D4]ODCB); HRMS (‡FAB): calcd for C80H15N2O2S ([M‡1]): 1067.0854,
found 1067.0814.
Open-cage fullerene derivatives 9, 10, and 11: A purple solution of
compound 5 (66 mg, 0.066 mmol) in CCl4 (65 mL) in a Pyrex flask was
irradiated by a high-pressure mercury lamp (500 W) from a distance of
20 cm for 6 h under air. The resulting brown solution was evaporated and
the residual black solid was dissolved in ODCB (3 mL). This was subjected
to preparative HPLC using a Cosmosil 5PBB column (10 Â 250 mm) eluted
with ODCB (flow rate, 2 mLminÀ1
) to afford open-cage fullerene
derivatives 9 (40 mg, 0.038 mmol, 60%, retention time: 8.7 min), 10
(21 mg, 0.020 mmol, 31%, retention time: 9.2 min), and 11 (1 mg,
0.001 mmol, 2%, retention time: 9.1 min), after nine recycles, all as brown
powders.
When the photoirradiation of 5 (41 mg, 0.041 mmol) was conducted in CS2
(50 mL) in the presence of trifluoroacetic acid (30 mL, 0.39 mmol) under
the same conditions as described above and the reaction mixture was
neutralized by addition of triethylamine (55 mL, 0.40 mmol) before
evaporation of the mixture, the separation by preparative HPLC gave 9
(22 mg, 0.021 mmol, 53%), 10 (4mg, 0.004mmol, 10%), and 11 (3 mg,
0.002 mmol, 6%).
Table 2. Crystallographic data for open-cage fullerene derivatives.
5 ¥ 3(C6H6)
15 ¥ 0.5(C7H8)
empirical formula
formula weight
crystal system
space group
a [ä]
b [ä]
c [ä]
a [8]
b [8]
C98H32N2
1237.26
triclinic
C83.5H18N2O2S
1113.06
9: M.p. >3008C; IR (KBr): nÄ ˆ 1747, 1700 cmÀ1 (C O); UV/Vis (CHCl3):
ˆ
triclinic
lmax (log e) ˆ 257 (5.11), 325 (4.69), 424 nm (3.70); 1H NMR (300 MHz,
CS2:[D6]acetone (7:1)): d ˆ 8.52 (m, 1H), 8.37 (m, 1H), 8.07 7.98 (m, 3H),
7.82 (m, 1H), 7.37 7.03 ppm (m, 8H); 13C NMR (100 MHz, CS2:CDCl3
(1:2)): d ˆ 196.04, 188.96, 167.84, 161.38, 149.53, 148.22, 148.18, 148.04,
147.43, 147.32, 147.29, 147.15, 146.92, 146.37, 145.99, 145.85, 145.67, 145.53,
145.44, 145.38, 145.32, 145.25, 145.20, 145.19, 144.96, 144.85, 144.67, 144.35,
144.20, 143.79, 143.71, 143.51, 142.87, 142.38, 142.17, 141.95, 141.57, 141.46,
141.11, 140.77, 140.56, 140.24, 140.13, 139.89, 139.73, 139.58, 139.27, 139.13,
138.96, 138.35, 137.30, 137.21, 137.16, 135.97, 135.92, 135.35, 132.90, 132.54,
131.08, 130.48, 129.96, 129.85, 129.78, 129.51, 128.88, 128.69, 128.37, 127.27,
127.17, 126.76, 123.13, 122.60, 74.97, 52.63 ppm; HRMS (‡FAB): calcd for
C80H15N2O2 ([M‡1]): 1035.1134, found 1035.1132.
≈
≈
P1
P1
13.750(2)
15.237(2)
15.498(2)
63.891(3)
71.078(4)
85.258(3)
2751.4(7)
2
10.0158(8)
13.1987(10)
17.5019(14)
98.638(2)
91.309(2)
98.317(2)
2261.0(3)
2
g [8]
V [ä3]
Z
T [K]
100(2)
100(2)
l [ä]
m [mmÀ1
0.71073
0.086
0.71073
0.142
]
q range [8]
limiting indices
1.55 25.50
À 14 ꢀ h ꢀ 16
À 18 ꢀ k ꢀ 18
À 18 ꢀ l ꢀ 18
1.493
1.82 25.60
À 11 ꢀ h ꢀ 12
À 15 ꢀ k ꢀ 16
À 21 ꢀ l ꢀ 13
1.635
10: M.p. >3008C; IR (KBr): nÄ ˆ 1748 cmÀ1 (C O); UV/Vis (CHCl3): lmax
ˆ
(log e) ˆ 259 (5.10), 324 (4.70), 423 nm (3.72); 1H NMR (300 MHz,
CS2:[D6]acetone (7:1)): d ˆ 8.58 (m, 1H), 8.20 8.17 (m, 2H), 8.02 (m,
1H), 7.93 (m, 1H), 7.37 7.21 ppm (m, 9H); 13C NMR (100 MHz,
CS2:CDCl3 (1:2)): d ˆ 195.31, 190.36, 165.18, 163.66, 151.71, 149.50,
148.38, 148.20, 147.48, 147.40, 147.34, 147.00, 146.37, 145.99, 145.96, 145.61,
145.55, 145.29, 145.22, 145.20, 145.12, 145.10, 144.98, 144.52, 144.39, 144.25,
144.11, 143.89, 143.78, 143.51, 143.05, 142.80, 142.43, 142.06, 141.80, 141.56,
141.49, 140.79, 140.69, 140.62, 140.07, 139.92, 139.63, 139.58, 139.49, 139.27,
138.96, 138.40, 138.00, 137.48, 137.23, 136.58, 136.55, 136.18, 135.51, 135.44,
134.73, 134.23, 133.33, 132.01, 131.91, 131.68, 129.35, 129.32, 128.95, 127.21,
126.87, 124.44, 122.15, 72.42, 61.58 ppm; HRMS (‡FAB): calcd for
C80H15N2O2 ([M‡1]): 1035.1134, found 1035.1151.
1calc [g cm3]
data/restraints/parameters
unique reflections
completeness [%]
absp correction
Rint
R1 [I > 2s(I)][a]
wR2 [I > 2s(I)][b]
R1 (all data)[a]
wR2 (all data)[b]
GOF on F 2
10201/0/957
10201
99.498.2
SADABS
0.0357
0.0613
0.1318
0.1093
0.1499
8357/34/829
8357
SADABS
0.0188
0.0559
0.1348
0.0792
0.1439
0.969
11: M.p. >3008C; UV/Vis (CHCl3): lmax (log e) ˆ 317 (4.64), 442 nm (3.52);
1H NMR (300 MHz, CS2:[D6]acetone (7:1)): d ˆ 8.64(m, 1H), 8.48 (m,
1H), 8.03 7.89 (m, 4H), 7.66 (m, 1H), 7.39 7.21 ppm (m, 7H); 13C NMR
0.931
[a] R1ˆSjjFc jÀjFo jj/SjFo j. [b] wR2ˆ{S[w(Fo2 ÀFc2†2]/S[w(Fo2†2]}1/2
.
1608
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1608 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 7

Open-Cage Fullerenes
1600 1609
X-ray structural analysis of 5 and 15: Single crystals of compounds 5 and 15
suitable for X-ray crystallography were obtained by slow evaporation of the
solutions in benzene and in toluene, respectively, over three days at room
temperature. Single-crystal X-ray data were collected on a Bruker SMART
diffractometer equipped with a CCD area detector using MoKa radiation.
Single-crystal diffraction data were collected at 100 K. All structure
solutions were obtained by direct methods and refined using full-matrix
least-squares with a Bruker SHELXTL (Version 5.1) Software Package.
The crystal parameters are summarized in Table 2. CCDC-197761 (5) and
CCDC-197762 (15) contain the supplementary crystallographic data for
this paper. These data can be obtained free of charge at www.ccdc.cam.a-
c.uk/conts/retrieving.html (or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax:(‡44) 1223-336-
033; or deposit@ccdc.cam.uk).
[13] L. T. Scott, M. M. Boorum, B. J. McMahon, S. Hagan, J. Mack, J.
Blank, H, Wegner, A. de Meijere, Science 2002, 295, 1500.
¬
¬
[14] a) M. Saunders, H. A. Jimenez-Vazquez, R. J. Cross, S. Mroczkowski,
D. I. Freedberg, F. A. L. Anet, Nature 1994, 367, 256; b) M. Saunders,
¬
¬
H. A. Jimenez-Vazquez, R. J. Cross, S. Mroczkowski, M. L. Gross,
D. E. Giblin, R. J. Poreda, J. Am. Chem. Soc. 1994, 116, 2193; c) K.
Yamamoto, M. Sanders, A. Khong, R. J. M. Grayson, M. L. Gross,
A. F. Benedetto, R. B. Weisman, J. Am. Chem. Soc. 1999, 121, 1591;
d) M. S. Syamala, R. J. Cross, M. Saunders, J. Am. Chem. Soc. 2002,
124, 6216, and references therein.
[15] S. Patchkovskii, W. Thiel, J. Am. Chem. Soc. 1998, 120, 556.
[16] G. Schick, T. Jarrosson, Y. Rubin, Angew. Chem. 1999, 111, 2508;
Angew. Chem. Int. Ed. 1999, 38, 2360.
[17] Y. Rubin, T. Jarrosson, G.-W. Wang, M. D. Bartberger, K. N. Houk, G.
Schick, M. Saunders, R. J. Cross, Angew. Chem. 2001, 113, 1591;
Angew. Chem. Int. Ed. 2001, 40, 1543.
[18] a) K. Komatsu, Y. Murata, N. Kato in Fullerenes 2000, Vol. 9,
Functionalized Fullerenes (Eds.: N. Martin, M. Maggini, D. M. Guldi),
Electrochemical Society, N. J. Pennington, 2000, p. 20; b) Y. Murata,
M. Suzuki, K. Komatsu, Chem. Commun. 2001, 2338; c) G. P. Miller,
M. C. Tetreau, Org. Lett. 2000, 2, 3091; d) G. P. Miller, M. C. Tetreau,
M. M. Olmstead, P. A. Lord, A. L. Balch, Chem. Commun. 2001, 1758.
[19] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr., R. E.
Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels,
K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R.
Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski,
G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V.
Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, M.
Head-Gordon, E. S. Replogle, J. A. Pople, Gaussian 98, Revision A.7,
Gaussian, Inc.: Pittsburgh PA, 1998.
Acknowledgement
This work was supported by the Grant-in-Aid for COE Research on
Elements Science (No. 12CE2005) from the Ministry of Education,
Culture, Sports, Science and Technology, Japan. Computation time was
provided by the Supercomputer Laboratory, Institute for Chemical
Research, Kyoto University.
[1] a) A. Hirsch, The Chemistry of Fullerenes, Thieme Verlag, Stuttgart,
1994; b) F. Diederich, C. Thilgen, Science 1996, 271, 317; c) F.
Diederich, R. Kessinger, Acc. Chem. Res. 1999, 32, 537; d) A. Hirsch,
Angew. Chem. 2001, 113, 1235; Angew. Chem. Int. Ed. 2001, 40, 1195;
e) A. Hirsch, Fullerenes and Related Structures, Topics in Current
Chemistry 199, Springer, Berlin, 1999; f) R. Taylor, Synlett 2000, 776.
[2] J. C. Hummelen, M. Prato, F. Wudl, J. Am. Chem. Soc. 1995, 117, 7003.
Another example of [70]fullerene having an 11-membered-ring orifice
and 8 phenyl groups has been reported. See P. R. Birkett, A. G. Avent,
A. D. Dawish, H. W. Kroto, R. Taylor, D. R. M. Walton, Chem.
Commun. 1995, 1869.
[20] K. Wolinski, J. F. Hilton, P. Pulay, J. Am. Chem. Soc. 1990, 112, 8251.
[21] For a comprehensive review of the quantum-mechanical determina-
tion of NMR chemical shifts, see: T. Helgaker, M. Jaszunski, K. Ruud,
Chem. Rev. 1999, 99, 293.
[22] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) P. Lee, W. Yang,
R. G. Parr, Phys. Rev.B 1988, 37, 785.
[3] For the derivatives of C59N, see: a) A. Hirsch, B. Nuber, Acc. Chem.
Res. 1999, 32, 795; b) U. Reuther, A. Hirsch, Carbon 2000, 38, 1539;
c) F. Hauke, A. Hirsch, Tetrahedron 2001, 57, 3697;
¬
[4] J. C. Hummelen, B. Knight, J. Pavlovich, R. Gonzalez, F. Wudl,
[23] J. P. Perdew, Y. Wang, Phys. Rev.B 1992, 45, 13244.
Science 1995, 269, 1554.
[5] M. Keshavarz-K, R. Gonzalez, R. G. Hicks, G. Srdanov, V. I. Srdanov,
[24] A. Sch‰fer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829.
[25] Several GIAO calculations have so far been made on various
fullerenes and their derivatives to give satisfactorily reliable results,
see ref. [10a, c] and also see: a) G. Sun, M. Kertesz, New J. Chem.
2000, 24, 741; b) M. B¸hl, A. Hirsch, Chem. Rev. 2001, 101, 1153, and
references therein; c) Y. Murata, M. Ito, K. Komatsu, J. Mater. Chem.
2002, 12, 2009; d) M. S. Meier, H. P. Spielmann, R. G. Bergosh, R. C.
Haddon, J. Am. Chem. Soc. 2002, 124, 8090.
¬
T. G. Collins, J. C. Hummelen, C. Bellavia-Lund, J. Pavlovich, F. Wudl,
K. Holczer, Nature 1996, 383, 14 7.
[6] a) M.-J. Arce, A. L. Viado, Y.-Z. An, S. I. Khan, Y. Rubin, J. Am.
Chem. Soc. 1996, 118, 3775; b) W. Qian, M. D. Bartberger, S. J. Pastor,
K. N. Houk, C. L. Wilkins, Y. Rubin, J. Am. Chem. Soc. 2000, 122,
8333.
[7] T.-Y. Hsiao, K. C. Santhosh, K.-F. Liou, C.-H. Cheng, J. Am. Chem.
Soc. 1998, 120, 12232.
[26] S. M. Cybulski, D. M. Bishop, Chem. Phys. Lett. 1993, 98, 8057.
[27] O. Walker, P. Mutzenhardt, P. Tekely, D. Canet, J. Am. Chem. Soc.
2002, 124, 865.
[28] The GIAO calculations at the all six levels of theory mentioned above
were conducted for 9, 10, 11, and 12, and calculations at three levels of
theory using the 6 311G** basis set were conducted for 15. Among
them, the results of GIAO-B3PW91/6 311G** calculations were
found to give the best fit to the observed spectra. See the Supporting
Information.
[8] a) H. Inoue, H. Yamaguchi, T. Suzuki, T. Akasaka, S. Murata, Synlett
2000, 1178; b) S. Iwamatsu, P. S. Vijayalakshmi, M. Hamajima, C. H.
Suresh, N. Koga, T. Suzuki, S. Murata, Org. Lett. 2002, 4, 1217.
[9] Y. Murata, N. Kato, K. Komatsu, J. Org. Chem. 2001, 66, 7235.
[10] a) Y. Murata, K. Komatsu, Chem. Lett. 2001, 896; b) H. Inoue, H.
Yamaguchi, S. Iwamatsu, T. Uozaki, T. Suzuki, T. Akasaka, S. Nagase,
S. Murata, Tetrahedron Lett. 2001, 42, 895; c) Y. Murata, M. Murata,
K. Komatsu, J. Org. Chem. 2001, 66, 8187.
[29] F. Toda, K. Tanaka, Chem. Lett. 1979, 1451.
[11] a) Y. Rubin, Chem. Eur. J. 1997, 3, 1009; b) Y. Rubin, Top. Curr. Chem.
1999, 199, 67; c) J.-F. Nierengarten, Angew. Chem. 2001, 113, 3061;
Angew. Chem. Int. Ed. 2001, 40, 2973.
[30] a) P. W. Stephens, D. Cox, J. W. Lauher, L. Mihaly, J. B. Wiley, P.-M.
Allemand, A. Hirsch, K. Holczer, Q. Li, J. D. Thompson, F. Wudl,
Nature 1992, 355, 331; b) H. Klos, I. Rystau, W. Sch¸tz, B. Gotschy, A.
Skiebe, A. Hirsch, Chem. Phys. Lett. 1994, 224, 333.
[12] a) S. W. McElvany, M. M. Ross, N. S. Goroff, F. Diederich, Science,
1993, 259, 1594; b) Y. Rubin, T. C. Parker, S. Pastor, S. Jalisatgi, C.
Boulle, C. L. Wilkins, Angew. Chem. 1998, 110, 1353; Angew. Chem.
Int. Ed. 1998, 37, 1226; c) Y. Tobe, N. Nakagawa, K. Naemura, T.
Wakabayashi, T. Shida, Y. Achiba, J. Am. Chem. Soc. 1998, 120, 4544;
d) Y. Tobe, N. Nakagawa, J. Kishi, M. Sonoda, K. Naemura, T.
Wakabayashi, T. Shida, Y. Achiba, Tetrahedron 2001, 57, 3629;
e) M. M. Boorum, Y. V. Vasil'ev, T. Drewello, L. T. Scott, Science
2001, 294, 828.
[31] The first reduction potential is lowered by 0.12 V compared with C60
:
M. Keshavartz-K, B. Knight, G. Srdanov, F. Wudl, J. Am. Chem. Soc.
1995, 117, 11371.
[32] T. Suzuki, Y. Maruyama, T. Akasaka, W. Ando, K. Kobayashi, S.
Nagase J. Am. Chem. Soc. 1994, 116, 1359.
[33] The detailed results on the calculations will be reported elsewhere.
Received: October 24, 2002 [F4527]
Chem. Eur. J. 2003, 9, No. 7
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1609 $ 20.00+.50/0
1609